WP4.D2 DRAFT PROCEDURES AND RESOURCE DOCUMENTS Subtask 1: Advanced Collectors Dissemination level: Public

Transcription

1 WP4.D2 DRAFT PROCEDURES AND RESOURCE DOCUMENTS Subtask 1: Advanced Collectors Dissemination level: Public August 2007 CONTENTS WP4-D2.1.A (3 pages) WP4-D2.1.B (10 pages) WP4-D2.1.C1 (4 pages) WP4-D2.1.C2 (5 pages) WP4-D2.1.D1 (33 pages) WP4-D2.1.D2 (5 pages) WP4-D2.1.E (6 pages) WP4-D2.1.F (4 pages) ANNEX 1 (41 pages) ANNEX 2 (13 pages) WP4-D2.1.G (5 pages) WP4-D2.1.H (21 pages) WP4-D2.1.I (4 pages) The task of WP4.D2 draft procedures and resource documents was divided into the following 9 subtasks: 1. advanced collectors 2. advanced stores 3. advanced controllers 4. combisystems 5. solar cooling 6. solar desalination 7. fluids 8. LCA (Life Cycle Assessment) 9. m² -> power and energy The resource documents of subtask 1 advanced collectors (subtask leader: SP) are included in this report. WP4-D2.1.a IAM dependencies - ITW WP4-D2.1.b ETC-testing - SP WP4-D2.1.c1 Exposure/ageing - Australian Stand. for stagn. WP4-D2.1.c2 Exposure/ageing, degradation modelling - ITW WP4-D2.1.d1 Qualific. test for absorber surface durability - IEA WP4-D2.1.d2 Tests & reqs. for polymeric materials - SP WP4-D2.1.e Models for un-glazed collectors - Uni. Kassel WP4-D2.1.f Air collectors, recommendations - arsenal Annex 1, Air collectors, draft standard - arsenal Annex 2, Air collectors, testing - arsenal WP4-D2.1.g Tracking concentrating collectors - ITW (INETI) WP4-D2.1.h Improv. acc. of FP collector test - DTU WP4-D2.1.i Collector annual output, inquiry - SP WP4-D2.1.j WP4-D2.1.k Final report of WP4.1 - SP Definitions and test procedure related to the incidence angle modifier - INETI, SP, ITW WP4-D2.1.J (144 pages) In total this Deliverable consists of 305 pages. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

2 TEST AND SIMULATION OF SOLAR THERMAL COLLECTORS WITH MULTI-AXIAL INCIDENT ANGLE BEHAVIOR Stephan Fischer Wolfgang Heidemann Hans-Müller Steinhagen Universität Stuttgart, Institut für Thermodynamik und Wärmetechnik (ITW) Pfaffenwaldring Stuttgart Germany fischer@itw.uni-stuttgart.de ABSTRACT The thermal performance of solar collectors depends among other things on the incident angle modifier for the beam irradiance. In case of flat plate collectors an isotropic behavior is assumed, whereas evacuated tubular collectors show a bi-axial dependency with respect to the incidence angle of the beam irradiance. In this case it can be distinguished between the longitudinal incident angle (along the tube) and the transversal incidence angle (perpendicular to the tube). However, more and more collectors show a dependency with respect to the incident angle that can not be characterized with the bi-axial approach. Examples are the Swedish MaReCo /1/, the French glass collector /2/ or evacuated tubular collectors with adjusted tubes. For those collector types the paper introduces the multiaxial incident angle modifier and illustrates exemplary simulation results for two different collector orientations. 1. INCIDENCE ANGLE MODIFICATION IN GENERAL The incidence angle θ is defined as the angle between the position vector of the sun and the vector perpendicular on the collector aperture. The dependency of the thermal performance is described by the incidence angle modifier. The incidence angle modifier IAM(θ) is defined by the ratio of the collector output at a given incidence angle q& (θ ) and the collector output at normal incidence q& ( θ = 0) (see eqn. 1). q& ( θ ) IAM( θ ) = (1) q& ( θ = 0) The behavior of flat plate collectors with respect to the angle of incidence of the beam radiation is assumed to be isotropic. According to the European Standard EN /3/ it is calculated according equation 2. 1 IAM ( θ ) = 1 b0 1 (2) cosθ The parameter b 0 is a collector parameter and depends basically on the incidence angle properties of the absorber and the transparent cover as well as on the geometry of the collector casing. Up to now there is no procedure documented within an international standard how to deal with bi-axial incidence angle behavior. However, with the publication of the reviewed EN solar collectors having a bi-axial behavior with respect to the angle of incidence of the beam radiation will be considered by a normative document. To do so the approach of McIntire /3/ is used (see eqn. 3). IAM θ ) = IAM ( θ, θ ) = IAM ( θ,0) IAM (0, θ ) (3) ( l t l t This approach calculates the bi-axial incidence angle modifier by multiplication of the incidence angle modifiers in longitudinal and transversal plane. Thereby it is assumed that the collector shows the same longitudinal incidence angle behavior no matter if the incident irradiance is coming from the north (θ l < 0) or the south (θ l > 0). The same is assumed for the transversal incidence angle modifier defining east by θ t < 0 and west by θ t > 0. Figure 1 shows a typical distribution of the longitudinal and transversal incidence angle modifier for a evacuated tubular collector.

3 incident angle modifier [-] longitudinal transversal incidence angle [ ] Fig. 1: Typical distribution of the incidence angle modifier of a evacuated tubular collector (2) An improvement of the thermal performance at inclined incident angle from the right (referring to figure 3) as long as no shading occurs (3) A reduction of the thermal performance at inclined incident angle from the left (referring to figure 3) As a result of points (2) and (3), the symmetrical incidence angle behavior perpendicular to the tube must be divided into two different areas. In previous tests and simulations of evacuated tube collectors, this fact has usually not been considered. To enable both, testing and simulation of these collector configurations the multi-axial incident angle modifier has been implemented in the TRNSYS type INTRODUCTION OF THE MULTI-AXIAL INCIDENT ANGLE MODIFIER An increasing number of evacuated tubular collectors are mounted horizontally on flat roofs (see figure 2) or vertical on facades. Fig. 3: Schematic cross section of three tubes with planar absorbers in conventional orientation (upper row) and turned absorber (lower row) 3. PARAMETER IDENTIFICATION AND SIMULATION To investigate the multi-axial incidence angle modifier the thermal efficiency of an evacuated tubular collector with a flat absorber (see figure 3) was determined according to EN Afterwards the tubes have been turned 30 to the east and the zero-loss efficiency η 0 and the incidence angle modifiers were determined again. Fig. 2: Flat roof mounted evacuated tubular collectors In this connection most of the time the tubes are orientated along the east-west axis. To increase the thermal efficiency of such collector installations the single tubes of the collector are adjusted in a way that the absorber is oriented towards in incident sun rays (see figure 3). The adjustment of the tubes has the following impact on the thermal behavior of the evacuated tubular collector: (1) A reduction in the optical efficiency at normal incidence due to the reduced projected absorber area and the inclined incident angle on the absorber. basic configuration 30 east oriented absorber η 0 [-] K θd [-] a 1 [W/(m²K)] a 2 [W(m²K²)] c eff [J/(m²K)] Table 1: Collector parameters

4 Table 1 shows the determined collector parameters for the two collector configurations. It was assumed that the heat loss coefficients a 1 and a 2 as well as the effective heat capacity c eff of the collector do not change due to the adjustment. Figure 4 shows the distribution of the transversal incidence angle modifier of the two collector configurations. transversal incidence angle modifier [-] 30 east orientated absorber flat absorber Fig. 4: Distribution of the transversal incidence angle modifier of the two collector configurations To determine the impact on the yearly energy gain of the collector a system simulation using TRNSYS with collector type 132 was carried out for both collector configurations. Two different collector orientations are simulated within a reference system: 1. The collector is mounted horizontally on a flat roof. The tube are orientated in east-west direction ( flat roof ). 2. The collector is mounted with a collector azimut of 60 west on a roof tilted 50 ( west roof ). Table 2 summarizes the calculated yearly collector gains for the four different system simulations. The collector configurations with the adjusted tubes results despite the significant reduction of the zero-loss efficiency in an increase of the yearly collector gain of 6% for the flat roof configuration and 4% for the west roof configuration. basic configuration transversal incidence angle [ ] 30 east oriented absorber flat roof 564 kwh/(m²year) 597 kwh/(m²year) west roof 593 kwh/(m²year) 614 kwh/(m²year) Table 2: Yearly collector gain for the four different system simulations. 4. SUMMARY The thermal behavior of certain collector types can not be modeled by a bi-axial incidence angle modifier to a sufficient extend. This is also true for individually adopted collector configurations like evacuated tubular collectors with adjusted tubes. To predict the yearly collector gain of these collector configurations to a reasonable extend the multi-axial incidence angle modifier was introduced. For parameter identification and system simulation the multiaxial incidence angle modifier has been implemented within the TRNSYS type NOMENCLATURE a 1 [W/(m²K)] heat loss coefficient a 2 [W/(m²K²)] temperature dependence of the heat loss coefficient b 0 [-] parameter for the characterization of the incident angle modifier of the beam irradiance c eff [J/(m²K)] effective heat capacity IAM(θ) [W/m²] incidence angle modifier for beam irradiance K θd [-] incident angle modifier for diffuse irradiance q& [W/m²] useful collector output η 0 [-] zero loss efficiency θ [ ] incident angle of the beam irradiance θ l [ ] in the longitudinal plane projected component of the incident angle of the beam irradiance θ t [ ] in the transversal plane projected component of the incident angle of the beam irradiance 6. REFERENCES 1. Karlson B. and Wilson G., MaReCo-design for horizontal, vertical or tilted installation, EuroSun 2000, Kopenhagen, Denmark 2. Robin J.M., Flament B. and Vasile C., A new solution for the architectural integration, EuroSun 2004, Freiburg, Germany 3. DIN EN (2001), Thermal solar systems and components solar collectors - Part 2: test methods. 4. McIntire W., Factored approximations for biaxial incident angle

5 W4.1 Resource document Recommendations on testing of evacuated tubular collectors (ETCs) Dissemination level: Public Author: Peter Kovacs, SP December 2006 CONTENTS INTRODUCTION Why present standards are not adapted to ETCs and why the present test procedures need to be revised with regard to ETCs. EXPERIENCE FROM RECENT ETC IMPLEMENTATION AND TESTING Description of the experience of some of the main actors in the field of ETC testing to give a background and to motivate the need for revision. SPECIFIC ASPECTS OF EVACUATED TUBULAR COLLECTORS ETCS Introduction to the various specifics of testing and quality assurance of testing solar ETC collectors. SUMMARY This report aims to put focus on evacuated tubular collectors (ETCs) and their specific features from the point of view of testing and quality assurance. ETCs have many advantages to flat plate collector and can therefore contribute significantly, much more than today, to the large scale introduction of solar heating technology in Europe. A prerequisite for this however is that their quality and performance can be determined and assured in the same manner as how it has been successfully done for flat plate collectors. In order to achieve this, the present test methods and quality assurance schemes need to be updated so that the specific characteristics of ETC are also taken into account. This report describes why it is important to update the present methods to make them better fit to cope with performance and quality testing of ETCs and how it can be done. Some new tests and some ideas about further research work in the field of ETCs are also presented. RECOMMENDATIONS Suggestions for revision of existing test methods for ETCs and for development of new methods. Recommendations for further activities. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

6 page 2 of 10 pages Table of contents SUMMARY...1 Table of contents Introduction Experience from recent ETC implementation and testing Market development Inquiry about the state of the art Specific aspects of evacuated tubular collectors ETCs Specific features relevant to all types of ETCs Single glass tubes Double glass tubes ETC with direct connection ETC with heat pipe connection Recommendations Recommendations for revision of present methods (EN :2006) Definitions Thermal performance testing- Efficiency testing Thermal performance testing- Thermal capacity Thermal performance testing- Stagnation temperature measurement Thermal performance testing- Specification of physical properties Quality testing- General Quality testing- High temperature resistance- and exposure test Quality testing- Mechanical loads Durability of reflector materials Quality testing- Impact resistance Documentation Proposals for new test methods Freeze testing of heatpipes Durability of glass to metal seals Recommendations for further work References...9 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

7 page 3 of 10 pages 1 Introduction Testing the thermal performance and quality of solar collectors has a relatively long history. Today s European test standards where developed on the basis of ISO- and Ashrae standards that originate from before Even though the first evacuated tubular collectors where already present at that time, the flat plate collector was the standard product and so it has been until around Therefore, almost all of the work related to development of test methods and quality criteria for solar collectors has been done with the flat plate collector in mind. Only to a minor extent the ETC and it s specific properties has been addressed. Recently, mainly due to the strong economic development in China, ETC has become more and more common in the global context. Mainly in China itself production in absolute numbers is amazing, but also in some European countries, ETC market shares have been growing rapidly. This has partly been due to promising cost/ performance ratios and the fact that ETC tend to perform better than flat plates under some circumstances. However in some cases it has become obvious that the low prices were also accompanied by low quality in different respects only this low quality wasn t always revealed due to inadequate or improper test methods. Therefore, today s test methods and requirements need to be updated and adapted to this somewhat reborn technology. Not only in order to create a fair competition between different collector types but also to give manufacturers and importers the proper tools to judge and further develop the quality and performance of ETCs. This way the technology will be able to contribute more significantly to the different European markets on the rise. Avoiding the risk of low quality, low cost products destroying the good reputation of Solar thermal technology is another important reason why ETCs need more attention in the test standards and in the quality assurance schemes. 2 Experience from recent ETC implementation and testing 2.1 Market development Apart from a very strong development in Sweden where market shares for ETCs, mainly imported from China, has grown from 5 to 25 % in a few years, only Ireland in Europe seem to have a large market share for ETCs. In most other countries the ETCs share of the total market is below 10% and has not been reported to grow. Due to a market share just above 10 % and a large total domestic market Germany has by far the largest number of ETCs sold in Europearound m 2 in Australia reports a slowly growing market for ETC but also of many products failing to enter the market due to low quality. In China ETCs are a true success story. The overall annual growth rate for solar thermal is close to 30%, annual sales around 15 million square meters and the market shares for ETCs are 80-90%, compared to 1997 when it was only around 35%. 2.2 Inquiry about the state of the art In 2005 an inquiry about ETC testing and quality aspects was performed within the NEGST project [5]. The purpose of the inquiry was to give a background to an assessment of the need for revised test procedures regarding performance- and quality testing of ETC:s. The inquiry addressed the partners of the NEGST and Keymark II projects but also relevant industry actors and test institutes not directly involved in these projects. The majority of questions were very test- specific but also general comments regarding need for revised test procedures, weaknesses in ETCs that should be assessed, present tests that are being done without justification etc. could be input. The questionnaire used in the inquiry was answered by ten laboratories, two Swedish importers of ETCs and by one Chinese manufacturer. The answers showed that most laboratories had quite a limited experience of ETC testing but also that several of the difficulties encountered in testing were common. The inquiry also turned out to raise more new questions than it answered and the proposals developed from the inquiry input may in many cases need further back up in NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

8 page 4 of 10 pages terms of scientific investigations in order to be clearly justified. Therefore the majority of the following recommendations, developed from the questionnaire input, should be seen as input to discussions in future standardisation work rather than as fully developed proposals for revisions. 3 Specific aspects of evacuated tubular collectors ETCs In the following some specifics aspects of ETCs are highlighted in order to give a better understanding and background of the new or redefined tests that may be required to assure their proper function and long lasting performance. As different generic types of ETCs are partly very differently constructed and thus also have different potential failure modes they are here addressed first with respect to their common features and then with respect to the characteristics of each generic type. 3.1 Specific features relevant to all types of ETCs The main features where ETCs in general differ significantly from flat plate collectors and thus require some particular attention with respect to performance- or quality testing and also regarding design requirements are summarized in the following table. Specific feature The comparatively low heat losses resulting in high stagnation- and maximum operation temperatures The non planar shape of the collector surface, either it is fitted with a reflector or not The frequent use of (external) reflector mirrors The fragile structure of the vacuum tubes The fact that the performance is heavily dependent on the quality (level, durability) of the vacuum Implication on testing or system design Difficult to determine efficiency at high temperatures with good accuracy Difficult to determine unambiguous stagnation temperature Special attention required in system design in order to avoid thermal stress on the heat transfer fluid Difficult to determine proper loads for mechanical load tests Bi- or multi axial incidence angle modifiers need to be determined in performance testing Highly exposed component having a high influence on the performance but not being assessed in present test standards Difficult to assess the long term effects on the collector output Impact resistance testing required in some regions Difficult to determine vacuum loss in connection to quality tests Difficult to assess the long term durability of the vacuum 3.2 Single glass tubes ETCs built up by single glass tubes used to be the most common type of the early ETCs but these tubes have definitely become less common at the test institutes in Europe from 2004 and on. This probably also reflects their limited market penetration. The only feature specific to single glass tubes mentioned by the respondents in the inquiry was the reliability of the glass/ metal seal, which is not specifically addressed in the presently used quality tests. IZES seems to be the one European test laboratory that has measured most single glass tubes. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

9 page 5 of 10 pages 3.3 Double glass tubes Double glass tubes or Dewar type tubes have become more common at some of the test institutes from 2004 and on. SP and ITW did the most tests on double glass tubes and UNSW also reports quite a large number of tests. In Austria, Greece, Canaries and Portugal only one or two products were tested in the last years. In China this is by far the most common type of glass tube used in ETCs. Some of the specific features of double glass tubes mentioned by the respondents in the inquiry were Effective capacity using calculation method according to EN is underestimated for ETCs built on double glass tubes The collector models presently used in performance testing and modelling are not sufficient to describe the thermal characteristics (time constant and thermal capacitance) of ETCs built up by double glass tubes Double glass tubes result in constructions having a comparatively large number of failure modes, in particular when equipped with a heat pipe and external reflector. This makes it more difficult to track reasons for low performance figures in testing and to safeguard a high and even quality of the product. Double glass tubes, when used in water in glass systems means new challenges for quality testing of gaskets etc but this is presently treated as a system reliability issue and therefore not further addressed in this context. 3.4 ETC with direct connection A critical aspect of ETCs with direct connection, i.e. with the heat transfer fluid (glycol/water mixture) circulating through the tubes is the thermal stress on this fluid. This is particularly relevant when the collector is fitted with a reflector resulting in stagnation temperatures in excess of 250 C. There are reports saying this can be dealt with by using a proper system design [2] but there are also reports where the conclusion is that this construction is very likely to cause problems. Potential problems could be avoided through special requirements on the installer documentation for this type of collectors or, in general terms, for any collector type that can reach stagnation temperatures higher than C. 3.5 ETC with heat pipe connection From the increasing numbers of double glass tubes, many are fitted with a heat pipe which can then in turn, be fitted to the manifold through a wet or a dry connection. The heat pipe construction can be sensitive in several ways: The amount and composition of the evaporating liquid The vacuum inside the metal pipe The material quality in the pipe and the design of the pipe and the bulb Bad design or faulty installation of dry connection resulting in low heat transfer capacity Risk for freezing Damage due to high temperatures (reflectors) Air pockets inside the bulb as a result of improper filling or material These possible failure modes together with the ones resulting from the metal fin inside double wall glass tube construction and the absorber itself makes variable quality and/ or energy performance much more likely for these collectors than for e.g. ordinary flat plate collectors. Furthermore, from the point of view of performance testing, dry out effects during testing under high irradiance should also be considered. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

10 page 6 of 10 pages 4 Recommendations In order to make the best out of ETC technology in the future, test methods and quality assurance measures need to be further developed. Furthermore, the knowledge about which the critical design parameters are and how they are affected by time and different kinds of stress need to be further developed. As an input to this work within standardization and research the following recommendations have been developed. 4.1 Recommendations for revision of present methods (EN :2006) Definitions Background material behind tubes (test rig) needs to be defined in the test standard when no reflector is present. Ideally a plain mat black background should be used, that will not contribute to light levels. Absorber area when reflector is used should be redefined. The definition according to EN will give efficiencies higher then Thermal performance testing- Efficiency testing Several laboratories reported a) that dry out effects can occur during testing of ETCs with heat pipes during high irradiance conditions and b) that the present collector model used in the standard was not able to accurately model the thermal capacitance and time constants of the collector (ETC with double glazing and heat pipe in particular) [4]. It should be discussed how the dry out effects are to be handled during testing and in the reporting. Regarding the inadequacies of the collector performance model, one way around the problem suggested is using the model presented in [9], however this seems to create other difficulties in the evaluation of measurements. Therefore, this is at present a case for further research Thermal performance testing- Thermal capacity The method available for calculating the thermal capacity ( ) of the collector has been reported to underestimate quite substantially the figures for double glass ETCs and should be reviewed Thermal performance testing- Stagnation temperature measurement Whether it refers to performance- or to quality testing, the determination of the stagnation temperature of ETCs is not adequately well defined in the present standard. Notes given in section are too vague. More precise recommendations on sensor positioning (Note 1) should be given, one for each of the most common designs of ETCs unless a common generally significant positioning can be defined. Alternatively, the method suggested in Note 2 could be made normative for ETCs. Ability to determine accurately the stagnation temperatures of ETCs may eventually lead to a more accurate performance prediction at high working temperatures (over 120 C) where today s figures are extrapolations from measurements at lower temperatures. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

11 page 7 of 10 pages Thermal performance testing- Specification of physical properties To serve better as a base for quality assurance of ETCs, Annex D should be further differentiated in order to describe better the physical characteristics of the collectors. For example most properties stated for the absorber in todays standard should also be stated for the heat transfer plate used in double glass tubes. A number of terms should furthermore either be generalised or specifically mentioned with regard to ETCs in addition to the ones that today refers explicitly to flat plate collectors, eg. tube spacing in addition to riser spacing. Also the properties of an optional reflector, often present with ETCs, should be declared Quality testing- General In order to reveal low quality products, in particular among ETCs with double glazings and heat pipes, it is recommended to introduce a test cycle for these collectors where the same collector is first measured for efficiency, then subjected to a (possibly revised, tougher) high temperatureand exposure test and then measured for efficiency once again. In order to save costs one of the efficiency tests could be limited to zero loss efficiency, but preferably also the eventual increase in heat losses should be assessed. This way the entire collector will be checked for any fault that may have occurred during the high temperature- and exposure tests and which are more probable to occur on these specific collector types than on other types as explained in previous sections. A second, simpler option for this type of test is presented in section of this document but in this case, only the reliability of the heat pipe it self will be checked Quality testing- High temperature resistance- and exposure test In Australian collector tests, a type of accelerated exposure test is achieved by means of exposing the whole collector to high irradiance and ambient temperature conditions for a period of 12 hours per day during ten days. A solar simulator with less stringent requirements on it s performance than for performance testing of collectors, and a temperature controlled chamber is used for the test [8]. The present European exposure test has been under a lot of debate, mainly due to it s inability to maintain uniform test conditions all over Europe and in the specific case of ETCs it is not considered very efficient in terms of revealing their weaknesses. It is therefore suggested that a new form for the exposure test is developed along the ideas of the Australian test. This new test could either be complemented by, or merged with the concluding high temperature test for heat pipes suggested in paragraph Quality testing- Mechanical loads Defining the proper load in mechanical load tests on ETCs is a tricky matter. What loads that are likely to create problems for the durability of ETCs are most probably also not completely the same kind of loads as the ones creating problems for flat plates. In Germany test labs decided together with the manufacturers to cancel tests for negative and positive pressure on collector cover as these tests are not considered to make sense for ETCs. However the issue of heavy snow loads might still be worth considering. It is recommended to follow the introduction (or collect already existing experience) of ETCs in snowy regions to find out if it s a serious problem worth taking into account for testing. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

12 page 8 of 10 pages In Sweden ETCs resistance to positive pressure has been tested for some time, with a requirement of 2000 Pa, hardly ever resulting in failure of the tubes themselves however sometimes revealing weaknesses in the fixings of the tubes. The method used has been more or less the same as the one described in of the EN however, an air pressure was used to create the load and a thick plastic film used to distribute the load evenly over the tubes. The load requirement was furthermore reduced by multiplying it with a factor equal to the ratio of the aperture to the aperture including space between tubes area, in general resulting in the normal requirement of 2000 Pa being reduced to around 1200 Pa. It is suggested to exclude ETCs without reflectors from the mechanical load requirements in and unless it is shown from recorded practical experiences that either of them are motivated. It is suggested to include ETCs with reflectors in the mechanical load requirements in and but in this case the reflector and not the glazing should be the subject of the requirements. A positive or a negative mechanical load can be applied to the reflector by means of an airtight box divided in two compartments by a plastic film also covering either side of the reflector surface and one of the compartments being pressurized by air Durability of reflector materials In addition to the mechanical load resistance, suggested to be tested in paragraph 4.1.8, the long term optical properties of the reflector is of course also of high interest. Some international standards (to be checked XXX) for assessment of this is available and it is recommended that one of them is incorporated in the standard as a normative requirement for ETCs that are tested and sold with an external reflector Quality testing- Impact resistance Positive impact resistance test is only possible with ice balls. Impact resistance using steel balls and ice balls are not comparable [4] (which is already spelled out in the standard) but furthermore ETCs are told not being able to withstand the impact of the steelballs. Therefore method 1 in 5.10 should be deleted. Method 2 may need a review [7] Documentation In order to reduce the risk for brake down of the heat transfer fluid in the collector loop that may occur when exposed to temperatures above 230 C, special considerations should be taken when designing the loops including such high efficient collectors. In 7.3 of EN (requirements on the installer manual) it should at least be added installation... to the paragraph Recommendations about the heat transfer media.. and furthermore reference [2] containing some practical design considerations, should be added in the same paragraph. 4.2 Proposals for new test methods Freeze testing of heatpipes Damaging of heatpipes due to freezing can result from improper composition of the working media in the heatpipe or from bad design of the metal tube (material quality, thickness, shape of lower end) and has been reported by several sources [3, 4, 10]. As breakage of the metal tube in the case of bad design often doesn t occur until after several freeze/ thaw cycles, the following procedure has been proposed: At least five samples of the product should be exposed to 100 cycles of NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

13 page 9 of 10 pages 1) 1 hour freeze time in -20 C or lower temperatures. Heat pipes should be sitting at a angle which is the harshest condition for them 2) 20 minutes in an oven at 150 C, Once every 10 cycles, in the oven for 20min to 200 C (or the actual stagnation temperature of the collector in case) thus mimicking a possible stagnation condition. Problems are shown either as a deformation at the bottom of the heatpipe (requirement on max. change in diameter needed) or as the formation of an air pocket which results in a cool area at the top of the heatpipe when exposed to heat at the bottom. The 100 cycles should be concluded by a 24 hour exposure at the stagnation temperature of the collector in case and then: 3) Turn the heat pipes upside down (tip down) in a container of hot (>50oC) water, at least 30cm deep. Swirl the heat pipes around to allow them to "start" and check to make sure they are working ok The product has failed if more than one heatpipe either forms an air pocket or shows a deformation higher than allowed. If one heatpipe fails, the test can be rerun with new samples Durability of glass to metal seals In order to assess the durability of the glass to metal seal in the types of ETCs where such seals exists either a vibration or a cyclic tension applied to metallic parts entering the glass tube may be considered. 4.3 Recommendations for further work In general the heat transfer mechanisms of the ETCs and how different parts of ETCs respond to different kinds of stress need to be better known. A call for a common research effort was distributed in late 2006 [6]. Two more specific topics that call for attention are: Any severe weaknesses in today s performance test methods/ models would be revealed through a project where the most common ETC types where tested for performance and then measured for a longer period to have a thorough comparison: measured to modelled performance. Problems with vacuum losses of the tubes are of course critical to the performance of ETCs. At present it is difficult to determine the vacuum loss and the tests performed in the standards are not able to verify the long time durability of the vacuum. 5 References [1] EN ,2:2006. Thermal solar systems and components- Solar collectors. Part 1: Requirements. Part 2: Test methods [2] Hausner, R. Report of IEA SH&C Annex 26, Stagnation behaviour of solar thermal systems [3] Michael Humphreys, Managing director, Apricus Solar Co. Ltd. China Personal communication [4] Johansson, S. Managing director, Intelliheat AB Sweden. Personal communication [5] Kovacs, NEGST working paper, Inquiry about needs for revised test procedures for evacuated tubular collectors, ETCs. [6] Kovacs, NEGST working paper, Call for a common research effort on ETCs NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

14 page 10 of 10 pages [7] Morrison, G. Working paper, Ice ball impact tests on evacuated tubes [8] Morrison, G. Excerpt from Australian/ New Zealand standard AS/NZS2712. Solar and heat pump water heaters: Design and construction- Stagnation test for collector and integral collector and container. [9] Streicher, E et al. ISES Performance model for solar thermal collectors taking into account degradation effects. [10] Sundquist,R. Managing director Exoheat AB, Sweden Personal communication NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

15 STAGNATION TEST FOR COLLECTORS AND INTEGRAL COLLECTOR AND CONTAINER (Normative) A.1 SCOPE This Appendix sets out two methods for assessing the ability of a collector or integral collector and container to withstand temperatures close to the maximum temperatures that it will encounter under some or all of the following conditions: (a) (b) (c) NOTE: Either of the two methods set out in this Appendix may be used, but it should be noted that Method 1 is not suitable for use with collectors with heat pipes. When empty during installation; When empty during its service life; and When full of water but not being used in peak summer conditions Such temperatures occur during periods of no useful heat removal from the collector with high solar radiation and ambient temperatures. NOTES: 1 In order to more realistically model extreme conditions, the test conditions have been changed from 1200 W/m 2 radiation and 40 C ambient temperature to 1100 W/m 2 and 50 C. Collectors compliant with the former conditions are deemed to comply with the new conditions. The conditions in ISO for high temperature resistance test Sunny C are also deemed to comply. The effective environmental temperature for New Zealand is 40 C for Method 1. For New Zealand the conditions in the ISO for high temperature resistance test Sunny B are acceptable. However, the collector must be marked in accordance with Clause Error! Reference source not found. Error! Reference source not found. if this test method is used. 2 The Method 2 test has a lower radiation level and ambient temperature to account for the fact that the thermal radiation level received by the collector from the simulator lights is higher than from the natural sky. A.2 APPARATUS C2.1 Method 1 NOTE: This method is not appropriate for use with collectors with heat pipes. The following apparatus is required (see also Figure C1). (a) A pumped heat transfer loop using a suitable heat transfer liquid, with the collector forming part of the loop. A suitable heat transfer fluid is one that will remain in its liquid state at the stagnation temperature and the maximum operating pressure of the collector. (b) (c) A stand on which the collector will face normal to the direct beam solar radiation at solar noon. A suitable stand is described in AS/NZS or ISO Thermometers or temperature-measuring devices to measure the temperature of any critical materials or heat-sensitive components during the test. C2.2 Method 2 The following apparatus is required: (a) A solar simulator of suitable size. (b) (c) A temperature-controlled chamber in which the collector may be placed when exposed to the simulator. Thermometers or temperature-measuring devices to measure the temperature of any critical materials or heat-sensitive components during the test.

16 A.3 PROCEDURE C3.1 Method 1 The procedure shall be as follows: NOTE: In this procedure, two approaches are made to determine the stagnation temperature (T s ). The first approach is detailed in Steps (a) and (b), and the second approach (which is a cross-check and correction) in Steps (c) and (d). (a) Test the collector for thermal performance in accordance with AS (b) From the constants supplied from Step (a) calculate, in accordance with the equation below, the stagnation temperature (Ts) when (i) (ii) (iii) the efficiency is zero; the total global radiation is 1100 W/m 2 ; and the effective environmental temperature is 50 C. The stagnation temperature (T s ) is calculated as follows: C3(1) (c) The values for the terms used in the equation are obtained by testing in accordance with AS/NZS As a cross-check for the stagnation temperature, proceed as follows: (i) (ii) Install the collector on the stand in accordance with the manufacturer s instructions and adjust its inclination to provide for maximum (clear sky) solar irradiation. Ensure all the air has been bled from the system. Measure the total global solar radiation (G) on the plane of the collector and the ambient temperature (T a ). Adjust the temperature of the fluid (T f ) entering the collector to C3(2) (iii) Measure the temperature difference (ΔT fluid ) of the fluid flowing through the collector and determine that the collector has reached steady state by plotting the collector outlet temperature versus time. If there is a temperature rise, increase the fluid temperature entering the collector by 5 C, or, if there is a temperature drop, reduce the fluid temperature entering the collector by 5 C. (iv) Measure G, T a, and the temperature difference across the collector. If there is still a rise or drop, repeat Step (iii). If the rise has become a drop or vice versa, then the true stagnation temperature has been saddled. (v) The correction to be applied to the predicted stagnation (T s ) is shown in Figure C2. (d) (e) Adjust the fluid temperature to T s plus the correction obtained in Step (c). Adjust the flow rate through the collector so that the average inlet temperatures and the average outlet temperature over the test period are both greater than the stagnation temperature (measured at least every 5 min). (f) Run the system at this flow rate and temperature for 12 h and then turn the pump off for 12 h. Continue this cycle of operation for 10 days. Ensure that the heat transfer fluid is at the calculated stagnation temperature prior to commencement of each cycle of operation. (g) (h) Visually inspect the collector daily and note any changes in its appearance. Terminate the test after 10 days or when there is evidence of structural or material deterioration that would impair the operation of the collector, whichever is sooner.

17 (i) Carry out a test of the thermal performance of the collector in accordance with AS/NZS , at a single test point using a fluid inlet temperature of at least 50 C or a value of (T f T e )/G T, which is half of the stagnation value at G T = 1000 W/m 2 and T a = 25 C; whichever is the lower. Compare the results of this test with those obtained in Paragraph C3.1(a). C3.2 Method 2 The procedure shall be as follows: (a) Test the collector for thermal performance in accordance with AS (b) (c) (d) (e) (f) (g) (h) (i) Install the collector on the stand in a temperature-controlled chamber and adjust its inclination so that it receives normal incident radiation from the solar simulator. Ensure that all the air has been bled from the system. Ensure that the collector is full of heat transfer fluid. It is not necessary to have any flow through the collector for this test method. Ensure that the air temperature adjacent to the collector is greater than 38 C, or, for the New Zealand only test, greater than 30 C, when the lamps are operating. Adjust the solar simulator output so that the average radiation measured at 6 uniformly distributed points on the collector is 1050 W/m 2 or greater, with less than 20% variation across the aperture. Solar spectral lamps as specified in AS/NZS shall be used. NOTE: Alternative arc lamps may be used; however, these lamps have a higher long wave radiation output and will result in a more severe test for some covers and glazing seals. Run the system under these conditions with the simulator being operated 12 h on and 12 h off for 10 days. Visually inspect the collector daily and note any changes in its appearance. Terminate the test after 10 days or when there is evidence of structural or material deterioration that would impair the operation of the collector, whichever is sooner. Carry out a test of the thermal performance of the collector in accordance with AS/NZS , at a single test point using a fluid inlet temperature of at least 50 C or a value of (T f T e )/G T, which is half of the stagnation value at G T = 1000 W/m 2 and T a = 25 C; whichever is the lower. Compare the results of this test with those obtained in Paragraph C3.2(a). A.4 REPORTING OF RESULTS The following results shall be reported: (a) The make and model identification of the system of which the collector forms a part. (b) Full details of the temperature measurements and the dates and duration of the test. (c) Details of the condition of the collector following the test with particular regard to (i) (ii) (iii) (iv) any structural failure; any burning, scorching or heat shrinkage; any effect likely to impair the serviceability of the collector; and any degradation in performance as a result of the test.

18 FIGURE C1 SCHEMATIC OF APPARATUS TO MEASURE EFFECTS OF PROLONGED STAGNATION TEMPERATURE (T s )

19 PERFORMANCE MODEL FOR SOLAR THERMAL COLLECTORS TAKING INTO ACCOUNT DEGRADATION EFFECTS Elke Streicher Stephan Fischer Wolfgang Heidemann Hans-Müller Steinhagen Universität Stuttgart, Institut für Thermodynamik und Wärmetechnik (ITW) Pfaffenwaldring Stuttgart Germany streicher@itw.uni-stuttgart.de ABSTRACT Ageing processes due to degradation factors like temperature may result in a decrease of the thermal efficiency of solar collectors. To be able to determine the energy output delivered by solar collectors throughout their lifetime, it is necessary to quantify the efficiency degradation. This can be achieved by accelerated ageing tests and the implementation of the characteristic degradation mechanisms in a suitable collector model. The paper explains how ageing processes due to temperature influence can be described with numerical models. A first draft of a performance model for solar thermal collectors that takes into account ageing processes has been developed on the basis of the existing collector model of the European Standard EN /1/. With this advanced model degradation mechanisms due to elevated temperatures can be considered. The model is implemented in the simulation programme TRNSYS and some exemplarity simulations are carried out. The results show the expected solar gain throughout an operation period of 20 years and the impact of ageing on the collector gain is demonstrated exemplarily. 1. INTRODUCTION Standardized test procedures are already available for the determination of thermal efficiency, reliability and durability of solar collectors /1/. Regarding environmental aspects, the determination of the energy payback time has become an important assessment tool. Up to now the yearly energy output of a solar collector is determined based on the results of a thermal performance test where the characteristic collector parameters are determined with new collectors. However, for an integral assessment of solar thermal collectors and systems, it is necessary to consider the change of performance with time due to degradation effects. In order to be able to calculate the overall energy output delivered by the solar collector during its lifetime, a collector model has been developed that takes into account degradation mechanisms. 2. DEGRADATION MECHANISMS A great variety of degradation mechanisms can be observed for solar thermal collectors. Nevertheless only few of them will be mentioned in the following. Regarding the selective absorber coating two degradation processes are of major importance: oxidation as a result of high temperatures and hydratisation/hydrolysis due to high humidity. Outgassing of the thermal insulation can result in a reduced transmittance of the transparent collector cover as well as in a deterioration of the thermal conductivity of the insulation material. Components made from polymeric materials are sensible to ultra-violet radiation, which activates photooxidation processes that may result in a loss of transmittance. Several investigations on the influence of the degradation factors temperature and UV-irradiation on component level have been made in the past years. IEA Task X /2/ examined the degradation behaviour of selective absorber coatings

20 with the result of a draft standard /3/ for accelerated ageing tests for solar absorber surfaces. Other investigations dealt with the ageing behaviour of transparent covers /4/. The results make it possible to forecast the degradation of individual components like the selective coating or the transparent cover. Nevertheless no statement can be made how these mechanisms affect the overall performance of the entire collector. Therefore an integral approach has been developed, that permits a forecast on the ageing process of the whole collector. This integral approach considers the influence of degradation processes on the collector performance parameters (τα), U 1 and U 2. That means that it is not necessary to verify the degree of degradation a certain temperature load would have for example on the absorption/emission coefficients of the absorber. But it is interesting how this certain temperature load will affect the collector parameters and consequently the collector performance. 3. COLLECTOR MODEL The major driving force of the degradation processes related to the thermal insulation and the selective coating is the absorber temperature. The possible absorber temperature has been increased considerably in the past few years through the ambition to achieve higher performance and higher fractional energy savings. Therefore in this first approach of implementing degradation effects in a collector model, we concentrated on the degradation caused by the factor absorber temperature. 3.1 One-node-model The already established collector models documented in national and international testing standards e.g. EN /1/ and ISO 9806 /5/ use a 1-node model (see figure 1). node 1 fluid GF' ( τα ) C fl Fig. 1: One-Node-Model 1-Node-Model q & 2 loss = a 1 ( ϑ fl,m ϑ amb ) + a 2 ( ϑ fl,m ϑ amb ) ϑ fl q & use m cp ( ϑfl,out ϑ = & A fl,in ) In this case all the capacities of the components of the collector are lumped together in one capacity node having fluid temperature. The 1-node collector model is characterized by the differential equation 1. d ϑ fl,m C fl = GF '( τα ) a 1( ϑ fl,m ϑ amb ) dt mc & p( ϑfl,out ϑ ) 2 fl,in (1) a 2 ( ϑ fl,m ϑ amb ) A Obviously the 1-node model does not distinguish between absorber temperature and the temperature of the heat transfer fluid. As for degradation of the selective surface the absorber temperature is the dominant factor, it is necessary to know the temperature of the absorber. Therefore a twonodes-model has to be defined. 3.2 Two-nodes-model For the time being the 2-nodes-model illustrated in fig. 2 is used. node 1 absorber node 2 fluid Fig. 2: Two-Nodes-Model d ϑ cfl dt fl,m 2-Nodes-Model G ( τα ) C abs C fl = (UA)( ϑ abs mc & p ( ϑfl,out ϑ ϑfl,m ) A ( UA)( ϑ abs ϑ fl,m) = G( τα ) U 1 ( ϑ abs ϑ amb ) 2 2 abs ϑ amb ) U ( ϑ q & 2 loss = U 1 ( ϑ abs ϑ amb ) + U 2 ( ϑ abs ϑ amb ) ϑ abs c abs d ϑ dt abs fl,in In this case the thermal capacity of the collector is subdivided into the capacity of the absorber C abs and the fluid C fl. The absorber node is heated by the incident solar irradiance G depending on the value of the transmittanceabsorptance product (τα). The heat losses of the absorber q& loss are based on the temperature difference between the absorber and the ambient. The heat transfer from absorber to fluid is determined by the heat transfer capacity rate (UA) and the temperature difference of the absorber and the fluid. Using differential equation 2 and 3 the 2-nodes-model has q & abs,fl ϑ fl q & use (UA) ( ϑ abs ϑ fl ) = A m cp ( ϑfl,out ϑ = & A fl,in ) ) (2) (3)

21 been implemented into a modified version of the TRNSYS type ACCELERATED AGEING TESTS As ageing is a long-lasting process, accelerated ageing tests have to be carried out in order to be able to predict the ageing behaviour of the collector. Accelerated ageing can be achieved by charging the collector with a higher dose of the maximal load than can be found under operation conditions. For most degradation processes caused by diffusion, chemical reactions or desorption the temperature dependence can be described with the Arrhenius correlation as shown in equation 4. E T 1 1 a n = exp (4) R Tref Tn The equation gives a correlation between the time t n and temperature T n at which the collector will experience the same degree of degradation as it does at the accelerated ageing test (t ref, T ref ). Whereas a n represents the acceleration of degradation according equation 5. ref a n = (5) t n ( τα ) = f ( ϑ (t)) U U 1 2 t The activation energy E T for the degradation has to be determined experimentally. Based on the results of the accelerated ageing tests and the above mentioned time transformation function a relation between real time ageing and accelerated ageing can be derived. Since degradation also depends on the individual material composition of the collector the ageing process will be different for every collector, and accelerated ageing tests have to be done for each collector separately. First a thermal performance test has to be carried out, where the three collector parameters (τα),u 1 and U 2 will be determined for a collector in new condition. Taking into account the temperature and time dependency of the collector parameters according to equation (6) to (8) within the 2- node-collector model the accelerated ageing can be started. = f ( ϑ abs abs (t)) = f ( ϑ (t)) abs (6) (7) (8) In this case a high temperature load for a definite time span is applied for accelerated ageing. Subsequently the three collector parameters are determined again in a thermal performance test and the deviation of the three collector parameters (τα), U 1 and U 2 with respect to the initial state is recorded. Now the degradation of the collector parameters is known for a certain impact and time span and accelerated ageing can be started again. This can be repeated as many times as required. 5. IMPLEMENTATION OF DEGRADATION EFFECTS IN THE COLLECTOR MODEL The Arrhenius correlation according to equation 4 has been implemented in the two-nodes-collector model in a way that for each simulation time step the degradation of the collector parameters (τα), U 1 and U 2 is determined. In order to account for the individual ageing behaviour of the collector the parameter list of TRNSYS type 132 was extended by the following parameters. - the activation energy E T - the temperature used during the accelerated aging test T ref - 10 possible defined times (t 0 to t 9 ) at which the accelerated aging test has been interrupted for the determination of the changed collector parameters (τα), U 1 and U 2-3 times 10 parameters describing the changes ( (τα), U 1 and U 2 ) in the collector parameters (τα), U 1 and U 2 corresponding to the times (t 0 to t 9 ) Fig. 3 shows as an example two possible degradation scenarios for the transmittance-absorptance product (τα). (τα τα) t [h] t 0 t 1 t 2 t 3 t 4 t 5 t 6 t 7 t 8 t 9 linear decline non-linear decline Fig. 3: Examples of possible results of the accelerated ageing tests: linear degradation of the collector parameter (τα) in comparison with non-linear degradation One scenario represents a linear degradation process with a continuous decrease of the parameter (τα) in comparison

22 with the initial values at t 0 = 0. The other scenario shows exemplarily a course of a strong degradation in the beginning that stops after some time. All parameters have to be determined by experiment whereas not all of the 10 possible supporting points of the degradation process have to be used. However the model can be used to analyze the thermal performance of a solar collector over its lifetime using virtual degradation courses. 6. SIMULATION RESULTS In the following it is shown with the help of an example how the model can be used to determine the energy output of the collector during its lifetime. The parameters used in the system simulation (table 1) and the course of degradation indicated in table 2 are an assumption and no real measured data, with the intention to show how the model works. The system simulation is carried out for a typical solar combisystem consisting of 15 m² collector area, a tank-intank store with a volume of 750 l, a daily hot water load of 200 l per day and a yearly space heating demand of 9090 kwh/m²a. Regarding weather data the test reference year of Würzburg was taken, a typical German city with moderate climate. During the simulation time that has been set to 20 years, the same weather data and load profile file have been used for each year. The initial collector parameters (τα), U 1 and U 2, the temperature at which the accelerated ageing tests have been carried out t ref as well as the activation energy E T for the ageing process are indicated in table 1. degradation process pattern the parameter U 2 has not much influence, this contribution was neglected and the deviation was put to zero. TABLE 2: ASSUMED DEGRADATION PROCESS PATTERN During the simulation the three collector parameters are modified continuously after each individual timestep in order to account for the aging effects. The programme calculates the respective absorber temperature in the timestep and determines the deviation of the initial collector parameters with the help of the implemented Arrhenius correlation and the provided degradation process pattern. As one result, the simulation gives out a summary with the collector parameters after a specified time span. In the present study the simulation time was set to 20 years. So the progress of the degradation as illustrated in figure 4 can be seen. (τα) [ ] Time of exposure after that collector test was carried out [h] (τα) U (τα) u U1 [W/(m²K)] TABLE 1: INPUT DATA USED FOR THE SIMULATION Input Data Unit years [a] 3.5 (τα) [-] 0.8 U 1 [W/(m²K)] 3.5 U 2 [W/(m²K²)] test temperature T ref [ C] 220 activation energy E T [J/mol] It is assumed that the accelerated ageing test of the collector was carried out at a test temperature of 220 C. The collector parameters are determined with a performance test in characeristic time steps between t 0 = 0 and t 3 = 100 hours exposure to test temperature in laboratory. Whereas the parameters at t 0 = 0 correspond with the parameters of the collector in new condition (table 1). Table 2 indicates the time after which the collector performance and the corresponding degradation of the collector parameters (τα) and U 1 was determined. As in the chosen Fig. 4: Impact of degradation process on collector parameters Interesting is the effect this degradation will have on the collector energy output Qcol. Figure 5 faces the yearly energy output of the collector without taking into account degradation (Qcol standard) with the yearly energy gain delivered by the present simulation with consideration of the degradation process. Qcol standard remains unchanged over the whole period, as no degradation was considered. Even though only an assumed degradation of the parameters was considered, the comparison of both simulations makes clear that the loss due to degradation might constitute a considerable part in the course of 20 years.

23 Qcol [kwh] Fig. 5: Comparison of yearly energy gain with and without consideration of degradation progress 6. SUMMARY The energy output of solar thermal collectors can be influenced considerably through degradation processes. In order to be able to calculate the change of performance with time due to degradation effects, the degradation processes have to be implemented in a collector model. A first draft of a collector performance model that takes into account degradation effects caused by temperature was implemented in TRNSYS type 132. Using this model the overall energy output during the whole lifetime of the collector can be predicted. Besides the determination of the overall energy output of the collector other interesting aspects can be investigated by using the model. System simulations can give information about: - Does an anti-stagnation strategy really result in a significantly reduced degradation of the collector? - Will the advantage of antireflective glazing prevail the disadvantage of higher thermal stresses over the lifetime? In order to be able to give reliable forecasts, additional investigations are necessary so that a uniform accelerated testing procedure can be developed. One important aspect, would be the establishment of a relationship between accelerated ageing in laboratory and ageing under operating conditions, which is presently part of the research work at ITW, Germany. 7. NOMENCLATURE Qcol standard Qcol with degradation years [a] a 1 [W/(m²K)] heat loss coefficient a 2 [W/(m²K²)] temperature dependent heat loss coefficient a n [-] acceleration due to increase of temperature a n = t ref /t n c p [J/(kgK)] specific heat capacity of the heat transfer fluid C abs [J/(m²K)] absorber heat capacity C fl [J/(m²K)] fluid heat capacity E T [J/mol] activation energy F [-] collector efficiency factor G [W/m²] hemispherical irradiance m& [kg/s] mass flow rate of the heat transfer fluid q& abs [W/m²] heat flux from the absorber to the heat transfer fluid q& loss [W/m²] heat losses a of the absorber q& use [W/m²] collector output Qcol [kwh/y] yearly collector output R [J/(mol K)] individual gas constant (R=8.314 J/(mol K)) t n [h] time during which the collector is exposed the temperature T n, resulting in the same degradation effect as under reference temperature t ref [h] time during which the collector is exposed to a reference test temperature T ref T n [K] any other temperature the collector is exposed during a certain time t n T ref [K] reference test temperature (UA) [W/K] heat transfer capacity rate between absorber and fluid U 1 [W/(m²K)] heat loss coefficient U 2 [W/(m²K²)] temperature dependent heat loss coefficient ϑ abs [ C] absorber temperature ϑ amb [ C] ambient temperature ϑ fl,in [ C] fluid inlet temperature ϑ fl,m [ C] mean fluid temperature ϑ fl,out [ C] fluid outlet temperature (τα) [-] transmittance-absorptance product 8. REFERENCES (1) EN 12975, Thermal solar systems and components Solar Collectors, June 2001 (2) IEA Task 10 Accelerated Life Testing of Solar Energy Materials, case study of selective solar absorber coatings materials for DHW systems, February (3) ISO/CD , Solar Energy, Materials for flat plate collectors, Qualification Test Procedures for Solar Absorber Surface Durability, (4) IEA Task 27, Adoption of General Methodology to Durability Assessment of Polymeric Glazing Materials, Gary Jorgensen, NREL, USA, 2003 (5) ISO :1994 Test methods for solar collectors Part 1: Thermal Performance of glazed liquid heating collectors including pressure drop, 1994

24 Date: Editor Bo Carlsson 1 IEA Solar Heating and Cooling Program Task 27 Performance of Solar Façade Components Project: Service life prediction tools for Solar Collectors Recommended qualification test procedure for solar absorber surface durability 1 Address: SP Swedish National Testing and Research Institute, P.O.Box 857, SE Borås, bo.carlsson@sp.se

25 Contents Page Foreword...iv Introduktion...v 1 Scope Normative references Terms and definitions Requirements and classification Test methods for assessing material properties as measure of absorber performance Sampling and preparation of test specimens Solar absorptance Thermal emittance Adhesion Tests for assessing the thermal stability of absorber surface Principle Apparatus Procedure for execution of high temperature tests Qualification procedure Tests for determining the resistance to condensed water of absorber surface Principle Apparatus Procedure for execution of constant condensation tests Test for determining absorber surface corrosion resistance to high humidity air containing sulphur dioxide Principle Apparatus Reference test specimen Procedure for execution of corrosion test in high humidity air containing sulphur dioxide Qualification procedure Report Annex A (normative) Procedure for determination of solar absorptance and thermal emittance of absorber surface A.1 Assessment of solar absorptance A.1.1 Scope A.1.2 Apparatus A.1.4 Evaluation of spectral absorptance values A.1.5 Evaluation of solar absorptance A.1.6 Crucial factors in the assessment of solar absorptance A.2. Assessment of thermal emittance A.2.1 Scope A.2.2 Apparatus A.2.3 Assessment of thermal emittance A.2.4 Crucial factors in the assessment of thermal emittance from spectral measurement Annex B (normative) Temperature and failure time characteristics in assessment of thermal stability of absorber surface... 19

26 Annex C (normative) Temperature/condensation and failure time characteristics together with qualification scheme in assessment of resistance to condensed water of absorber surface...23 Annex D (informative) Suitable designs for test apparatus to be used in the qualification testing of solar absorber surface...26 Annex E (informative) Bibliography...28

27 Foreword This recommended qualification procedure is primarily based on the results of work performed on solar collector absorber surfaces within the framework of the International Energy Agency Program on Solar Heating and Cooling. The IEA work on solar collector absorber surfaces is reported in references given in Annex E.

28 Introduction To effectively select, use and maintain a material in a given application, its degradation under service conditions must be predicted prior to use. Preferably, the durability of the material should be expressed quantitatively in terms of an expected service life. Durability in this case is the ability of a material to withstand deterioration caused by all external factors in the environment, which may influence the performance of the material under service conditions. Service life is defined as the period of time after installation during which specific material properties important for the performance of the material meet or exceed minimum acceptable values. The service life of a material is, thus, not solely dependent on its physical and chemical properties, but also on its performance requirement in the application considered, and on the external environmental factors, which influence performance under service conditions. In design work, the important question is if a specific material can be expected to have a service life longer than a certain value, the so called design service life; the latter dictated by life cost considerations taking into account the total system. Service life assessment may be based on feed back data from practise or on results from so called qualification or acceptance durability tests. The present recommended qualification procedure for solar absorber surface durability is based on the conduct of a series of short term durability tests. During a test the optical performance of the absorber surface tested is determined by measuring its solar absorptance and thermal emittance. From the loss in optical performance of the absorber surface, its failure time in the test performed is assessed and compared with the shortest acceptable failure time set by the design service life of the absorber. Design service life, performance requirement defining failure time in terms of loss in optical performance, classification of type and levels of environmental stress are set under the assumption, that the absorber surface tested will be installed in a flat plate solar collector for use in domestic hot water systems. The recommended qualification procedure may favourably be used in the development and validation of new kinds of absorber surfaces. From the results of tests, it can be concluded whether it is likely that an absorber surface tested may meet the requirement for an acceptable service life also in practise. The recommended durability testing procedure has proved to give results in fairly good agreement, both qualitatively and quantitatively, with what has actually been observed on absorber surfaces tested for longer time periods in solar collectors working under typical domestic hot water system conditions.

29 Recommended qualification test procedure for solar absorber surface durability 1 Scope This recommended procedure specifies a scheme of short term durability tests to be used in the qualification of absorber surfaces for the intended use in single glazed flat plate solar collectors for domestic hot water production. An absorber surface is considered qualified if it meets the requirement of design service life of 25 years. The maximum loss in the optical performance of the absorber surface, defining its service life, is fixed at a level corresponding to a reduction in solar domestic hot water system performance, i.e. solar fraction, of 5 % in relative sense. In the qualification procedure, the environmental factors considered the most important for the absorber surface durability are limited to temperature, humidity, and atmospheric corrosivity, the latter expressed in terms of concentration of sulphur dioxide in high humidity air. The program of short term durability tests is accordingly restricted to the simulation of three types of absorber surface degradation processes: a) high temperature degradation, b) degradation by the action of condensed water on the absorber surface, and c) degradation caused by high humidity in air containing a small concentration of sulphur dioxide as an airborne pollutant. To quantify the expected environmental stress on the absorber surface determining its service life, microclimate data are utilized representing typical service conditions for absorbers in single-glazed flat plate collectors for domestic hot water production. For interpretation of the test results, time-transformation functions are used to relate intensities of the environmental factors under service conditions to that of the test conditions for the short-term durability tests. The requirement on the service life of the absorber surface may then be converted into acceptable failure times in the short-term tests. To conclude whether a specific absorber surface is qualified or not, requires a scheme of short-term tests of each category to be executed. The recommended procedure is applicable for qualification of all kinds of absorber surface materials designed for flat plate collectors for the intended use of domestic hot water production. It is particularly suited for qualification of electroplated and sputtered selective absorber coatings for which the procedure was originally developed. If the absorber surface consists of an organic coating, e.g. a selective paint system, the effect of UV-degradation on optical performance of coating should be considered. 2 Normative references The following referenced documents are indispensable for the application of this recommended procedure. For dated references, only the edition cited applies. For undated references, the latest edition of the referenced document (including any amendments) applies. ISO Paints and varnishes - Pull-off test for adhesion ISO orrosion of metals and alloys - Corrosivity of atmospheres - Determination of corrosion rate of standard specimens for the evaluation of corrosivity ISO Corrosion tests in artificial atmosphere at very low concentrations of polluting gas(es)

30 ISO Corrosion of metals and alloys - Removal of corrosion products from corrosion test specimens ISO :1992 Solar Energy - Reference solar spectral irradiance at the ground at different receiving conditions. Part 1: Direct normal and hemispherical irradiance for air mass Terms and definitions For the purpose of this recommended qualification procedure, the following definitions apply. 3.1 Design service life The time period of exposure under service conditions after installation during which the absorber surface is expected to meet the performance requirement. 3.2 Failure time The time period of exposure in the test during which the absorber surface meets the performance requirement. 3.3 Solar absorptance α s Fraction of solar radiation absorbed by absorber surface determined from measurement as is described in Annex A. 3.4 Thermal emittance ε Near normal thermal emittance at 100 C determined from measurement as is described in Annex A. 3.5 Performance criterion function PC Quantity used for expressing the change in optical performance of absorber surface defined as: PC = - α s + 0,50 ε (1) The change in the solar absorptance α s = α s,t - α s,i (2) with α s,t equal to the value of the solar absorptance at the actual time of the test or at service, and with α s,i equal to the initial value of solar absorptance Τhe change in the thermal emittance ε = ε t - ε i (3) with ε t equal to the value of the thermal emittance at the actual time of the test or at service and with ε i equal to the initial value of thermal emittance. NOTE 1 This performance criterion function is primarily based on location averaged values of the performance of typical solar domestic hot water systems; see reference [1] in Annex E. But, in the IEA Task 10 testing procedure referred to a slightly different definition of PC is used, namely PC = - α s ε. Investigations made by the IEA MSTC group, however, showed that a weighting factor of 0.5 for the thermal emittance is more appropriate; see reference [6] in Annex E. 2

31 4 Requirements and classification 4.1 For classification of the durability of the absorber surface, the following performance requirement shall apply: PC = - α s + 0,50 ε < 0,05 (4) NOTE 1 Higher values for the PC function may be used if considered more appropriate. PC < 0,10 should mean that the optical performance of absorber surface and, thus also the performance of solar domestic hot water system, is allowed to be reduced to a level equal to 90% of its original value during the design service life time period. For further details; see reference [1] in Annex E 4.2 Before durability testing of an absorber surface, all test specimens, sampled and prepared as described in subclause 5.1, shall be characterized with respect to their value for solar absorptance, determined as described in subclause 5.2, and their value for thermal emittance, determined as described in subclause 5.3. To be qualified for testing, the set of test specimens shall have a standard deviation in the determined values for solar absorptance of less than 0,01 and for the determined values of thermal emittance a standard deviation less than 0, For coated absorber surfaces, three extra test specimens shall be prepared and the adhesion of coating on those test specimens be assessed, as is described in subclause 5.4. For the absorber to be qualified for testing, the adhesion of coating shall be > 0,15 MPa for all test specimens. NOTE 3 If considered accurately enough, the adhesion of the coating may be assessed by a more simple method. The method of ISO 2409 Paints and varnishes - Cross-cut test may be used and the requirement for satisfactory adhesion be set at the degree of 0 or 1. Alternatively some suitable method in ISO 2819 Metallic coatings on metallic substrates - Electrodeposited and chemically deposited coatings - Review of methods available for testing adhesion may be used. 4.4 For an absorber surface to be qualified with respect to its thermal stability, the procedure of durability testing as is described in clause 6 shall be applied. An absorber surface with sufficient thermal stability shall meet the requirement for test results as is specified in clause For an absorber surface to be qualified with respect to its resistance to condensed water when used in a non-airtight solar collector with more or less uncontrolled ventilation of air in the solar collector, the procedure of durability testing as is described in clause 7 shall be applied. An absorber surface with sufficient resistance to condensed water shall meet the requirements for test results as are specified in subclause For an absorber surface to be qualified with respect to its resistance to degradation caused by sulphur dioxide as an airborne pollutant, the procedure of durability testing as is described in clause 8 shall be applied. An absorber surface may be qualified for use in two classes of solar collectors; the two classes representing different severity classes as regards atmospheric corrosivity. Solar collector of type A: Airtight solar collector or solar collector with controlled ventilation of air in the space between the absorber surface and the cover plate. At the top and at the bottom of the frame of collector, it should be equipped with ventilation holes. The atmospheric corrosivity at the bottom part of the collector under service conditions may typically correspond to a corrosion rate of zinc of 0,1 g/m 2 per year. 3

32 Solar collector of type B: Non-airtight solar collector with more or less uncontrolled ventilation of air in the solar collector. The atmospheric corrosivity at the bottom part of the collector under service conditions corresponds to a corrosion rate of zinc of 0.3 g/m 2 per year. An absorber surface with sufficient resistance to degradation caused by sulphur dioxide in high humidity air, either regarding only a type A collector or regarding both type A and type B collectors, shall meet the requirements for test results as are specified in subclause Test methods for assessing material properties as measure of absorber performance 5.1 Sampling and preparation of test specimens For durability testing, test panels with an absorber surface area of preferably 50x50 mm shall be prepared. Sampling from larger pieces of absorber plate shall be made in such a way that variation in the optical properties between the different test specimens is minimized. For execution of the complete programme of durability tests of this recommended procedure, a minimum of 21 test panels are required. For absorbers with coatings three extra test panels are also required for assessment of the adhesion of coating. 5.2 Solar absorptance Determine the value of the solar absorptance for each of the 21 test panels from reflectance measurements as is described in Annex A.1. For the complete set of test panels calculate also the mean value and standard deviation of solar absorptance. 5.3 Thermal emittance Determine also the value for the thermal emittance for each of the 21 test panels as is described in Annex A.2. For the complete set of test panels calculate also the mean value and the standard deviation of thermal emittance. 5.4 Adhesion For coated absorber surfaces assess the adhesion of coating by making use of the three extra test panels prepared for this purpose. If the adhesion of coating is to be determined according to ISO 4624, use the general method for testing both rigid and deformable substrates when selecting test assembly and use test cylinders with a diameter of 20 mm. If a simpler method for assessment of adhesion is used, see subclause 4.3, proceed as described in the relevant standard. 6 Tests for assessing the thermal stability of absorber surface 6.1 Principle High-temperature ageing is frequently used in many technical application areas for the assessment of thermal stability of materials. A high temperature accelerates all kinds of processes, normally leading to an increased rate of degradation of materials. For a selective absorber coating composed of small metal particles, a high temperature enhances oxidation of metal decreasing mainly the absorptance of coating When installed in a single flat plate collector for domestic hot water production, an absorber surface is exposed to a temperature, which may vary greatly; in the extreme case from -20 C up to more than 200 C. As a measure of the level of thermal load, the effective mean temperature, T eff, during one year of service for an absorber surface is here used. It is defined by the following expression: 4

33 exp ( - E T T eff 1 Tmax ) = R Tmin exp (- R ET T -1 ) f (T) dt (3) f(t) is yearly based frequency function for service temperature of absorber surface in solar collector, meaning the time fraction of a year when service temperature is in the interval T to T+dT. T max is the maximal service temperature of absorber surface in collector in K. T min is the minimal service temperature of absorber surface in collector in K. E T is the Arrhenius activation energy expressing the temperature dependence of a thermal degradation reaction of absorber surface. R is the ideal gas law constant equal to 8,314 J/ K, mole The yearly based frequency function f(t) is determined by the external climatic load acting on the solar collector and the optical properties of the solar absorber surface and the glazing. In this recommended procedure it is assumed that the solar collector is under operating conditions for 11 months of a year and then producing tap water during daytime when the solar collector temperature exceeds 40 C. For one month of a year, during summer, the solar collector is under stagnation conditions. From a thermal ageing point of view it is only during the sunny days when the solar collector is under stagnation that the temperature load on the solar absorber surface will result in significant thermal degradation. The reference thermal load or temperature frequency function for one year representing service conditions in this recommended procedure corresponds therefore to 30 sunny days when the solar collector is under stagnation; see Figure B1 in Annex B. The temperature load acting on the solar absorber surface depends also on the optical properties of the absorber surface and so does the maximum solar absorber surface temperature during stagnation conditions. The maximum solar absorber surface temperature is in this recommended procedure determined from the solar absorptance and the thermal emittance by use of interrelations shown in Table B1 in Annex B. The reference thermal load in terms of an effective mean temperature is thereafter calculated as a function of the activation energy for thermal degradation by making use of the maximum absorber surface temperature. NOTE 1 If found more appropriate another temperature frequency function may be used to represent service conditions. The new effective mean temperature corresponding to a specific activation energy may be calculated by use of equation (3) To assess the thermal stability of the absorber surface, short-term tests, enhancing thermal degradation of the absorber surface at a constant high temperature, are used. To convert the design service life of 25 years into a shortest acceptable failure time, y R, for a constant temperature test to be executed at the temperature T R, the following time-transformation function is used: y R = 25 exp (- where E T R ( T 1 eff - T 1 R )) (4) T eff in K is the effective mean temperature of absorber surface defined by equation (3). The effective mean temperature will vary with the activation energy and so will also the shortest acceptable failure time for a specific constant temperature test To conclude whether an absorber surface is qualified or not, the results from at least two different constant temperature tests are needed unless the optical performance of solar absorber surface tested is unaffected during the first test. The qualification scheme and the conditions for the temperature tests are given in Figures B2 - B4 and Table B2, respectively. 5

34 6.2 Apparatus Testing chamber to be used for assessing the thermal stability of the absorber surface shall be constructed so that: 1) Constant temperature tests can be executed up to a temperature of at least 300 C. 2) The temperature of the test is monitored by sensing devices in the chamber so that it reflects the true temperature of the test panels. NOTE 1 Testing chambers with circulating air heating are recommended in favour of those based on radiative heating, because the temperature difference between the sensor and the test panels will be less due more uniform temperature condition in testing chambers of the former type. NOTE 2 When radiative furnaces are used, the temperature of the test panels and the temperature of the temperature sensor strongly depend on the radiation exchange with the heater and therefore also on their optical properties. It is in this case preferable to measure the temperature of the test panel for heating control. 3) The temperature is maintained at level of ± 1 C after stabilized conditions have been reached after start of test. NOTE 3 Even a symmetrical variation around the set temperature results in a higher effective mean temperature with respect to thermal degradation. 4) The temperature in the chamber is so uniform that the variation between absorber specimens tested simultaneously is within the range of ± 1 C; see Note 1 and 2. 5) During cooling down of chamber after high temperature exposure, the rate of temperature decrease shall be at least 10 C/min (from 200 C to 100 C). If the chamber does not meet this requirement, the test panels shall to be taken out of the chamber immediately after the specified testing time has been reached; see subclause Instruments for measuring of optical properties of absorber surfaces complying with the requirements as are specified in Annex A Tensile tester and test cylinders for measurement of adhesion of absorber coating in accordance with ISO 4624 as described in subclause 5.4. If a simpler method for assessment of adhesion is used, see subclause 4.3, use equipment complying with the requirements given in the relevant standard for assessment of adhesion. 6.3 Procedure for execution of high temperature tests Select three test panels of absorber surface with known solar absorptance and thermal emittance and qualified for testing according to subclause Increase the temperature of the testing chamber to the specified level of test. After this temperature has been reached, place the test panels of room temperature in the testing chamber Keep the test panels at this temperature level for the specified time period of test After the specified time period of test or of interruption for measurement of extent of degradation, decrease the temperature of the testing chamber down to room temperature at a mean rate of minimum -10 C/ min. during first phase of cooling down to 100 C below the specified test temperature. NOTE 4 If the testing chamber does not meet the requirement for minimum rate of cooling down as specified above, the test panels shall be taken out of the testing chamber immediately after the specified time period of test has been reached. The hot test panels shall after they have been taken out of the testing chamber be placed on a thermally insulating material to minimize damages, which may result from the thermal chock the test panels are exposed to during cooling down. 6

35 6.3.5 Determine the solar absorptance and the thermal emittance of test panels as is described in Annex A Calculate from the change in solar absorptance and thermal emittance of test panels, the value of the PC function for each panel tested by use of equation (1), see clause 3. Calculate also the mean value of the PC function for the different test panels If the test was interrupted only for measurement of extent of degradation, reintroduce the test panels after measurement into the climatic chamber after stabilized test conditions at the specified levels have been confirmed. 6.4 Qualification procedure Determine from the mean values of the solar absorptance and the thermal emittance of the solar absorber surface to be tested the expected maximum absorber surface temperature T max by use of Table B1 in Annex B; see also Figure B2 in Annex B Determine from the T max value the temperature T 1 of the first test to be performed making use of Table B With a set of three test panels perform test, as described in clause 6.3, at T 1 and measure α s and ε after the testing times 36, 75, 150, 300, and 600 h or till PC = α s + 0,50 ε > 0,05 is reached. Introduce the time t 1, which is the last testing/measuring time with PC < 0, If PC 0,01 after t 1 = 600 h the absorber surface is qualified with respect to its thermal stability if it meets also the requirement on satisfactory adhesion as specified in subclause If PC > 0,05 at t 1 = 300 h or PC > 0,01 after t 1 = 600 h check whether the absorber surface meets the adhesion requirement as specified in subclause 4.3. If the requirement on adhesion is met proceed as described in subclause If the requirement on adhesion is not met the solar absorber surface is not qualified with respect to its thermal stability unless adhesion measurements performed on the solar absorber surface after a testing time corresponding to PC = 0,05 in a new test show that the adhesion requirement is met. Then, proceed as described in subclause Use Table B2 to determine the T 3 value which corresponds to the previously determined T max value. Determine also from Table B2 the testing time t 3 which corresponds to the previously determined t 1 value; see also Figure B3 in Annex B With a new set of three test specimens perform a test at T 3 for a time period of t 3 and measure α s and ε to determine PC If PC (T 3, t 3 ) PC (T 1, t 1 ) the absorber surface is qualified with respect to its thermal stability If PC > 0,05 at t h check whether the absorber surface meets the adhesion requirement as specified in subclause 4.3. If the requirement on adhesion is not met the solar absorber surface is not qualified with respect to its thermal stability unless adhesion measurements performed on the solar absorber surface after a testing time corresponding to PC = 0,05 in a new test show that the adhesion requirement is met. Then, proceed as described in subclause Use Table B2 to determine the T 2 value which corresponds to the previously determined T max value. Determine also from Table B2 the testing time t 2 which corresponds to the previously determined t 1 value; see also Figure B4 in Annex B With a new set of three test specimens perform a test at T 2 for a time period of t 2 and measure α s and ε to determine PC. 7

36 If PC (T 2, t 2 ) PC (T 1, t 1 ) the absorber surface is qualified with respect to its thermal stability. 7 Tests for determining the resistance to condensed water of absorber surface 7.1 Principle High humidity and the effect of moisture and condensed water on materials may initiate many kind of degradation reactions. Selective absorber coatings composed of inorganic oxides may sometimes undergo hydratization reactions increasing the thermal emittance of coating. High humidity must prevail for electrochemical corrosion to occur causing oxidation of metal and as a result the absorptance of coating may decrease As the casing of a flat plate collector is usually ventilated, this means that the absorber surface is in contact with the ambient air. Humid air from the ambient therefore enters the collector and sometimes the temperature of the collector inside is so low related to the humidity level that condensation of water takes place. As condensation of water appears in most of today s solar collectors, some collectors are not even rain tight, this means, an absorber surface has to resist periods of exposure in very humid atmospheres during its service life. An absorber surface should therefore to be qualified according to this procedure resist a humidity load representative for a non-airtight solar collector with more or less uncontrolled ventilation of air in the solar collector. The yearly time fraction, when the relative air humidity in the gap between absorber and cover plate exceeds 99 %, is used as a measure of the severity of the humidity at service conditions. The severity depends, however, also on the effective mean temperature during such time periods of high humidity. The effective mean temperature during high humidity conditions is defined as in equation (3), see subclause In this case equation (3) will, accordingly, contain the parameters given below with the following meaning: f H (T) is the yearly based frequency function for service temperature of absorber surface in the solar collector when the relative humidity level exceeds 99%, meaning the time fraction of a year when service temperature is in the interval T to T+dT and the relative humidity level exceeds 99%. T H,max in K is the maximal service temperature of the absorber surface in the solar collector, when the relative humidity level exceeds 99%. T H,min in K is equal to 0 C, as below this temperature ice is formed on the surface of absorber. E H,T is the Arrhenius activation energy expressing the temperature dependence for a possible degradation reaction of absorber surface caused by the action of condensed water. In Figures C.1 in Annex C, the function f H, (T) of this recommended procedure is shown. NOTE If found more appropriate other humidity/ temperature data may be used to represent service conditions. The new effective mean temperatures for specific activation energy values may be calculated by use of equation (3) To assess the resistance to condensed water of absorber surface, short-term tests at different temperatures of absorber surface are performed. To obtain constant condensation of water on the surface of absorber during testing, the surrounding air is kept at a temperature 5 C above the temperature of absorber surface and at a relative humidity of 95%. To convert the design service life, set at 25 years, into a shortest acceptable failure time for a short term test, equation (4), see subclause 6.1.4, is also here applied. In Figure C.2 in Annex C, the shortest acceptable failure time, as function of activation energy, is shown for a series of constant condensation tests To conclude whether an absorber surface is qualified or not, the results from at least two different constant condensation tests performed at two different temperatures are needed. For this recommended procedure the conduct of tests at absorber surface temperatures of 40 C and 30 C or at 40 C and 60 C are contained. 8

37 7.2 Apparatus Climatic chamber to be used shall be constructed so that: 1) Climatic conditions ranging from room temperature and 50% RH up to at least 65 C and 95% RH can be obtained, controlled and monitored during test. 2) The temperature can be maintained at a level of ± 1 C and the relative humidity at a level of ± 3 % RH relative to the specified climatic conditions during test Liquid-cooled sample holder for temperature control of test panels in climatic chamber constructed so that: 1) The test panels can be fixed to it so that the test panels will be electrically insulated from each other and the sample holder. 2) The test panels will be in good thermal contact via the sample holder to the heat transfer liquid, which are used for cooling and control of the temperature of the test panel by aid of a thermostatic bath. 3) The test panels will be positioned at an angel of 45 relative to the horizontal plane. 4) The temperature of sample holder can be measured for control of test temperature. This can preferably be made by use of Pt 100 silicon foil sensor which adheres to the surface of the sample holder. A suitable sample holder made of aluminium and which can be used for the purpose of this test is described in Annex D. NOTE The crucial point in ensuring a high reproducibility of the condensation tests is the careful mounting of the test panels on the sample holder. A thin layer of an electrically insulating heat sink compound in combination with a foil for distance control, made of e.g. Teflon, can preferably be used to guarantee good thermal contact. The arrangement made means that bimetallic corrosion will be prevented too Thermostatic bath for temperature control of the liquid-cooled sample holder complying with the requirement of a temperature constancy over time for the sample holder of ± 0,5 C Instruments for measuring of optical properties of absorber surfaces complying with the requirements as are specified in Annex A Tensile tester and test cylinders for measurement of adhesion of absorber coating in accordance with ISO 4624, see subclause Procedure for execution of constant condensation tests Select three test panels of absorber surface with known solar absorptance, thermal emittance, and qualified for testing according to subclause Adjust the temperature of the climatic chamber at a level of 5,0 C above the specified test temperature for test panels and set the relative humidity level of climatic chamber at 95% RH. Adjust also the temperature of the thermostatic bath so that the temperature of the sample holder placed in the climatic chamber will reach the specified test temperature After stabilized conditions have been obtained, fix the test panels to the sample holder. The testing time period starts when condensation of water is first observed on the surface of the test panels After the specified time period of the test or of interruption for measurement of extent of degradation, take out the test panels from the climatic chamber and remove gently the excess of condensed water on the surface of the test panels by a clean water absorbing paper. 9

38 7.3.5 Condition the test panels under normal laboratory climatic conditions for at least 2 hrs. Determine the solar absorptance and the thermal emittance of the test panels as are described in Annex A Calculate from the change in solar absorptance and thermal emittance of the test panels, the value of the PC function for each panel tested by use of equation (1), see clause 3. Calculate also the mean value of the PC function for the different test panels If the test was interrupted only for measurement of extent of degradation, reintroduce the test panels after measurement into the climatic chamber after stabilized test conditions at the specified levels have been confirmed. 7.4 Qualification procedure Perform a constant condensation test, involving exposure of the test panels at 40 C with interruptions for measurement of extent of degradation after 80 hrs, 150 hrs, 300 hrs, and 600 hrs as described in subclause Dependent on the PC mean value obtained after 600 hrs of test proceed as follows: If a coated absorber surface is tested check first whether the coating is qualified with respect to adhesion according to and then 1) If PC 0,015 the absorber surface is qualified with respect to its resistance to condensed water 2) If 0,015 < PC 0,05 proceed to subclause ) If PC > 0.05 after 80 hrs of testing, perform a new test at 40 C for 40 hrs. If PC > 0.05 after 150 hrs of testing, 300 hrs of testing, and 600 hrs of testing, perform new tests at 40 C for 115 hrs, 225 hrs, and 450 hrs, respectively. Perform the new test without any interruptions for measurements until after the test is completed. After PC value has been determined after test, proceed to subclause If a coated absorber surface is tested, measure, as described in subclause 5.4, the adhesion of coating on all test panels after that a PC mean value > 0,05 for the first time is obtained in the 40 C constant condensation test or after 600 hrs of testing. If the adhesion requirement as specified in subclause 4.3 is not met the absorber surface is not qualified with respect to its resistance to condensed water After that the mean value of PC has been found > 0,05 and a new 40 C constant condensation test has been performed, estimate from the results by interpolation, the failure time, or in other words the testing time period, which should correspond to PC = 0,05. Use preferably, for this estimation a numerical or a graphical procedure, assuming PC plotted versus the testing time period can be represented by a polynomial series of an order which give the best fit to the test data obtained Use the estimated value of failure time for the 40 C test and determine the corresponding or lowest acceptable activation energy from the 40 C curve in Figure C.2 of Annex C Use this value for the activation energy and determine, from the 30 C curve in Figure C.2 in Annex C, the corresponding or shortest acceptable failure time for the absorber surface in a test to be performed at 30 C Perform, as described in subclause 7.3, a 30 C constant condensation test for a testing time period corresponding to the shortest acceptable failure time period according to subclause Dependent on the PC mean value obtained after the 30 C test, the following shall be concluded: 1) If PC > 0,05, the absorber surface is not qualified with respect to its resistance to condensed water. 2) If PC 0,05, the absorber surface is qualified with respect to its resistance to condensed water. 10

39 7.4.9 If after 600hrs of test at 40 C 0,015 < PC 0,05 according to subclause execute a third kind of constant condensation test, involving exposure at 60 C for a testing time period of 85 hrs, as described in subclause Dependent on the PC value obtained after the 60 C test, the following shall be concluded: 1) If PC after 85 hrs in the 60 C test > PC after 600 hrs in the 40 C test, the absorber surface shall be qualified with respect to its resistance to condensed water 2) If PC after 85 hrs in the 60 C test PC after 600hrs in the 40 C test, it can not be concluded whether the absorber surface shall be considered qualified or not and, therefore, a more comprehensive investigation on its resistance to condensed water is recommended The qualification procedure to be used is schematically shown in Figure C.3 in Annex C. 8 Test for determining absorber surface corrosion resistance to high humidity air containing sulphur dioxide 8.1 Principle Many air borne pollutants, such as sulphur dioxide present in air as a trace substance, accelerates highly, as well known, electrochemical corrosion of most metallic materials at high humidity As a solar collector exchanges air with the ambient, this means also that air borne pollutants will be transported from the ambient into the collector and the absorber surface. Airborne pollutants may, therefore, influence the longterm optical performance of an absorber by promoting corrosion attacks on the metallic substrate. Air borne pollutants may also cause loss in the optical performance of selective absorber coatings pigmented with small metallic particles, due to oxidation/corrosion of the metal particles Due to the complex nature of degradation caused by air pollutants, atmospheric corrosivity, as defined from exposure of standard panels of metals as described in ISO 9226, is here used. More precisely, the corrosion rate of zinc is taken as a measure for severity of environmental stress on absorber surface in collectors and sulphur dioxide is considered the dominating air borne pollutant as regards degradation of absorber surface. As described in subclause 4.6, two severity classes are here used related to two kinds of collectors for which corrosion rates for zinc are defined. NOTE 1 If found more appropriate other levels of atmospheric corrosivity may be used to represent service conditions. The new levels of atmospheric corrosivity may be expressed also in terms of corrosion rates of other metal than zinc, if found more appropriate To assess the atmospheric corrosion resistance of the absorber surface, exposure in high-humidity air containing small concentrations of sulphur dioxide (1 ppm) is adopted. To convert the design service life of 25 years into shortest acceptable failure times relevant for the two severity classes defined, the principle of comparative testing is applied. This means that the time transformation function for degradation of optical performance of absorber surface is assumed to be the same as the one for the corrosion of zinc. NOTE 2 In the study of nickel pigmented anodized aluminium absorber coatings as described in reference [1] of Annex E, it was roughly assumed that the time transformation function for degradation of the nickel-pigmented anodized aluminium coatings was the same as for the corrosion of zinc. The reasonableness of this assumption was supported by results from measurements of deposited amounts of sulphur dioxide on this kind of coatings both during laboratory tests, involving exposure in high humidity air containing sulphur dioxide, and during in-service tests of absorbers in collector; see also reference [4] in Annex E To conclude whether an absorber surface is qualified or not, the results from one test, involving exposure in circulating air of a relative humidity of 95% RH, temperature of 20 C, and with a volume fraction of sulphur dioxide of 10-6, are needed. The test is essentially performed as described in ISO

40 8.2 Apparatus Climatic cabinet with inner chamber and gas flow system, shall comply with the requirements of ISO An example of a suitable design is shown in Annex D. The equipment used for testing shall be constructed so that: 1) The inner chamber and gas flow system consist of inert materials, e.g. Teflon or glass, to avoid or minimize adsorption of sulphur dioxide on surfaces other than of that of the test panels. 2) The air flow and sulphur dioxide injection system are designed to ensure uniform test conditions in the inner chamber or working space of the cabinet. NOTE1 In the most common design of test equipment, the test atmosphere in the working space is obtained by continuously introducing the necessary quantity of sulphur dioxide into a damp air flow to obtain the required concentration. Sulphur dioxide and conditioned air are mixed outside the cabinet. The conditioned air is taken from the outer chamber of the climatic cabinet. The air flow after injection of sulphur dioxide is then mixed with a flow of recirculated test atmosphere and the resulting gas flow admitted into the inner chamber or working space of the cabinet. Half of the flow of the test atmosphere through the inner chamber may be recirculated. To ensure uniform test conditions in the working space, the test atmosphere is normally supplied to the working space from the bottom and the outlet is placed at the top. Perforated plates are placed in front of the openings to assure uniform air flow through the working space. 3) Uniformity of temperature in the working space shall be better than ± 1 C and uniformity of relative humidity better than ± 3 %. In terms of corrosivity, as expressed in terms of corrosion rates of standard metals, the uniformity shall be not less than 5%. NOTE 2 The uniformity of the test conditions in the working space may be checked regularly by exposing a number of copper coupons, placed at different positions in the working space during sulphur dioxide exposure. The differences in weight change of the metal coupons indicate if the uniformity of test conditions is within specified range. 4) The damp air flow shall be within the tolerance for the specified temperature ± 1 C and relative humidity ± 3% and the linear flow rate of air shall be in the range of 1 mm/s to 5 mm/s. The damp air flow shall be free of water droplets or aerosols. NOTE 3 - In the most common design of test equipment, the air is introduced to the outer chamber of the cabinet after filtration and purification by activated charcoal and a particulate filter. The sulphur dioxide gas may taken from a pressurized cylinder filled with high-purity sulphur dioxide gas at a volume fraction of 10-3 in high-purity nitrogen gas. 5) For exposure of test panels in the working space, specimens holder shall be used so the test panels do not shield one another or disturb the uniformity of the flow of air across the chamber. 6) The temperature, relative humidity, and concentration of sulphur dioxide in the air flow at the outlet of the working space is monitored so that they reflect the true test conditions for the test panels Instruments for measuring of optical properties of absorber surfaces complying with the requirements as are specified in Annex A Tensile tester and test cylinders for measurement of adhesion of absorber coating in accordance with ISO 4624; see subclause Reference test specimen For measurement of corrosivity in the working space during testing, standard test panels of zinc shall be used. The test panel of zinc may preferably has a dimension of 50 mm x 100 mm x 1 mm, and shall have an impurity level at or lower than 0,5 %. 12

41 Before testing the test panels shall carefully be cleaned with a hydrocarbon solvent in order to remove all marks of dirt, oil, or other foreign matter capable of influencing the result from the corrosion rate determination. After drying the panel shall be possible to weigh to the nearest 0,1 mg. After testing the corrosion products on the zinc panel shall be removed as described in ISO 8407 and the mass loss of the metal be determined. Use for removal of corrosion products a solution with a mass fraction of 5% of acetic acid in distilled water. During chemical removal of corrosion products work at room temperature with cleaning cycles with a length of around 2 min. Express mass loss of metallic zinc in mg / m Procedure for execution of corrosion test in high humidity air containing sulphur dioxide Select three test panels of absorber surface with known solar absorptance, thermal emittance, and qualified for testing according to subclause Prepare the climatic cabinet for test by firstly adjusting temperature to 20 C, air humidity to 95% RH and air flow rate at a selected value between 1 mm/s and 5 mm/s. After stable conditions have been reached, adjust the sulphur dioxide gas flow to the specified level so that the sulphur dioxide volume fraction in the inlet air flow to the working space will be at a level of When stabilized conditions have been reached also after this step, open the door to cabinet and place the test and reference panels quickly in the working space. After a testing time period not more than 5% of the specified time period of test, the concentration of sulphur dioxide in the outlet air flow from the working space shall not be less than 90% of that in the inlet flow of air. If the concentration of sulphur dioxide in the outlet air flow is lower, this probably means the total area of the test panels in the cabinet is too large. During the test, check the exposure conditions regularly and, if necessary, make adjustments to the specified levels After the specified time period of test or interruption for measuring the extent of degradation in optical performance of the test panels, take out the test panels from the working space of the cabinet and place them in a desiccator over silica gel for at least 2 hrs. Determine the solar absorptance and the thermal emittance of the test panels as are described in Annex A Calculate from the change in solar absorptance and thermal emittance of the test panels, the value of the PC function for each panel tested by use of equation (1), see clause 3. Calculate also the mean value of the PC function for the different test panels If the test was interrupted only for measurement of extent of degradation, reintroduce the test panels after the measurement into the working space after stabilized test conditions at the specified levels have been confirmed. 8.5 Determination of shortest acceptable failure times in test by use of reference test specimens If the corrosion rate of zinc, under the test conditions specified in subclause 8.4, is not known from previous exposures in the test equipment used, proceed as follows Perform a corrosion test as described in subclause 8.4 with three pairs of reference test panels of zinc with known initial mass, see subclause 8.3. Make interruptions of the test after 90 hrs, 180 hrs and 360 hrs and take out from cabinet at each interruption of test one pair of zinc panels for determination of mass loss in metallic zinc caused by corrosion during test, see subclause Assume the mass loss in metallic zinc versus the testing time period is linear and determine by least square fitting, the mean corrosion rate of zinc, r Zn, during the test conditions specified in subclause 8.4. Express r Zn in mg / m 2, h Use the mean corrosion rate of zinc and determine the following shortest acceptable failure times in hrs of the test according to subclause 8.4 as: 1) For a type A solar collector, see clause 4: Shortest acceptable failure time, t f,a = 2.5 / r Zn 13

42 2) For a type B solar collector, see clause 4: Shortest acceptable failure time t f,b = 7,5 / r Zn 8.6 Qualification procedure Perform a corrosion test as specified in subclause 8.4 for t f,b hrs, including also one interruption of the test for measurement of extent of degradation after t f,a hrs. If only qualification for the type A solar collector severity class is required complete the test after t f,a hrs If a coated absorber surface is tested, measure, as described in subclause 5.4, also the adhesion of the coating on all the panels after the test. If the requirement on adhesion as specified in subclause 4.3 is not met the absorber surface is not qualified with respect to long-term performance Dependent on the PC mean value obtained after the two testing times of subclause 8.6.1, the following shall be concluded regarding corrosion resistance of absorber surface: 1) If after t f,a hrs of test, PC < 0,05, the absorber surface is qualified for the type A solar collector severity class, if, when coated, it also fulfils the requirement of subclause ) If after t f,b hrs of test, PC < 0,05 too, the absorber surface is qualified also for the type B solar collector severity class, if, when coated, it also fulfils the requirement of subclause Report The test report shall give the following information. a) Reference to this recommended procedure b) The type and designation of the tested product of absorber surface c) Any deviations from the prescribed testing method d) Method for assessing adhesion if appropriate and used requirement for acceptable adhesion e) Test results f) Testing laboratory g) Dates for start and completion of tests 14

43 Annex A (normative) Procedure for determination of solar absorptance and thermal emittance of absorber surface A.1 Assessment of solar absorptance A.1.1 Scope This procedure specifies a method for determination of the directional solar absorptance α s, for near-normal incidence, i.e. 8 to 10 to the surface normal, from spectral directional absorptance α(λ) values in the spectral range of 0.32 µm µm. The spectral directional absorptance α(λ) values are determined from spectral (near-normal) directional/hemispherical reflectance ρ(λ) values measured on opaque samples of solar absorber surfaces at room temperature. A.1.2 Apparatus Spectral (near-normal) directional/hemispherical reflectance ρ(λ) values shall be determined by photometric integration by use of a spectrophotometer equipped with an integrating sphere. The wall of sphere shall be coated with a highly and diffusely reflecting coating, e.g. BaSO 4. In the photometric integration the specular component of reflected radiation has to be included. The sphere shall be designed so that the sample is part of sphere wall during the measurement. The detector shall be positioned so that it is shielded against radiation received directly from the sample. The geometry of the sphere shall preferably be so that the comparison method can be used, i.e. sample and reference are simultaneously part of the sphere wall. NOTE 1 When using the substitution method, i.e. sample and reference are alternately covering the measuring port, the sphere error must be corrected for by measuring the respectively corresponding brightness of the sphere wall. For calibration purposes, freshly pressed BaSO 4 powder or a diffusely reflecting white tile shall be used as reference. A.1.3 Measurement of reflectance values Solar absorptance shall to be assessed by use of the weighted ordinates method; see A.1.5. The reflectance shall be measured at al least 40 wavelengths in the range 0,32 µm - 2,50 µm. A.1.4 Evaluation of spectral absorptance values Calculate the near normal/hemispherical spectral reflectance of sample, ρ (λ), by use of the following expression: ρ (λ) = (R s (λ) / R r (λ) ). ρ r (λ) (A.1 ) where R s (λ) is the recorded reflectance value of sample. 15

44 R r (λ) is the recorded reflectance value of reference. ρ r (λ) is the near-normal / hemispherical reflectance of reference. NOTE 2 - Values for the spectral near normal / hemispherical reflectance of reference can in case of freshly pressed BaSO 4 powder be found in the literature, see for example reference [7] in Annex E. If a diffusely reflecting white title is used as reference, ρ r (λ) values should be available from manufacturer or be determined by own measurement using freshly pressed BaSO 4 powder as a reference. As no correction of possible distortion of the measured result when a specular component of the reflected radiation exists, the spectral directional absorptance α(λ) value are calculated as: α(λ) = 1 - ρ (λ) (A.2) A.1.5 Evaluation of solar absorptance The solar absorptance, α s, is calculated from the following expression:: α s = ( n n (α(λ) S λi )/ i=1 i=1 ( S λi λ i (A.4) where the set of λ i are the selected measuring wavelengths. λ i is the respective wavelength interval. S λi is the spectral solar irradiance according to ISO 9845, correctly summed over the respective wavelength interval. A.1.6 Crucial factors in the assessment of solar absorptance When assessing the solar absorptance of absorber surface according to this procedure, attention shall be paid to: 1) Possible anisotropy of the samples, i.e. the samples shall be marked to allow mounting of the samples at the measuring port using the same orientation and ensuring that the same areas of samples are measured. When establishing degradation-over-time relationships in an durability test, measurements of extent of degradation at different testing times should be performed on one and the same sample. 2) The different weighting of different directions of polarisation by the monochromators may distort the measurement, although this is not a great effect for near-normal incidence. A.2. Assessment of thermal emittance A.2.1 Scope This procedure specifies methods for determination of 1) total directional emittance ε n (100 o C) for near-normal incidence 16

45 2) spectral near-normal directional emittance ε (λ), and 3) total hemispherical emittance ε h (100 o C). These quantities are elaborated from measurements of 1) total hemispherical/directional reflectance ρ (100 o C) for thermal radiation incident from a black-body radiator of 100 o C, yielding ε n (100 o C), or 2) total directional emittance ε (100 o C) of heated samples, yielding ε n (100 o C), and 3) spectral directional / hemispherical reflectance ρ(λ) firstly yielding the directional spectral emittance ε n (λ) and by weighted integration with a Planck-distribution for a black body radiator at a chosen temperature, T, yielding finely ε n (Τ) Whenever the reflection method is used, transmittance has to be zero within spectral range measured. The spectral range is 2.0 µm µm. The angle of incidence or emission is restricted to near-normal, i.e. 8 to 10 to the surface normal. The temperature of sample is 100 o C in radiometric measurements on heated samples, and is room temperature when using the reflection method. A.2.2 Apparatus For determination of emittance quantities from spectral measurements, use for photometric integration a spectrophotometer equipped with diffusely reflecting gold coating as sphere wall. In photometric integration the specular component of reflected radiation shall be included. The requirements on sphere design is the same as described in subclause A.1.2. For calibration purposes, the following references are recommended: a) For broadband measurements, use reference Nextel Velvet Coating 2010 black (American indication: Nextel C10) with ε n (100 o C) = 0.95 and ε h = 0.90 NOTE 1 Emittance values refer to measurements made by J. Lohrengel, PTB Braunschweig, Germany. NOTE 2 For Devices and Services Emissiometer Model AE, use the references delivered by manufacturer with the instrument. c) For spectral measurements, as reference Labsphere diffusely reflecting gold coating "Infragold". NOTE 3 Recommended reference with calibration certificate available from Labsphere Inc., North Sutton, NH 03260, USA A.2.3 Assessment of thermal emittance a) For assessment based on broad-band measurements, use the procedure given by manufacturer of measuring instrument. b) For assessment based on spectral measurements in the wavelength range 2.0 µm µm, use the following procedure. Calculate the near normal/hemispherical spectral reflectance of sample, ρ (λ), by use of the following expression: 17

46 ρ (λ) = (R s (λ) / R r (λ) ). ρ r (λ) (A.5) where R s (λ) R r (λ) ρ r (λ) is the recorded reflectance value of sample. is the recorded reflectance value of reference. is the near-normal / hemispherical reflectance of reference. As no correction of possible distortion of the measured result when a specular component of the reflected radiation exists, calculate the spectral directional emittance, ε(λ), as: ε (λ) = 1 - ρ (λ) (A.6) Convolute the total emittance from spectral measurements, ε (100 o C), with the aid of the Planck function P λ for a black-body radiator with temperature of 100 o C, as: ε (100 o C) = ( ε(λ) P λ dλ)/ ( P λ dλ) (A.6) 0 0 NOTE Values for the Plank function can be found for example in most handbooks in physics. A.2.4 Crucial factors in the assessment of thermal emittance from spectral measurement When assessing the thermal emittance of absorber surface from spectral measurements, attention shall be paid to the same factors as are described in subclause A

47 Annex B (normative) Temperature and failure time characteristics in assessment of thermal stability of absorber surface T absorber [ C] :00 04:00 08:00 12:00 16:00 20:00 24:00 time [h] f [h] T abs [ C] Figure B.1 Reference thermal load for assessment of thermal stability of absorber surface (a) Measured solar absorber surface temperatures at stagnation condition during one day of clear sky conditions (b) Corresponding temperature frequency function for 30 days with clear sky conditions. The temperature profile of the selective absorber surface (α s = 0.94 and ε = 0.06 ) was measured in a commercial flat plate collector installed in Freiburg/Germany, facing south with a tilt angle of 45. The profile was recorded 25 th of August 1997 with a maximum global radiation of about 930W/m², measured in the collector plane. The maximum stagnation temperature was 184 C. 19

48 Table B.1 Relation between the maximum solar absorber surface temperature T max and the optical properties of the solar absorber surface to be tested. The line entitled AR α" has to be used for solar absorber surfaces that will be used in solar collectors with antireflective glazing leading to higher stagnation temperatures. 20

49 Table B.2 Test conditions for the different accelerated temperature tests used in the qualification of solar absorber surface T max ( C) T 1 ( C) T 2 ( C) t 2 (h) (t 1 =18 h) t 2 (h) (t 1 =36 h) t 2 (h) (t 1 =75 h) t 2 (h) (t 1 =150 h) T 3 ( C) t 3 (h) (t 1 =300 h) t 3 (h) (t 1 =600 h) <

50 Measure α s and ε and determine T max from Table B1 Use Table B2 to determine the T 1 value which corresponds to the determined T max value Perform test at T 1 and measure α s and ε after the testing times 36, 75, 150, 300, and 600 h or till PC = α s + 0,50 ε > 0,05 is reached. Introduce the time t 1, which is the last testing time PC < 0,05, then (a) (b) (c) If PC 0,01 after t 1 = 600 h the absorber surface is qualified if the adhesion after testing 0.15 MPa If PC > 0,05 at t 1 = 300 h or PC > 0,01 after t 1 = 600 h check that the adhesion 0.15 MPa and if so perform additional tests in accordance with Figure B3 If PC > 0,05 at t h check that the adhesion 0.15 MPa and if so perform additional tests in accordance with Figure B4 Figure B2 Qualification scheme for testing the thermal stability of solar absorber surfaces Use Table B2 to determine the T 3 value which corresponds to the previously determined T max value. Determine also from Table B2 the testing time t 3 which corresponds to the previously determined t 1 value Perform a test at T 3 for a time period of t 3 and measure α s and ε to determine PC If PC (T 3, t 3 ) PC (T 1, t 1 ) the absorber surface is qualified Figure B3 Continuation of the qualification scheme in Figure B2 for the (b) option Use Table B2 to determine the T 2 value which corresponds to the previously determined T max value. Determine also from Table B2 the testing time t 2 which corresponds to the previously determined t 1 value Perform a test at T 2 for a time period of t2 and measure α s and ε to determine PC If PC (T 2, t 2 ) PC (T 1, t 1 ) the absorber surface is qualified Figure B4 Continuation of the qualification scheme in Figure B2 for the (c) option 22

51 Annex C (normative) Temperature/condensation and failure time characteristics together with qualification scheme in assessment of resistance to condensed water of absorber surface hours/ year Figure C.1 Reference humidity/condensation yearly load used in this recommended procedure. The temperature frequency function represents the conditions when RH 99 during the reference year. Table C.1 Numerical data for the reference temperature frequency function f given in Figure C.1 Temp.( o C) f (h/year) Temp.( o C) f (h/year) Temp.( o C) f (h/year) Temp.( o C) f (h/year) -5 0, , , ,75-4 0, , , ,83-3 2, , , , , , , , , , , , , , , , , , , , , ,09 23

52 Figure C.2 Shortest acceptable failure time for absorber surface in different condensation tests given as a function of the activation energy for the degradation reaction. The failure time given corresponds to a service life with PC < 0,05 of 25 years. 24

53 Initial test (i1) Perform a test at a test panel temperature of 40 C and measure the extent of degradation after 80 hrs, 150 hrs, 300 hrs, and 600 hrs PC > 0.05 PC 0.05 check if adhesion 0.15 MPa 1) No The absorber surface is not qualified PC The absorber surface is qualified Yes < PC 0.05 Perform final test (f1) at 60 C for 85 hrs PC > 0.05 Perform initial test (i2) at 40 C, for a period as given in 2) below PC (f1) PC (i1) The absorber surface is qualified PC (f1) PC (i1) It cannot be concluded whether the absorber surface shall be considered qualified 3) Use the PC value obtained in initial test i2 to calculate the test conditions for the final test f2 as described in 4) below. Perform test f2 and determine the mean PC value The absorber surface is qualified if the adhesion 0.15 MPa 1) No PC 0.05 Yes The absorber surface is not qualified 1) If a tape test is used to check the adhesion between the coating and the substrate, the adhesion between the tape and the coating should be better than 0.15 MPa. 2) If PC > 0.05 after 80 hrs of testing, perform a new test at 40 C for 40 hrs. If PC > 0.05 after 150 hrs of testing, 300 hrs of testing, and 600 hrs of testing, perform new tests at 40 C for 115 hrs, 225 hrs, and 450 hrs, respectively. Perform test i2 without any interruptions for measurements until after complete test. 3) A more comprehensive investigation of the resistance to moisture is recommended 4) Estimate by interpolation, the testing time, which should correspond to PC = Determine the lowest acceptable activation energy on the 40 C curve in Figure 5 and also the corresponding testing time for a test f2 at 30 C. Figure C3 Test procedure for qualification of the resistance to condensed water of an solar absorber surface 25

54 Annex D (informative) Suitable designs for test apparatus to be used in the qualification testing of solar absorber surface Figure D.1 Schematic drawing of suitable sample holder to be used for condensation testing. Samples are fastened to the PTFE-coated surface of the cooling block by aid of a heat sink compound. 26

55 1. Climate chamber 2. External control 3. Inner chamber (Teflon walls) 4. Lead-through, inner chamber 5. Lead-through, climate chamber 6. Inlet for circulation 7. Outlet for circulation 8. Analysis point (movable) 9. Temperature sensor 10. Intake of conditioned air 11. Outlet from system 12. Mixing tube 13. Circulation pump 14. Evacuation pump 15. Gas analyzer 16. Connection to logger 17. Flow meters 18. Regulation valves 19. Shut off valve 20. Magnetic valve 21. Gas bottle 22. Intake of clean air to climatic chamber 23. Particle filter 24. Active coal filter 25. Air inlet Figure D2 Climatic cabinet suitable for use in corrosion tests involving exposure in high humidity air containing sulphur dioxide 27

56 Annex E (informative) Bibliography [1] Accelerated Life Testing of Solar Energy Materials - Case study of some selective solar absorber for DHW systems; A Technical report of Task X Solar Materials Research and Development of the International Energy Agency Solar Heating and Cooling Program; B. Carlsson, U. Frei, M. Köhl, K. Möller; SP-Report 1994:13, ISBN [2] Qualification Test Procedure for Solar Absorber Surface Durability, Carlsson, B.; Möller, K.; Köhl, M.; Frei, U.; Brunold, S. (2000). Solar Energy Materials and Solar Cells 61, [3] Accelerated Life Testing of Solar Absorber Coatings: Testing Procedure and Results.Brunold, S.; Frei, U.; Carlsson, B.; Möller, K.; Köhl, M.; (2000). Solar Energy 68, [4] Comparison between Predicted and Actually Observerved In-Service Degredation of a Nickel Pigmented Anodized Aluminium Absorber for Solar DHW Systems Carlsson, B.; Möller, K.; Frei, U.; Brunold, S.;Köhl, M.; (2000). Solar Energy Materials and Solar Cells 61, [5] Round Robin on Accelerated Life Testing of Solar Absorber Surface Durability. Brunold, S.; Frei, U.; Carlsson, B.;Möller, K.; Köhl, M.; (2000). Solar Energy Materials and Solar Cells 61, [6] Advanced procedure for the assessment of the lifetime of solar absorber coatings, M. Köhl, M. Heck, S. Brunold, U. Frei, B. Carlsson and K. Möller, Solar Energy Materials and Solar Cells (in press) [7] A study of the reflection factor of usual photometric standards in the near infrared; L. Morren, G. Vandermeersch, and P. Antoine; Lighting Res. and Techn. 4 (1972),

57 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 1 (5) Leo Spilg Polymeric materials in solar collectors Test methods and technical requirements Department of Chemistry and Materials Technology Borås 2004 This is a translation from the Swedish original document. In the event of any dispute as to the content of the document, the Swedish text shall take precedence.

58 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 2 (5) Index 1 Scope 3 2 Assumptions/tests Thermo-oxidative degradation Weatherability Chemical resistance 4 3 Requirements Absorption Transmittance Mechanical characteristics 5 4 Material composition 5 5 Report 5

59 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 3 (5) 1 Scope This SP-method relates to plastic- and rubber components in solar collectors. The present issue focus primarily on absorbers, reflectors, cover glazing and frames. The method can however with certain adjustments be applied to other polymeric material components, such as pipe systems and sealings. The purpose of the method is to ensure a 15-year lifetime of the components by lifetime analysis of included materials regarding mechanical characteristics, absorption and transmittance. The method is also intended to, in combination with other requirements, be used as a basis for P-marking of solar collectors. 2 Assumptions/tests The relevant material characteristics are degraded by among other things the following: - thermo-oxidative degradation. - solar radiation (foremost UV radiation). - water/humidity (possibly acid rain). - chemicals, foremost products used in heat transfer fluids. 2.1 Thermo-oxidative degradation Refers in this case to material degradation caused by time under the influence of atmospheric oxygen. The degradation caused can be expected to influence mechanical characteristics, absorption and transmittance. As the speed of this process is depending on temperature, the lifetime of a material at a certain temperature of use can be assessed by accelerated ageing at a higher temperature followed by an evaluation of the relevant characteristic. The choice of time and temperature/temperatures for this accelerated ageing can be complicated, and is governed by material type, possible existing knowledge or data for the material in question, expected temperature levels during use etc. The time/temperature correlation during the ageing will therefore in general be specific for each sample/product. Expected temperature levels during use will primarily be determined according to SP-method 3860 (only in Swedish). The evaluation is performed regarding mechanical characteristics, absorption and transmittance. The test method, especially for the mechanical characteristics, is obviously depending on the shape of the available test sample. Normally an impact or a tensile test is performed.

60 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 4 (5) 2.2 Weatherability Weatherability includes effects of solar radiation (primarily UV-radiation), rain/moisture and other environmental elements such as for example acid rain (sulphuric acid). The durability of materials is tested by exposure in a Weatherometer according to ISO , method A, as follows: - Black standard temperature: 65 ±3 C - Relative humidity: 50 ±5 % - Light intensity ( nm): 60 ±6 W/m² - Spray cycle: 102/18 min (dry interval between spraying 102 min followed by 18 min of spraying) - Time of exposure: 3000 hours The evaluation is performed regarding mechanical characteristics, solar absorption α s and/or solar transmission τ s. 2.3 Chemical resistance The characteristics of polymeric materials are in many cases degraded under the influence of chemicals, water and humidity, especially at elevated temperatures. The necessity to perform a durability test regarding this must be judged from case to case, depending on the anticipated type of chemicals, earlier experience and the operational characteristics of the solar collector. During the test, the materials are exposed to the chemicals in question during a time, and at a temperature relevant to the expected operational characteristics. Evaluation is performed regarding mechanical characteristics.

61 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 5 (5) Requirements 3.1 Absorption The solar absorptance, α s, is measured before and after ageing of the absorber. The solar absorptance is determined by acquiring a reflectance spectrum of the absorber surface between 200 and 2500 nm using a UV-VIS-NIR-spectrometer equipped with an integrating sphere attachment. The solar absorptance is calculated by convoluting the reflectance spectrum and the solar spectrum (air mass 1.5). The reduction in solar absorptance must not exceed 5 % in absolute terms after ageing. 3.2 Transmittance The solar transmittance, τ s, is measured before and after ageing of the cover glazing. The solar transmittance is determined by acquiring a transmittance spectrum of the cover glazing between 200 and 2500 nm using an UV-VIS-NIR-spectrometer equipped with an integrating sphere attachment. The solar transmittance is calculated by convoluting the transmittance spectrum and the solar spectrum (air mass 1.5). The reduction in solar transmittance must not exceed 10 % in absolute terms after ageing. 3.3 Mechanical characteristics Normally a reduction of 50 % of the measured characteristic is accepted after ageing relative the unaged material. 3 Material composition If the result is to be used as a basis for P-marking of solar collectors, an identification of the material is performed by a suitable method, for example IR-spectrofotometry or thermal analysis. The composition of type-tested materials may not be changed (a new composition requires a new test) Only raw materials with stated manufacturer may be used. Unspecified or re-processed materials may not be used without specific permission. 4 Report The test report should include the following: a. Test laboratory. b. Purpose of the test. c. Reference to this test method. d. Identification and description of the material tested. e. Test conditions. f. Test results. g. Any deviations from this method.

62 INVESTIGATION AND COMPARISON OF DIFFERENT MODEL EQUATIONS FOR UNGLAZED COLLECTORS Klaus Vajen Elimar Frank Christopher Jipp Universität Kassel Institut für Thermische Energietechnik Kassel (Germany) Abstract Different model equations have been proposed in literature for the description of the thermal behaviour of uncovered collectors. Most of the models are derived from the energy balance at the absorber surface. In the research study presented in this paper, the impact of different approaches to derive a single one-node model-equation on the calculated thermal performance has been investigated. Physically reasonable heat-transfer coefficients the impacts have been investigated of different approaches to derive a single one-node model-equation as well as model equations given in technical norms have been applied. The equation given in (EN 12975, 2002) was shown to be less suitable for the modelling of uncovered collectors, whereas equations suggested in (Perers 1987) and (Vajen et al. 2003) deliver more accurate results for varying boundary conditions. Further investigations are necessary concerning the sensitivity of different parameters on the prediction of the useful solar energy gains. Introduction To describe the thermal behaviour of uncovered solar collectors different numerical models can be used. They should be able to deliver the correct thermal power gained by the collector for given measured or generated boundary conditions like weather, fluid inlet temperature and mass flow rate if the parameters of the model have been determined theoretically or experimentally. Starting point of most of the models is the energy balance of the absorber surface. Usually, the physical formulation in eq. (1) is used for the derivation of a model equation with one thermal node: Tf q& use + q& b + ceff = hi ( Tp Tf ) t = q& q& q& q& with ( ) use p o i abs sky conv cond (1) q& = v& c ρ T T (2) T ( T T ) 2 f = o + i (3) q& = α G (4) abs q & =ε σ (T T ) (5) 4 4 sky eff p sky q& abs Here, is described with a simple relation, but eq. (4) could easily be extended by an incidence angle modifier. Following a common approach, the convective heat transfer is assumed to increase linearly with the wind speed:

63 ( ) ( Uc0 Uc1 vw ) ( Tp Ta ) q& = U T T conv c p a = + ( ) b b f a (6) q& = U T T (7) describes the heat transfer on the bottom of the collector. Since the average surface temperatures of the back of the absorber and of the underlying roof are difficult to measure, these temperatures are often approximated by T f and T a, respectively. Further model assumptions are: 1) The flow through all absorber channels connected in parallel is evenly distributed. 2) Heat transport in flow direction takes place only due to the fluid flow and not e.g. by conduction inside the absorber material. 3) The difference of the upper absorber surface temperature and the corresponding fluid temperature is assumed to be independent of the position in fluid flow direction. This is equivalent to a linear increase of the temperatures of fluid and absorber surface. Thus, the validity of the models is limited to sufficiently high specific flow rates. 4) All heat capacities are lumped together in a single node, represented by the average fluid temperature. 5) The heat transfer is modelled quasi-stationary and the heat transfer coefficients are assumed to be constant for the specific absorber ranges and the time intervals taken into account. 6) Furthermore, the model covers the standard operation temperatures and does not include e.g. heat gains due to condensation. Implicitly a specific arrangement of the temperatures is assumed: Td Ta Ti T o (8) With the model described in eqs. (1-7), the collector outlet temperature T o can not be predicted directly, even if all boundary conditions (G, dv/dt, T i, T sky ) and parameters (c p, ρ, α, ε, h i, U's, c eff ) are known, because the average absorber surface temperature T p is very difficult to measure and thus usually unknown. To determine T p, an iterative algorithm has to be applied: T p has to be changed until both sides of eq. (1) are fulfilled. On the other hand, several one-node models for uncovered collectors are described in literature (e.g. (Perers 1987, Vajen et al 1999, EN 12795, 2002, Vajen et. al 2003)), each consisting of a single equation which can be evaluated directly without any iterations. Besides the (ISO 9806, 1995) and (Rockendorf et al. 2001) these models are based on the energy balance of the absorber surface as described in eq. (1). To end up with a single model equation further assumptions are made, especially regarding a linearity of the radiant heat transfer between absorber surface and the surroundings. The model errors due to different assumptions are investigated in the following sections, taking into account varying boundary conditions. The iterative evaluation of eq. (1) uses the least assumptions and can reproduce the thermal behaviour of the collector most precisely. Thus, in the following the model is taken as a reference for the comparison of the accuracy of the different approaches. Derivation of the single model-equations Alternative to the iterative calculation the average surface temperature can be approximated by the average fluid temperature plus a constant temperature difference: T = T + C (9) p f In this way, the energy balance can be calculated directly. Under real conditions, the temperature difference between T p and T f varies. As already mentioned, it is difficult to measure the absorber plate temperature, whereas the fluid temperature is known in most cases. Thus, the terms including T p in the second part of eq. (1) are extended to temperature differences (T p -T f ) which can be eliminated later. E.g., the convective heat transfer term, of. eq. (6), can be extended as follows: q = ( U + U v ) ( T T ) + ( T T ) & conv c0 c1 w p f f a (10) The main differences between the simplified models are contained in the modelling of the radiant heat exchange between absorber surface and the sky. Here, the cubic part of the temperature difference is linearized and approximated by a model parameter. The temperature differences can be expressed in different ways. In (EN 12795, 2002), eq. (5) is substituted by the following equation: ( ) ( ) ( ) 2 2 ( T T ) ( T T ) ( T T ) ( T T ) 4 4 ( T T ) q & = ε σ T T + T T + T T sky eff p f f a a sky = ε σ eff p a p a p f f a + ε σ eff a sky (11) ε eff is an effective emission-coefficient that includes the properties of the absorber-material and the geometric structure of the absorber surface. In eq. (10), T p -T f can be separated and eliminated afterwards. However, T p 2 still remains in eq.(11). Therefore, for a defined operation point with constant values of T p and T a, a heat transfer parameter has to be defined which is assumed to be constant: 2 2 ( ) ( ) U : =ε σ T + T T + T (12) pa eff p a p a

64 Hence: ( ) ( ) 4 4 ( T T ) q& sky = Upa Tp Tf + Tf T a + (13) ε σ eff a sky (Vajen et al. 2003) extended eq. (11) only by T f instead of T f and T a which leads to 2 2 ( ) ( ) ( ) 4 4 ( T T ) q& sky =εeff σ Tp + Tf Tp + Tf Tp Tf + (14) ε σ eff f sky with a heat transfer parameter defined as 2 2 ( ) ( ) U : =ε σ T + T T + T (15) pf eff p f p f and finally 4 4 ( ) ( ) q& = U T T +ε σ T T (16) sky pf p f eff f sky For a direct linearization of eq. (10) in T p -T sky, as it was carried out by (Perers 1987) and (Vajen et al 1999), the extension delivers 2 2 ( ) ( ) ( ) 2 2 ( T T ) ( T T ) ( T T ) q& sky =εeff σ Tp + Tsky Tp + Tsky Tp Tf + (17) with ε σ + + eff p sky p sky f sky 2 2 ( ) ( ) U : =ε σ T + T T + T (18) ps eff p sky p sky and hence ( ) ( q& sky = Ups Tp Tf + Tf Tsky ) (19) Here, the term with the forth power is eliminated completely which leads to a lower sensitivity regarding measured data of the infrared irradiance. Furthermore, ε eff is integrated in U ps so that this linearization leads to a lower number of model parameters. After eliminating (T p -T f ) and rearranging eq.(1), described in detail in (Vajen et al. 2003), the model equation for the power delivery of the collector can be evaluataed. For the approach of (EN 12795, 2002) one receives for example q& use h hi Upa Uc0 Uc1 v w i = h hi Upa Uc0 Uc1 v w ( Upa Uc0 Uc1 vw ) ( Tf Ta ) i ( T T ) i εeff σ a sky hi+ Upa+ Uc0+ Uc1 v w b h α G + U Tf Ceff t ( T ) f T a (20) In this equation one faces an aggregation of parameters (h i, U x ) and boundary conditions (v w ). Therefore, another approximation has to be carried out: A factor F * is introduced which accumulates several parameters, depending on the different mathematical transformations described above. F * is somehow equivalent to the well known collector-efficiency factor F, but in contrary to F, F * does not include the heat losses at the rear side of the absorber. In the model equation a transformation is made to separate the terms depending on the wind speed. The factor F * can now be approximated as follows: h h U U U v i Fpa = h + U + U + U v h i pa c0 c1 w i pa c0 c1 w i Assuming - a wind speed of 2 m/s, - h i = 200 W/(m²K), - U c0 = 2 W/(m²K) and - U pa = 3 W/(m²K) (21) this approximation leads to a relative deviation of about 1%. However, the accuracy of the approximation is a function of the wind speed and it can not be assumed in general that the precondition U c1 v w << h i is valid. Fig. 1 illustrates the relative deviation between the two terms in eq. (21). Thus, a seasonal average deviation of 1 to 3% due to the approximation has to be expected. For different locations with differing wind speed distributions the deviation may be significantly smaller or higher. relative deviation in % Vw in m/s Fig. 1: Relative deviation of the terms of eq. (21) as a function of the wind speed. Parameters: h i = 200 W/(m²K), U pf =3 W/(m²K), U c0 = 2 W/(m²K), U c1 = 7 Ws/(m³K)

65 For the other models (Vajen et al and 2003) the error due to the approximation in eq. (21) is similar to the one above. The final equation for the useful energy gain of the model (EN 12795, 2002) results to: q& use = hi Uc0 Upa α hi G Uc1 α h vw G i hi Uc0 Upa + + hi hi 2Uc0 2Upa U v T hi 2 c1 i ( Uc0 Upa ) Ub ( Tf Ta ) ( T ) c1 w f a U + v T T h ( ) 2 w f a hi Uc0 Upa 4 4 εeff σ a sky hi U +ε σ C ( T T ) 4 4 ( ) c1 eff vw Ta Tsky hi eff Tf t (22) In (EN 12975, 2002) the parameters are bundled together to constants c 1 to c 6 which can not be interpreted physically any more. Furthermore, a term equivalent to v w ² (T f - T a ) is missing, whereas a term equivalent to (T f -T a )² appears without physical reason. Additional errors occur due to the different approximations of U x describing the radiant heat exchange in eq. (14) and (19) which affects the accuracy of the calculated useful system power. For T p =280 K and T sky, ref = 290 K, a comparison of eq. (19) and (5) by eq. (23) 4 4 ( Tp Tsky ) Ups ( Tp Tsky ) ε σ ( Tp Tsky) ε σ err = 100 rel,sky 4 4 (23) delivers a deviation of more than 0,5 % / K. Thus, already a variation of T sky by 5 K leads to an error in the model of about 2,5%. Regarding the other model equations (14) and (16) the deviation is similar. Sensitivity analysis of the behaviour of the models To get an impression of the behaviour of the different models, a sensitivity analysis has been carried out regarding varying boundary conditions. For this purpose, reference values of boundary conditions and parameters have been defined (Tab. 1). Further, T sky, T a, T i and G have been varied. The deviations have been calculated with respect to the physical model, eq. (1), which has been solved iteratively and serves as a reference model. The results are plotted in the following figures. Here, deviation always means the relative error of the useful power gain of the linearized model equations, compared with the value calculated by the reference model. Relative deviation in % Relative deviation in % Relative deviation in % Relative deviation in % 1,0 0,5 0,0-0,5-1,0-1,5-2,0 Tf+C Lin-TpTa -2, ,0 0,5 0,0-0,5-1,0-1,5-2,0 Lin-TpTf Lin-TpTsky Ti in C Lin-TpTa Lin-TpTsky Lin-TpTf -2, ,0 0,5 0,0-0,5-1,0-1,5-2,0 Lin-TpTsky Tf+C Ta in C Tf+C -2, ,0 0,5 0,0-0,5-1,0-1,5-2,0 Lin-TpTa Tsky in K Lin-TpTf Lin-TpTa -2, Fig. 2: Deviation of Lin-TpTf q& use Lin-TpTsky G in W/m² Tf+C for the different models, compared with the iteratively solved physical model, eq. (1) Lin-TpTa: eq. (12), Lin-TpTf: eq.(15), Lin-TpTsky: eq. (18), Tf+C: eq. (9)

66 Tab. 1: Boundary conditions for the model tests T i G T 293,15 K (20 C) 800 W/m² a 291,15 K (18 C) T sky 280,0 K (6,85 C) v w 2 m/s ε eff 0,8 α 0,95 h i 200 W/(m²K) C eff 4,6 Wh/(kgK) U c0 1,0 W/(m²K) U c1 3,0 Ws/(m³K) v& 4 m³/h As shown in fig. 2, the simplified model equations tend to underestimate the power prediction of the reference model. This is not necessarily the case if the parameters of the absorber have been fitted instead of being defined. Furthermore, obviously model eq. (9), T p =T f +C, is not suitable. A variation of T i or G does not seem to be critical for the three remaining models. Major deviations occur for the model containing the varied boundary condition in the linearized term. As expected, the maximal deviations occur if linearization is carried out for boundary conditions. Especially for the model Lin-TpTa, eq. (13) the highest deviations appear for different ambient temperatures. Therefore, based on these investigations, the models Lin-TpTsky, eq. (19), (Perers 1987), or Lin-TpTf, eq. (16), (Vajen et al. 2003), seem to be most suitable. Simply varying only one factor (e.g. T sky or G) doesn t allow an adequate final conclusion. It has to be investigated, how model parameters (e.g. U x ) can be fitted to measured data and how the system behaves under changing boundary conditions. Conclusions and Outlook The models of (Perers 1987) and (Vajen et al. 2003) delivered the most accurate predictions of the useful power gain for uncovered solar collectors, whereas the equation based on (EN 12975, 2002) showed major deviations, especially for varying ambient temperatures. The pure equation in the European norm could not be included in the comparison, because it contains terms which can not be derived from the energy balance, eq. (1). The ENequation might be proper for covered collectors with T cove r T a, but it seems to be less suitable for uncovered collectors. However, a general statement regarding the quality of different models cannot be delivered yet. The investigations and comparisons carried out were based on physically reasonable heat transfer parameters in the model equations. A comparison with parameters fitted to measured values might lead to different results. The linearized parameters U x implicitly contain material properties and boundary conditions, i.e. the weather conditions during the collector tests. Thus, further investigations are necessary concerning the sensitivities of different parameters on the model s result and the possibility to split U x in material properties and boundary conditions. Compared to the other models taken into account, the approach of (Perers 1987) contains one parameter less. However, this parameter reduction could be achieved as well with a different formulation of (Vajen et al. 2003), if U pf is split into ε σ and (T p +T f )(T p ²+T f ²), because T p and T f are nearly constant in the usual range of operating conditions of uncovered collectors. Nevertheless, it can be recommended to carry out an iterative determination of T p on the basis of eq. (1). Nomenclature c p Wh/(kgK) specific heat capacity h i W/(m²K) inner heat transfer coefficient q& b W/m² backside heat transfer q& conv W/m² convective heat losses q& sky W/m² radiant heat losses q& use W/m² useful power gain T a K ambient temperature T f K average fluid temperature T K fluid inlet-temperature i T K fluid outlet- o temperature T p K average surfacetemperature of the absorber T sky K sky temperature U b W/(m²K) back heat transfer coefficient U c, U c0 W/(m²K) convective heat transfer U c1 Ws/(m³K) coefficients U pa, U ps W/(m²K) radiant heat transfer coefficients

67 v w m/s average wind speed α Absorption coefficient ε eff effective emission coefficient ref reference Acknowledgements The authors like to thank especially the following institutions for financial and logistical support: Volkswagen- Stiftung which supports this research, The Center of the Problems of Unconventional and Renewable Resources of Energy in Bishkek (Kyrgyzstan), Kyrgyz Technical University, and the company Bishkek Teploenergo. References (1) EN :2002 (2002): Thermal solar systems and components, Part 2: Test methods (2) ISO (1995): Test methods for solar collectors Part 3: Thermal performance of unglazed liquid heating collectors (sensible heat transfer only) including pressure drop (3) Perers, B., Performance Testing of unglazed collectors, Report for IEA Task III, Studsvik Energy, Sweden, 1987 (4) Rockendorf, G., Sillmann, R., Bethe, T., Köln, H., Solare Freibadbeheizung, Absorberprüfung und Testergebnisse Anlagen Planung und Betrieb, Institut für Solarenergieforschung, Hameln / Emmerthal, 2001 (5) Vajen, K., Krämer, M., Orths, R., Boronbaev, E.K., Solar Absorber System for Preheating Feeding Water for District Heating Nets, Proc. ISES Solar World Congress, , Jerusalem, Vol. III, pp , 1999 (6) Vajen, K., Frank, E., Krämer, M., Extended Model Equations for Uncovered Collectors, ISES Solar World Congress, Gothenburg, 2003

68 W4.1 Resource document Recommendations on testing of solar air collectors Dissemination level: Public Author: Josef Buchinger, arsenal research Reviewer: xxxxxxxx January 2007 CONTENTS INTRODUCTION Introduction, why testing of solar air collectors is important and what has been done so far. RECOMMENDATIONS Recommendations for further activities to achieve a standard for testing. Technical suggestions for testing of solar air collectors. SUMMARY Solar air systems are a promising technology in the active use of solar energy for heating. Differently to solar liquid systems, air systems so far have not entered the market with significant rates. As main obstacle for a wide dissemination of solar air systems appears lacking information as well as lack of confidence on how these systems will perform. Planners and architects as well as end consumers need trustworthy parameters and facts to start applying and investing in this technology. For this reason testing of the respective components is essential. Up to now, no widely accepted standard exists for air collectors. Within NEGST WP4, a draft-standard for testing solar air collectors based on existing standards has been developed and is presented. Recommendations for further work to be done on sections of this standard are given. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6 File: WP4-D2.1.f- recommendations for testing of solar air collectors 2.0.doc

69 page 2 of 4 pages 1 Introduction Solar air collectors are not wide spread so far. As main obstacle for a wide dissemination appears lacking information as well as lack of confidence on how these systems will perform. Testing of the respective components is therefore essential. Such tests should be reproducible and acknowledged, and therefore commonly standardised in e.g. EN-Standards. Up to now no common standard exists for solar air collectors. Compared to liquid collectors, the measuring procedures for solar air collectors need more expenditure. Some standardised testing procedure exists so far: The Italian Standard UNI 8937 (Norma Italiana 1987) The US-American Standard ASHRAE (ASHRAE) The UNI 8937 only gives an idea of how testing of air collectors can be carried out, but does not touch the specific problems of solar air systems in its 12 pages. The ASHRAE is already a usable standard but since it describes procedures for testing of both liquid and air collectors it does take into account all possible variations of solar air collectors and leaves some uncertainties. As mentioned earlier the market for solar air collectors is weak; hence the current situation is that the air collector industry is not interested in supporting the definition of a standard (ANNEX 2). Therefore the NEGST project offered an excellent opportunity to start drafting a standard and compile recommendations on further work required. 2 Recommended draft standard for testing of solar air collectors For achieving comparable and reproducible test results, the conditions for testing must be defined carefully. Based on the EN for testing of solar liquid collectors, the ASHRAE , UNI 8927 and experience in solar air collector testing of arsenal research (Fechner 1999; Selke 2005), the definition of a draft for a European Standard for testing air collectors has been brought on the way. The actual status can be found in ANNEX 1. This draft describes procedures for testing reliability and performance of glazed solar thermal air collectors operated either in open or closed loop systems with either positive or negative pressure applied. The focus so far has been laid on the adoption of the sections describing the thermal performance tests, especially steady-state outdoor testing. Issues of reference temperatures for calculation of the thermal performance have been defined and are open for discussion. This draft so far does not include procedures for unglazed solar thermal systems and is open for a revision with regards to air collectors made of polymeric or organic materials. On the following topics the adoption of current standards (e.g. EN or ASHRAE ) is further required and foreseen to be included in the next version of this draft: Stagnation temperature Time constant Pressure drop Incidence angel modifier (IAM) Since the thermal performance of solar air collectors is dependent on the mass flow through the collector, testing procedures of the thermal performance as a function of the mass flow is already described in the current draft. Hence, it is planned to integrate the testing procedure for identification of the pressure drop into the procedures for testing thermal performance. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

70 page 3 of 4 pages 2.1 Open questions With regards to the resource document (ANNEX 2) from June 2006 on this issue and to the draft standard for testing of solar air collectors (ANNEX 1) the following questions are to be answered on the way to an acceptable standard for testing of solar air collectors Internal pressure tests for collectors (See Section 5.2 in ANNEX 1) Considering the fact that i.e. 6 bar air-pressure behind a glass wall might be very dangerous, tests for high pressure resistance will need special precautions. Therefore, first of all, the following question should be answered: How relevant are pressure tests for air collectors? Leading to the next questions: Are there any air tight or pressure resistant air collectors on the market? Is there a chance that relevant over pressure can develop under real conditions within an air collector? In case no positive answer on the questions can be found and due to the fact that leakage is a known problem for air collectors (a test procedure on leakage can be found in section 6.) plus several collectors are build with an open absorber (i.e. transpired air collectors) so that pressure tests do not make sense anyway, the internal pressure test for air collectors may be dropped Fluid inlet temperature range Air collectors can be designed either for direct heating to the space or in combination with heat exchangers or heat pumps. This could further mean that the inlet temperature range of such air collectors operated in an open loop could start at -20 C. To test such an air collector under real conditions it would require that the complete testing environment needs to be cooled to an ambient temperature of e.g. -20 C, especially if the collector operates at negative gauge pressure and is supposed to suck in ambient air. Therefore open questions are: How relevant are these testing conditions for air collectors? Is it applicable to chill down the testing environment to -20 C? 2.2 Required new testing procedures Wind dependency For testing collectors it is essential to find out the wind dependency of a collector. Some collectors, especially uncovered, but also collectors with the air flow directly under the cover are strongly dependent on wind. Recommendations for installation of the wind simulation can be found in the existing standard for testing water collectors. Only for uncovered collectors the direction of the wind can be influential as well. It is therefore recommended to conduct further research on the impacts of the surrounding air speed and redefine the section Surrounding air speed of the current draft or even define new test procedures to determine the wind dependency in order to have a standard covering the needs of all stakeholders. 3 Recommendations for next steps It is recommended to initiate an additional activity of producers, planners, architects, research and testing institutes to investigate thoroughly the different aspects of testing. In cooperation with the German air collector producer Grammer Solar a workshop on solar air collectors will be organised in Austria this autumn If relevant stakeholders will get together, next important steps towards working out the standard for air collectors can be discussed. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

71 page 4 of 4 pages 4 References ASHRAE (2003). Methods of Testing to Determine the Thermal Performance of Solar Collectors. ANSI/ASHRAE Standard Buchinger J. (2006). NEGST - W4.1 - Resource document, Recommendations on testing of solar air collectors. Vienna, arsenal research. Fechner H. (1999). IEA TASK 19 Solar air systems - Investigation on Series Produced Solar Air Collectors - Final Report. Vienna, arsenal research - Department of Renewable Energy. Norma Italiana (1987). Collettori solari piani ad aria - Determinazione del rendimento termico. UNI Selke T. (2005). Studie zur Darstellung des technischen und wirtschaftlichen Marktpotentials des Solar-Luftkollektors der BAYER AG. Wien, arsenal research: 73. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

72 SolarPACES 13th Symposium on Concentrating Solar power and Chemical Energy Technologies, June , Seville, Spain EFFICIENCY TESTING OF PARABOLIC TROUGH COLLECTORS USING THE QUASI-DYNAMIC TEST PROCEDURE ACCORDING TO THE EUROPEAN STANDARD EN Stephan Fischer 1, Eckhard Lüpfert 2, Hans Müller-Steinhagen 1,2 1 Universität Stuttgart, Institut für Thermodynamik und Wärmetechnik (ITW) Pfaffenwaldring 6, D Stuttgart 2 DLR Deutsches Zentrum für Luft- und Raumfahrt, Institut für Technische Thermodynamik Solarforschung, Köln, Germany Abstract The collector efficiency of a parabolic trough collector prototype has been tested according to the European Standard EN The Standard includes, apart from the well known steady state parameters, an incident angle modifier for diffuse irradiation and an effective collector thermal capacity. The addition of these two collector parameters allows the evaluation of continuous measurements over several hours even under irradiance fluctuations and changing sun position. Keywords: Parabolic Trough Collector, Testing, Standards, Thermal Efficiency Introduction The thermal performance of a solar collector is of major interest to all parties, e.g. designer, investor, operator and last but not least collector manufacturer, involved in the setup of a solar thermal system. In order to be able to compare the thermal performance of different collectors a standardized test method must be available. Standardized test methods have been published in international normative documents for decades 1,2. These Standards are well accepted for the test of flat plate collectors and evacuated tubular collectors. However if it comes to collectors with a significant concentration ratio the stipulated use of the hemispherical solar irradiance as reference irradiance does not meet the requirement for the performance characterization anymore. To overcome this difficulty the concentrating community uses the direct irradiance as reference irradiance together with the test procedures 1,2 to characterize the thermal performance of concentrating collectors. This non normative approach leads to a variety of collector models as well as differing nomenclatures and methodologies. The first attempt to standardize these different approaches was done by all mayor institutions involved in the performance testing of tracking concentrating collectors 3. With the implementation of the European Standard EN an alternative test method under so called quasi-dynamic conditions has been introduced. This test method, in contrast to previous ones takes into account direct irradiance as well as diffuse irradiance and thus permits the performance measurement of tracking concentrating collectors. page 1 / 5

73 EUROPEAN STANDARD FIRST DRAFT NORME EUROPEENNE EUROPÄISCHE NORM UDC Descriptors: English version Thermal solar systems and components Solar air collectors Test methods

74 draft standard for testing of solar air collectors based on EN and ASHRAE Contents Page 1 SCOPE REFERENCES TERMS AND DEFINITIONS SYMBOLS AND UNITS RELIABILITY TESTING OF LIQUID HEATING COLLECTORS General Internal pressure tests for collectors Objective Apparatus and procedure Test conditions High-temperature resistance test Objective Apparatus and procedure Test conditions Results Exposure test Objective Apparatus and procedure Test conditions Results External thermal shock test Objective Apparatus and procedure Test conditions Results Internal thermal shock test Objective Apparatus and procedure Test conditions Results Rain penetration test Objective Apparatus and procedure

75 5.7.3 Test conditions Results Mechanical load test Positive pressure test of the collector cover Negative pressure test of fixings between the cover and the collector box Negative pressure test of collector mountings Impact resistance test Objective Apparatus and procedure Test conditions Results Final inspection Test report THERMAL PERFORMANCE TESTING OF AIR HEATING COLLECTORS Collector mounting and location General Collector mounting frame Tilt angle Collector orientation outdoors Shading from direct solar irradiance Diffuse and reflected solar irradiance Thermal irradiance Surrounding air speed Instrumentation Solar radiation measurement Thermal radiation measurement Temperature measurements Measurement of collector liquid flow rate Measurement of surrounding air speed Pressure measurements Humidity Measurement Elapsed time Instrumentation/data recorders Collector area Collector fluid capacity Test installation

76 draft standard for testing of solar air collectors based on EN and ASHRAE General consideration Heat transfer fluid Test ducts Pump and flow control devices Air-Preconditioning Apparatus Humidity Ratio Outdoor steady-state efficiency test Test installation Preconditioning of the collector Test conditions Test procedure Measurements Test period (steady-state) Presentation of results Computation and presentation of collector efficiency and thermal performance Steady-state efficiency test using a solar irradiance simulator General The solar irradiance simulator for steady-state efficiency testing Test installation Preconditioning of the collector Test procedure Measurements during tests in solar irradiance simulators Test period Test conditions Computation and presentation of results Determination of the effective thermal capacity and the time constant of a collector Incidence angle modifier Determination of the pressure drop across a collector Determination of the leakage rate Test Apparatus Presentation of results

77 Foreword This document has been prepared by Josef Buchinger as a resource document for WP4.1 of the NEGST project as recommendations on testing of solar air collectors. It is based on the EN (2002), the ASHRAE and experiences made during various testing of air collectors, mainly within the IEA SHCP Task 19 Solar Air Systems. This draft describes procedures for testing reliability and performance of glazed solar thermal air collectors operated either in open and closed loop systems with either positive or negative pressure applied. Whereby the focus on the work so far has been laid on the adoption of the sections describing the thermal performance tests, especially steady-state outdoor testing. This draft so far does not include procedures for unglazed solar thermal systems and is open for a revision with regards to air collectors made of polymeric or organic materials. Further the complete topic of testing incident angle modifiers is not covered by this document. Annexes as in the basis documents are not included in this draft and have to be adapted or developed newly where required. Introduction This standard specifies test methods for determining the ability of a solar air collector to resist the influence of degrading agents. It defines procedures for testing collectors under well-defined and repeatable conditions. This standard also provides test methods and calculation procedures for determining the steadystate thermal performance of glazed air heating solar collectors. It contains methods for conducting tests outdoors under natural solar irradiance and natural and simulated wind and for conducting tests indoors under simulated solar irradiance and wind. This standard will also provides methods for determining the thermal performance of unglazed air heating solar collectors. For unglazed absorbers, readily fabricated modules with a specific size are seldom used. Therefore, during the test, it is to check, that a realistic flow pattern and flow velocity is used. 5

78 draft standard for testing of solar air collectors based on EN and ASHRAE Scope This European Standard specifies test methods for validating the durability, reliability and safety requirements for air heating collectors as specified in EN xxxx-1. It is not applicable to those collectors in which the thermal storage unit is an integral part of the collector to such an extent that the collection process cannot be separated from the storage process for the purpose of making measurements of these two processes. This standard applies to nonconcentrating solar air collectors in which a fluid enters the collector through a single inlet and leaves the collector through a single outlet. Collectors containing more than one inlet and more than one outlet may be tested according to this standard provided that the external piping or ducting can be connected so as to provide effectively a single inlet and a single outlet. Collectors that are custom built (built in; e.g. roof integrated collectors that do not compose of factory made modules and are assembled directly on the place of installation) cannot be tested in their actual form for durability, reliability and thermal performance according to this standard. Instead, a module with the same structure as the ready collector may be tested. The module gross area shall be at least 2m 2. The test is valid only for larger collectors, than the tested module. 2 References This European Standard incorporates, by dated or undated reference, provisions from other publications. The normative references are cited at the appropriate places in the text and the publications are listed hereafter. For dated references, subsequent amendments to or revisions of any of these publications apply to this European Standard only when incorporated in it by amendment or revision. For undated references the latest edition of the publication referred to applies. ISO 9060 Solar energy - Specification and classification of instruments for measuring hemispherical solar and direct solar radiation. ISO Test methods for solar collectors Part 1: Thermal performance of glazed liquid heating collectors including pressure drop. ISO Test methods for solar collectors Part 2: Qualification test procedures. ISO : 1995 Test methods for solar collectors - Part 3: Thermal performance of unglazed liquid heating collectors (sensible heat transfer only) including pressure drop. ISO 9846 Solar energy - Calibration of a pyranometer using a pyrheliometer. ISO 9847 Solar energy - Calibration of field pyranometers by comparison to a reference pyranometer. ISO/TR 9901 Solar energy - Field pyranometers - Recommended practice for use. EN ISO 9488 Solar Energy - Vocabulary (ISO 9488:1999) EN :2000 Thermal solar systems and components - Solar collectors - Part 1: General requirements 6

79 EN 12911:2000 Windows and doors : Resistance to wind load Test method 3 Terms and definitions For the purpose of this standard, the Terms and definitions given in EN ISO 9488 apply. 4 Symbols and units a 1 heat loss coefficient at (T m - T a )=0 Wm -2 K -1 a 2 temperature dependence of the heat loss coefficient Wm -2 K -2 A A absorber area of collector m 2 A a aperture area of collector m 2 A G gross area of collector m 2 AM optical air mass b u collector efficiency coefficient (wind dependence) m -1 s b o constant for the calculation of the incident angle modifier b 1 heat loss coefficient at (T m - T a )=0 Wm -2 K -1 b 2 collector efficiency coefficient Wsm -3 K -1 c 1 heat loss coefficient at (T m - T a )=0 Wm -2 K -1 c 2 temperature dependence of the heat loss coefficient Wm -2 K -2 c 3 wind speed dependence of the heat loss coefficient Jm -3 K -1 c 4 sky temperature dependence of the heat loss coefficient Wm -2 K -1 c 5 effective thermal capacity J m -2 K -1 c 6 wind dependence in the zero loss efficiency sm -1 c f specific heat capacity of heat transfer fluid Jkg -1 K -1 C effective thermal capacity of collector JK -1 D date YYMMDD E L longwave irradiance (λ >3µm) Wm -2 E β longwave irradiance on an inclined surface outdoors Wm -2 E s longwave irradiance Wm -2 F radiation view factor F collector efficiency factor G hemispherical solar irradiance Wm -2 G* hemispherical solar irradiance Wm -2 G'' net irradiance Wm -2 G b direct solar irradiance (beam irradiance) Wm -2 7

80 draft standard for testing of solar air collectors based on EN and ASHRAE G d diffuse solar irradiance Wm -2 LT local time h K θ incidence angle modifier K θ b incidence angle modifier for direct radiation K θ d incidence angle modifier for diffuse radiation m thermally active mass of the collector kg. m mass flow rate of heat transfer fluid kgs -1 Q. useful power extracted from collector W Q. L power loss of collector W t time s t a ambient or surrounding air temperature C t dp atmospheric dew point temperature C t e collector outlet (exit) temperature C t in collector inlet temperature C t m mean temperature of heat transfer fluid C t s atmospheric or sky temperature C t stg stagnation temperature C T absolute temperature K T a ambient or surrounding air temperature C T * m reduced temperature difference ( = (t m t a )/G*) m 2 KW -1 T s atmospheric or equivalent sky radiation temperature K U measured overall heat loss coefficient of collector, with reference to T * m Wm -2 K -1 U L overall heat loss coefficient of a collector with uniform absorber temperature t m Wm -2 K -1 u surrounding air speed ms -1 V f fluid capacity of the collector m 3 p pressure difference between fluid inlet and outlet Pa t time interval s T temperature difference between fluid outlet and inlet(t e - t in ) K α 8 solar absorptance β tilt angle of a plane with respect to horizontal degrees γ azimuth angle degrees ε hemispherical emittance ω solar hour angle degrees

81 θ angle of incidence degrees Φ latitude degrees λ wavelength µm η collector efficiency, with reference to T * m η o zero-loss collector efficiency (η at T * m = 0), reference to T * m σ Stefan-Boltzmann constant Wm -2 K -4 ρ density of heat transfer fluid kgm -3 τ c collector time constant s τ transmittance (τα) e effective transmittance-absorptance product (τα) ed effective transmittance-absorptance product for diffuse solar irradiance (τα) en effective transmittance-absorptance product for direct solar radiation at normal incidence (τα) eθ effective transmittance-absorptance product for direct solar radiation at angle of incidence θ NOTE 1 In the field of solar energy the symbol G is used to denote solar irradiance, rather than the generic symbol E for irradiance. NOTE 2 C is often denoted (mc) e in basic literature (see also Annex H) NOTE 3 For more information about thermal performance coefficients (parameters) c1 to c6, see Annex H. 9

82 draft standard for testing of solar air collectors based on EN and ASHRAE General 5 Reliability testing of liquid heating collectors The detail of numbers of collectors and sequences used to carried out qualifications tests detailed in the list below (table1) shall be given in the report. For some qualification tests, a part of the collector may have to be tampered with in some way, for example a hole may have to be drilled in the back of the collector to attach a temperature sensor to the absorber. In these cases care should be taken to ensure that any damage caused does not affect the results of subsequent qualification tests, for example by allowing water to enter into a previously raintight collector. Table 1 - Test List Subclause Test 5.2 Internal pressure 5.3 High-temperature resistance 5.4 Exposure 2) 5.5 External thermal shock 3) 5.6 Internal thermal shock 3) 5.7 Rain penetration 4) 5.8 Freeze resistance 5) 5.9 Mechanical load 5.10 Impact resistance Thermal performance 1) For organic absorbers, the high-temperature resistance test shall be performed first in order to determine the collector stagnation temperature needed for the internal pressure test. 2) The high temperature and exposure test shall be carried out on the same collector 3) The external and internal thermal shock tests may be combined with the exposure test or the hightemperature resistance test. 4) The rain penetration test shall be carried out only for glazed collectors. 5) The freeze resistance test shall be carried out only for collectors claimed to be freeze resistant. 6) The Thermal performance test shall be carried out on a collector that had not been used for other tests. 1), 2) 5.2 Internal pressure tests for collectors Objective The absorber shall be pressure-tested (see ) to assess the extent to which it can withstand the pressures, which it might meet in service while operating at elevated temperature. The tests shall be carried out at elevated temperatures, because the pressure resistance of an air collector 10

83 with organic parts such as sealings may be adversely affected as its temperature is increased. One of the methods described in through may be chosen Apparatus and procedure General The apparatus consists of a pneumatic pressure source, and a means of heating the absorber to the required test temperature. The characteristics of a solar irradiance simulator shall be the same as those of the simulator used for efficiency testing of liquid heating solar collectors. A temperature sensor shall be attached to the absorber to monitor its temperature during the test. The sensor shall be positioned at two-thirds of the absorber height and half the absorber width. It shall be fixed firmly in a position to ensure good thermal contact with the absorber. The sensor shall be shielded from solar radiation. The pressure in the absorber shall be raised in stages as specified in , and the absorber shall be inspected for swelling, distortion or rupture after each increase in pressure. The pressure shall be maintained while the absorber is being inspected. For safety reasons, the collector shall be encased in a transparent box to protect personnel in the event of explosive failure during this test High temperature pneumatic pressure test The absorber may be pressure-tested using compressed air, when heated by either of the following methods: a) heating the whole collector in a solar irradiance simulator (see figure A.5); b) heating the whole collector outdoors under natural solar irradiance (see figure A.5). The compressed air supply to the absorber shall be fitted with a safety valve and a pressure gauge having a standard uncertainty better than 5% Test conditions Temperature For air collectors the test temperature shall be the maximum temperature which the absorber will reach under stagnation conditions. The reference conditions given in table 2 shall be used. The calculations employed to determine the test temperature are included in Annex C and shall either: - use measured collector performance characteristics, or - extrapolate from average values, measured in the high-temperature resistance test (see 5.3.3), of the global solar irradiance (natural or simulated) on the collector plane, the surrounding air temperature and the absorber temperature. Table 2 - Climate reference conditions to determine test temperatures for internal pressure test of organic absorbers Climate parameter Value for all climate classes 11

84 draft standard for testing of solar air collectors based on EN and ASHRAE Global solar irradiance on collector plane, G in W/m Surrounding air temperature, ta in C Pressure The test pressure shall be 1.5 times the maximum collector operating pressure specified by the manufacturer. For absorbers made of organic materials, the pressure shall be raised to the test pressure in equal stages of 20 kpa (approximately) and maintained at each intermediate pressure for 5 min. The test pressure shall then be maintained for a least 1 h Results The collector shall be inspected for leakage, swelling and distortion. The results of the inspection shall be reported. Full details of the test procedure used, including the temperature, intermediate pressures and test periods used, shall be reported with the test results. 5.3 High-temperature resistance test Objective This test is intended to assess rapidly whether a collector can withstand high irradiance levels without failures such as glass breakage, collapse of plastic cover, melting of plastic absorber, or significant deposits on the collector cover from outgassing of collector material Apparatus and procedure The collector shall be tested outdoors, or in a solar irradiance simulator. A schema for testing is shown in figure A.6. The characteristics of the solar irradiance simulator to be used for the high-temperature resistance test shall be those of the solar irradiance simulator used for efficiency testing of liquid heating solar collectors. The collector shall be mounted outdoors or in a solar simulator, and shall not be filled with fluid. One of its fluid pipes shall be sealed to prevent cooling by natural circulation of air, but the other shall be left open to permit free expansion of air in the absorber. A temperature sensor shall be attached to the absorber to monitor its temperature during the test. The sensor shall be positioned at two-thirds of the absorber height and half the absorber width. It shall be fixed firmly in a position to ensure good thermal contact with the absorber. The sensor shall be shielded from solar radiation. The test shall be performed for a minimum of 1 h after steady-state conditions have been established, and the collector shall be subsequently inspected for signs of damage as specified in Test conditions The set of reference conditions given in table 3 or conditions resulting in the same collector temperature according to eqn. C.3, shall be used for all climate classes. Table 3 - Climate reference conditions for high-temperature resistance test Climate parameter Value for all climate classes 12

85 Global solar irradiance on collector plane, G in W/m 2 >1000 Surrounding air temperature, ta in C Surrounding air speed in m/s < Results The collector shall be inspected for degradation, shrinkage, outgassing and distortion. The results of the inspection shall be recorded together with the 5 min average values of solar irradiance (natural or simulated) on the collector plane, surrounding air temperature and speed, and absorber temperature (and the pressure of the special fluid in the absorber, if that method is used) recorded during the test. 5.4 Exposure test Objective The exposure test provides a low-cost reliability test sequence, indicating (or simulating) operating conditions which are likely to occur during real service and which also allows the collector to "settle", such that subsequent qualification tests are more likely to give repeatable results Apparatus and procedure The collector shall be mounted outdoors (see figure A.7), but not filled with fluid. One of the fluid openings shall be sealed to prevent cooling by natural circulation of air, while the other shall be left open to permit free expansion of air in the absorber. The air temperature shall be recorded to an uncertainty of 1 K and the global irradiance on the plane of the collector recorded using a pyranometer of class I or better in accordance with ISO Irradiation and mean air temperature values shall be recorded every 30 min and rainfall shall be recorded daily. The collector shall be exposed until the test conditions have been met. At the end of the exposure, a visual inspection shall be made for signs of damage as specified in Test conditions The set of reference conditions given in table 4 shall be used. The collector shall be exposed until at least 30 days (which need not be consecutive) have passed with the minimum irradiation H shown in table 4. The irradiation is determined by recording irradiance measurements using a pyranometer. The collector shall also be exposed for at least 30 h to the minimum irradiance level G given in table 4, as recorded by a pyranometer, when the surrounding air temperature is greater than the value shown in table 4 or conditions resulting in the same collector temperature according to eqn. C.3. These hours shall be made up of periods of at least 30 min. NOTE In regions where these conditions cannot be met during certain periods of the year, the 30-h exposure to high irradiance levels (table 4) can be conducted in a solar irradiance simulator having characteristics identical to those of a simulator used for efficiency testing of liquid heating solar collectors. The 30-h exposure test should be conducted after the collector has completed at least 10 days, but no more than 15 days, of the exposure to the minimum irradiation level (table 4). If the external and internal thermal shock tests are combined with the exposure test, the first 13

86 draft standard for testing of solar air collectors based on EN and ASHRAE external and internal shocks shall be caused during the first 10 of the 30 h defined above, and the second during the last 10 of the 30 h. Table 4 - Climate reference conditions for exposure test as well as for external and internal thermal shock tests Climate parameter Value for all climate classes Global solar irradiance on collector plane, G in W/m Global daily irradiation on collector plane, H in MJ/m 2 14 Surrounding air temperature, ta in C 10 NOTE Values given are minimum values for testing Results The collector shall be inspected for damage or degradation. The results of the inspection shall be reported together with a record of the climatic conditions during the test, including daily irradiation, surrounding air temperature and rain. 5.5 External thermal shock test Objective Collectors may from time to time be exposed to sudden rainstorms on hot sunny days, causing a severe external thermal shock. This test is intended to assess the capability of a collector to withstand such thermal shocks without a failure Apparatus and procedure The collector shall be mounted either outdoors or in a solar irradiance simulator. One of its fluid pipes shall be sealed to prevent cooling by natural circulation of air, while the other shall be left open to permit free expansion of air in the absorber (see figure A.8). A temperature sensor may be optionally attached to the absorber to monitor its temperature during the test. The sensor shall be positioned at two-thirds of the absorber height and half the absorber width. It shall be fixed firmly in a position to ensure good thermal contact with the absorber. The sensor shall be shielded from solar radiation. An array of water jets shall be arranged to provide a uniform spray of water over the collector. The collector shall be maintained under a high level of solar irradiance for a period of 1 h before the water spray is turned on. It is then cooled by the water spray for 15 min before being inspected. The collector shall be subjected to two external thermal shocks Test conditions The set of reference conditions given in table 4 shall be used. The specified operating conditions shall be: - solar (or simulated solar) irradiance G greater than the value shown in table 4. - surrounding air temperature ta greater than the value shown in table 4. The water spray shall have a temperature of less than 25 C and a flow rate in the range 0,03 kg/s to 0.05 kg/s per square metre of collector aperture. 14

87 If the temperature of the water which first cools the collector is likely to be greater than 25 C (for example if the water has been sitting in a pipe in the sun for some time), then the water shall be diverted until it has reached a temperature of less than 25 C before being directed over the collector Results The collector shall be inspected for any cracking, distortion, condensation or water penetration. The results of the inspection shall be reported. The measured values of solar irradiance, surrounding air temperature, absorber temperature (if measured), water temperature and water flow rate shall also be reported. 5.6 Internal thermal shock test Objective Collectors may from time to time be exposed to a sudden intake of cold heat transfer fluid on hot sunny days, causing a severe internal thermal shock, for example, after a period of shutdown, when the installation is brought back into operation while the collector is at its stagnation temperature. This test is intended to assess the capability of a collector to withstand such thermal shocks without failure Apparatus and procedure The collector shall be mounted either outdoors or in a solar irradiance simulator (see figure A.9). One of its fluid pipes shall be connected via a shutoff valve to the heat transfer fluid source and the other shall be left open initially to permit the free expansion of air in the absorber and also to permit the heat transfer fluid to leave the absorber (and be collected). A temperature sensor may be optionally attached to the absorber to monitor its temperature during the test. The sensor shall be positioned at two-thirds of the absorber height and half the absorber width. It shall be fixed firmly in a position to ensure good thermal contact with the absorber. The sensor shall be shielded from solar radiation. The collector shall be maintained under a high level of solar irradiance for a period of 1 h before it is cooled by supplying it with heat transfer fluid for at least 5 min until the absorber temperature drops below 50 C. The collector shall be subjected to two internal thermal shocks Test conditions The set of reference conditions given in table 4 shall be used. The specified operating conditions shall be: - solar (or simulated solar) irradiance G greater than the value shown in table 4; - surrounding air temperature ta greater than the value shown in table 4. or conditions resulting in the same collector temperature according to eqn. C.3 The heat transfer fluid shall have a temperature of less than 25 C. The recommended fluid flow rate is at least 0.02 kg/s per square metre of collector aperture (unless otherwise specified by the manufacturer). 15

88 draft standard for testing of solar air collectors based on EN and ASHRAE Results The collector shall be inspected for any cracking, distortion or deformation. The results of the inspection shall be reported. The measured values of solar irradiance, surrounding air temperature, absorber temperature (if measured), heat transfer fluid temperature and heat transfer fluid flow rate shall also be reported. 5.7 Rain penetration test Objective This test is applicable only for glazed collectors and is intended to assess the extent to which glazed collectors are substantially resistant to rain penetration. They shall normally not permit the entry of either free-falling rain or driving rain. Collectors may have ventilation holes and drain holes, but these shall not permit the entry of drifting rain Apparatus and procedure General The collector shall have its fluid inlet and outlet pipes sealed (unless hot water is circulated through the absorber, see ), as shown in figure A.10, and be placed in a test rig at the shallowest angle to the horizontal recommended by the manufacturer. If this angle is not specified, then the collector shall be placed at a tilt of 30 to the horizontal. Collectors designed to be integrated into a roof structure shall be mounted in a simulated roof and have their underside protected. Other collectors shall be mounted in a conventional manner on an open frame or a simulated roof. The collector shall be sprayed on exposed sides, using spray nozzles or showers Detection of ingress of water The collector shall be mounted and sprayed as explained above while the absorber in the collector is kept warm (minimum 50 C). This can be done either by circulating hot air at about 50 C through the absorber or by exposing the collector to solar radiation. The penetration of water into the collector shall be determined by inspection (looking for water droplets, condensation on the cover glass or other visible signs) and by one of the following methods: a) by weighing the collector (standard uncertainty better than 5 gr/m 2 collector area); or b) by means of humidity measurement (standard uncertainty better than 5%) or c) by means of measuring the condensation level. The heating up of the collector should be started before the spraying of the water in order to ensure that the collector box is dry before testing. In cases of collectors having wood in the backs (or other special cases), the laboratory must take all necessary measures during the conduction of the test so that the final result will not be influenced or altered by the special construction of the collector Test conditions The collector shall be sprayed with water at a temperature lower than 30 C with a flow rate of more than 0.05 kg/s per square metre of sprayed area. The duration of the test shall be 4 h Weighing method If the weighing method is chosen, the collector shall be put on the scale before the start of the test on three consecutive occasions. The weights recorded shall not vary by more than ±5 gr/m 2 collector area. 16

89 Humidity measurement method When measuring the penetration of water into the collector by means of humidity measurement, an absolute humidity sensor is placed in the air gap between absorber and glazing. Collector and sensor are connected to a hot fluid loop for at least five hours before the rain is switched on in order to stabilise. When testing outdoors, in order to minimize disturbances of the measurement, the collector shall be shaded during the whole test. The humidity shall be monitored from five hours before the raining till at least five hours after the raining. Ingress of water might also be detected at a later stage, during the test Final Inspection (Clause 5.11) Condensation level method If the condensation level method is chosen, the penetration of water is determined by measuring the condensation level on the cover glass and by measuring the water that come out of the collector when tipping it. The heating of the collector shall be started at least 30 minutes before the spaying of water and shall continue until it can be ensured that the collector box is dry before testing. This shall be done by circulating hot water (or other fluid) above 50 o C through the absorber before but also during the complete test. The water will thereafter condense on the inside of the glazing, which is being cooled by cold water on the outside. After 2 hours an intermediate inspection of condensation on the cover glass shall be done in order to facilitate the reporting of the places where water penetrates. After finishing the spraying the inspection of condensation should be done after a short time for ventilating, in order to distinguish collectors with good ventilation qualifications that are without accumulation of humidity inside the collector. However, the inspection should be done within one minute after finishing the spraying before the collector will make any temperature changes. To ensure that no water has penetrated the collector box without forming condensation on the glazing, the collector shall be tipped on all four sides in turn after the test is terminated. The collector shall not be exposed by solar radiation Results The collector shall be inspected for water penetration. The results of the inspection, i.e. the extension of water penetration and the places where water penetrated shall be reported. 5.8 Mechanical load test Positive pressure test of the collector cover Objective This test is intended to assess the extent to which the transparent cover of the collector is able to resist the positive pressure load due to the effect of wind and snow Apparatus and procedure The collector shall be placed horizontally on an even ground. On the collector a foil shall be laid and on the collector frame a wooden or metallic frame shall be placed, high enough to contain the required amount of gravel or similar material (see figure A.12). The gravel, preferably type 2-32 mm, shall be weighed in portions and distributed in the frame so that everywhere the same load is created (pay attention to the bending of the glass), until the wanted height is reached. 17

90 draft standard for testing of solar air collectors based on EN and ASHRAE The test can also be carried out installing the collector in accordance with and loading the cover using suction cups, gravel or other suitable means (e.g. water). As a further alternative, the necessary load may be created by applying an air pressure on the collector cover. In this case, apparatus in accordance with EN can be used Test conditions The test pressure shall be increased in steps of 100 Pa to the recommended maximum test pressure, which shall be at least 1000 Pa or optional load test above 1000 Pa up to the value as specified by manufacturer or national requirements Results The pressure at which any failure of the collector cover occurs shall be reported together with details of the failure. If no failure occurs, then the maximum pressure which the collector sustained shall be reported Negative pressure test of fixings between the cover and the collector box Objective This test is intended to assess the extent to which the fixings between the collector cover and collector box are able to resist uplift forces caused by the wind Apparatus and procedure The collector shall be installed horizontally on a stiff frame by means of its mounting fixtures. The frame which secures the cover to the collector box shall not be restricted in any way. A lifting force which is equivalent to the specified negative pressure load, shall be applied evenly over the cover. The load shall be increased in steps up to the final test pressure. If the cover has not been loosened at the final pressure, then the pressure may be stepped up until failure occurs. The time between each pressure step shall be the time needed for the pressure to stabilise. Either of three alternative methods may be used to apply pressure to the cover: - Method (a): The load may be applied to the collector cover by means of a uniformly distributed set of suction cups (see figure A.13). - Method (b): For collectors which have an almost airtight collector box, the following procedure may be used to create a negative pressure on the cover (see figure A.14). Two holes are made through the collector box into the airgap between the collector cover and absorber, and an air source and pressure gauge are connected to the collector airgap through these holes. A negative pressure on the cover is created by pressurising the collector box. For safety reasons the collector shall be encased in a transparent box to protect personnel in the event of failure during this test. - Method (c): The load may be created by applying a negative air pressure on the collector cover. In this case, apparatus in accordance with EN can be used. During the test, the collector shall be visually inspected and any deformations of the cover and its fixings reported. The collector shall be examined at the end of the test to see if there are any permanent deformations. Test method (a) is not designed to check the strength of the collector mounting fixtures. If a glass cover fails before the fixings which hold the cover to the collector box, then the pressure at which the failure occurred shall be noted. If the cover fails again then it shall be concluded that the collector is inadequately designed and the test need not be continued. 18

91 Test conditions The test pressure shall be increased in steps of 100 Pa to the recommended maximum test pressure, which shall be at least 1000 Pa or optional load test above 1000 Pa up to the value as specified by the manufacturer or national requirements Results Any deformations observed during the inspection shall be reported together with the pressure at which any failure of the cover or cover fixings was observed. Details of the failures shall also be reported. If no failure occurs, then the maximum pressure which the collector sustained shall be reported Negative pressure test of collector mountings Objective Solar collectors are generally installed on a roof or on the ground by means of mounting brackets and supporting frames. This test is intended to assess the extent to which the mounting brackets, supporting frames and fixing points can withstand the uplift forces caused by the wind Apparatus and procedure The collector shall be installed horizontally on a stiff frame by means of the mounting fixtures supplied by the manufacturer. A lifting force shall be applied to the collector either by applying a negative pressure uniformly to the top of the collector frame and cover using the method (a) described in , or by applying a positive pressure uniformly over the back of the collector by means of air pressure in large air bags (see figure A.14). The pressure shall be increased in steps up to the final test pressure. If the mounting fixtures have not failed at the final test pressure, then the pressure may be stepped up until failure occurs. The time between each pressure step shall be the time needed for the pressure to stabilise Test conditions The test pressure shall be increased in steps of 100 Pa to the recommended maximum test pressure, which shall be at least 1000 Pa or optional load test above 1000 Pa up to the value as specified by manufacturer or national requirements Results The pressure at which any failure of the mounting fixtures or fixing points occurs shall be reported together with details of the failure. If no failure occurs, then the maximum pressure which the collector sustained shall be reported. 5.9 Impact resistance test Objective This test is intended to assess the extent to which a collector can withstand the effects of heavy impacts caused by hailstones Apparatus and procedure General The testing of the solar collector to determine its impact resistance can be done by one of two methods, i.e. by using steel balls or ice balls. 19

92 draft standard for testing of solar air collectors based on EN and ASHRAE Method 1 The collector shall be mounted either vertically or horizontally on a support (see figure A.15). The support may be stiff enough so that there is negligible distortion or deflection at the time of impact. Steel balls shall be used to simulate a heavy impact. If the collector is mounted horizontally then the steel balls are dropped vertically, or if it is mounted vertically then the impacts are directed horizontally by means of a pendulum. In both cases, the height of the fall is the vertical distance between the point of release and the horizontal plane containing the point of impact. The point of impact shall be no more than 5 cm from the edge of the collector cover, and no more than 10 cm from the corner of the collector cover, but it shall be moved by several millimetres each time the steel ball is dropped. A steel ball shall be dropped onto the collector 10 times from the first test height, then 10 times from the second test height, etc. until the maximum test height is reached (as specified by the manufacturer). The test is to be stopped when the collector sustains some damage or when the collector has survived the impact of 10 steel balls at the maximum test height. NOTE This method does not correspond to the physical effect of hailstones as the deformation energy absorbed by the ice particles is not being considered Method 2 The apparatus consists of the following equipment: a) Moulds of suitable material for casting spherical ice balls of the required diameter (25 mm). b) A freezer, controlled at -10 C ± 5 C. c) A storage container for storing the ice balls at a temperature of -4 C ± 2 C. d) A launcher capable of propelling an ice ball at the specified velocity (as specified by the manufacturer), within ± 5 %, so as to hit the collector within the specified impact location. The path of the ice ball from the launcher to the collector may be horizontal, vertical or at any intermediate angle. e) A rigid frame for supporting the collector, with the impact surface normal to the path of the projected ice ball; the support shall be stiff enough so that there is negligible distortion or deflection at the time of impact. f) A balance for determining the mass of an ice ball to a standard uncertainty of ± 2 %. g) An instrument for measuring the velocity of the ice ball to a standard uncertainty of ± 2 ms -1. The velocity sensor shall be no more than 1 m from the surface of the collector. As an example, figure A.16 shows in schematic form a suitable apparatus comprising a horizontal pneumatic launcher, a vertical collector support and a velocity meter which measures electronically the time it takes the ice ball to traverse the distance between two light beams. The testing procedure shall be the following: a) Using the moulds and the freezer, make sufficient ice balls of the required size for the test, including some for the preliminary adjustment of the launcher. b) Examine each one for cracks, size and mass. An acceptable ball shall meet the following criteria: - no cracks visible to the unaided eye; - diameter within ± 5% of the ball (25 mm); 20

93 - mass within ± 5% of the ball (25 mm). c) Place the balls in the storage container and leave them there for at least 1 h before use. d) Ensure that all surfaces of the launcher likely to be in contact with the ice balls are near room temperature. e) Fire a number of trial shots at a simulated target in accordance with step g) below and adjust the launcher until the velocity of the ice ball, as measured with the velocity sensor in the prescribed position, is within ± 5% of the required hailstone test velocity. f) Install the collector at room temperature in the prescribed mount, with the impact surface normal to the path of the ice ball. g) Take an ice ball from the storage container and place it in the launcher. Take aim at the impact location and fire. The time between the removal of the ice ball from the container and impact on the collector shall not exceed 60 s. The point of impact shall be no more than 5 cm from the edge of the collector cover, and no more than 10 cm from the corner of the collector cover, but it shall be moved by several millimetres each time the ice ball is launched. An ice ball shall be launched onto the collector 10 times; the test shall be stopped when the collector sustains some damage or when the collector has survived the impact of 10 ice balls Test conditions If the test is conducted according to method 1, the steel ball shall have a mass of 150 g ± 10 g and the following series of test heights shall be used: 0.4 m, 0.6 m, 0.8 m, 1,0 m, 1.2 m, 1.4 m, 1.6 m, 1.8 m and 2.0 m. If the test is conducted according to method 2, the ice ball shall have a diameter of 25 mm ± 5%, a mass of 7,53 g ± 5 % and its velocity shall be 23 m/s ± 5 % Results The collector shall be inspected for damage. The results of the inspection shall be reported, together with the height from which the steel ball was dropped (if method 1 is used) and the number of impacts which caused the damage. NOTE As test method 2 is closer to reality, this method ( ) is preferable Final inspection When the full test sequence has been completed, the collector shall be dismantled and inspected. All abnormalities shall be reported and accompanied by a photograph Test report The format sheets given in annex B shall be completed for each test, together with the introductory format sheet (B.1) reporting a summary of main results, including the test methods. 21

94 draft standard for testing of solar air collectors based on EN and ASHRAE Thermal performance testing of air heating collectors 6.1 Collector mounting and location General The way in which a collector is mounted will influence the results of thermal performance tests. Collectors to be tested shall therefore be mounted in accordance with to Full-size collector modules shall be tested, because the edge losses of small collectors may significantly reduce their overall performance Collector mounting frame The collector shall be mounted in the manner specified by the manufacturer. The collector mounting frame shall in no way obstruct the aperture of the collector, and shall not significantly affect the back or side insulation, unless otherwise specified (for example, when the collector is part of an integrated roof array). Collectors designed to be mounted directly on standard roofing material may be mounted over a simulated roof section. In case of roof integrated collectors, a model consisting of a small scale collector placed on an artificial roof should be prepared for the purpose of the tests. If mounting instructions are not specified, the collector shall be mounted on an insulated backing with a quotient of the materials thermal conductivity to its thickness of 1 Wm -2 K -1 ± 0.3 Wm -2 K -1 and the upper surface painted matt white and ventilated at the back. NOTE Example material suited for the insulated backing is 30 mm of polystyrene foam. The collector shall be mounted such that the lower edge is not less than 0.5 m above the local ground surface. Currents of warm air, such as those which rise up the walls of a building, shall not be allowed to pass over the collector. Where collectors are tested on the roof of a building, they shall be located at least 2 m away from the roof edge. The performance of some forms of unglazed solar collectors is a function of module size. If the collector is supplied in fixed units of area greater than 1 m2 then a sufficient number of modules shall be linked together to give a test system aperture of at least 3 m2. If the collector is supplied in the form of strips the minimum built-up module area shall be 3 m2 (gross area) Tilt angle Collectors may be tested at tilt angles, as recommended by manufacturers or specified for actual installations. Otherwise the collector shall be tested at tilt angles such that the incidence angle with direct solar radiation θ is less than 30 or at angles of tilt such that the incidence angle modifier varies by less than ±2 % from normal incidence. Before deciding on a tilt angle it may be necessary to check the incidence angle modifier at two angles prior to commencing the tests. NOTE For most unglazed collectors, the influence of tilt angle and radiation incidence angle on collector efficiency is small and unglazed collectors are commonly installed at low inclinations. However care should be taken to avoid air locks at low inclinations Collector orientation outdoors The collector may be mounted outdoors in a fixed position facing the equator, but this will result in the time available for testing being restricted by the acceptance range of incidence angles. A 22

95 more versatile approach is to move the collector to follow the sun in azimuth, using manual or automatic tracking Shading from direct solar irradiance The location of the test stand shall be such that no shadow is cast on the collector during the test Diffuse and reflected solar irradiance For the purposes of analysis of outdoor test results, solar irradiance not coming directly from the sun's disc is assumed to come isotropically from the hemispherical field of view of the collector. In order to minimize the errors resulting from this approximation, the collector shall be located where there will be no significant solar radiation reflected onto it from surrounding buildings or surfaces during the tests, and where there will be no significant obstructions in the field of view. Not more than 5 % of the collector's field of view shall be obstructed, and it is particularly important to avoid buildings or large obstructions subtending an angle of greater than approximately 15 to the horizontal in front of the collectors. The reflectance of most rough surfaces such as grass, weathered concrete or chippings is usually low enough so no problem is caused during collector testing. Surfaces to be avoided in the collector's field of view include large expanses of glass, metal or water. In most solar simulators the simulated beam approximates direct solar irradiance only. In order to simplify the measurement of simulated irradiance, it is necessary to minimize reflected irradiance. This can be achieved by painting all surfaces in the test chamber with a dark (low reflectance) paint Thermal irradiance The performance of some collectors is particularly sensitive to the levels of thermal irradiance. The temperature of surfaces adjacent to the collector shall be as close as possible to that of the ambient air in order to minimize the influence of thermal radiation. For example, the outdoor field of view of the collector shall not include chimneys, cooling towers or hot exhausts. For indoor and simulator testing, the collector shall be shielded from hot surfaces such as radiators, airconditioning ducts and machinery, and from cold surfaces such as windows and external walls. Shielding is important both in front of and behind the collector. The major difference between indoor and outdoor testing of unglazed collectors is the long wave thermal irradiance. The relative long wave radiation in a simulator shall not be higher than ±50 Wm -2 (typically -100 Wm -2 for outdoor conditions) Surrounding air speed The performance of many collectors is sensitive to the surrounding air speeds. In order to maximize the reproducibility of results, collectors shall be mounted such that air can freely pass over the aperture, back and sides of the collector. The mean surrounding air speed, parallel to the collector aperture, shall be between the limits specified in section Where necessary, artificial wind generators shall be used to achieve these air speeds. Collectors designed for integration into a roof may have their backs protected from the wind; if so, this shall be reported with the test results. The performance of unglazed collectors is sensitive to air speed adjacent to the collector. In order to maximize the reproducibility of results, unglazed collectors shall be mounted such that air can freely pass over the front side of the collector, and exposed back and sides of the collector. 23

96 draft standard for testing of solar air collectors based on EN and ASHRAE The average surrounding air speed at a distance of 100 mm above and parallel to the collector aperture shall cover the range 0 ms -1 to 3,5ms -1 subject to the tolerance specified in table B1. If these conditions cannot be achieved under natural conditions then an artificial wind generator shall be used. If a wind generator is used the turbulence level shall be in the range of 20 % to 40 % to simulate natural wind conditions. The turbulence level shall be checked at the leading edge of the collector 100 mm above the collector surface. The turbulence level shall be monitored using a linearised hot wire anemometer with a frequency response of at least 100 Hz. If the absorber is not mounted directly on a roof or a sheet of backing material, the air speed shall be controlled and monitored on the front and back of the absorber. 6.2 Instrumentation Solar radiation measurement The solar radiation measurement shall be handled in accordance with EN Chapter Solar radiation measurement Thermal radiation measurement Measurement of long wave irradiance A pyrgeometer mounted in the plane of the collector shall be used to measure global long wave radiation Precaution for effects of temperature gradient The pyrgeometer used during the tests shall be placed in the same plane as the collector absorber and allowed to equilibrate for at least 30 minutes before measuring Precautions for effects of humidity and moisture The pyrgeometer shall be provided with a means of preventing accumulation of moisture that may condense on surfaces within the instrument and effect its reading. An instrument with a desiccator that can be inspected is required. The condition of the desiccator shall be observed prior to and following each daily measurement sequence Precautions for effect of short wave heating The influence of short wave solar heating effects should be minimised Calibration interval The pyrgeometer shall be calibrated within 12 month preceding the tests, in accordance with ISO A change of more than 5% over a year period shall warrant the use of more frequent calibration or replacement of the instrument. If the instrument is damaged in any significant manner, it shall be recalibrated or replaced. All calibrations shall be performed with respect to the World Radiometric Reference (WRR) Scale Temperature measurements General Three temperature measurements are required for solar collector testing. These are the fluid temperature at the collector inlet, the fluid temperature at the collector outlet, and the ambient air temperature. The required accuracy and the environment for these measurements differ, and hence the sensor for temperature measurement and associated equipment may be different Measurement of heat transfer fluid inlet temperature (t in ) Required accuracy 24

97 The temperature of the heat transfer fluid at the collector inlet shall be measured to an uncertainty of 0.2 K, but in order to check that the temperature is not drifting with time, a very much better resolution of the temperature signal to ±0.04 K is required. NOTE This resolution is needed for all temperatures used for collector testing (i.e. over the range 0 C to 100 C) which is a particularly demanding accuracy for recording by data logger, as it requires a resolution of one part in or a 12-bit digital system Mounting of sensors The determination of the mean temperature in air flows is critical, several layers of different air temperatures are often close adjacent, therefore a specific mixing device optimized according to fluid dynamic experiences - at the outgoing duct just before the sensors and a sophisticated arrangement of temperature-sensors are necessary. If thermocouples are used to measure the temperatures, thermocouple grids shall be fabricated with thermocouples located as shown in Fig. 1Fig. 2. There shall be a minimum of eight thermocouples in a grid in the air inlet test duct and in the outlet test duct. Thermocouples in the grid shall be located at the center of equal cross-sectional or concentric areas, as illustrated in Fig. 1Fig. 2 and Fig. 3 Fig. 1: Schematic of the thermophile arrangement used to measure the temperature difference across the solar collector. Fig. 2: Schematic of equal areas thermocouple grid. Minimum of eight junctions located at the center of equal cross- 25

98 draft standard for testing of solar air collectors based on EN and ASHRAE sectional areas are connected in parallel to obtain an average reading. All thermocouple sets must have leads of identical length. 26 Fig. 3: Distribution of thermocouples in round duct for equal cross-sectional areas grid. When using other temperature sensors than thermocouples a similar arrangement of the sensors within the ducts is required. The sensor for temperature measurement shall be mounted at no more than 200 mm from the collector inlet, and insulation shall be placed around the ducts both upstream and downstream of the sensor. If it is necessary to position the sensor more than 200 mm away from the collector, then a test shall be made to verify that the measurement of fluid temperature is not affected Determination of heat transfer fluid temperature difference ( T) The difference between the collector outlet and inlet temperatures ( T) shall be determined to a standard uncertainty of <0.05 K. Standard uncertainties approaching 0.02 K can be achieved with modern well-matched and calibrated transducers, and hence it is possible to measure heat transfer fluid temperature differences of 1 K or 2 K with a reasonable accuracy. Delta-T sensors shall be calibrated in the relevant flow range and temperature range, using the same fluid Measurement of surrounding air temperature (t a ) Required accuracy The ambient or surrounding air temperature shall be measured to a standard uncertainty of 0.5 K Mounting of sensors For outdoor measurements the sensor shall be shaded from direct and reflected solar radiation by means of a white-painted, well-ventilated shelter, preferably with forced ventilation. The shelter itself shall be shaded and placed at the midheight of the collector but at least 1 m above the local ground surface to ensure that it is removed from the influence of ground heating. The shelter shall be positioned to one side of the collector and not more than 10 m from it. If air is forced over the collector by a wind generator, the air temperature shall be measured in the outlet of the wind generator and checks made to ensure that this temperature does not deviate from the ambient air temperature by more than ±1 K Measurement of collector liquid flow rate The standard uncertainty of the liquid flow rate measurement shall be within ±1.5 % of the measured value, in mass per unit time. The flowmeter shall be calibrated over the range of fluid flowrates and temperatures to be used during collector testing.

99 6.2.5 Measurement of surrounding air speed The measurement of surrounding air speed shall be handled in accordance with EN Chapter Measurement of air speed. For unglazed collectors the measurement of surrounding air speed shall be handled in accordance with Chapter of EN Pressure measurements Pressure-measuring stations shall have four externally manifolded pressure taps, as shown in Fig. 4. The pressures in the test circuit and the pressure drop across the solar collector shall be measured using static pressure tap holes and either a manometer or a differential-pressure transducer. The edges of the holes on the inside surface of the duct shall be free of burrs. The hole diameter shall not exceed 40% of the wall thickness or 1.6 mm. Provision shall be made for determining the absolute pressure of the entering transfer fluid. The static pressure drop across an air collector and static pressure upstream or downstream of the collector shall be determined with instruments that have an accuracy of ±2.5 Pa. Fig. 4: Schematic representation of the measurement of pressure drop across the solar collector. Measuring stations shall be provided upstream and downstream of the collector, as illustrated in Fig. 5. For collectors tested under negative gauge pressure, the collector inlet gauge pressure shall be below the atmospheric pressure by at least 124 Pa minimum or the maximum allowable operating pressure specified by the manufacturer, whichever is smaller. For collectors tested under positive pressure, the gauge pressure at the collector discharge shall be 124 Pa minimum or the maximum allowable operating pressure specified by the manufacturer, whichever is smaller Humidity Measurement When air is used as the heat transfer fluid, its moisture content is needed for the correct determination of the density and specific heat of the air. The humidity ratio W n shall be measured to an accuracy of ±0.005 (kg water/kg dry air). 1 Humidity measurement should be made in accordance with ASHRAE Standard Elapsed time Elapsed time shall be measured to a standard uncertainty of 0.2 %. 1 ASHRAE

100 draft standard for testing of solar air collectors based on EN and ASHRAE Instrumentation/data recorders In no case shall the smallest scale division of the instrument or instrument system exceed twice the specified standard uncertainty. For example, if the specified standard uncertainty is 0.1 K, the smallest scale division shall not exceed 0.2 C. Digital techniques and electronic integrators shall have an standard uncertainty equal to or better than 1.0 % of the measured value. Analog and digital recorders shall have an error equal to or better than 0.5 % of the full-scale reading and have a time constant of 1s or less. The peak signal indication shall be between 50 % and 100 % of full scale. The input impedance of recorders shall be greater than 1000 times the impedance of the sensors or 10 MΩ whichever is higher Collector area The collector area (absorber, gross or aperture) shall be measured to a standard uncertainty of 0.3 %. Area measurements shall take place at a collector temperature of (20 ± 10) C and under operating pressure if the absorber is made of organic material Collector fluid capacity The fluid capacity of the collector shall be determined by calculation, which is based on the geometrical circumstances. 6.3 Test installation General consideration An example of test configurations for testing solar air collectors are shown in Fig. 5. It is schematic only, and are not drawn to scale. Fig. 5: Example of an open test loop Heat transfer fluid The heat transfer fluid used for collector testing is air. The specific heat capacity and density of the fluid shall be known to within ±1% over the range of fluid temperatures used during the tests. These values are given for air in annex L Test ducts The air ducts between the solar collector and the pressure-measuring station, upstream and downstream of the collector, shall be of the same cross-sectional dimension. The cross-sectional area of these ducts in the pressure-measuring section shall be equal in size to the inlet discharge opening of the collector, whichever is smaller. The air flow pattern inside the collector is very important for a correct assessment of the performance. The air flow pattern inside the collector (especially the partition close to the inlet) mainly depends on the connection between ducting system and collector. In standardised tests only a single collector module is tested, which might not comply with the mode of installation in praxis. 28

101 To reach an even air flow pattern throughout the collector special distribution ducts at the inlet and outlet should be used for each collector tested. By means of boxes with perforated metal sheets at inlet and outlet a well distributed air flow can be achieved, which means that an even air flow from the centre-line of the collector to the edges from entrance to outlet exists. The air ducts used in the collector loop shall be suitable for operation at temperatures up to 95 C. Duct lengths shall generally be kept short. In particular, the length of piping between the outlet of the fluid temperature regulator and the inlet to the collector shall be minimized, to reduce the effects of the environment on the inlet temperature of the fluid. This section of pipe shall be insulated to ensure a rate of heat loss of less than 0.2 WK -1, and shall be protected by a reflective weatherproof coating. Ducts between the temperature sensing points and the collector (inlet and outlet) shall be protected with insulation and reflective (for outdoor measurements also weatherproof) covers to beyond the positions of the temperature sensors, such that the calculated temperature gain or loss along either pipe portion does not exceed ±0.01 K under test conditions. NOTE 1 Filters should be placed upstream of the flow measuring device and the pump, in accordance with normal practice (a nominal filter size of 200 µm is usually adequate) Pump and flow control devices The fluid pump shall be located in the collector test loop in such a position that the heat from it which is dissipated in the fluid does not affect either the control of the collector inlet temperature or the measurements of the fluid temperature rise through the collector. With some types of pump, a simple bypass loop and manually controlled needle valve may provide adequate flow control. Where necessary, an appropriate flow control device may be added to stabilize the mass flow rate. The pump and flow controller shall be capable of maintaining the mass or volume flow rate through the collector stable to within ±1.5 % 2 despite temperature variations, at any inlet temperature chosen within the operating range Air-Preconditioning Apparatus The preconditioning apparatus shall control the dry-bulb temperature of the transfer medium entering the solar collector to within ±1.0 C of the desired test values at all times during the test period. Since the rate of energy collection in the collector is deduced by measuring instantaneous values of the fluid inlet and outlet temperatures, it follows that small variations in inlet temperature could lead to errors in the rates of energy collection deduced. It is particularly important to avoid any drift in the collector inlet temperature. Its heating and cooling capacity shall be selected so that the dry-bulb temperature of the air entering the preconditioned apparatus may be raised or lowered to the required amount to meet the applicable test conditions in Section and Humidity Ratio When air is the transfer fluid and the test panel is operated at a negative pressure, the humidity ratio of the test fluid shall be equal to the humidity ratio of the air surrounding the test panel. 2 ASHRAE

102 draft standard for testing of solar air collectors based on EN and ASHRAE Outdoor steady-state efficiency test Test installation The collector shall be mounted in accordance with the specifications given in 6.1.1, and coupled to a test loop as described in The heat transfer fluid shall flow from the bottom to the top of the collector, or as recommended by the manufacturer. The air flow pattern inside the collector is very important for a correct assessment of the performance. The air flow pattern inside the collector (especially the partition close to the inlet) mainly depends on the connection between ducting system and collector. In standardised tests only a single collector module is tested, which might not comply with the mode of installation in praxis. To reach an even air flow pattern throughout the collector special distribution ducts at the inlet and outlet should be used for each collector tested. By means of boxes with perforated metal sheets at inlet and outlet a well distributed air flow can be achieved, which means that an even air flow from the centre-line of the collector to the edges from entrance to outlet exists Preconditioning of the collector The collector shall be visually inspected and any damage recorded. The collector aperture cover shall be thoroughly cleaned. If moisture is formed on the collector components, then the heat transfer fluid shall be circulated at approximately 80 C for as long as is necessary to dry out the insulation and collector enclosure but at least for 30 minutes. If this form of preconditioning is carried out, then it shall be reported with the test results. The fluid shall be inspected for entrained particles, by means of a transparent tube built into the fluid loop pipework. Any contaminants shall be removed. If the customer wants only a performance test and not the qualification tests, the empty collector shall be exposed to irradiation for 5 hours at the level of more than 700 Wm Test conditions At the time of the test, the total solar irradiance at the plane of the collector aperture shall be greater than 700 Wm -2. NOTE 1 If the manufacturer has limitations on operation with respect to maximum irradiance but not less than 800 Wm -2, this can be requested with the test. That maximum value should be clearly reported. The angle of incidence of direct solar radiation at the collector aperture shall be in the range in which the incident angle modifier for the collector varies by no more than ±2 % from its value at normal incidence. For single glazed flat plate collectors, this condition will usually be satisfied if the angle of incidence of direct solar radiation at the collector aperture is less than 20. However, much lower angles may be required for particular designs. In order to characterize collector performance at other angles, an incident angle modifier may be determined (see 6.1.7). Where diffuse solar irradiance is less than 30 %, its influence may be neglected. The collector shall not be tested at diffuse irradiance levels of greater than 30 %. The average value of air speed parallel to the collector aperture, taking into account spatial variations over the collector and temporal variations during the test period, shall be 3 ms -1 ± 1 ms -1 Measurements of fluid temperature difference of less than 1 K shall not be included in the test results because of the associated problems of instrument error. 30

103 6.4.4 Test procedure General The collector shall be tested over its operating temperature range and over a range of mass flow rates as specified below. Tests hall be done under clear sky conditions in order to determine its efficiency characteristic. If test conditions permit, an equal number of data points shall be taken before and after solar noon for each fluid inlet temperature. The latter is not required if the collectors are moved to follow the sun in azimuth and altitude using automatic tracking. During a test, measurements shall be made as specified in These may then be used to identify test periods from which satisfactory data points can be derived Fluid inlet temperature range If the fluid inlet temperature range is specified by the manufacturer, data points, shall satisfy the requirements given below and be obtained for at least three fluid inlet temperatures spaced evenly over the operating temperature range of the collector. If possible, one inlet temperature shall be selected such that the mean fluid temperature in the collector lies within ±3 K of the ambient air temperature, in order to obtain an accurate determination of η o. If no specification by the manufacturers are provided it is recommended to have at least three evenly spaced fluid inlet temperature within the following ranges, with regards to the system the collector is predominantly designed for: closed loop systems: 0 to 60 C open loop systems: -20 to +40 C Fluid flow rate range Unless the range of fluid flow rate is specified by the manufacturer, the fluid flow rate shall be set to the values equally distributed between 20 to 120 kg/h/m² absorber area. It shall be held stable to within ±2 % of the set value during each test period, and shall not vary by more than ±3 % of the set value from one test period to another. In some collectors the recommended fluid flow rate may be close to the transition region between laminar and turbulent flow. This may cause instability of the internal heat transfer coefficient and hence variations in measurements of collector efficiency. In order to characterize such a collector in a reproducible way, it may be necessary to use a higher flow rate, but this shall be clearly stated with the test results. NOTE: In the transition regime, the flow rate should first be set high (turbulent) and then reduced to the setpoint value. This will prevent transition from laminar to turbulent during the measurements Measurements The following data shall be measured: the gross collector area A G, the absorber area A A and the aperture area A a ; the global solar irradiance at the collector aperture the global long wave radiation at the collector aperture 31

104 draft standard for testing of solar air collectors based on EN and ASHRAE the diffuse solar irradiance at the collector aperture (only outdoors) the angle of incidence of direct solar radiation (alternatively, this angle may be determined by calculation) the surrounding air speed parallel to the collector aperture the surrounding air temperature the dew point temperature of the surrounding air the temperature of the heat transfer fluid at the collector inlet the temperature of the heat transfer fluid at the collector outlet the dew point temperature of the heat transfer fluid at the collector inlet the dew point temperature of the heat transfer fluid at the collector outlet the mass flow rate of the heat transfer fluid at the collector inlet the mass flow rate of the heat transfer fluid at the collector outlet Test period (steady-state) The test period for a steady state data point shall include a pre-conditioning period of at least four times the time constant of the collector (if known), or not less than 10 minutes (if time constant is not known), with the correct fluid temperature at the inlet, followed by a steady state measurement period of at least 4 times the time constant of the collector (if known), or not less than 5 minutes (if time constant is not known). A collector is considered to have been operating in steady-state conditions over a given measurement period if none of the experimental parameters deviate from their mean values over the measurement period by more than the limits given in table 5. To establish that a steady state exists, average values of each parameter taken over successive periods of 30 s shall be compared with the mean value over the measurement period. Table 5 - Permitted deviation of measured parameters during a measurement period Parameter Permitted deviation from the mean value (Global)Test solar irradiance ± 50 Wm -2 Surrounding air temperature ± 1 K Fluid mass flow rate ± 2 % Fluid temperature at the collector inlet ± 0.1 K Presentation of results The measurements shall be collated to produce a set of data points which meet the required test conditions (see ), including those for steady-state operation. These shall be presented using the data format sheets given in Annex D Computation and presentation of collector efficiency and thermal performance General The instantaneous efficiency of a solar collector, operating under steady-state conditions, is defined as the ratio of the actual useful extracted power to the solar energy intercepted by the collector. The actual useful power extracted, Q., is calculated from: 32

105 Q & = mc & T f (3) A value of c f corresponding to the mean fluid temperature shall be used. If m& is obtained from volumetric flow rate measurement, then the density shall be determined for the temperature of the fluid in the flow meter Solar energy intercepted by the collector Provided that the angle of incidence is less than 20, the use of an incident angle modifier, as described in 6.7, is not required for single glazed flat plate collectors. The solar energy intercepted is A G where the area is A A when referred to the absorber area of the collector and A a when referred to the aperture area of the collector, and the collector efficiency is. Q mc & f ( te tin ) η = = (4) AG AG Presentation of the thermal performance subject to reference temperature The efficiency η depends on the operation conditions of the collector. It decreases with increasing temperatures, because of the increasing heat losses. Important is, how to define the operation conditions of the collector given by the temperature difference between the overall collector temperature t c and ambient t a. An efficiency curve can be drawn in dependency of a certain reference temperature which corresponds to the collector temperature t c. In a physically correct way one has to take a weighted mean temperature t c of the whole collector box, but in the measuring practice this is not practicable. The instantaneous efficiency shall be presented graphically as a function of either the inlet temperature t in, the outlet temperature t e and a so called mean collector temperature t m which can be calculated as the arithmetic mean value between inlet and outlet temperature. Hence the physical mean temperature t c. of the collector is often much closer to the outlet temperature t e than to the arithmetical mean temperature t m it is recommended that to use the outlet temperature t e for the presentation of the collector efficiency. The value of G to be used for the presentation of second-order fits shall be 800 Wm -2. The test conditions shall be recorded on the data format sheets given in annex D Presentation of thermal performance subject to the mean collector temperature The reduced temperature difference T m *. When the mean temperature of the heat transfer fluid t m is used, where T t m = tin + 2 the reduced temperature difference is calculated as: T * m t = m t G The efficiency is then calculated as: a (5) (6) 33

106 draft standard for testing of solar air collectors based on EN and ASHRAE U t t mc t t L ( m a ) & f ( e in ) η m = FR τ α = G (7) AG Presentation of thermal performance subject to the outlet temperature The outlet temperature t e U t t mc t t L ( e a ) & f ( e in ) η e = Fo τ α = G (8) AG F o is the collector heat removal factor in relation to t e - the collector outlet temperature and η e is the efficiency when the outlet temperature is taken as reference. F 0 accounts for the fact that the absorber temperature is not the same as the outlet air collector temperature neither in the horizontal direction nor vertical Presentation of thermal performance subject to the inlet temperature When using the inlet temperature t in the efficiency is calculated as: 34 U t t mc t t L ( in a ) & f ( e in ) η in = FR τ α = G (9) AG Presentation of thermal performance subject to leakage While equations (7-9) characterize the thermal performance of air-heating collectors, significant leakage of the heat transfer medium is likely and different flow rates in the inlet and exit streams must be taken into account in measuring the useful energy collected. Defining the leakage mass flow rate positive for in-leakage, i.e., m & L = m& m& (10) e in the useful energy collected may be determined for both test modes as follows Testing with negative gauge pressure In this configuration with leakage ( m& L > 0), air at ambient temperature is drawn into the heated airstream in the collector. The actual useful energy gain in this situation is Q & = m& h m& h + m& h ) (11) e e ( in in L a where the bracketed term is the total incoming enthalpy flow. Neglecting effects of moisture transfer between different airstreams and assuming constant specific heat, Equation 11 can be expressed in terms of measured quantities as Q& = m& ec f ( te tin) + ( m& in m& in) c f ( tin ta ) For convenience in subsequent expressions for systems with significant leakage, Equations 4 and 12 may be combined to define an effective heat transfer fluid flow rate for air-heating systems, i.e., (12) ( tin ta ) m& = m& e + ( m& e m& in ) (13) ( t t ) e in which can be used for calculations in Sections 7, 8 and 9. Equation 7, 8 and 9 can be used as the basis for plotting measured efficiencies versus (t in t a )/G provided t in is taken as the massweighted mean temperature of the in-leakage and inlet flow rates Testing with positive gauge pressure In this configuration with leakage ( m& L < 0), heated air escapes from the collector to the environment with an attendant loss of useful energy. In addition to the exit enthalpy flow, the

107 collector also introduces an infiltration of ambient air into the load at a flow rate equal to the collector leakage. Therefore, the collector supplies the enthalpy flow m & h + m& ) h ) to the load. e e ( L a With the inlet enthalpy flow rate of m& ihi an enthalpy balance for this case again gives Equation 11 (with due regard for the sign of m& L ). Consequently, Equations 7 to 9 apply also when testing under positive gauge pressure Conversion of thermal performance test characteristics The equations shall be presented in terms of the aperture area A a, as well as in terms of the absorber area A A. The following basic conversions shall be used: η 0 A = η a 1 = a A a 2 = a A 1a 0 a 2a A A A A a A A A a A a A (9) (10) (11) 6.5 Steady-state efficiency test using a solar irradiance simulator General The performance of most collectors is better in direct solar radiation than in diffuse and at present there is little experience with diffuse solar simulation. This test method is therefore designed for use only in simulators where a near-normal incidence beam of simulated solar radiation can be directed at the collector. In practice it is difficult to produce a uniform beam of simulated solar radiation and a mean irradiance level has therefore to be measured over the collector aperture The solar irradiance simulator for steady-state efficiency testing A simulator for steady-state efficiency testing shall have the following characteristics: The lamps shall be capable of producing a mean irradiance over the collector aperture of at least 700 Wm -2. Values in the range 300 Wm -2 to 1000 Wm -2 may also be used for specialized tests, provided that the accuracy requirements given in table 5 can be achieved and the irradiance values are noted in the test report. The mean irradiance over the collector aperture shall not vary by more than ±3 % during a test period. At any time the irradiance at a point on the collector aperture shall not differ from the mean irradiance over the aperture by more than ±15 %. The spectral distribution of the simulated solar radiation shall be approximately equivalent to that of the solar spectrum at optical air mass

108 draft standard for testing of solar air collectors based on EN and ASHRAE Where collectors contain spectrally selective absorbers or covers, a check shall be made to establish the effect of the difference in spectrum on the (τα) product for the collector. If the effective values of (τα) under the simulator and under the optical air mass 1.5 solar radiation spectrum differ by more than ± 1 %, then a correction shall be applied to the test results. 3µ m τ( λ)α( λ)g( λ)dλ 0,3µ m Effective( τα) = 3µ m (12) G( λ)dλ 0,3µ m Measurement of the solar simulator's spectral qualities shall be in the plane of the collector over the wavelength range of 0.3 µm to 3 µm and shall be determined in bandwidths of 0.1 µm or smaller. For certain lamp types, i.e. metal halide designs, it is recommended that the initial spectral determination be performed after the lamps have completed their burn-in period. The amount of infrared thermal energy at the collector plane shall be suitably measured (measurements in the wavelength range above about 2.5 µm if possible, but starting not beyond 4 µm) and reported (see 6.2.2). The thermal irradiance at the collector shall not exceed that of a blackbody cavity at ambient air temperature by more than 5 % of total irradiance. The collimation of the simulator shall be such that the angles of incidence of at least 80 % of the simulated solar irradiance lie in the range in which the incident angle modifier of the collector varies by no more than ± 2 % from its value at normal incidence. For typical flat plate collectors, this condition usually will be satisfied if at least 80 % of the simulated solar radiation received at any point on the collector under test shall have emanated from a region of the solar irradiance simulator contained within a subtended angle of 60 or less when viewed from any point. NOTE 1 Additional requirements concerning collimation apply to measurement of the incident angle modifier (see ). The irradiance shall be monitored during the test and shall not vary by more than ± 3 % during the test period. The method used for measuring the irradiance during the test period shall produce values of mean irradiance which agree with those determined by spatial integration to within ±1%. NOTE 2 The spectral distribution of the lamps (indoors) and of the sky (outdoors) can and do lead to very wide discrepancies in spectrally selective absorbers or covers Test installation Collector mounting and location requirements outlined in 6.1 shall be followed. A wind generator shall be used with a solar simulator to produce an air flow in accordance with Preconditioning of the collector The procedure outlined in shall be followed. 36

109 6.5.5 Test procedure The collector shall be tested over its operating temperature range in approximately the same way as specified for outdoor testing (see 6.4.4). However, eight test points shall be adequate for testing in solar simulators provided that at least four different inlet temperatures are used, and adequate time is allowed for temperatures to stabilize. One inlet temperature should lie within ± 3 K of the ambient air temperature, if possible. During a test, measurements shall be made as specified in These may then be used to identify test periods from which satisfactory data points can be derived Measurements during tests in solar irradiance simulators General Measurements shall be made as specified in Measurement of simulated solar irradiance NOTE Simulated solar irradiance usually varies spatially over the collector aperture as well as varying with time during a test. It is therefore necessary to employ a procedure for integrating the irradiance over the collector aperture. Time variations in irradiance are usually caused by fluctuations in the electricity supply and changes in lamp output with temperature and running time. Some lamps take more than 30 min to reach a stable working condition when warming up from cold. Pyranometers may be used to measure the irradiance of simulated solar radiation in accordance with Alternatively, other types of radiation detector may be used, provided they have been calibrated for simulated solar radiation. Details of the instruments and the methods used to calibrate them shall be reported with the test results. The distribution of irradiance over the collector aperture shall be measured using a grid of maximum spacing 150 mm, and the spatial mean deduced by simple averaging Measurement of thermal irradiance in simulators The thermal irradiance in a solar simulator is likely to be higher than that which typically occurs outdoors. It shall therefore be measured to ensure that it does not exceed the limit given in The mean thermal irradiance in the collector test plane shall be determined whenever changes are made in the simulator which could affect the thermal irradiance, and at least annually. The mean thermal irradiance in the collector test plane and the date when it was last measured shall be reported with collector test results Ambient air temperature in simulators The ambient air temperature ta in simulators shall be measured, taking the mean of several values if necessary sensors shall be shielded in order to minimize radiation exchange. The air temperature in the outlet of the wind generator shall be used for the calculations of collector performance Test period The test period may be determined in the same way as for outdoor steady-state testing. The more stable environment of an indoor test facility may allow steady-state conditions to be maintained more easily than outdoors, but adequate time shall still be allowed to ensure proper steady-state operation of the collector as specified in

110 draft standard for testing of solar air collectors based on EN and ASHRAE Test conditions The test conditions described in for outdoor testing shall be observed with the following additions: - The thermal irradiance in the plane of the collector aperture shall not exceed that from a blackbody cavity at ambient air temperature by more than 5 % of the total irradiance. - The air issuing from the wind generator shall not differ in temperature from ambient air by more than ±1 K Computation and presentation of results The analysis presented in for outdoor testing is also applicable to solar simulator tests, and the results shall be presented on the format sheets shown in annex D. 6.6 Determination of the effective thermal capacity and the time constant of a collector Work in progress 6.7 Incidence angle modifier Work in progress 6.8 Determination of the pressure drop across a collector Work in progress 6.9 Determination of the leakage rate The testing of the solar collector to determine its air leakage rate can be conducted to generate a leakage curve that can be used to compare the relative leakage of different collectors. Air leakage due to the test loop can have a significant effect on the results when testing a collector. Air leakage of the test loop, excluding the collector, shall be less than one-half of 0.5% of the manufacturer's recommended operating flow rate or m³/min, whichever is greater, at 249 Pa and shall be determined by a static air leakage test. The test for measuring collector leakage rate will utilize equipment that determines volumetric leakage rate with an accuracy to ±3% of reading. Measurement of collector pressure will be made with a manometer that meets the accuracy provisions of Section The collector is allowed to come to thermal equilibrium with the ambient air temperature. Using ambient air, a leakage curve is determined by either evacuating or pressurizing the collector using a calibrated flow-measuring device and by measuring the pressure difference between the solar collector and the ambient. The leakage test will be performed at negative gauge pressure for collectors that will be thermally performance tested at negative pressure and the test performed at positive gauge pressure for collectors that will be performance tested under positive pressure. Sufficient data points of collector pressure and leakage should be measured to allow accurate interpolation of collector leakage at operating pressures between 0 and the manufacturer's maximum published operating pressure, whichever is higher. For those cases where there is no published maximum pressure, the collector shall be tested at an operating pressure between 0 and 250 Pa. At least four data points will be taken at collector pressures between these levels. A leakage curve will be drawn and reported along with the actual data points. An example of a representative leakage curve is shown in Fig

111 Fig. 6: Example of a leakage curve for a flat-plate air collector for positive internal pressure with outward leakage Test Apparatus Air leakage shall be minimized by sealing all joints except those that are part of the collector or the manufacturer's collector assembly. The portion of the duct loop that requires testing will include all sections of ductwork that contain temperature-, pressure-, or flow-measuring stations used for measuring collector performance and transition fittings/ducts used for collector hookup. This includes the inlet duct from before the flow-measuring station to the collector and the outlet duct from the collector discharge point to the outlet of the flowmeasuring station. The applicable sections of the test loop may be capped off and tested separately or together to determine air leakage. Fig. 7: Schematic of apparatus used for measuring air leakage in air collectors. The recommended leak check apparatus is shown in Fig. 7. This apparatus utilizes an orifice or nozzle mounted in a section of straight pipe, a motor and blower, and a flow control damper. Pressure taps are installed on either side of the orifice and are connected to a manometer for flow measurement. 39

112 draft standard for testing of solar air collectors based on EN and ASHRAE Presentation of results Work in progress graphs for eta t eta t mass flow t-mass flow eta mass flow pressure drop leakage rate 40

113

114 SolarPACES 13th Symposium on Concentrating Solar power and Chemical Energy Technologies, June , Seville, Spain In the frame work of a collaborative research project (Solar Heat for Industrial Processes) of the IEA Solar Heating and Cooling Program (Task 33) and the IEA SolarPACES Program (Task 4) a parabolic trough collector has been tested according to this test method under quasi dynamic conditions. For the purpose of this work the test identity was eliminated by introducing an arbitrary scale factor. Collector model The collector output is modeled with 6 parameters using the following equation 4. Q& 2 dϑ = η K K G c c c m b θ G b η d d ϑm ϑa ϑm ϑa A θ ( ) + θ 1 ( ) 2 ( ) dt In contrast to the Standard 1,2 the hemispherical irradiance G is divided into the direct G b and diffuse G d parts. For both irradiances an incident angle modifier is used. K θb (θ) being a function of the angle of incidence of the direct irradiance and the constant K θd for the diffuse irradiance. The conversion factor η 0 is the efficiency of the collector at ambient temperature under steady state conditions. The thermal losses are modeled by a 2 nd order polynomial approach, c 1 and c 2 being the heat loss coefficients corresponding to the temperature difference between the mean fluid and ambient temperature and the square of the temperature difference respectively. The effective collector capacity c 3 accounts for the transient behavior of the solar collector and permits measurements under changing levels of irradiance. The introduced effective thermal capacity permits continuous measurements even under scattered cloud conditions. Collector test A prototype collector test was carried out on the test facility of the German Aerospace Center (DLR) in Cologne. The test facility allows for testing up to a temperature of 250 C. Two axis normal tracking (K θb = 1) was active throughout all sequences of the test. In order to operate the collector array at different conditions five test sequences have been used covering clear sky conditions as well as scattered clouds. The mean fluid temperature varied from close to ambient up to 175 C. The length of the test sequences varied between four and seven hours. Table 1 summarizes the conditions of the five test sequences used for the parameter identification. Test sequence Duration [min] Mean fluid temp [ C] Sky condition Clear sky Scattered clouds Clear sky Mainly clear sky Clear sky Table 1: Test sequences used for parameter identification Figures 1 and 2 show the direct irradiance G b, diffuse irradiance G d and the specific collector output P col per aperture area during two test sequences (data in arbitrary scale, to eliminate original test identity). page 2 / 5

115 SolarPACES 13th Symposium on Concentrating Solar power and Chemical Energy Technologies, June , Seville, Spain 1000 irradiance, collector output [W/m²] time [h] Gb Gd Pcol Figure 1: Test sequence 2, unstable irradiance 1000 irradiance, collector output [W/m²] time [h] Gb Pcol Gd Figure 2: Test sequence 5, on a clear day Parameter identification and results For the evaluation of the measured data Multiple Linear Regression (MLR) as the parameter identification tool is foreseen 4. MLR uses a fast, non-iterative matrix method. However, other algorithms, mainly used for non-linear models, lead to the same results and will be allowed as page 3 / 5

116 SolarPACES 13th Symposium on Concentrating Solar power and Chemical Energy Technologies, June , Seville, Spain parameter identification tool in the next review of the Standard. A comparison of the MLR method and the iterative method has been published 7. The advantage of the iterative method is a high flexibility with respect to the input data as well as to the collector model. For this study the DF program 5 was used. It uses the Levenberg Marquardt algorithm 6 for the parameter identification process.. Table 2 shows the parameter set determined from five test data series. In Figure 3 the measured and calculated collector output for test sequence 1 is plotted. The dynamics of the measured collector output are very well described by the five collector parameters. η 0 K θd c 1 c 2 c 3 [-] [-] [W/(m²K)] [W/(m²K²)] [J/(m²K)] Table 2: Determined collector parameter collector output [W/m²] time [h] measured modeled Figure 3: Measured and modeled collector output of test sequence 2 Conclusion A parabolic trough collector prototype has been efficiency tested according to EN Standard using the test method under quasi dynamic conditions. This test method allows varying ambient conditions and continuous measurements over the day. This is possible because a collector model is used that takes into account the effective collector capacity as well as the diffuse irradiance on the aperture plane. The results show a very good agreement between measured and modelled collector output. An overall testing time of 5 days only in part under clear sky conditions, was sufficient to extract the relevant collector performance parameter set. With the use of the European Standard EN a performance testing not only for flat plate or evacuated tubular collectors but also for all tracking and concentrating collectors is possible. page 4 / 5

117 SolarPACES 13th Symposium on Concentrating Solar power and Chemical Energy Technologies, June , Seville, Spain Nomenclature Symbol Unit Description A m² Area b m Collector width C geo - Geometric concentration ratio b/πd c 1 W/(m²K) Heat loss coefficient at (t m t a ) = 0 c 2 W/(m²K²) Temperature dependence of the heat loss coefficient c 3 kj/(m²k) Effective thermal capacity dϑ m /dt K/s Time derivative of the mean fluid temperature d m Absorber tube diameter G W/m² Hemispherical solar irradiance G b W/m² Direct (beam) irradiance G d W/m² Diffuse irradiance K θb (θ) - Incident angle modifier for beam irradiance K θd - Incident angle modifier for diffuse irradiance P col W Useful output power Q W Useful output power η 0 - Conversion factor ϑ a C Ambient temperature ϑ m C Mean fluid temperature θ Incident angle of the beam irradiance References 1 ASHRAE 93-77, Methods of Testing to determine the thermal performance of solar collectors, American Society of Heating, Refrigeration and Air Conditioning Engineers. New York, ISO 9806:1994, Test methods for solar collectors - Part 1: Thermal performance of glazed liquid heating collectors including pressure drop, Part 2: Qualification test procedures 3 Lüpfert E, Herrman U, Price H, Zarza E, Kistner R, Towards standard performance analysis for parabolic trough collector fields, Proceeding SolarPaces Conference Oxaca, EN :2001, Thermal solar systems and components Solar collectors. Part 2: Test methods, CEN Brussels, Spirkl W, Dynamic SDHW system Testing, Program Manual, Sektion Physik der Ludwig- Maximilians Universität München, Press W, Teukolsky SA, Vetterling WT, and Flannery BP, Numerical Recipes, second Edition. Cambridge University press, Fischer S, Heidemann W, Müller-Steinhagen H, Perers B, Collector parameter identification iterative methods versus multiple linear regression, ISES Solar World Congress, Gothenburg, page 5 / 5

118 W4.1 Resource document Recommendations on testing of solar air collectors Dissemination level: Public Author: Josef Buchinger, arsenal research Reviewer: Peter Kovacs, SP June 2006 CONTENTS INTRODUCTION Introduction, why testing of solar air collectors is important and why standards are lacking so far. EXPERIENCE AND ACTIVITIES OF ARSENAL RESEARCH Description of the experience and activities of one of the main actors in the field to gain an insight and background knowledge of what has happened in testing of solar air collectors so far. ASPECTS OF TESTING SOLAR AIR COLLECTORS Introduction into the various technical problems of testing solar air collectors. RECOMMENDATIONS Recommendations for further activities to achieve a standard for testing. Technical suggestions for testing of solar air collectors. ACKNOWLEDGEMENTS FOR EARLIER REPORTS Hubert Fechner, arsenal research SUMMARY Solar Air systems are a further promising technology in active using solar energy for heating. Differently to solar liquid systems, air systems so far have not entered the market with significant rates. As main obstacle for a wide dissemination of solar air systems appears lacking information as well as lack of confidence on how these systems will perform. Planners and architects as well as end consumers need trustworthy parameters and facts to start applying and investing in this technology. Testing of the respective components is therefore essential but up to now no such widely accepted standard exists for air collectors. Based on the experience of arsenal research the technical problems such as calculation and presentation of the thermal performance are introduced and it is pointed out that testing of solar air systems in general is difficult and many technical aspects still need to be investigated. It is recommended to start serious activities which will ascertain these aspects and to gain a profound base for drafting a standard procedure for testing the performance and quality of solar air collectors. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6 File: WP4-D2.1.f-ANNEX-2-Testing of solar air collectors 1.1.doc

119 page 2 of 13 pages Table of contents SUMMARY...1 Table of contents Introduction Experience and activities of arsenal research IEA Task 19: Solar air systems Further activities Latest activities Aspects of testing solar air collectors Thermal performance subject to reference temperature Presentation of Efficiency Curves Closed loop system Open loop system Temperature rise Recommendations Need for further investigation Testing Conditions General Testing Features Symbols, Units and Abbreviations References...13 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

120 page 3 of 13 pages 1 Introduction Solar air collectors are not wide spread so far. As main obstacle for a wide dissemination appears lacking information as well as lack of confidence on how these systems will perform. Testing of the respective components is therefore essential. Such tests should be reproducible and acknowledged and are therefore commonly standardised in e.g. European Standards. So far no such standard exists for air collectors. Development of a standard could be done by research institutes familiar with testing and standardisation, but require resources. Those resources could be made available from the industry or from public funds. As mentioned in the beginning the market for solar air collectors is weak; hence the current situation is that the air collector industry is not interested in supporting the definition of a standard. Further, to gain national public funds the authorities require contribution from national industry. Unfortunately the situation in Austria is such that a research institute (arsenal research) is willing to start the process of standardisation but is lacking of local support by the industry since there is none in the field of air collectors and therefore lacking financial support. Achieving high standards of air collector measurements is not a simple task; Generally, measuring of air-temperatures and air mass flows requires much higher effort for gaining satisfactory accuracy s. Moreover, leakage, the air flow pattern inside the collector and the much lower heat transfer from the absorber to the heat transfer medium are further complex affects. The assembling of the air system components, the way how the components are connected, how the system is operated are all very decisive factors for the efficiency of the whole air system. Compared to liquid collectors, the measuring procedure for solar air collectors needs even more expenditure; no satisfactory standardised testing procedure exists so far. The Italian Standard UNI only gives an idea of how testing of air collectors can be carried out, but does not touch the specific problems of solar air systems in its 12 pages. Starting a standardisation process for testing solar air collectors has been already discussed in the Technical Committee 180 of the International Standardisation Organisation (ISO), but work is still resting. 1 UNI 8937: Collettori solari piani ad aria Determinazione del rendimento termico, Norma Italiana, NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

121 page 4 of 13 pages 2 Experience and activities of arsenal research In order to start solving the problems of solar air collectors arsenal research has had some small and major projects starting from investigating solar air systems up to activities within this project to start a self financed group on defining a standard for testing of solar air collectors. 2.1 IEA Task 19: Solar air systems In order to pool the experience in designing air systems for space heating, the International Energy Agency (IEA) initiated a five year project: Within Task 19 Solar Air Systems of the Solar Heating and Cooling Programme more than twenty experts from nine countries, coordinated by the Operating Agent Arch. Robert S. Hastings worked together. One part of this task was an investigation on series produced solar air collectors, done by arsenal research in 1999 (Fechner 1999). Seven long time proven products as well as prototypes from seven different countries, mainly from Europe but also from Canada and Australia have been tested. The main topics of development, investigation and research during this project have been: Development of a steady state testing procedure for solar air collectors, suited for all types Discussion on physically correct and proper efficiency presentations Development of different performance descriptions adequate for all common operation modes A comparison of available series-produced products Investigation of the technical behaviour of different types of air collectors Recommendations for an optimised utilisation of solar air collectors Recommendations for improvements of tested products Adaptation of the existing solar-laboratory-facilities for testing solar air collectors 2.2 Further activities In recent years arsenal research has tried to gain further development projects in this field. One very promising activity was the thoroughly investigation and testing of prototypes of an air collector totally made of polymeric materials (Selke 2005). 2.3 Latest activities Initiated by the solar air collector producer Grammer Solar GmbH arsenal research has tried to assemble producers and testing institutions to define a draft for a European Standard of testing air collectors. A small survey beyond actual producers of solar air collectors has been undertaken to get an impression on the need for a standard and the willingness to support a project to define such a testing procedure. The findings were that only the initiator had a significant interest where as other producers either did not feel the need for such a standard since they are making their business even without any standards or did not want to invest in such a poor dogs or question mark (regarding the definitions used in SWOT analysis) product. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

122 page 5 of 13 pages 3 Aspects of testing solar air collectors The function of solar-air collectors is to efficiently transfer the energy of the sun (radiated to the earth by short-wave rays) into thermal energy. To assess the quality of this conversion process under different operation conditions is the aim of collector-testing. With these results customers and planners should be enabled to compare different products, calculate or simulate the outcome of systems and plan and design systems. The current problem in testing air collectors are missing standards for: the definition of reference temperature for the different types of air collectors presentation of efficiency curves or other performance indicators for best usability and significance testing procedures, instrumentation and sensor arrangements for different types of air collectors procedures to avoid condensation and monitor humidity As stated before, measuring of air temperatures and air mass flows requires higher effort for gaining comparable accuracies. Moreover, leakage, the air flow distribution inside the collector and the much lower heat transfer from the absorber to the heat-transfer-medium are further complex affects. Opposite liquid solar collectors the efficiency of solar air collectors is strongly influenced by the actual mass flow rate inside the collector due to the often rather low heat transfer between absorber and air. This heat transfer is highly dependent on the air speed. It is, therefore, often difficult/impossible to extrapolate from tests of small modules of solar air collectors in test rigs to larger solar air collector arrays as the air flow pattern might be different. In the following some aspects are presented in detail to exemplify the problems of testing solar air collectors. 3.1 Thermal performance subject to reference temperature Optical features (absorption and emittance of the absorber, transmittance of the cover), materials used (absorber material, cover material, frame, insulation) and constructing characteristics (mainly the airflow-principle and the effective heat transfer area) of the collector are of basic importance for the efficiency. However, the respective operation condition of the collector is decisive as well and the efficiency decreases with increasing temperatures within the collector because of the increasing heat losses. The efficiency of a solar-(air)-collector is defined as the ratio of useful gain of the collector to the respective solar performance of the sun G T at the collector reference area A C. Q& u Q& u Efficiency (eta): η = = Q& Q& u = m& c p ( To Ti ) A G sol c T As collector reference area can be considered: aperture area, absorber area or gross area. A general equation for solar collector performance is based on (Hottel 1958) and (Bliss 1959): η = F τ 0 0 α U L ( T Q& o sol T a ) NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

123 page 6 of 13 pages where T a is the ambient temperature and F 0 is the collector heat removal factor in relation to T 0 - the collector outlet temperature and η 0 is the efficiency when the outlet temperature is taken as reference. F 0 accounts for the fact that the absorber temperature is not the same as the outlet air collector temperature neither in the horizontal direction nor vertical. It appears that the efficiency η depends on the operation conditions of the collector. It decreases with increasing temperatures, because of the increasing heat losses. Important is, how to define the operation conditions of the collector given by the temperature difference between the overall collector temperature T K and ambient T a. An efficiency curve can be drawn in dependency of a certain reference temperature which corresponds to the collector temperature T K. In a physically correct way one has to take a weighted mean temperature T K of the whole collector box, but in the measuring practice this is not practicable. That is why three temperatures are for choice: The inlet temperature (T i ), the outlet temperature (T O ) and a so called mean collector temperature (T m ) which can be calculated as the arithmetic mean value between inlet and outlet temperature. Efficiency curves of a solar collector corresponding to the three possible reference temperatures for a constant mass flow rate are shown below: Fig. 1: Efficiency related to different reference temperatures (Fechner 1999) For solar liquid collector it is the custom to present the efficiency related to the mean collector temperature (T m ) representative for the heat losses of the collector. For liquid collectors, where the temperature difference between inlet and outlet is very small (normally less than 10 K) and the heat transmission from the absorber to the fluid is high, the arithmetical mean value (T m ) is in fact very close to the physical mean temperature (T K ) of the collector. For air collectors the difference between inlet and outlet can be up to 30K or 40K dependent on the mass flow. Also important is the amount of heat transmission from the absorber to the fluid, which is for air collectors usually not that high. Due to these effects there will be no longer a linear increase of the fluid temperature along the collector plate and the arithmetical mean value (T m ) is often not representative for the heat losses of the collector. Measurements indicate that the physical mean temperature (T K ) of the collector is often much closer to the outlet temperature (T O ) than to the arithmetical mean temperature (T m ). Therefore the presentation of the collector efficiency curve using the outlet temperature (T O ) often seems to be the best solution. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

124 page 7 of 13 pages 3.2 Presentation of Efficiency Curves Solar air collectors usually operate in two different ways: 1. closed loop system 2. open loop system It is recommended to use separate ways to present the efficiency curves for the different systems Closed loop system In a closed loop system the inlet temperature can be sometimes much higher than the ambient temperature. For this mode of operation the presentation of efficiency ηversus (T o -T a )/G is adequate. Fig. 2: Efficiency related to Outlet temperature (Fechner 1999) NOTE: Be aware of the fact that the efficiency values close to y-axis would only appear, if the inlet temperature is below ambient, because if T o = T a, T i is always below T a. The maximum efficiency which will therefore appear in practice (if operating with air with at least ambient temperature) can be seen from the second curve (efficiency vs. massflow rate) Open loop system If the collector operates in an open loop, it always sucks in air with ambient temperature (T i = T a ), then a presentation of the efficiency η versus the mass flow rate is better suited for engineering purposes. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

125 page 8 of 13 pages Fig. 3: Efficiency versus mass flow rate (Fechner 1999) NOTE: Pay attention to the fact, that both the increased heat transfer and the fact that the higher the mass flow rates the lower the mean temperature of the collector (which is responsible for the heat losses) makes the conversation process more efficient at higher mass flow rates. 3.3 Temperature rise Another important aspect for planners is the temperature-rise. For direct heating you often need a certain level of the outlet-temperature. In such a case, optimisation towards high efficiency is no longer practicable. The diagram below shows the temperature rise for different irradiance levels. Fig. 4: Temperature rise vs. mass flow rate (Fechner 1999) NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

126 page 9 of 13 pages 4 Recommendations 4.1 Need for further investigation As it can be seen from the results and experience described above many aspects in testing solar air collectors still need further investigation, research and discussion. It seems that to get out of the vicious circle where the solar air collectors are at the moment, it requires an initiative beyond this project to solve the problems of missing standards and guidelines for planners and architects. It is therefore recommended to initiate an additional activity of producers, planners, architects, research and testing institutes to investigate thoroughly the different aspects of testing. To describe the relevant basics, define procedures for testing and find the best solution of presenting results, so that everybody is enabled to compare solar air collectors and calculate their performance within a solar air system. In cooperation with the German air collector producer Grammer Solar a workshop on solar air collectors will be organised in Austria this autumn. If relevant stakeholders will get together it might be a next step towards working out a standard for air collectors. Still, their further activities will need the support by national and international funding institutions and the industry. 4.2 Testing Conditions As an entry point and until further results are available the following recommendations for testing conditions derived from the experience made under the IEA TASK 19 can be made. For achieving comparable and reproducible test results, the conditions for testing must be defined carefully. Basis, whenever reasonable, are the conditions written in the standards for testing of liquid collectors (EN ). Especially the radiation elements remain the same. In the following the recommendations made earlier in (Fechner 1999) are updated and summarised. From liquid systems we know that judging a solar system is often to much concentrated on the thermal performance of the collector; other features of the system like control strategy, mounting of temperature sensors, connecting the modules, storages, insulation matters and many other questions should also be considered carefully. Further Investigations seems to be necessary in the general issue of presenting the thermal energy output of air collectors. The problem with reference area well known from liquid collectors as well as the problem of the reference temperature is open for further discussions General Testing Features Facilities for indoor measurements Artificial radiation source IR-shielding device simulating the cold sky temperature Surrounding air speed generated by fans (wind simulation) Automatic electro-pneumatic driven facility for measuring the global radiation Temperature precision control for stability of the inlet temperature of about ±0.05K Cooling machines for constant ambient temperature NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

127 page 10 of 13 pages Measurement In- and outlet temperature Surrounding air temperature Mass flow rate: orifice plates in combination with high precision pressure gauges Global radiation: Pyranometer Devices for controlling and stabilising the voltage supply Barometric pressure measurement Surrounding air speed measurement Dew point measurement Static pressure measurement at inlet and outlet Tests Efficiency testing (temperature rise, mass flow rate, specific heat) Wind dependence Leakage rate Pressure drop Stagnation temperature Time constant Temperature distribution in the air channel Buoyancy effects Tilt angle Tilt angle for testing differs according to the typical installation and should always be arranged according to the manufacturer Mass flow rates Mass flow rates recommended for testing can be ranging from 20 to 100 kg/h/m² if no special arrangements with the manufacturer are taken Leakage For an accurate measuring process 2 fans are needed, one at the inlet and one at the outlet. In order to minimise the leakage rate the mean static pressure inside the collector should be equal to the atmospheric pressure. Realistic conditions can be simulated by arranging the fan according to the producer s recommendations either before or after the collector Mass flow measurements Due to possible leakages it is recommended to test the mass flow in the inlet ducts as well as in the outlet ducts. Moreover leakages in the duct system will be found easily if the mass flow is measured twice Air-flow pattern The air flow pattern inside the collector is very important for a correct assessment of the performance. The air flow pattern inside the collector (especially the partition close to the inlet) mainly depends on the connection between ducting system and collector. In standardised tests only a single collector module is tested, which might not comply with the mode of installation in praxis. To reach an even air flow pattern throughout the collector special distribution ducts at the inlet and outlet should be used for each collector tested. By means of boxes with perforated metal sheets at inlet and outlet a well distributed air flow can be achieved, which means that an even air flow from the centre-line of the collector to the edges from entrance to outlet exists. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

128 page 11 of 13 pages Temperature measurements While correct measurements of the inlet temperature should be of no problem, the determination of the outlet temperature is critical. Several layers of different air temperatures are often close adjacent, therefore a specific mixing device optimized according to fluid dynamic experiences - at the outgoing duct just before the sensors and a sophisticated arrangement of temperaturesensors are necessary Wind simulation For testing collectors it is essential to find out the wind dependency of a collector. Some collectors, especially uncovered, but also collectors with the air flow directly under the cover are strongly dependent on wind. Recommendations for installation of the wind simulation can be found in the existing standard for testing water collectors. Only for uncovered collectors the direction of the wind can be influential as well Humidity The humidity should be monitored during the tests. Care should be taken for effects of condensation, since for some testing points inlet temperatures far below ambient can be necessary. Any eventual condensation inside the collector or the circuit must be removed while heating the collector circuit at 70 C for a period of at least 30 minutes Irradiation level Although irradiation levels used at solar air applications are sometimes low, for testing levels starting with about 700 W/m² are recommended. Testing at lower irradiation levels should be avoided to ensure a significant temperature rise between the inlet and outlet and further accuracy of measurements Conditioning For testing the collector with different air temperatures, the preparation of differently conditioned air is needed. To attain foreseen stationary/stable conditions, the test circuit must be preconditioned for 15 minutes at the test temperature so as to verify that the inlet temperature is well within the conditions. For indoor-testing an enclosed climatic chamber with reasonable size and temperatures between some degrees below zero and up to 60 C is recommended. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

129 page 12 of 13 pages 5 Symbols, Units and Abbreviations A C c p Collector reference area (normally: aperture area) [m2] Specific heat [Jkg-1K-1] η Efficiency F 0 collector efficiency factor G T Global solar irradiance on tilted surface (further on only G) [Wm-2] G Global solar irradiance [Wm-2] m Mass flow rate [kgh-1] m i Mass flow rate at collector inlet [kgh-1] m L Leakage Mass flow rate [kgh-1] m o Mass flow rate at collector outlet [kgh-1] Q sol Performance irradiated by the sun at the collector reference area [W] Q u Useful gain of the collector [W] T a Ambient temperature [ C] T am Measured absorber temperature [ C] T i Inlet temperature [ C] T K Collector temperature (physical collector-mean-temperature) [ C] T m Mean Collector temperature (arithmetic mean value between inlet and outlet temperature) [ C] T O Outlet temperature [ C] U L Collector overall heat loss coefficient α Solar absorptance ε Hemispherical emittance τ Transmittance NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

130 page 13 of 13 pages 6 References Bliss R. (1959). "The Derivation of Several Plate Efficiency Factors Useful in the Design of Flatplate Solar-heat Collectors." Solar Energy Vol. 3: pp Fechner H. (1999). IEA TASK 19 Solar air systems - Investigation on Series Produced Solar Air Collectors - Final Report. Vienna, arsenal research - Department of Renewable Energy. Hottel H., Whillier, A. (1958). Evaluation of Flat Plate Collector Performance. Trans. Conf. On the use of Solar Energy, Tucson, Arizona, U.S.A. Selke T. (2005). Studie zur Darstellung des technischen und wirtschaftlichen Marktpotentials des Solar-Luftkollektors der BAYER AG. Wien, arsenal research: 73. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

131 New Generation of Solar Thermal Systems CONTENTS 1. INTRODUCTION 2. INVESTIGATED SOLAR COLLECTORS 3. INFLUENCE OF SOLAR COLLECTOR FLUID VOLUME FLOW RATE 4. INFLUENCE OF THE METHOD TO DETERMINE THE MEAN SOLAR COLLECTOR FLUID TEMPERATURE 5. INFLUENCE OF THE METHOD TO DETERMINE THE SPECIFIC HEAT OF THE SOLAR COLLECTOR FLUID 6. INFLUENCE OF THE TEMPERATURE LEVELS USED IN THE TEST 7. INFLUENCE OF THE WEATHER CONDITIONS USED IN THE TEST 8. CONCLUSIONS REFERENCES WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR SUMMARY DTU, January 2007 The test method of the standard EN (European Committee for Standardization, 2004) is used by European test laboratories to determine the efficiency of solar collectors. The aim of this work is to present an evaluation of the test method for a 12.5 m² flat plate solar collector panel from Arcon Solvarme A/S. CFD (Computational Fluid Dynamics) simulations, calculations with a solar collector simulation program SOLEFF and thermal experiments are carried out in the investigation. The influence of flow nonuniformity on the efficiencies of the solar collector is elucidated for different volume flow rates and weather conditions. The influences of the method to determine the mean solar collector fluid temperature, the approximation used to determine the specific heat of the solar collector fluid, the temperature levels used in the tests and the weather conditions on the collector efficiency are investigated. Based on the investigations, it is concluded that with a volume flow rate between 6.0 l/min and 10.0 l/min (corresponding to 0.48 and 0.80 l/min per m 2 solar collector) the collector investigated will have the best efficiency. If a solar collector fluid volume flow rate of 25.0 l/min (corresponding to 2.0 l/min per m 2 solar collector) is used in the test of collector efficiency, it is recommended that the maximum temperature level used in the tests is not higher than the maximum operation temperature of the collector. Further, if the solar collector efficiency for low flow rates are measured in future test methods there is a need to change the method used to determine the mean solar collector fluid temperature.

132 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 2 of Introduction The thermal performance of flat-plate collectors is strongly related to the flow distribution through the absorber tubes /Duf91/. The more uniform the flow distribution, the higher the collector efficiency. However, it is shown by numerous investigations that uniform flow distribution is not always present in solar collectors /Jon94/,/Wan90/, /Wei02/, /Fan05/. The flow distribution through the tubes are influenced by the design of the collectors, the collector tilt and operating conditions of the collector such as volume flow rate and properties of solar collector fluid. The aim of this work is to theoretically and experimentally investigate the flow and temperature distribution in a solar collector panel with an absorber consisting of horizontal fins. CFD simulations are used in the determination of the collector efficiency at different collector fluid flow rates. Based on the CFD calculated efficiencies, the collector efficiency expression is obtained by means of regression. The influence of collector fluid flow rate on the flow distribution and on the collector efficiency will be elucidated. In the test method of the standard EN /Eur04/, the mean solar collector fluid temperature in the solar collector, T m is determined by the approximate equation: T m = ( Tin + Tout ) 2, where T in is the inlet temperature to the collector and T out is the outlet temperature from the collector. The specific heat of the solar collector fluid is a function of the temperature of the fluid. In the test method the specific heat of the solar collector fluid is an approximation for each measuring period determined as a constant equal to the specific heat of the solar collector fluid at the temperature T m. The power produced by the solar collector in a steady state test period is determined by the product of the specific heat, the mass flow rate and the temperature increase of the solar collector fluid. The solar collector efficiency is determined by measurements at different temperature levels. Based on these efficiencies, an efficiency equation is determined by regression analysis. In the test method, there are no requirements on the ambient air temperature and the sky temperature. This work will present an evaluation of the test method for a 12.5 m² flat plate solar collector panel from Arcon Solvarme A/S. The investigations will elucidate: How the collector fluid volume flow rate is influencing the flow distribution and the collector efficiency. How the mean solar collector fluid temperature T m is underestimated by the approximate equation in the test standard and how the collector efficiency equation is influenced by the underestimation of T m. The dependence of the volume flow rate is shown. How the use of the approximate specific heat of the solar collector fluid is influencing the collector efficiency. How the temperature levels used in the tests are influencing the collector efficiency expression. How the measured collector efficiency is influenced by the weather conditions such as the ambient air temperature and the sky temperature. 2. Investigated solar collectors The investigated solar collectors are 12.5 m² solar collector panels, type HTU and HT from Arcon Solvarme A/S, designed for medium and large solar heating systems. Fig. 1 shows the design of the HTU solar collector. The HTU solar collector is tested side-by-side with the similar collector HT, which includes a Teflon foil between the absorber and the cover glass. Each collector consists of two manifolds, one dividing and one combining manifold, and 16 parallel connected horizontal fins in a U type configuration, see Fig. 2. The well known Sunstrips are used as fins and the collector is equipped with a low iron antireflection treated NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

133 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 3 of 21 glass cover. A propylene glycol/water mixture is used as solar collector fluid. Properties of the mixture and their dependences on temperature for a 40% (weight %) glycol solution are as follows /Fur97/: Density, [kg/m 3 ] Dynamic viscosity, [kg/(ms)] Specific heat, [J/(kgK)] Thermal conductivity, [W/(mK)] where T is fluid temperature, [ C]. 2 ρ = * T * T T µ = *( ) C p = * T * T λ = * T 2 Fig. 1. Design and dimensions of the investigated HTU collector Fig. 2. A schematic illustration of the HTU collector configuration. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

134 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 4 of 21 The solar collector fluid is forced to circulate through the collector panel by a pump. The circulating flow rate, in the range of l/min, is measured using a QEC type Clorius flow meter. The flow distribution through the tubes is difficult to measure directly, therefore it is evaluated by temperature measurement just before the fluid enters the combining manifold. Copper-constantan thermo couples, type TT, attached to the back of the tubes are used for the temperature measurements. The positions of the measurement points are schematically shown in Fig. 2 as circles in the collector panel. Since the absorber tube wall has a higher temperature than the fluid inside, small corrections are made to the temperature measurements to get the fluid temperatures. The corrections are determined based on the difference between the measured solar collector fluid outlet temperature and the mean of the measured absorber tube wall temperatures. Solar irradiance on the collector panel is measured using a calibrated pyranometer type CM11 from Kipp & Zonen. The data collection and control program IMPVIEW is used to measure the volume flow rate and temperature as well as the temperature increase over the collector and the ambient temperature and solar irradiance during the steady state test periods. 3. Influence of solar collector fluid volume flow rate Investigations on how the test conditions and the approximate methods used in the test method to determine the collector efficiency will influence the efficiency and the efficiency expression for the HTU and HT collector will be described in the following sections. A 40% propylene glycol/water mixture is assumed to be the solar collector fluid. A collector tilt of 40 is assumed. 3.1 CFD model description The flow distribution through the absorber tubes is investigated theoretically with CFD calculations. A simplified model is built using the CFD code Fluent 6.1 /Flu03/, where the manifolds and the absorber strips are fully modelled, while the existence of the collector casing is represented by a heat loss from the absorber strips to the surrounding air. Due to the large difference in the dimension of the absorber tube length (5.790 m) and the tube hydraulic diameter ( m), a refined grid distribution is needed in the cross section of the tube. Fig. 3 shows grid distribution for the absorber strips and the manifolds. View A shows grid distribution at the vertical cut-plane of both the absorber strips and the manifold. View B and View C show grid setup at the cross-section of the manifold and the absorber strips respectively. Positions of View plane A, B and C are schematically shown in Fig. 2. As shown in Fig.3, the absorber strips including the tubes and the absorber fins are meshed with hexahedron cells, while the manifolds are meshed with a mixed mesh of tetrahedron and wedge cells. Since flow distribution through the tubes is of interest, the mesh density of the tubes (1.4E-8 m 3 /cell on average) and the manifolds (1.1E-8 m 3 /cell on average) are higher than that of the absorber fins (6.9E-8 m 3 /cell on average), which results in about 0.8 million mesh cells for the whole model. The flow distribution through the absorber tubes is investigated with heat transfer and buoyancy effect considered. The conductive heat transfer of absorbed solar energy along the absorber fin and from the absorber fin to the tube walls as well as the convective heat transfer from the tube walls to the collector fluid are included in the CFD calculation. The absorber strips are considered to be surrounded by air. The heat loss from the absorber strips is strongly related to the air flow conditions inside the panel, which are quite complicated and varied for different operating conditions. This makes determination of the heat loss from the absorber strips difficult. In the CFD calculations, the heat loss from the absorber strips is assumed to take place only by means of convection. The convective heat loss can be determined by means of the heat loss coefficient and the air temperature inside the collector. The heat loss coefficient is determined by thermal measurements and the air temperature in the collector is obtained from a solar collector calculation program, SOLEFF /Ras96/. The influence of the collector fluid volume flow rate on the air temperature distribution inside the collector panel is included in the calculations. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

135 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 5 of 21 View A View B View C Fig. 3. Grid setup of the solar collector model. Steady state CFD simulations are performed with Boussinesq approximation for buoyancy modelling. The PRESTO and second order upwind method are used for discretization of the pressure and the momentum equations respectively. The SIMPLE algorithm is used to treat the pressure-velocity coupling (Fluent Inc., 2003). One simulation takes approximately 12 hours for a computer with 3 GHz CPU frequency and 1G memory. The fluid flow through the i th tube (counted from the top) is characterized by a parameter β i, defined as follows: Q i i = Q 0 β (1) where Q i is the volume flow rate through the i th tube, while Q 0 is the overall volume flow rate for all the tubes. For an ideal, uniform flow distribution through 16 tubes, β i equals to 1/16. However, this is not always the case in the solar collector. Instead, strong deviations from uniform flow distribution have been observed in this work as well as in the numerous investigations of other collectors in the literature. Φ, a relative flow non-uniformity parameter, is introduced to quantify the flow maldistribution. Φ 16 2 ( β i 1/16) 1 is defined as: i= 1 (2) Φ = 16 ( 1 ) 100% 16 From its definition, it can be seen that the relative flow non-uniformity parameter is similar to the root-mean-square or standard deviation generally used in the statistical treatment of experiment data, except it is divided by the mean value (1/16). NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

136 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 6 of CFD calculated flow and temperature distribution The flow and temperature distribution through the 16 parallel absorber tubes are investigated under different operating conditions. The parameters investigated are volume flow rates: from 2.7 l/min to 25.0 l/min; solar collector fluid inlet temperatures from 20 C to 100 C. Fig. 4 shows CFD calculated flow and temperature distribution in the strips just before the fluid enters the combining manifold. The solar collector has an inlet flow rate of 2.7 l/min and an inlet fluid temperature of 19.4ºC. The solar irradiance is 883 W/m 2 and the ambient temperature is 30.0ºC. It can be seen that the flow distribution through the tubes is not uniform. Fluid flow rate in the strip increases from the top to the bottom. At the top strip 1, the flow rate is 0.11 l/min, while at the bottom strip 16 the flow rate is 0.23 l/min, approx. twice of that at the top strip. The fluid temperature decreases from the top to the bottom. At the top strip 1, the fluid temperature can be as high as 86.6ºC compared with a temperature of 54.6ºC at the bottom strip. The temperature of the tube wall is approx. 5 K higher than the solar collector fluid temperature. It should be pointed out that all the temperatures are calculated by averaging all the computational cells at the cross section with cell masses weighted. The temperature distribution of the collector panel in the middle of the absorber is presented in Fig. 5. The uneven flow distribution through the tubes results in an uneven temperature distribution. At the top of the collector and close to the outlet, the fluid and the absorber fin temperature are high, while the temperature is lower at the bottom of the collector. The outlet fluid temperature is 64.5ºC. As can be seen from Fig. 4, the minimum and the maximum temperature of the fluid before it enters the combining manifold is 54.6ºC and 86.6ºC respectively, correspondingly 9.9 K lower and 22.1 K higher than the outlet temperature. The temperature difference between the absorber fin and the collector fluid can be seen in the temperature profile Temperature, ºC Temperature of tube wall Fluid temperature Volume flow rate Flow rate 2.71 l/min T inlet =19.4ºC T outlet=64.3ºc T ambient=21.7ºc Solar irradiance G 883 W/m Volume flow rate, l/min 10 0 top bottom Strip number, i th Fig. 4. Temperature and volume flow rate distributions in the strips just before the fluid enters the combining manifold NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

137 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 7 of 21 Fig. 5. CFD calculated temperature distribution ( C) in the middle of the absorber The CFD model was used to calculate heat and fluid flow in the collector panel at different collector fluid flow rates. The investigations are carried out with a solar collector with a slope of 40, a solar irradiance of 1000W/m 2 and an ambient temperature of 30ºC. The investigated parameters include: Collector volume flow rates 3.3 l/min, 4.0 l/min, 6.0 l/min, 10.0 l/min and 25.0 l/min and inlet fluid temperatures 20ºC, 40ºC, 60ºC, 80ºC and 100ºC. The flow and temperature distribution through the absorber tubes just before the fluid enters the combining manifold are shown in Fig. 6 and Fig. 7 respectively. The flow and temperature distribution are calculated with an inlet temperature of 60ºC. Two driving forces control the flow distribution through the absorber tubes: The buoyancy force and the frictional drag force. For low flow rates such as 3.3 l/min and 4.0 l/min, the buoyancy effect is significant, which tends to circulate the collector fluid clockwise in the collector panel, thus decreasing the flow rate in the upper absorber tubes and increasing the flow rate in the lower absorber tubes. For high flow rates such as 25.0 l/min, the frictional drag force tend to be more significant, therefore the flow rate in the top tubes are larger due to less friction loss compared with the fluid flow through the bottom tubes. The temperature distribution of the collector fluid at the end of the absorber tubes is shown in Fig. 7. When the flow rate decreases from 25.0 l/min to 3.3 l/min, the tilt of the temperature distribution curve will increase dramatically. When the volume flow rate is 4.0 l/min, the temperature at the upper part of the collector will have a sharp increase, which will get worse with a decreasing flow rate. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

138 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 8 of Flow rate 3.3 l/min Flow rate 4.0 l/min Flow rate 6.0 l/min Flow rate 10.0 l/min Flow rate 25.0 l/min Uniform flow distribution Φ = 40.2% Φ = 23.9% Φ = 6.9% β i 0.06 Φ = 2.9% Φ = 7.8% Collector tilt angle 40, solar irradiance 1000 W/m 2, T ambient 30 C Inlet fluid temperature 60 C top bottom Strip number, i th Fig. 6. The flow distribution through the absorber tubes at different collector volume flow rates. 160 Fluid temperature, C Flow rate 3.3 l/min Flow rate 4.0 l/min Flow rate 6.0 l/min Flow rate 10.0 l/min Flow rate 25.0 l/min Collector tilt angle 40, solar irradiance 1000 W/m 2 T ambient 30 C, Inlet fluid temperature 60 C top bottom Strip number, i th Fig. 7. The fluid temperature distribution through the absorber tubes at different collector volume flow rates. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

139 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 9 of Collector efficiency Measurements are carried out with the HTU collector at different collector volume flow rates and fluid inlet temperatures and under different weather conditions. Collector efficiencies are determined according to the test standard /Eur04/. The measured collector efficiencies are compared to the CFD calculations. Fig. 8 shows the comparison between the measured and the calculated efficiencies. The collector mean temperature is calculated by averaging inlet and outlet fluid temperatures. There is a good agreement between the measured and calculated efficiencies with a disagreement of less than 4% point. The calculated temperature increase over solar collector is very close to the measured one with a difference of less than 0.50 K. The uneven flow and temperature distribution will decrease the collector efficiency. Fig. 9 shows the CFD calculated collector efficiency at different collector volume flow rates. Boiling is not considered in the CFD calculations. The Reynolds number of the fluid flow in the tube will be in the range of for a volume flow rate between 3.3 l/min and 25.0 l/min if a solar collector fluid of 40% glycol water mixture and a mean collector fluid temperature of 60ºC are used. Therefore a laminar flow model is used in the CFD calculations. Based on the CFD calculated efficiencies, a regression analysis is carried out to find the efficiency expressions of the collector at an incidence angle of 0º operating at different fluid flow rates (see the following 2 Tm Ta ( Tm T a) table 1): η = η0 a1 * a2 * (3) G G Table 1: collector efficiency expressions of the collector at an incidence angle of 0º and at different fluid flow rates Flow rate 3.3 l/min 4.0 l/min 6.0 l/min 10.0 l/min 25.0 l/min η a 1 W/m 2 K a 2 W/m 2 K It can be seen that collectors at flow rates of 6.0 l/min and 10.0 l/min give the highest efficiency, while collectors at flow rates of 3.3 l/min and 25.0 l/min gives lower efficiency. This is due to the uneven flow distribution when the flow rate is too high or too low. The relative flow nonuniformity for flow rates of 3.3 l/min and 25.0 l/min are 40.2% and 7.8% respectively. If the flow rate is low enough, the flow nonuniformity will dramatically increase resulting in an increased collector mean temperature and a decreased collector efficiency. For high flow rates ( >10.0 l/min), the flow nonuniformity will also increase resulting in a decreased collector efficiency. Besides the flow distribution problems, the air temperature distribution inside the collector panel, which varies for different collector volume flow rates, will also influence collector efficiency. That is the reason why the collector efficiency is relatively low at the flow rate of 25.0 l/min. However, if the flow rate is higher than 25.0 l/min, the fluid flow in the tubes will normally be turbulent. For such high flow rates, the collector efficiency will have a sharp increase because the heat transfer coefficient from the tube wall to the fluid will increase dramatically as the fluid flow transits from laminar to turbulent region. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

140 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 10 of Case 2 Case 1 CFD Measurement 60 Case 3 Efficiency, % Case 4 Case 6 Case (Tm-Ta)/G, [Km2/W] Fig. 8. Measured and calculated collector efficiencies Case 1: Inlet flow rate 2.7 l/min, T inlet =19.4ºC, T outlet =64.3ºC T ambient =21.7ºC, G=883 W/m 2 Case 2: Inlet flow rate 3.4 l/min, T inlet =20.0ºC, T outlet =57.6ºC T ambient =24.5ºC, G=866 W/m 2 Case 3: Inlet flow rate 5.0 l/min, T inlet =41.2ºC, T outlet =69.1ºC T ambient =27.8ºC, G=1000 W/m 2 Case 4: Inlet flow rate 4.9 l/min, T inlet =68.7ºC, T outlet =89.9ºC T ambient =27.8ºC, G=919 W/m 2 Case 5: Inlet flow rate 10.2 l/min, T inlet =85.9ºC, T outlet =93.4ºC T ambient =25.3ºC, G=822 W/m 2 Case 6: Inlet flow rate 24.3 l/min, T inlet =86.7ºC, T outlet =91.3ºC T ambient =22.7ºC, G=1018 W/m 2 Collector efficiency, [-] CFD calculation - flow rate 3.3 l/min CFD calculation - flow rate 4.0 l/min CFD calculation - flow rate 6.0 l/min CFD calculation - flow rate 10.0 l/min CFD calculation - flow rate 25.0 l/min Regression - flow rate 3.3 l/min Regression - flow rate 4.0 l/min Regression - flow rate 6.0 l/min Regression - flow rate 10.0 l/min Regression - flow rate 25.0 l/min (T m -T a )/G, [Km 2 /W] Fig. 9. the CFD calculated collector efficiency and the efficiency expression at different collector fluid flow rates The flow and temperature distribution through the absorber tubes in a HTU solar collector panel is investigated experimentally and theoretically. Results show that the flow and temperature are NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

141 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 11 of not uniformly distributed. The uneven distribution is controlled by two driving forces: The buoyancy force and the frictional drag force. For low flow rates (<6.0 l/min, corresponding to 0.48 l/min per m 2 solar collector), the buoyancy effect is significant. The flow and temperature distribution get worse with the decrease of collector volume flow rate, resulting in a decreased collector efficiency. For higher flow rates (> 10.0 l/min, corresponding to 0.80 l/min per m 2 solar collector), the friction drag force tend to be more dominant, the flow distribution get worse with the increase of collector fluid flow rate. This will result in a decreased collector efficiency as long as the flow in the tubes is laminar. It is concluded that with a volume flow rate between 6.0 l/min and 10.0 l/min the collector will have the best efficiency. 4. Influence of the method to determine the mean solar collector fluid temperature CFD calculations have been carried out in order to determine the flow distribution as well as the solar collector fluid temperatures in the HTU collector for different conditions /Fan05/. The CFD model has been validated by temperature measurements. By means of the calculations the mean solar collector fluid temperature in the solar collector can be determined. Figure show results from the calculations. In these calculations, the solar irradiance is 1000 W/m² and the ambient air temperature is 30 C. From Fig. 10 it can be seen that for a low flow rate there are large temperature differences inside the collector and that the fluid temperatures are much higher in the upper strips than in the lower strips of the collector. The temperature profile in the strips from the inlet side to the outlet side is not linear. Fig. 11 shows that the mean solar collector temperature is higher than (T in +T out )/2 which in the test method is used as the mean solar collector fluid temperature, due to the nonlinear temperature profile in the strips. For decreasing flow rate the underestimation of the mean solar collector fluid temperature is increasing T ambient 30 C, solar irradiance 1000 W/m Fluid temperature, C Strip Strip 06 Strip Strip 16 Outlet side Inlet side Distance from the combining manifold, m Fig. 10. Calculated solar collector fluid temperatures inside 4 strips: The top strip, strip no. 6 from the top, strip no. 10 from the top and the bottom strip for an inlet temperature of 40 C and a flow rate of 2.9 l/min. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

142 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 12 of Temperature difference, K CFD: collector tilt angle 40, solar irradiance 1000 W/m 2 T ambient 30 C, collector fluid 40% glycol/water mixture Collector flow rate 3.3 l/min Collector flow rate 4 l/min Collector flow rate 6 l/min Collector flow rate 10 l/min Collector flow rate 25 l/min Inlet temperature, C Fig. 11. Difference between mean solar collector fluid temperature determined by CFD calculations and (T in +T out )/2 for different flow rates and inlet temperatures. Fig. 12 and Fig. 15 show efficiency data points and efficiency expressions for different collector fluid flow rates. The efficiency expressions are obtained by regression based on the data points. From Fig. 12 it is seen that for a low flow rate of 3.3 l/min the use of (T in +T out )/2 as the mean solar collector fluid temperature results in a too low efficiency, especially at high temperature levels. Yearly thermal performance of the HTU collector is calculated with the two efficiency expressions: one is determined by the approximated mean solar collector fluid temperature (T in +T out )/2; the other is determined by the real solar collector fluid mean temperature. Collector efficiency, [-] Mean fluid temperature T m =(T inlet +T oulet )/2 η = *(T m -T a )/G *(T m -T a ) 2 /G Real mean fluid temperature η = *(T m -T a )/G Flow rate 3.3 l/min, solar irrdiance 1000 W/m 2, T ambient 30 C (T m -T a )/G, [Km 2 /W] Fig. 12. Efficiency expression based on T mean as the real mean solar collector fluid temperature and as (T in +T out )/2 for a flow rate of 3.3 l/min. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

143 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 13 of Fig. 13 shows yearly thermal performance of the HTU collector as a function of a constant mean solar collector fluid temperature throughout the year. It can be seen that the thermal performance of the collector will be underestimated if the collector efficiency is determined by the approximation (T in +T out )/2, especially for high solar collector mean temperatures. Fig. 14 shows the relative thermal performance of the two HTU solar collectors as presented in Fig. 13. The underestimation of yearly thermal performance by the approximation method to determine collector fluid mean temperature is varying from 1% to 49% for a mean solar collector fluid temperature between 10 and 100 C. Further, if the difference between the real mean solar collector fluid temperature and the approximate mean solar collector fluid temperature, (T in +T out )/2 is high, it is not likely that the normal used equation for the collector efficiency for all weather conditions is suitable for determining the collector efficiency. Therefore the test method is not suitable for low flow rates Mean fluid temperature Real mean fluid temperature T m =(T inlet +T oulet )/2 Yearly thermal performance, [kwh/(year*m 2 )] Mean solar collector fluid temperature, [ C] Fig. 13 Yearly thermal performance of the HTU collector for efficiency expressions determined in different ways. The calculations are based on the weather data from the Danish Design Reference Year Mean fluid temperature Real fluid temperature T m =(T inlet +T oulet )/ Performance ratio, [-] Solar collector fluid mean temperature, [ C] Fig. 14. Relative thermal performance of the HTU collector for efficiency expressions determined in different ways. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

144 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 14 of From fig. 15, it is seen that the error caused by the approximate equation is of no importance if the flow rate is 6 l/min. It can be concluded that the normally used approximation T ( T + T ) 2 m = only will result in a wrong efficiency expression if the flow rate is lower than 6 l/min, corresponding to 0.48 l/min per m² collector. For low flow rates the approximation will result in too low efficiencies especially at high temperature levels. In the test method a recommended flow rate of 1.2 l/min per m² collector is normally used. Consequently, the approximation does not require any changes of the test method. Changes will only be needed if low flow rates will be used in future test methods. in out Collector efficiency, [-] Flow rate 6.0 l/min, solar irrdiance 1000 W/m 2, T ambient 30 C Mean fluid temperature T m =(T inlet +T oulet )/2 η = 0, *(T m -T a )/G Real mean fluid temperature η = 0, *(T m -T a )/G (T m -T a )/G, [Km 2 /W] Fig. 15. Efficiency expression based on T mean as the real mean solar collector fluid temperature and as (T in +T out )/2 for a flow rate of 6.0 l/min. 5. Influence of the method to determine the specific heat of the solar collector fluid The specific heat of a 40 % propylene glycol/water mixture is determined by /Fur97/: C p ( T ) = * T * T², [J/kg *K] where T is the fluid temperature, [ C] Toutlet The power produced by the solar collector is found by: Q 1 = ρ * v * C p( T ) dt, [W] Tinlet where ρ is the density of the fluid at the temperature of the fluid in the flow meter, [kg/m 3 ] and v is the volume flow rate, [m 3 /s] The test method makes use of the following approximate equation by calculation of the power from the solar collector: Q 2 = ρ * v * C p,, [W] T where C, is the specific heat of the fluid at the temperature (T in +T out )/2, [J/kg *K] p T Table 2 shows measured collector efficiencies found by both methods for different temperature levels and flow rates. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

145 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 15 of Table 2. Collector efficiencies calculated by two different methods to determine the specific heat of the solar collector fluid for the HTU collector. Flow T inlet T outlet T mean Q 2 η 2 Q 1 η 1 rate l/min C C C W % W % From the table it is concluded that the error introduced by the approximate method used to determine the specific heat of the solar collector fluid is insignificant. 6. Influence of the temperature levels used in the test Thermal measurements have been carried out to investigate how the temperature levels used in the test will influence the efficiency expression. Fig. 16 shows results of the measurements with the HTU collector and a similar collector, type HT which includes a Teflon foil between the absorber and the cover glass. Five temperature levels (mean solar collector fluid temperature) are used in the test: Group 1 (29 C), group 2 (45 C), group 3 (64 C), group 4 (84 C) and group 5 (94 C). The collector fluid volume flow rate is 25.0 l/min, corresponding to 2.0 l/min per m 2 collector. The full curves show efficiency expressions of the HT and HTU solar collectors determined by means of regression analysis based on the temperature levels 1, 2, 3, 4, while the dashed curves show the efficiency expressions determined based on the temperature levels 1, 2, 3, 5. It can be seen that the efficiency expression determined by the high temperature levels has a higher heat loss coefficient than that determined by the temperature levels 1, 2, 3, 4. This is due to the fact that there is a sharp decrease of the measured collector efficiency at the high temperature level 5, caused by boiling in one of the strips. The boiling is only discovered by means of temperature sensors placed inside the solar collector. Efficiency [%] Solar radiation 975 W/m 2 Regression HT based on group 1, 2, 3, 4 Regression HTU based on group 1, 2, 3, 4 Regression HT based on group 1, 2, 3, 5 Regression HTU based on group 1, 2, 3, 5 Measurement HT Measurement HTU 5 Regression based on group 1, 2, 3, 4 Regression based on group 1, 2, 3, 5 HT η = *(T m -T a )/G HTU η = *(T m -T a )/G *(T m -T a ) 2 /G HT η = *(T m -T a )/G *(T m -T a ) 2 /G HTU η = *(T m -T a )/G *(T m -T a ) 2 /G (T m -T a )/G, [Km 2 /W] Fig. 16. Efficiency expressions of the HT and HTU solar collectors determined by tests at different temperature levels. Yearly thermal performance of the HT collector is calculated with the two efficiency expressions: One determined based on the temperature level 1, 2, 3, 4 and one determined based on the NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

146 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 16 of temperature levels 1, 2, 3, 5. The collector is facing south and the collector tilt is 40. The calculations are based on the weather data from the Danish Design Reference Year. Fig. 17 shows yearly thermal performance of the HT collector as a function of a constant mean solar collector fluid temperature throughout the year. The yearly thermal performance calculated with the efficiency expression determined by 1, 2, 3, 5 is lower than that with the efficiency expression determined by temperature levels 1, 2, 3, 4 except for a mean solar collector fluid temperature less than 10 C. Fig. 18 shows the relative thermal performance of the HT solar collector defined as the ratio of the yearly thermal performance of the solar collector with an efficiency determined at the temperature levels in question during the collector tests and the yearly thermal performance of the solar collector with an efficiency determined at the temperature levels 1, 2, 3 and 4 during the collector tests. It can be seen that the HT collector with efficiency expression determined at the high temperature levels will reduce thermal performance by 5% - 15% comparable to the efficiency expression determined at temperature levels 1, 2, 3 and 4 for a mean solar collector fluid temperature from 45 C to 84 C Yearly thermal performance, [kwh/(year*m 2 )] Based on efficiency determined by group 1, 2, 3, 4 Based on efficiency determined by group 1, 2, 3, Mean solar collector fluid temperature, [ C] Fig. 17. Yearly thermal performance of the HT collector for efficiency expressions determined at different temperature levels Relative performance, [-] Based on efficiency determined by group 1, 2, 3, 4 Based on efficiency determined by group 1, 2, 3, Mean solar collector fluid temperature, [ C] Fig. 18. Relative thermal performance of the HT collector for efficiency expressions determined at different temperature levels. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

147 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 17 of Fig. 19 shows yearly thermal performance of the HTU collector as a function of a constant mean solar collector fluid temperature throughout the year. The yearly thermal performance calculated with the efficiency expression determined by 1, 2, 3, 5 is lower than that with the efficiency expression determined by temperature levels 1, 2, 3, 4 except for a mean solar collector fluid temperature between 15 and 55 C. Fig. 20 shows the relative thermal performance of the HTU solar collector defined as the ratio of the yearly thermal performance of the solar collector with an efficiency determined at the temperature levels in question during the collector tests and the yearly thermal performance of the solar collector with an efficiency determined at the temperature levels 1, 2, 3 and 4 during the collector tests. It can be seen that the HTU collector with efficiency expression determined at the high temperature levels will reduce thermal performance by up to 7% comparable to the efficiency expression determined at temperature levels 1, 2, 3 and 4 for a mean solar collector fluid temperature from 55 C to 84 C. It can be concluded that for a solar collector operating with a solar collector fluid temperature no larger than 84 C, the collector efficiency and the thermal performance of the collector will be underestimated by up to 15% for HT collector and up to 7% for HTU collector if a high temperature level like 94 C is used in the test to determine the collector efficiency. It is therefore recommended that the maximum temperature level used in the tests is not higher than the maximum operation temperature of the collector Yearly thermal performance, [kwh/(year*m 2 )] Based on efficiency determined by group 1, 2, 3, 4 Based on efficiency determined by group 1, 2, 3, Mean solar collector fluid temperature, [ C] Fig. 19. Yearly thermal performance of the HTU collector for efficiency expressions determined at different temperature levels. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

148 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 18 of Relative performance, [-] Based on efficiency determined by group 1, 2, 3, 4 Based on efficiency determined by group 1, 2, 3, Mean solar collector fluid temperature, [ C] Fig. 20. Relative thermal performance of the HTU collector for efficiency expressions determined at different temperature levels. 7. Influence of the weather conditions used in the test Calculations with the program SOLEFF /Ras96/ are carried out in order to determine the collector efficiency for the HTU collector and the HT collector for different weather conditions. A collector fluid volume flow rate of 25.0 l/min, corresponding to 2.0 l/min per m 2 collector, is used in the calculations. Fig. 21 shows the calculated efficiencies for the HTU collector for different ambient air temperatures and sky temperatures. The sky temperature is assumed to be 15 K lower than the air temperature. It is seen that the efficiency for high temperature levels is decreased for increased temperature levels of the ambient air and the sky. Fig. 22 shows the calculated yearly thermal performance of the HTU collector without Teflon foil as a function of a constant mean solar collector fluid temperature throughout the year. The calculations are carried out with weather data from the Danish Design Reference Year based on two efficiency expressions: One determined by means of collector tests with an ambient air temperature of 15 C and a sky temperature of 0 C and one determined by means of collector tests with an ambient air temperature of 30 C and a sky temperature of 15 C. Fig. 23 shows the calculated relative performance of the HTU collector as a function of the mean solar collector fluid temperature. The relative performance is the ratio between the yearly thermal performance of the collector determined with an efficiency expression determined at the ambient air/sky temperatures in question and the yearly thermal performance of the collector determined with an efficiency expression tested at an ambient air temperature of 15 C and a sky temperature of 0 C. For solar collector fluid temperature levels between 15 C and 85 C the difference between the yearly thermal performances of the collector determined by means of the tests at the different weather conditions is lower than 3%. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

149 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 19 of 90 Tambient=15 C, Tsky=0 C η = 0, *(T m -T a )/G *(T m -T a ) 2 /G 80 Tambient=30 C, Tsky=15 C η = 0, *(T m -T a )/G *(T m -T a ) 2 /G 70 Efficiency, [%] HTU collector without Teflon G = 800 W/m (T m -T a )/G, [Km 2 /W] Fig. 21. Calculated efficiency for the HTU collector for different ambient air and sky temperatures for a solar irradiance of 800 W/m² and an incidence angle of Yearly thermal performance, [kwh / (year * m 2 )] HTU collector without Teflon Tambient=15 C, Tsky=0 C Tambient=30 C, Tsky=15 C Mean solar collector fluid temperature, [ C] Fig. 22. Yearly thermal performance of the HTU collector for efficiency expressions determined with two different ambient and sky temperatures. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

150 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 20 of Relative performance, [-] HTU collector without Teflon Tambient=15 C, Tsky=0 C Tambient=30 C, Tsky=15 C Mean solar collector fluid temperature, [ C] Fig. 23. Relative thermal performance of the HTU collector at different mean solar collector fluid temperatures. The calculations are carried out with the HT solar collector as well. It is shown that for solar collector fluid temperature levels between 15 C and 100 C the difference between the yearly thermal performances of this collector determined by means of the tests at the different weather conditions is lower than 4%. Consequently, taken the measuring accuracy into consideration, the weather conditions used in the tests will not significantly influence the calculated thermal performance of the collector. 8. Conclusion Investigations on the suitability of collector test methods have been carried out for two flat plate collectors from Arcon Solvarme A/S: The HTU collector without a Teflon layer and the HT collector with a Teflon layer. The flow and temperature distribution through the absorber tubes in a HTU solar collector panel is investigated experimentally and theoretically. Results show that the flow and temperature are not uniformly distributed. The uneven distribution is controlled by two driving forces: The buoyancy force and the frictional drag force. For low flow rates (<6.0 l/min, corresponding to 0.48 l/min per m 2 solar collector), the buoyancy effect is significant. The flow and temperature distribution get worse with the decrease of collector volume flow rate, resulting in a decreased collector efficiency. For higher flow rates (> 10.0 l/min, corresponding to 0.80 l/min per m 2 solar collector), the friction drag force tend to be more dominant, the flow distribution get worse with the increase of collector fluid flow rate. This will result in a decreased collector efficiency as long as the flow in the tubes is laminar. It is concluded that with a volume flow rate between 6.0 l/min and 10.0 l/min the collector will have the best efficiency. For the HT solar collector operating at a collector fluid volume flow rate of 25.0 l/min (corresponding to 2.0 l/min per m 2 solar collector) and at solar collector fluid temperatures no larger than 84 C, the thermal performance of the collector will be underestimated by up to 15% for HT collector and up to 7% for HTU collector if a high temperature level like 94 C is used in the test to determine the collector efficiency. It is therefore recommended that the maximum temperature level used in the tests is not higher than the maximum operation temperature of the collector. Further, if the solar collector efficiency for low flow rates is measured in future test methods and if the collector fluid volume flow rate is less than 0.48 l/min per m 2 solar collector, there is a need to change the test method. The weather conditions like the ambient air NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

151 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR JAN page 21 of temperature and the sky temperature used in the tests will not significantly influence the calculated thermal performance of the collector. The error introduced by the approximate method used to determine the specific heat of the solar collector fluid is insignificant. References /Duf91/ /Jon94/ /Wan90/ /Wei02/ /Fan05/ /Eur04/ Duffie J. A. and Beckman W. A. (1991): Solar Engineering of Thermal Processes, 2nd edition, John Wiley & Sons, New York, pp Jones G. F., Lior N. L. (1994): Flow distribution in manifolded solar collectors with negligible buoyancy effects, Solar Energy 52 (3), Wang X. A., Wu L. G. (1990): Analysis and performance of flat-plate solar collector arrays, Solar Energy 45 (2), Weitbrecht V., Lehmann D., Richter A. (2002): Flow distribution in solar collectors with laminar flow conditions, Solar Energy 73 (6), Fan J., Shah L. J., Furbo S. (2005): Flow distribution in a solar collector panel with horizontal fins, Proceedings of ISES 2005, Orlando, USA, European Committee for Standardization, (2004): Thermal solar systems and components Solar collectors Part 2: Test methods, EN /Fur97/ /Flu03/ /Ras96/ Furbo S. (1997): Varmelagre til solvarmeanlæg, Institut for Bygninger og Energi, Technical University of Denmark, pp Fluent Inc. (2003): Fluent release 6.1, 10 Cavendish Court, Lebanon, NH USA. Rasmussen P. B. Svendsen S. (1996): SolEff Program til beregning af solfangeres effektivitet. Brugervejledning og generel programdokumentation, Thermal Insulation Laboratory, Technical University of Denmark. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

152 (4) New Generation of Solar Thermal Systems Calculation of collector energy output: Inquiry results and proposal for a common procedure A common European procedure for calculation of the (annual) energy output from solar thermal collectors tested according to EN has been discussed in the Solar Keymark II and NEGST projects. This inquiry gives an overview of the different methods currently used in Europe (and overseas) for this purpose and thus form a basis for a proposal on a common procedure. Calculation of the collector energy output is in the first place interesting if the results are connected to the subsidies that may be at hand, but also in the absence of any subsidies, collector energy output will serve a purpose as consumer information. The questionnaire was distributed to all main markets within Europe, to Australia and to the USA. The following countries provided some input: Sweden Portugal Germany Denmark Austria The Netherlands Spain (Canary Islands) UK Greece Australia USA Summary of results From the European countries that answered the questionnaire only Sweden and Germany seem to have nationwide subsidy schemes that are somehow based on collector output figures. In Sweden the subsidy is directly related to the kwh output whereas in Germany the energy output has to reach a predefined level in order to qualify for subsidies. In Portugal, system output calculated with a tool developed by INETI is used as a requirement for subsidies as well as it is used as input to EPBD calculations. In France, collector- and system output is presently used side by side in connection to subsidies but the movement is towards system output. All respondents but The Netherlands (TNO) was in favour of a common European calculation method for collector (or system) energy output based on test data from individual tests, but also

153 (2) TNO sees the need for a simple calculation procedure in order to assess larger systems in relation to the EPBD. In the following, the inputs from all countries are summarised. Sweden have a subsidy scheme for most solar thermal systems running since The subsidy is based on collector energy output calculated by means of a collector model in the Minsun or Trnsys softwares. The calculations are made for constant collector average temperature and are based on a Stockholm 1986 climate and collector test data according to EN Portugal: System output calculated with the INETI-developed software Solterm is the basis for governmental subsidies as well as for performance figures used in the framework of the new building regulations, the Thermal Performance Building Code. The calculations use specific weather data for each municipality in Portugal. Portugal considers that the method referred as Collector Based Methodology, Col in the document distributed by Uwe Brechlin (18 May 200&) and prepared By Jan Erik Nielsen, is a good approach. Germany uses a common approach for output calculations based on a TRNSYS system model and Wurzburg climate data. Collector output from a (reference-) system that fulfils fsol=0,4 is reported and compared to the limit of 525 kwh/m 2 annually. This method is used since 1996 and there are no plans for changes at the moment. Denmark have no subsidies for solar heating installations for the time being but are ahead in the development of solar energy calculations related to the implementation of the EPBD, where the solar savings are calculated using a method based on the draft pren standard. Collectors approved in the voluntary national product certification scheme are also calculated according to the same procedure, using a reference DHW system. A Danish design reference year are used in the calculations that were established in Austria: In connection to subsidies there is no method to calculate the energy output of a collector/system applied at the moment in Austria. For the national implementation of the EPBD there is a different method (OIB Leitfaden) proposed which is currently in the approval procedure. It is expected to be implemented in Currently this method will not be linked to any subsidies but might be in future in combination with the EPBD. The Austrian label Umweltzeichen, in operation since 2003, include a simple f-chart simulation wich is used to determine the energy payback of the collector. Australia suggests the German RAL-UZ 73 could be a basis for an European standard since it is disseminated and accepted. In the Netherlands there are no national subsidies for the time being. Performance figures focus on system output rather than collector output. DHW system output is based on EN 2976 tests and combisystem output is based on DC tests. The former can be converted into a figure used in declarations according to the EPBD. TRY De Bilt is used for all calculations. The Netherlands are in favour of system output calculations, mainly to be used as input to the EPBD building assessments. Suggests a simple method, with energy yields on the lower mean level compared to expectations, inviting system innovations is suggested as a common European approach.

154 (2) In Spain most of the subsidies directed to solar thermal energy (national level) were given by IDAE (National energy agency) until Since 2006 this competence has been transferred to regional governments which will manage the subsidies in the future. No national, but some regional subsidies are reported to be based on system output in Supports a common European calculation method and suggest to use the base of a well known software as Trnsys. The UK have no subsidies for solar heating installations based on collector- or system performance, but are calculating the solar system performance for use in the EPBD building assessments. These solar savings are calculated using a method with some correspondence to the draft pren standard. UK suggest to use EN results for collector and pr CEN TS for system. In Greece a subsidy scheme exist for large solar heating systems and the subsidies are then based on the collector efficiency at a specified point of operation. The option of system output based subsidies is currently going on in Greece. Australia: Uses a Renewable Energy Credits (REC) scheme for solar water heaters. Based on tests on collector and store (heat loss), 4 Australian climates and TRNSYS system simulations. Used since The present standard (AS4234 ) is currently under revision. This process is being drafted to be ISO Using TRNSYS has lead the industry to innovate and develop considerably better products. The detailed modelling approach used in Australia has resulted in significant innovation in the industry. Australia express a need for a consistent approach. It would be best in a standard rather than in the text of some regulations. USA: SRCC rating and certification, in operation since many years, is the basis for federal tax credits. There are also various incentives in individual states. The SRCC rating gives the collector output for three different type days and five different operating conditions. Motives for a collector based output calculation Main assumption for discussing a common method is that we re dealing with the collector as a component in this case. Subsidies in some European countries are based on collector output and in several other countries test data converted into annual output figures are used for rating collectors. It is assumed that a common procedure for these calculations will increase the competition on the market and contribute to an open market for collectors and systems. The objective of WP 5 to improve the Keymark scheme for collectors in order to make it more attractive to both industry and authorities would thus be partly achieved with the establishment of a common calculation procedure for collectors. Furthermore, common methods for calculating system output already exist in the EN and in prcen TS The Keymark II project does not have to have an opinion on weather system- or collector output is the most desirable basis for subsidies. Even if state of the art would remain unchanged or develop towards more of system performance based subsidies, a common method for collector energy output would still be justified. Desirable features of a common method for collector energy output:

155 (2) -Should be part of a standard (EN 12975) as an informative annex when ready -Easy to perform but enough sophisticated to take into account specific features of most collectors in the market. -Based on weather data from 3-4 reference locations (EN 12976) -Relating to the standard for EPBD calculations pren Relating to the procedure for m 2 to kwh conversion and IEA world statistics. Proposal 1 Method according to the German procedure but based on a 3-4 European climates whereof one (Wurzburg) will be the base case. Proposal 2 Method according to the principles outlined in the pren but with the model further developed in order to take into account e.g. biaxial incidence angle dependencies and second order heat losses. Calculate collector output for 3-4 locations and for three constant inlet temperatures. Arguments are in favour of proposal 1 as it is a well established method, in operation since several years and due to the fact that it already covers a large part of the European market.

156 New Generation of Solar Thermal Systems CONTENTS INTRODUCTION How the work for WP 4.1 was defined, the focus areas that were chosen and how the outputs are intended to be used. COLLECTOR COMPONENTS Durability of absorber coatings, reflector materials and anti reflective coatings. Ongoing work and proposed standards. QUALITY TESTING OF ETC:S A review of the present EN from the perspective of testing ETCs. EXPOSURE TESTING OF COLLECTORS A review of two approaches to exposure testing and a proposal for an alternative method in the present EN PERFORMANCE TESTING OF FLAT PLATE COLLECTORS Research on the accuracy of performance test results due to different boundary conditions and approximations in EN PERFORMANCE TESTING OF UNGLAZED SOLAR COLLECTORS Proposals for revision of EN PERFORMANCE TESTING OF SOLAR AIR COLLECTORS A draft standard is presented WP 4.1 D2, D3 Future standards for advanced collectors Dissemination level: Public Authors : Peter Kovács SP, Josef Buchinger and Dieter Gottwald arsenal research, Simon Furbo DTU, Stephan Fisher ITW, Bouzid Khebchache CSTB SUMMARY Reviewer: Maria Joao Carvalho, INETI April 2007 This report presents an overview of the work carried out in WP 4.1 Future standards for advanced collectors. A number of different inputs and activities are referred and those developed within the project are attached as annexes to this report. The work is in most parts providing inputs to ongoing and future work of CEN TC 312 Thermal solar systems and components. Furthermore recommendations for research work needed in order to pave the way for new standards are presented. The work started with a survey among research institutions and manufacturers active in the solar thermal field, to identify items of interest for future standards development. Based on the priorities that this survey revealed, a new work item has been proposed to CEN TC 312/ WG1. As a result, the upcoming revision of the EN will deal with the durability of collector components, methods for quality testing of evacuated tubular collectors and with improved exposure testing of collectors as well as with performance testing of unglazed collectors and calculation of annual collector energy output. These items have been further elaborated within the NEGST project and the result is summarized in the following. In addition, a draft standard for performance testing of solar air collectors and research work carried out to check the validity of boundary conditions and approximations used in performance testing according to EN is presented. DETERMINATION OF IAM AND APPLICATION OF PRESENT STANDARD TO TRACKING AND CONCENTRATING COLLECTORS A resource document for the application of EN GENERAL CONSIDERATIONS FOR ALL TYPES OF COLLECTORS Miscellaneous items of interest for an improved usefulness of EN ANNEXES REFERENCES

157 WP 4.1: ADVANCED COLLECTORS - 25/06/ Introduction This report summarizes the work related to future standards for advanced collectors and collector materials carried out in the NEGST project. The work started with an inquiry that was addressed to industries and research institutes in Europe in which the respondents were asked to prioritize among and comment on different potential working areas within this field. The inquiry revealed a particular interest in improved exposure tests for collectors, m 2 to energy conversion 1, improved characterization of incidence angle dependencies and methods for accelerated testing and determination of optical properties for absorber- and reflector materials. Results of the inquiry and a proposal of documents and procedures that was developed based on these are presented in Annex 1. The respective inputs to this work have been planned so as to integrate ongoing research with the interests indicated in this inquiry. Nevertheless other inputs, receiving lower interest in the inquiry have also been input here as to provide an up to date picture of the status in this field. Furthermore some items had to be dropped even though they attained a lot of interest, as the resources of the project did not allow for any major research work to be undertaken. As the solar energy field is growing rapidly at present, the conditions and requirements for standardisation are also quickly changing. It is therefore reasonable to assume that some of the products and techniques that were of no interest yesterday will be on top of the list tomorrow. With regard to the ongoing work in CEN TC 312, "Thermal solar systems and components", the work presented here have it s focus pointed either at the present standard for solar thermal collectors, EN /CEN06a/, or on new standards to be developed covering related products or sub components. In this respect, the work can be divided in three categories 1. Inputs that are directly proposing revisions of EN Inputs that are intended as resource documents either to be referenced in EN as a support to it s interpretation and practice or to be used as drafts for future standards 3. Inputs that are pointing out further work needed in order to improve the efficiency, accuracy or relevance of the present standard As extensive discussion, editing and redrafting will always be the case before any of this work comes in print in a revised or new standard; it should not be a major drawback that the different inputs are not strictly belonging to one or the other category. On the contrary, they more often comprise more than one of the categories 1-3. As a direct result of this work, and based on the priorities that this survey revealed, a new work item has been proposed to CEN TC 312/ WG1 /CEN06b/. Thus, in the upcoming revision of EN 12975, the following topics will be considered: a) clarification of the application of the present standards to tracking and/or concentrating collectors, b) unglazed collectors: refined performance test conditions and prediction, improved sky temperature measurement, c) collector components - requirements and test methods, d) quality tests for evacuated tubes, e) improved exposure - accelerated ageing test of collectors, f) annual collector energy output. 1 In order to make solar thermal energy more visible in energy statistics and promotional contexts it has been decided to develop procedures for conversion of figures of installed square meters of collectors into peak power and annual energy yields NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

158 WP 4.1: ADVANCED COLLECTORS - 25/06/ Collector components Here, the state of the art in durability testing of solar absorber coatings, anti reflective coatings, reflector materials and polymer components of solar collectors is presented. One method dealing with absorber surfaces developed within the IEA framework is considered ready for inclusion in an EN standard whereas for the other items, new drafts are still under development. 2.1 Quality tests for solar absorbers The absorber itself, in the collector, is directly or indirectly subjected to a number of tests in the present standard for collector testing EN Requirements for reliability are also defined. However, for the long term durability of the absorber or more specifically, the absorber coating, there are no requirements. Considering the rapid and continuously ongoing development of new materials, coatings etc. and the increasing specialization among manufacturers, it is assumed that manufacturers of absorbers could benefit from methods that can predict a long service life. Standardised methods and requirements would also benefit their clients, the collector manufacturers, who would then be able to strengthen quality requirements on their suppliers. In the case of absorber coatings, such methods do exist. At present they are referred to only in the reference list of EN (as ref. 15). This reference points at a report from IEA SH&C Annex 10 that describes methodology and requirements to assess the lifespan and optical properties of selective absorber coatings. This work has recently been reviewed and updated within the framework of IEA SH&C Annex 27 Performance of solar façade components. The new document: Recommended qualification test procedure for absorber surface durability /CAR04/ describes tests applicable to organic and inorganic coatings. The tests are designed so as to predict a 25 year life of the coating, which in this context is equal to a reduction in solar fraction of 5% (relative) in a solar water heating system resulting from the deteriorated properties of the surface. In addition, the document provides methods for determining the optical (absorptance and emittance) and mechanical (adhesivity) properties of the surface coating which can be useful in determining the initial characteristics as well. It is recommended to CEN TC 312 that the described document is either summarized in a short (informative) annex of EN and included among the references or further developed into a draft standard. In the latter case it should be included among normative references of Quality tests for polymers in solar collectors Polymer materials have so far only been used to a limited extent in solar thermal applications. In low temperature applications such as pool heating the introduction has been very successful and in general without problems related to the materials. On the contrary, in medium and high temperature applications where polymers were tried to replace inorganic materials, it has in general failed. As polymers definitely have many potential advantages to offer in solar thermal applications compared to traditional materials, it will be useful to researchers and manufacturers to have a set of common tools and methods to assess their properties and suitability for more demanding applications. A forum for development of such tools is the recently initiated Annex 39 within the IEA SH&C program; Polymeric materials for solar thermal applications. This work is still in the starting phase but it can provide an extensive input to this field and it can thus be expected to bring about a new interest from major industrial players. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

159 WP 4.1: ADVANCED COLLECTORS - 25/06/2007 One starting point for this work, and for related work in CEN TC 312 focusing at testing and evaluation of polymer materials in collectors could be a Swedish method and requirement specification /Spi04/ developed in parallel to the assessment of an innovative medium temperature all-polymer collector in The specification is presented in Annex Quality testing of reflector materials and anti reflective coatings Reflectors and anti reflective coatings of cover materials are being used more and more frequently in different collector- and PV-designs as a cost efficient way of improving the performance. For the same reasons as for polymers, it would be desirable if methods to assess the life expectancy of these components were available. Pre normative work is ongoing in IEA SH&C Annex 27 that should result in some practically applicable methods. At present no standardised methods are available for this purpose. 3 Performance and quality testing of Evacuated tubular collectors ETCs Most of the work related to development of test methods and quality criteria for solar collectors has been done with the flat plate collector as the reference. Only to a minor extent has the ETC and it s specific properties been addressed. In order to have ETCs contribute significantly to the large scale introduction of solar heating technology in Europe, quality and performance should be determined in tests that are also taking the specific characteristics of ETCs into account. How this can be done is outlined in Annex 3 and briefly summarized in the following. It is recommended that CEN TC 312/WG1 considers including these proposed changes in the present standard. 3.1 Background- Inquiry on ETC testing ETC collectors today are synonumous to a remarkable development of the chinese solar thermal market. In ten years their market shares have grown from 35 to 85 %. The total annual sales of collectors are around 15 million m 2 and growing by an annual 30%. In Europe ETCs have not been the same success so far but their shares are increasing. In 2005 an inquiry about ETC testing and quality aspects was performed within the NEGST project among in all 15 institutes, manufacturers and importers. The purpose of the inquiry was to give a background to an assessment of the need for revised test procedures regarding performance- and quality testing of ETC:s. The inquiry turned out to raise a number of new questions. However as a result, a number of proposals were developed from the inquiry inputs. 3.3 Quality testing ETCs Focusing to a large extent on the Dewar type ETC with heat pipes, some new tests and/ or revisions of the present quality tests are proposed: A revised version of the exposure test, possibly with efficiency measurements carried out on the same collector before and after the exposure (This would of course apply to all types of collectors, not only ETCs) Mechanical load tests that are not relevant to ETCs should be removed but in the case of external reflectors, they need to be further developed Assessment of the durability of reflector materials could be useful to manufacturers as well as to consumers NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

160 WP 4.1: ADVANCED COLLECTORS - 25/06/2007 The impact resistance test of EN should be reviewed and revised New methods for freeze testing of heat pipes and for testing the durability of the glass- metal seal in single glass ETCs are proposed 3.4 Performance testing of ETCs Some issues related to the performance testing of ETCs were also highlighted in the inquiry. The determination of the heat transfer capacity rate according to EN is not unambiguous and need to be revised. The same goes for the determination of stagnation temperature. Furthermore some suggestions regarding definitions and reporting related to ETCs are elaborated. 4 Exposure testing of solar collectors As already mentioned regarding ETCs, a new approach for exposure testing, which would of course also be applicable to other collector types, has been proposed. The background is that the present European exposure test has been under a lot of debate, mainly due to it s inability to maintain uniform test conditions when applied in different parts of Europe. As CSTB in France since long have had stricter exposure tests than what is required by EN 12975, they have carried out some investigations related to this issue. An FMEA approach has been suggested as a complement to the (one year) outdoor exposure test. This is described in further detail in Annex 4. Adopting the french method with a one year exposure test was discussed in NEGST as a way to deal with the shortcomings of the present method. It was then concluded that such a long lasting test would not be accepted by most manufacturers and furthermore, it would not solve the basic problem of unreproducible test conditions in different locations. Instead, an Australian method for exposure testing has been proposed as a starting point for a new European method. In short, this method is carried out by exposing the collector to a fixed, maximum stagnation temperature for ten twelwe hour periods during ten days. A method similar to this could be an improvement from the present methodology in two or three different ways: The same test conditions could be achieved irrespective of where in Europe the test was carried out The time required for testing could be reduced compared to the present test Depending on how the specific test conditions are designed, the test could despite it s shorter duration be harder and thus more likely to reveal any failures compared to today s tests, at least in countries with less favourable irradiance conditions Apart from australian test laboratories, there is no recorded experience from the application of this method. CSTB is about to do their first tries with it during In order to be able to propose a method based on the principles of the Australian method to CEN TC 312 more experience of it s practical application is needed. European laboratories are therefore encouraged to initiate tests accordingly. The Solar Keymark II project offers the best platform for planning such work. The Australian method is described in Annex 5. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

161 WP 4.1: ADVANCED COLLECTORS - 25/06/ Performance testing of flat plate collectors Even though performance tests for collectors has been worked out on the basis of the flat plate collector s characteristics it is not obvious that it gives full justice to a certain collector under all allowable test conditions. Moreover, some more extreme operating conditions that may result from new control strategies might result in performance deviations compared to results achieved during standardized test conditions. To bring more light to this issue, an investigation was carried out at DTU in Denmark where measurements and CFD (Computational Fluid Dynamics) simulations were performed on two different large (12.5 m²) flat plate solar collectors. A general conclusion is that the present test conditions do not need to be changed as long as the measured efficiencies are not to be representative for operating conditions (flow rates) below approximately 0,5 l/m 2 /min. If that is to change however, also the test conditions may need to be modified. This should in any case first need to be confirmed by furter experiments on other collector types. The investigations carried out at DTU focused on five different characteristics of the test method: Influence on the efficiency of the volume flow rate during tests (sometimes resulting in nonuniform flow distribution in the collector) Influence on the efficiency by the method to determine the mean collector fluid temperature Influence on the efficiency by the approximate method to determine the specific heat of the heat transfer fluid (water/ glycol) Influence on the efficiency by the temperature levels used in the tests Influence on the efficiency by the weather conditions during the test Based on the investigations, it is concluded that for this collector type, the efficiency will vary with the flow rate due to non uniform temperature- and flow distribution in the absorber. However considering normal operating conditions, the effect is not considered to be large enough to suggest changed test conditions for the standard. This conclusion is also valid regarding the present method of determining collector mean temperature. It is furthermore concluded that the error introduced by the approximate method used to calculate heat transfer fluid specific heat is so small that it can be ignored. Due to the fact that boiling can occur in parts of collectors of this type at high fluid temperatures it is recommended not to exceed the maximum intended operating temperature during the tests. Regarding the influence of different weather conditions during the tests it was concluded that these will not significantly influence the calculated thermal performance of the collector. The DTU report is appended in Annex 6. 6 Performance testing of unglazed solar collectors Unglazed collectors are likely to have a growing market ahead in Europe as sales are still far from their successful implementation in the USA and the use of swimming pools is steadily increasing. Apart from the basic pool heating application, combinations of these collectors and heat pumps have also been presented as a feasible concept. In the latter case it is likely that the collector will operate at temperatures below the dew point and thus it will have a significant part of is energy production coming from condensation heat. This is not taken into account in the present test standard EN Introducing a new correction term in the basic quasi dynamic NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

162 WP 4.1: ADVANCED COLLECTORS - 25/06/2007 collector equation and including relative humidity in the measurements during performance testing would probably be enough to take this effect into account. However, so far this is not known to have been verified anywhere. With regard to the present methods for testing unglazed collectors, arsenal research have proposed some changes in the standard, which are mainly aimed at improving the accuracy of efficiency tests on unglazed collectors. The proposed changes are: Lowered limits for allowed long wave irradiation during indoor efficiency tests on unglazed collectors Removal of correction for long wave irradiation when collector is inclined Introduction of an alternative method for measuring the sky temperature indoors Changing the ratio ε/α for unglazed collectors whose properties are not known. Further research on the issue of testing and modelling the utilization of condensation heat in unglazed collectors is recommended. It is also recommended that CEN TC 312/WG1 considers including these proposed changes in the present standard. The changes proposed by arsenal research are summarized in a report which is appended in Annex 7. 7 Performance testing of solar air collectors Solar air collectors are not wide spread so far. The main obstacles for a wide dissemination appears to be lacking information as well as lack of confidence on how these systems will perform. The same explanation is also valid for most types of solar dryers, where a solar air collector of some kind is the most central component. In developing countries in Africa, Asia and Latin America solar drying is emerging as a simple but promising technology that can bring new income opportunities and significantly reduce the amount of wasted products in agriculture and farming. Testing of the respective components is therefore essential. Such tests should be reproducible and widely acknowledged and this is here proposed to be achieved through the introduction of a new European Standard. So far no such standard exists for air collectors. Compared to liquid collectors, the measuring procedures for solar air collectors need more expenditure. Some standardised testing procedure exists so far: The Italian Standard UNI 8937 (Norma Italiana 1987) The US-American Standard ASHRAE (ASHRAE) The UNI 8937 only gives an idea of how testing of air collectors can be carried out, but does not touch the specific problems of solar air systems. The ASHRAE is already a usable standard but since it describes procedures for testing of both liquid and air collectors it does not take into account all possible variations of solar air collectors and leaves some uncertainties. Starting a standardisation process for testing solar air collectors has also been discussed in the Technical Committee 180 of the International Standardisation Organisation (ISO), but no work has been initiated so far. As arsenal research has been involved in most of the international research that has been carried out within this field in the past ten years and also did a number of laboratory tests on air collectors during this period, the NEGST project offered an excellent opportunity to start drafting a standard and compile recommendations on further work required. A draft-standard for testing NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

163 WP 4.1: ADVANCED COLLECTORS - 25/06/2007 solar air collectors based on existing standards has therefore been developed and is presented in Annex 8. Recommendations for further work to be done on sections of this standard are given in a resource document that is presented in Annex 9. This draft describes procedures for testing reliability and performance of glazed solar thermal air collectors operated either in open or closed loop systems with either positive or negative pressure applied. The focus so far has been laid on the adoption of the sections describing the thermal performance tests, especially steady-state outdoor testing. A workshop on solar air collectors will be organised in Austria during autumn If relevant stakeholders can get together, the next important steps towards working out a European standard for solar air collectors can be discussed. The options for developing this draft into a mature standard should also be discussed in CEN TC Determination of incidence angle modifiers and application of present standards to tracking and concentrating collectors One important parameter that needs to be determined in order to enable an accurate prediction of the collector performance is the incidence angle modifier (IAM). The NEGST project inquiry mentioned in the introduction identified a need for further elaboration of the definitions and the calculation- and test procedures related to this parameter. The resulting resource document elaborated within the NEGST project is attached in Annex 10. It defines existing collector types into one of the three categories isotropic, bi-axial and multi-axial behaviour with respect to the direction of incidence. It also explains how the corresponding IAMs shall be calculated and how the collectors shall be tested when determining the dependency on incidence angle of the thermal performance. The resource document furthermore describes how the present standard EN shall be applied in order to deliver relevant results also for concentrating and tracking collectors. The main conclusion is that testing of tracking and concentrating tracking collectors can be accurately carried out using the procedure and conditions according to EN 12975, section 6.3 under consideration of the guidelines elaborated in the resource document (Annex 10). CEN TC 312/ WG 1 is recommended to integrate the resource document into EN As the application of EN 12975, section 6.3., to tracking and concentrating collectors has not been fully experimentally validated yet it is recommended that performance testing accompanied by performance prediction and long term performance measurements are carried out on at least the following collector types: CPC type collectors with low concentration ratios Tracking concentrating collectors 9 General considerations for all types of collectors For solar thermal collectors using liquid as a cooling media, i.e. collectors being tested on a regular basis according to established standards, some common features are currently under consideration or should be considered in order to optimize the testing and the use of test results. The most crucial of these items are therefore briefly mentioned below. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

164 WP 4.1: ADVANCED COLLECTORS - 25/06/ Conversion of m 2 to power and energy In order to make solar thermal energy more visible in energy statistics and promotional contexts it has been decided to develop procedures for conversion of figures of installed square meters of collectors into peak power and annual energy yields. The conversion to power quickly gained overall acceptance after it was introduced in 2005 and a draft procedure for the conversion to energy is one of the outcomes or NEGST WP Standardized calculation of annual collector energy yield from efficiency test results A standardised procedure for calculation of the annual collector energy yield based on the performance parameters resulting from efficiency tests and a reference climate is now under development in the Solar Keymark II project /Sol07/. The calculation outlined in this procedure, giving the energy yield at three pre determined temperatures, will mainly facilitate performance comparisons for potential buyers. It should ideally also provide a basis for comparison of test results achieved through quasi steady state and quasi dynamic testing respectively. This would make it possible to more accurately determine the limitations of the quasi steady state method and the agreements between these two methods. It is recommended that CEN TC 312/ WG 1 closely follow this development and that the procedure when finalized is included as an annex to EN Adoption of common simulation tools to state of the art in testing Performance testing of solar thermal collectors according to EN paragraph 6.3, the so called quasi dynamic method, is a relatively new approach. To be able to fully benefit from the diversified results that the quasi dynamic test approach gives it is essential that the collector models that are used in the most common simulation programs are compatible with the model used in the standard. As the collector models used in most simulation programs today are not fully compatible with the quasi dynamic model, including multi axial incidence angle modifier, it should be given high priority to develop such models. 9.4 Simplified procedure for determination of stagnation temperature Determining accurate and representative stagnation temperatures as required from the EN quality test can sometimes be a difficult task. As this temperature is mainly used as a guide for selection of materials being part of the construction it should not necessarily have to be determined from very accurate direct measurements. Furthermore it is not obvious that direct measurements in one point on the collector will give an accurate result of the highest temperatures the construction will be exposed to. A way around this problem could be to determine the stagnation temperature based on calculation from the collector model parameters. A safety factor of degrees should then be added as the efficiency is determined under windy conditions and thus it will under estimate the stagnation temperature. The proposed method would save some efforts and costs in testing without significantly reducing, and in some cases even improving, the quality of the results. It is therefore recommended that CEN TC 312/ WG 1 take it into consideration. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

165 WP 4.1: ADVANCED COLLECTORS - 25/06/2007 Annexes Annex 1 Compiled responses WP4.1 Collectors questionnaire Annex 2 Polymeric materials in solar collectors- Test methods and technical requirements Annex 3 Performance and quality testing of Evacuated tubular collectors Annex 4 Accelerated ageing tests of solar collectors Annex 5 Australian exposure test methods Annex 6 Flat plate collectors Annex 7 Unglazed collectors Annex 8 Performance testing of solar air collectors-draft standard Annex 9 Recommendations on testing of solar air collectors Annex 10 Incidence angle modifier measurements and application of EN to tracking and concentrating collectors References /Car04/ /CEN06a/ /CEN06b/ /Sol07/ /Spi04/ Carlsson B (2004) Recommended qualification test procedure for absorber surface durability. IEA Solar Heating and Cooling Program Annex 27 Performance of Solar Façade Components. Project: Service life prediction tools for Solar Collectors EN ,2:2006 Thermal solar systems and components-solar collectors CEN TC 312/ WG1 N 333 E Resolutions from the 9:th plenary meeting of CEN TC 312. Solar Keymark II project (2007) Spilg L (2004) Polymeric materials in solar collectors- Test methods and technical requirements. SP-Method SP Technical Research Institute of Sweden NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

166 WP 4.1 D2, D3 Compiled results from questionnaire on priorities for collector standards and methods Author : Peter Kovacs February 2006 Product/ technology Type and ID of standard covering the product Rating if interest from representatives of Research and Industry (Industry input both from inquiry and from ESTIF workshop in November 2004 with appr. 25 companies represented) + Serious interest - No serious interest R Subject for further research General aspects concerning all types of collectors -EN Solar Keymark Scheeme rules Furthermore, some specific comments Research Industry R II I + IIIIII III - III II Further development can be done e.g. on m 2 to energy conversion improved exposure/ accelerated ageing test certification issues characterisation of incidence angle dependencies (non flat-plates) French type ageing: Present ageing test is not very sophisticated (+) for m 2 to energy (+) for m 2 to energy Improved exposure/ accelerated ageing test. To develop a short exposure test would be valuable for test institutes as well as for manufacturers. The focus should be on points which cause problems during lifetime of the collector (eg. UV radiation, aging of coating and glass,..) Characterisation of incidence angle dependencies: Some investigations for all non flat plate collectors High interest in improved exposure/acc. ageing and IAM Test conditions and performance prediction can be further refined not only for flat plate collectors but for all types of collector designs Special interest for the assessment of a common methodology for the conversion of m 2 in annual useful energy. Further work concerning compatibility between steady-state and quasidynamic methods NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission NEGST NEGST NEW NEW GENERATION OF OF SOLAR SOLAR THERMAL THERMAL SYSTEMS SYSTEMS is a project financed by the European Commission

167 ANNEX 1 Product/ technology Ordinary flat plate collectors Type and ID of standard covering the product -EN 12975: Perf. and qualification testing. -Solar Keymark Scheeme rules: Product certification page 2 of 5 pages Rating if interest from representatives of Research and Industry (Industry input both from inquiry and from ESTIF workshop in November 2004 with appr. 25 companies represented) + Serious interest - No serious interest R Subject for further research Furthermore, some specific comments Research Industry R + IIIIII III - III IIII Test conditions and performance prediction can be further refined. Especially concerning the accuracy of the (test)results Improvements in the fields of: Scheme rules, performance prediction and short lasting quality testing would be preferable Air collectors Italian norm for performance testing UNI 8937:1987 French UEAtc directive from 1986 Research Industry ESTIF workshop R I + I - IIIIIIII IIIIIII - No CEN or ISO standard available. Some prenormatory work done in IEA SH&C annex 19 and in the CSTG-group. Currently there is no serious interest in this technology. Nevertheless a recovery could come from the coupling with desiccant cooling systems Agreement of test institutes on test conditions for air collectors based on existing material + From the point of view of standardisation for cooling systems (INETI) Product/ technology Stationary CPC technology Type and ID of standard covering the product EN 12975: Perf. and qualification testing. Solar Keymark Scheeme rules: Product Rating if interest from representatives of Research and Industry (Industry input both from inquiry and from ESTIF workshop in November 2004 with appr. 25 companies represented) + Serious interest - No serious interest R Subject for further research Furthermore, some specific comments Research Industry ESTIF workshop R + III I - IIIIII IIIIII - NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

168 ANNEX 1 page 3 of 5 pages certification Some of the work on Tracking/concentrating collectors done in IEA SH&C Annex 33 SHIP (below) will also be applicable on these collectors No serious interest for normative work but research and development is needed since this technology is promising for driving absorption/adsorption chillers and desalination thermal processes. These collectors have been tested like flat-plates. The only difference to be accepted is that diffuse energy collected is 1/C of the total available, thus affecting the efficiency value for comparison with flatplates (C=1) Although these collectors are already part of EN12975 there is some work needed to clarify test conditions for performance testing qdt and steady state as well as for the durability tests Tracking/ concentrating collectors Research Industry ESTIF workshop R III IIIIII R + III - IIII I Product/ technology Uncovered low temp. collectors Type and ID of standard covering the product EN 12975: Perf. and qualification testing. Solar Keymark Scheeme rules: Product certification Test conditions, nomenclature and performance prediction is at present being further developed within IEA SH&C Annex 33 SHIP Although these collectors are already part of EN12975 there is some work needed to clarify test conditions for performance testing qdt and steady state as well as for the durability tests. Remarks apply here even stronger Rating if interest from representatives of Research and Industry (Industry input both from inquiry and from ESTIF workshop in November 2004 with appr. 25 companies represented) + Serious interest - No serious interest R Subject for further research Furthermore, some specific comments Research Industry ESTIF workshop R + IIII I + - IIIII IIIIII Interesting for combination with heat pump technology Test conditions and performance prediction can be further refined. Improvement of existing part of EN as well as performance prediction Research is warranted on materials and durability NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

169 ANNEX 1 Hybrid/ Combined solar thermal and PV collectors Research Industry ESTIF workshop R IIIIII IIIIIII R + - III page 4 of 5 pages Both performance and qualification standards exist for both products separately. No existing combinations known. Especially ECN is dealing with this issue in NL Product/ technology Vacuum tube collectors/ vacuum technology Type and ID of standard covering the product EN 12975: Perf. and quality testing. Solar Keymark Scheeme rules: Product certification Rating if interest from representatives of Research and Industry (Industry input both from inquiry and from ESTIF workshop in November 2004 with appr. 25 companies represented) + Serious interest - No serious interest R Subject for further research Furthermore, some specific comments Research Industry ESTIF workshop R I + IIIII II - IIIII IIII - Methods to determine quality and long term properties of vacuum and heat pipe efficiency could be very useful. Could be possible, need precision The remarkable growth of national industry motivates a renewed interest in this technology. Adaptation to specific characteristics and requirements, especially for quality tests Absorber surfaces Qualification procedures ISO/DIS EN provide informative guidelines but no quantitative requirements on long term durability Research Industry ESTIF workshop R I + IIIIIII IIIIII + - II Common European standards on components of collectors (e.g. selective surfaces, transparent insulation, reflectors and new generation materials) will help product development and quality management in the production line and might be included in existing standards to increase quality information. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

170 ANNEX 1 page 5 of 5 pages These are very important points with good likelihood for improvements Great interest by the manufacturers. Especially for methods to measure (ε) and (α), and secondary for durability testing of coatings Product/ technology Polymer absorbers and covers Reflector materials Type and ID of standard covering the product EN provide informative guidelines but no quantitative requirements on long term durability EN provide informative guidelines but no quantitative requirements on long term durability Rating if interest from representatives of Research and Industry (Industry input both from inquiry and from ESTIF workshop in November 2004 with appr. 25 companies represented) + Serious interest - No serious interest R Subject for further research Furthermore, some specific comments Research Industry ESTIF workshop R II III + III II + - IIII II These are very important points with good likelihood for improvements For quantitative requirements on long term durability referring to methods for measuring transmission coefficient (τ) for covers Research Industry ESTIF workshop R I I + IIIIIII IIII + - II II Research on the long term behaviour and accelerated ageing of reflectors has been published recently. Prenormatory work on this subject is going on in IEA SH&C Annex 27 Antireflective coatings - Research Industry ESTIF workshop R II III + III III + - IIII I Research on the long term behaviour needed? Interesting point for research; also on performance and costs! Long term aging & durability NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

171 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 1 (5) Leo Spilg Polymeric materials in solar collectors Test methods and technical requirements Department of Chemistry and Materials Technology Borås 2004 This is a translation from the Swedish original document. In the event of any dispute as to the content of the document, the Swedish text shall take precedence.

172 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 2 (5) Index 1 Scope 3 2 Assumptions/tests Thermo-oxidative degradation Weatherability Chemical resistance 4 3 Requirements Absorption Transmittance Mechanical characteristics 5 4 Material composition 5 5 Report 5

173 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 3 (5) 1 Scope This SP-method relates to plastic- and rubber components in solar collectors. The present issue focus primarily on absorbers, reflectors, cover glazing and frames. The method can however with certain adjustments be applied to other polymeric material components, such as pipe systems and sealings. The purpose of the method is to ensure a 15-year lifetime of the components by lifetime analysis of included materials regarding mechanical characteristics, absorption and transmittance. The method is also intended to, in combination with other requirements, be used as a basis for P-marking of solar collectors. 2 Assumptions/tests The relevant material characteristics are degraded by among other things the following: - thermo-oxidative degradation. - solar radiation (foremost UV radiation). - water/humidity (possibly acid rain). - chemicals, foremost products used in heat transfer fluids. 2.1 Thermo-oxidative degradation Refers in this case to material degradation caused by time under the influence of atmospheric oxygen. The degradation caused can be expected to influence mechanical characteristics, absorption and transmittance. As the speed of this process is depending on temperature, the lifetime of a material at a certain temperature of use can be assessed by accelerated ageing at a higher temperature followed by an evaluation of the relevant characteristic. The choice of time and temperature/temperatures for this accelerated ageing can be complicated, and is governed by material type, possible existing knowledge or data for the material in question, expected temperature levels during use etc. The time/temperature correlation during the ageing will therefore in general be specific for each sample/product. Expected temperature levels during use will primarily be determined according to SP-method 3860 (only in Swedish). The evaluation is performed regarding mechanical characteristics, absorption and transmittance. The test method, especially for the mechanical characteristics, is obviously depending on the shape of the available test sample. Normally an impact or a tensile test is performed.

174 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 4 (5) 2.2 Weatherability Weatherability includes effects of solar radiation (primarily UV-radiation), rain/moisture and other environmental elements such as for example acid rain (sulphuric acid). The durability of materials is tested by exposure in a Weatherometer according to ISO , method A, as follows: - Black standard temperature: 65 ±3 C - Relative humidity: 50 ±5 % - Light intensity ( nm): 60 ±6 W/m² - Spray cycle: 102/18 min (dry interval between spraying 102 min followed by 18 min of spraying) - Time of exposure: 3000 hours The evaluation is performed regarding mechanical characteristics, solar absorption α s and/or solar transmission τ s. 2.3 Chemical resistance The characteristics of polymeric materials are in many cases degraded under the influence of chemicals, water and humidity, especially at elevated temperatures. The necessity to perform a durability test regarding this must be judged from case to case, depending on the anticipated type of chemicals, earlier experience and the operational characteristics of the solar collector. During the test, the materials are exposed to the chemicals in question during a time, and at a temperature relevant to the expected operational characteristics. Evaluation is performed regarding mechanical characteristics.

175 Method Description SP-METHOD 2884 Polymer Technology Issue 1 Issued by Issued Leo Spilg Approved by: Page 5 (5) Requirements 3.1 Absorption The solar absorptance, α s, is measured before and after ageing of the absorber. The solar absorptance is determined by acquiring a reflectance spectrum of the absorber surface between 200 and 2500 nm using a UV-VIS-NIR-spectrometer equipped with an integrating sphere attachment. The solar absorptance is calculated by convoluting the reflectance spectrum and the solar spectrum (air mass 1.5). The reduction in solar absorptance must not exceed 5 % in absolute terms after ageing. 3.2 Transmittance The solar transmittance, τ s, is measured before and after ageing of the cover glazing. The solar transmittance is determined by acquiring a transmittance spectrum of the cover glazing between 200 and 2500 nm using an UV-VIS-NIR-spectrometer equipped with an integrating sphere attachment. The solar transmittance is calculated by convoluting the transmittance spectrum and the solar spectrum (air mass 1.5). The reduction in solar transmittance must not exceed 10 % in absolute terms after ageing. 3.3 Mechanical characteristics Normally a reduction of 50 % of the measured characteristic is accepted after ageing relative the unaged material. 3 Material composition If the result is to be used as a basis for P-marking of solar collectors, an identification of the material is performed by a suitable method, for example IR-spectrofotometry or thermal analysis. The composition of type-tested materials may not be changed (a new composition requires a new test) Only raw materials with stated manufacturer may be used. Unspecified or re-processed materials may not be used without specific permission. 4 Report The test report should include the following: a. Test laboratory. b. Purpose of the test. c. Reference to this test method. d. Identification and description of the material tested. e. Test conditions. f. Test results. g. Any deviations from this method.

176 WP4.1 Resource document Performance and quality testing of evacuated tubular collectors Dissemination level: Public Author: Peter Kovacs, SP Reviewer: Maria Joao Carvalho April 2007 CONTENTS INTRODUCTION Why present standards are not adapted to ETCs and why the present test procedures need to be revised with regard to ETCs. EXPERIENCE FROM RECENT ETC IMPLEMENTATION AND TESTING Description of the experience of some of the main actors in the field of ETC testing to give a background and to motivate the need for revision. SPECIFIC ASPECTS OF EVACUATED TUBULAR COLLECTORS ETCS Introduction to the various specifics of testing and quality assurance of testing solar ETC collectors. SUMMARY This report aims to put focus on evacuated tubular collectors (ETCs) and their specific features from the point of view of testing and quality assurance. ETCs have many advantages to flat plate collector and can therefore contribute significantly, much more than today, to the large scale introduction of solar heating technology in Europe. A prerequisite for this however is that their quality and performance can be determined and assured in the same manner as how it has been successfully done for flat plate collectors. In order to achieve this, the present test methods and quality assurance schemes need to be updated so that the specific characteristics of ETC are also taken into account. This report describes why it is important to update the present methods to make them better fit to cope with performance and quality testing of ETCs and how it can be done. Some new tests and some ideas about further research work in the field of ETCs are also presented. RECOMMENDATIONS Suggestions for revision of existing test methods for ETCs and for development of new methods. Recommendations for further activities. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

177 ANNEX 3 page 2 of 10 Table of contents SUMMARY...1 Table of contents Introduction Experience from recent ETC implementation and testing Market development Inquiry about the state of the art Specific aspects of evacuated tubular collectors ETCs Specific features relevant to all types of ETCs Single glass tubes Double glass tubes ETC with direct connection ETC with heat pipe connection Recommendations Recommendations for revision of present methods (EN :2006) Definitions Thermal performance testing- Efficiency testing Thermal performance testing- Thermal capacity Thermal performance testing- Stagnation temperature measurement Thermal performance testing- Specification of physical properties Quality testing- General Quality testing- High temperature resistance- and exposure test Quality testing- Mechanical loads Durability of reflector materials Quality testing- Impact resistance Documentation Proposals for new test methods Freeze testing of heatpipes Durability of glass to metal seals Recommendations for further work References...10 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

178 ANNEX 3 page 3 of 10 1 Introduction Testing the thermal performance and quality of solar collectors has a relatively long history. Today s European test standards where developed on the basis of ISO- and Ashrae standards that originate from before Even though the first evacuated tubular collectors where already present at that time, the flat plate collector was the standard product and so it has been until around Therefore, almost all of the work related to development of test methods and quality criteria for solar collectors has been done with the flat plate collector in mind. Only to a minor extent the ETC and it s specific properties has been addressed. Recently, mainly due to the strong economic development in China, ETC has become more and more common in the global context. Mainly in China itself production in absolute numbers is amazing, but also in some European countries, ETC market shares have been growing rapidly. This has partly been due to promising cost/ performance ratios and the fact that ETC tend to perform better than flat plates under some circumstances. However in some cases it has become obvious that the low prices were also accompanied by low quality in different respects only this low quality wasn t always revealed due to inadequate or improper test methods. Therefore, today s test methods and requirements need to be updated and adapted to this somewhat reborn technology. Not only in order to create a fair competition between different collector types but also to give manufacturers and importers the proper tools to judge and further develop the quality and performance of ETCs. This way the technology will be able to contribute more significantly to the different European markets on the rise. Avoiding the risk of low quality, low cost products destroying the good reputation of Solar thermal technology is another important reason why ETCs need more attention in the test standards and in the quality assurance schemes. 2 Experience from recent ETC implementation and testing 2.1 Market development Apart from a very strong development in Sweden where market shares for ETCs, mainly imported from China, has grown from 5 to 25 % in a few years, only Ireland in Europe seem to have a large market share for ETCs. In most other countries the ETCs share of the total market is below 10% and has not been reported to grow. Due to a market share just above 10 % and a large total domestic market Germany has by far the largest number of ETCs sold in Europearound m 2 in Australia reports a slowly growing market for ETC but also of many products failing to enter the market due to low quality. In China ETCs are a true success story. The overall annual growth rate for solar thermal is close to 30%, annual sales around 15 million square meters and the market shares for ETCs are 80-90%, compared to 1997 when it was only around 35%. 2.2 Inquiry about the state of the art In 2005 an inquiry about ETC testing and quality aspects was performed within the NEGST project /Kov05/. The purpose of the inquiry was to give a background to an assessment of the need for revised test procedures regarding performance- and quality testing of ETC:s. The inquiry addressed the partners of the NEGST and Keymark II projects but also relevant industry actors and test institutes not directly involved in these projects. The majority of questions were very test- specific but also general comments regarding need for revised test procedures, weaknesses in ETCs that should be assessed, present tests that are being done without justification etc. could be input. The questionnaire used in the inquiry was answered by ten laboratories, two Swedish importers of ETCs and by one Chinese manufacturer. The answers showed that most laboratories had quite a limited experience of ETC testing but also that several of the difficulties encountered in testing were common. The inquiry also turned out to raise more new questions than it answered and the proposals developed from the inquiry input may in many cases need further back up in NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

179 ANNEX 3 page 4 of 10 terms of scientific investigations in order to be clearly justified. Therefore the majority of the following recommendations, developed from the questionnaire input, should be seen as input to discussions in future standardisation work rather than as fully developed proposals for revisions. 3 Specific aspects of evacuated tubular collectors ETCs In the following some specifics aspects of ETCs are highlighted in order to give a better understanding and background of the new or redefined tests that may be required to assure their proper function and long lasting performance. As different generic types of ETCs are partly very differently constructed and thus also have different potential failure modes they are here addressed first with respect to their common features and then with respect to the characteristics of each generic type. 3.1 Specific features relevant to all types of ETCs The main features where ETCs in general differ significantly from flat plate collectors and thus require some particular attention with respect to performance- or quality testing and also regarding design requirements are summarized in the following table. Specific feature The comparatively low heat losses resulting in high stagnation- and maximum operation temperatures The non planar shape of the collector surface, either it is fitted with a reflector or not The frequent use of (external) reflector mirrors The fragile structure of the vacuum tubes The fact that the performance is heavily dependent on the quality (level, durability) of the vacuum (Also relevant for flat plate collectors with vacuum- or inert gas filling) Implication on testing or system design Difficult to determine efficiency at high temperatures with good accuracy (also relevant for high performing flat plate collectors) Difficult to determine unambiguous stagnation temperature Special attention required in system design in order to avoid thermal stress on the heat transfer fluid (also relevant for high performing flat plate collectors) Difficult to determine proper loads for mechanical load tests Bi- or multi axial incidence angle modifiers need to be determined in performance testing Highly exposed component having a high influence on the performance but not being assessed in present test standards Difficult to assess the long term effects on the collector output Impact resistance testing may be required in some regions (but then probably for flat plate collectors as well) Difficult to determine vacuum loss in connection to quality tests Difficult to assess the long term durability of the vacuum 3.2 Single glass tubes ETCs built up by single glass tubes used to be the most common type of the early ETCs but these tubes have definitely become less common at the test institutes in Europe from 2004 and on. This probably also reflects their limited market penetration. The only feature specific to single glass tubes mentioned by the respondents in the inquiry was the reliability of the glass/ NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

180 ANNEX 3 page 5 of 10 metal seal, which is not specifically addressed in the presently used quality tests. IZES seems to be the one European test laboratory that has measured most single glass tubes. 3.3 Double glass tubes Double glass tubes or Dewar type tubes have become more common at some of the test institutes from 2004 and on. SP and ITW did the most tests on double glass tubes and UNSW also reports quite a large number of tests. In Austria, Greece, Canaries and Portugal only one or two products were tested in the last years. In China this is by far the most common type of glass tube used in ETCs. Some of the specific features of double glass tubes mentioned by the respondents in the inquiry were Effective capacity using calculation method according to EN /CEN04/ is underestimated for ETCs built on double glass tubes The collector models presently used in performance testing and modelling are not sufficient to describe the thermal characteristics (time constant and thermal capacitance) of ETCs built up by double glass tubes Double glass tubes result in constructions having a comparatively large number of failure modes, in particular when equipped with a heat pipe and external reflector. This makes it more difficult to track reasons for low performance figures in testing and to safeguard a high and even quality of the product. Double glass tubes, when used in water in glass systems means new challenges for quality testing of gaskets etc but this is presently treated as a system reliability issue and therefore not further addressed in this context. 3.4 ETC with direct connection A critical aspect of ETCs with direct connection, i.e. with the heat transfer fluid (glycol/water mixture) circulating through the tubes is the thermal stress on this fluid. This is particularly relevant when the collector is fitted with a reflector resulting in stagnation temperatures in excess of 250 C. There are reports saying this can be dealt with by using a proper system design /Hau02/ but there are also reports where the conclusion is that this construction is very likely to cause problems. Potential problems could be avoided through special requirements on the installer documentation for this type of collectors or, in general terms, for any collector type that can reach stagnation temperatures higher than C. 3.5 ETC with heat pipe connection From the increasing numbers of double glass tubes, many are fitted with a heat pipe which can then in turn, be fitted to the manifold through a wet or a dry connection. The heat pipe construction can be sensitive in several ways: The amount and composition of the evaporating liquid The vacuum inside the metal pipe The material quality in the pipe and the design of the pipe and the bulb Bad design or faulty installation of dry connection resulting in low heat transfer capacity Risk for freezing Damage due to high temperatures (reflectors) Air pockets inside the bulb as a result of improper filling or material NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

181 ANNEX 3 page 6 of 10 These possible failure modes together with the ones resulting from the metal fin inside double wall glass tube construction and the absorber itself makes variable quality and/ or energy performance much more likely for these collectors than for e.g. ordinary flat plate collectors. Furthermore, from the point of view of performance testing, dry out effects during testing under high irradiance should also be considered. 4 Recommendations In order to make the best out of ETC technology in the future, test methods and quality assurance measures need to be further developed. Furthermore, the knowledge about which the critical design parameters are and how they are affected by time and different kinds of stress need to be further developed. As an input to this work within standardization and research the following recommendations have been developed. 4.1 Recommendations for revision of present methods (EN :2006) Definitions Background material behind tubes (test rig) needs to be defined in the test standard when no reflector is present. Ideally a plain mat black background should be used, that will not contribute to light levels. Absorber area when reflector is used should be redefined. The definition according to EN will give efficiencies higher then Thermal performance testing- Efficiency testing Several laboratories reported a) that dry out effects can occur during testing of ETCs with heat pipes during high irradiance conditions and b) that the present collector model used in the standard was not able to accurately model the thermal capacitance and time constants of the collector (ETC with double glazing and heat pipe in particular) /Kov05/. It should be discussed how the dry out effects are to be handled during testing and in the reporting. Regarding the inadequacies of the collector performance model, one way around the problem suggested is using the model presented in, however this seems to create other difficulties in the evaluation of measurements as it is using the absorber temperature as an input. Therefore, this is at present a case for further research Thermal performance testing- Thermal capacity The method available for calculating the thermal capacity ( ) of the collector has been reported to underestimate the figures for double glass ETCs. The quasi dynamic method on the other hand is said to overestimate the thermal capacity. The two node collector model mentioned in may provide a solution to this /Str05/. The method referred in annex G3 is reported to give unreliable results /Eis04/. To summarize, the whole procedure for determining thermal capacity of collectors should be reviewed Thermal performance testing- Stagnation temperature measurement NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

182 ANNEX 3 page 7 of 10 Determining accurate and representative stagnation temperatures, whether measured on ETCs or on flat plate collectors can be a difficult task. A way around this problem could be to determine the stagnation temperature based on calculation from the efficiency coefficients. A safety factor of degrees should then be added as the efficiency is determined under windy conditions and thus it will under estimate the stagnation temperature. Such a method would save some efforts and costs in testing without significantly reducing the quality of the results. Whether it refers to performance- or to quality testing and if to be kept, the determination of the stagnation temperature of ETCs is not adequately well defined in the present standard. Notes given in section are too vague. More precise recommendations on sensor positioning (Note 1) should be given, one for each of the most common designs of ETCs unless a common generally significant positioning can be defined. Alternatively, the method suggested in Note 2 could be made normative for ETCs Thermal performance testing- Specification of physical properties To serve better as a base for quality assurance of ETCs, Annex D should be further differentiated in order to describe better the physical characteristics of the collectors. For example most properties stated for the absorber in todays standard should also be stated for the heat transfer plate used in double glass tubes. A number of terms should furthermore either be generalised or specifically mentioned with regard to ETCs in addition to the ones that today refers explicitly to flat plate collectors, eg. tube spacing in addition to riser spacing. Also the properties of an optional reflector, often present with ETCs, should be declared Quality testing- General In order to reveal low quality products, in particular among ETCs with double glazings and heat pipes, it is recommended to introduce a test cycle for these collectors where the same collector is first measured for efficiency, then subjected to a (possibly revised, tougher) high temperatureand exposure test and then measured for efficiency once again. In order to save costs one of the efficiency tests could be limited to zero loss efficiency, but preferably also the eventual increase in heat losses should be assessed. This way the entire collector will be checked for any fault that may have occurred during the high temperature- and exposure tests and which are more probable to occur on these specific collector types than on other types as explained in previous sections. A second, simpler option for this type of test is presented in section of this document but in this case, only the reliability of the heat pipe it self will be checked Quality testing- High temperature resistance- and exposure test In Australian collector tests, a type of accelerated exposure test is achieved by either of two alternative methods /Mor06/. 1. By heating a fluid that is circulated in the collector by means of a heater and natural irradiance to a calculated maximum stagnation temperature and keeping it at this temperature for 12 hours per day during ten days. This method is not applicable to heat pipe collectors. 2. By means of exposing the collector to high irradiance and high ambient temperature for a period of 12 hours per day during ten days. A solar simulator with less stringent NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

183 ANNEX 3 page 8 of 10 requirements on its performance than for performance testing of collectors, and a temperature controlled chamber is used for the test. The present European exposure test has been under a lot of debate, mainly due to it s inability to maintain uniform test conditions all over Europe and in the specific case of ETCs it is not considered very efficient in terms of revealing their weaknesses. It is therefore suggested that a new form for the exposure test is developed along the ideas of the Australian test. This new test could either be complemented by, or merged with the concluding high temperature test for heat pipes suggested in paragraph Quality testing- Mechanical loads Defining the proper load in mechanical load tests on ETCs is a tricky matter. What loads that are likely to create problems for the durability of ETCs are most probably also not completely the same kind of loads as the ones creating problems for flat plates. In Germany test labs decided together with the manufacturers to cancel tests for negative and positive pressure on collector cover as these tests are not considered to make sense for ETCs. However the issue of heavy snow loads might still be worth considering. It is recommended to follow the introduction (or collect already existing experience) of ETCs in snowy regions to find out if it s a serious problem worth taking into account for testing. In Sweden ETCs resistance to positive pressure has been tested for some time, with a requirement of 2000 Pa, hardly ever resulting in failure of the tubes themselves however sometimes revealing weaknesses in the fixings of the tubes. The method used has been more or less the same as the one described in of EN however, an air pressure was used to create the load and a thick plastic film used to distribute the load evenly over the tubes. The load requirement was furthermore reduced by multiplying it with a factor equal to the ratio of the aperture to the aperture including space between tubes area, in general resulting in the normal requirement of 2000 Pa being reduced to around 1200 Pa. It is suggested to exclude ETCs without reflectors from the mechanical load requirements in and unless it is shown from recorded practical experiences that either of them is motivated. It is suggested to include ETCs with reflectors in the mechanical load requirements in and but in this case the reflector and not the glazing should be the subject of the requirements. A positive or a negative mechanical load can be applied to the reflector by means of an airtight box divided in two compartments by a plastic film also covering either side of the reflector surface and one of the compartments being pressurized by air. It is quite obvious that snow loads in reality are to a great extent carried by the tubes in front of the collector and therefore the same positive load as that applied to a flat plate collector cannot be applied on the reflector. The positive load applied needs therefore be downscaled in a suitable way. One source /Abr07/ suggests the use of wood pellets or similar material to load the collector as it will somehow behave like snow in terms of load distribution Durability of reflector materials In addition to the mechanical load resistance, suggested to be tested in paragraph 4.1.8, the long term effects on optical properties of the reflector is of course also of high interest. Methods for assessment of this are under development, e.g. in IEA SH&C Annex 27. It is recommended to follow this development and, if the solar industry sees the need for it, incorporate an NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

184 ANNEX 3 page 9 of 10 accelerated ageing test in the standard as a normative requirement for ETCs that are tested and sold with an external reflector Quality testing- Impact resistance Positive impact resistance test is only possible with ice balls. Impact resistance using steel balls and ice balls are not comparable /Kov05/ (which is already spelled out in the standard) but furthermore ETCs are told not being able to withstand the impact of the steelballs. Therefore method 1 in 5.10 should be deleted. Method 2 may need a review /Mor05/ and practical feedback regarding results from hailstorms on ETCs as well as on flat plate collectors would be useful. As one or two broken tubes in an ETC module is far less critical than a broken glass in a flat plate collector and there are no reports showing that either type should be hail resistant there are probably no good reasons to require impact resistance tests only for ETCs Documentation In order to reduce the risk for brake down of the heat transfer fluid in the collector loop that may occur when exposed to temperatures above 230 C, special considerations should be taken when designing the loops including such high efficient collectors. In 7.3 of EN (requirements on the installer manual) it should at least be added installation... to the paragraph Recommendations about the heat transfer media.. and furthermore reference /Hau02/ containing some practical design considerations, should be added in the same paragraph. 4.2 Proposals for new test methods Freeze testing of heatpipes Damaging of heatpipes due to freezing can result from improper composition of the working media in the heatpipe or from bad design of the metal tube (material quality, thickness, shape of lower end) and has been reported by several sources /Hum06/, /Joh06/, /Sun06/. As breakage of the metal tube in the case of bad design often doesn t occur until after several freeze/ thaw cycles, the following procedure has been proposed: At least five samples of the product should be exposed to 100 cycles of 1) 1 hour freeze time in -20 C or lower temperatures. Heat pipes should be sitting at a angle which is the harshest condition for them 2) 20 minutes in an oven at 150 C, Once every 10 cycles, in the oven for 20min to 200 C (or the actual stagnation temperature of the collector in case) thus mimicking a possible stagnation condition. Problems are shown either as a deformation at the bottom of the heatpipe (requirement on max. change in diameter needed) or as the formation of an air pocket which results in a cool area at the top of the heatpipe when exposed to heat at the bottom. The 100 cycles should be concluded by a 24 hour exposure at the stagnation temperature of the collector in case and then: 3) Turn the heat pipes upside down (tip down) in a container of hot (>50oC) water, at least 30cm deep. Swirl the heat pipes around to allow them to "start" and check to make sure they are working ok NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

185 ANNEX 3 page 10 of The product has failed if more than one heatpipe either forms an air pocket or shows a deformation higher than allowed. If one heatpipe fails, the test can be rerun with new samples Durability of glass to metal seals In order to assess the durability of the glass to metal seal in the types of ETCs where such seals exists either a vibration or a cyclic tension applied to metallic parts entering the glass tube may be considered. 4.3 Recommendations for further work In general the heat transfer mechanisms of the ETCs and how different parts of ETCs respond to different kinds of stress need to be better known. A call for a common research effort was distributed in late 2006 /Kov06/. Two more specific topics that call for attention are: Any severe weaknesses in today s performance test methods/ models would be revealed through a project where the most common ETC types where tested for performance and then measured for a longer period to have a thorough comparison: measured to modelled performance. Problems with vacuum losses of the tubes are of course critical to the performance of ETCs. At present it is difficult to determine the vacuum loss and the tests performed in the standards are not able to verify the long time durability of the vacuum. 5 References /CEN06/ /Eis04/ /Hau02/ /Hum06/ /Joh06/ /Kov05/ /Kov06/ /Mor05/ /Mor06/ /Str05/ /Sun06/ /Abr07/ EN ,2:2006. Thermal solar systems and components- Solar collectors. Part 1: Requirements. Part 2: Test methods Eisenmann, W et al. Conference paper Eurosun On the determination of the effective thermal capacitance of solar collectors Hausner, R. Report of IEA SH&C Annex 26, Stagnation behaviour of solar thermal systems Humphreys, M. Managing director, Apricus Solar Co. Ltd. China Personal communication Johansson, S. Managing director, Intelliheat AB Sweden. Personal communication Kovacs, P. NEGST working paper, Inquiry about needs for revised test procedures for evacuated tubular collectors, ETCs. Kovacs, P. NEGST working paper, Call for a common research effort on ETCs Morrison, G. Working paper, Ice ball impact tests on evacuated tubes Morrison, G. Excerpt from Australian/ New Zealand standard AS/NZS2712. Solar and heat pump water heaters: Design and construction- Stagnation test for collector and integral collector and container. Streicher, E et al. ISES Performance model for solar thermal collectors taking into account degradation effects. Sundquist, R. Managing director Exoheat AB, Sweden Personal communication. Abrecht, S. Project manager Paradigma Gmbh. Germany Personal communication. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

186 Accelerated ageing test of solar collector Dissemination level: Public Authors: Rodolphe Morlot / Bouzid Khebchache, CSTB Reviewer: Peter Kovacs February 2006 CONTENTS: INTRODUCTION Description of main objectives. ONE YEAR EXPOSURE TEST - Simulation Correspondence with the test of one year ageing of the collector Summary of the simulation results FAILURE MODE EFFECT AND CRITICALLY ANALYSIS CONCLUSIONS REFERENCES SUMMARY "Reliability" and "durability" terms gather the notion related to the maintenance in time of the functional characteristics of a product. Reliability applies more to the object taken as a whole, whereas durability is related to the materials which make it up. The use of new materials for solar thermal collectors, increasingly more powerful, but could lead to problems (with medium term) of reliability and durability of the solar components. The principal constraints imposed on the solar collectors are the temperature, the pressure and the atmospheric agents (rain, UV, freezing...). Consequently, with the aim of knowing if a test of one year ageing in natural exposure of a solar collector were relevant but also representative of the normal functioning of the collector during 15 to 20 years, we have chosen to practice a simulation on a solar water-heater and a combined solar system. With this intention, we developed a model with TRNSYS in order to be able to evaluate the value and the frequencies of the temperatures reached for the different components of the collector. The second part of the document deals with an original methodology based on the use and adaptation of the Failure Modes, Effects and Analysis (previously used in the aeronautical and automotive domains). This methodology, would identify the failure modes (exhaustive search of the behaviours, degradations and failures of elements), their causes and effects, taking into account the potential problems and errors found on a solar collector. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

187 ANNEX 4 page 2 of 23 pages 1. INTRODUCTION "Reliability" and "durability" terms gather the notion related to the maintenance in time of the functional characteristics of a product. Reliability applies more to the object taken as a whole, whereas durability is related to the materials which make it up. The use of new materials for thermal solar systems, increasingly more powerful, lead to problems (with medium term) of reliability and durability of the solar components. This is why the CSTB, has realized for several years, studies on the topic of corrosion, reliability and durability of the collectors. These studies have mainly aimed at materials and components making up the system (held with the temperature and UV, life-time behaviour...), with tests in laboratories, and an active participation in international working groups on this field (International Energy Agency (IEA), Collector Testing Group of the European Community, UEAtc). The constraints of operation of these systems (temperatures and pressures reached...), the modes of failures (defects in the processes of construction or maintenance, behaviour in time of the system) and the analysis of links between these different elements and of their impacts on the aptitude of the system to provide functions for which it is conceived (mechanical held, thermal performances...). From our point of view, a series of tests on each separated components of the solar collector (absorber, transparent cover...) does not seem to be the most representative way. By the same one, the test of a small-scale solar collector was abandoned, because it is partially answering to the manufacturer s problematic wishing to test their product but also because the test results measured on this small sample are not easily representative and able to be extrapolated on a collector of standard size. A solar collector must be made up on the basis of a suitable choice of materials in order to resist to high constraints, but must also answer to problems such as the mechanical resistance, watertightness, the corrosion resistance, the frost resistance. These kind of potential problems are examined in the UEAtc Directives and the standard EN , in which several sequences of tests are presented, like tests in real size in stagnation and natural exposure on a whole collector. Moreover, the method recommended in EN , imposes an exposure test over 30 days that we regard as rather relevant in a procedure of operating requirement of the collector but not adapted to the forecast of durability. Indeed, if this test of exposure appears necessary to control the behaviour of the collector in severe conditions, it does not make it possible to consider an extrapolation of the long-term behaviour. This is the reason why we think that an ageing test by a natural exposure of the solar collector during one year is more representative of reality, and thus the tested collector sudden of requests close to those we could await from a normal functioning on 15 to 20 years. The first part of this draft presents the justification of this choice of test, with the supply of elements allowing an analysis for a discussion on the subject. In complement, the second part of this document presents a method of analysis used by the CSTB with the aim of preventing the potential failures of a product. This method, which is entitled FMEA (Failure Modes Effect and Analysis) was carried out partly on a solar collector, in order to wonder about the behaviour of materials subjected to the functioning constraints reigning inside the collector. This method could be used to check test results obtained after ageing but also in order to make a finer and more detailed analysis of results obtained. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

188 ANNEX 4 page 3 of 23 pages 2. ACCELERATED AGEING TEST OF SOLAR COLLECTOR 2.1 THE ONE YEAR NATURAL EXPOSURE TEST REPRESENTATIVITY Simulation The principal constraints imposed on the solar collectors are the temperature, the pressure and the atmospheric agents (rain, UV, freezing...). Consequently, with the aim of knowing if a test of one year ageing in natural exposure of a solar collector were relevant but also representative of the normal functioning of the collector during 15 to 20 years, we have chosen to practice a simulation on a solar water-heater and a combined solar system. With this intention, we developed a model with TRNSYS, in order to be able to evaluate the value and the frequencies of the temperatures reached for the different components of the collector Solar domestic water heater (SDWH) Two configurations were retained for the dimensioning of the systems studied: Configuration 1: a collector aperture area of 4 m², for a daily load volume of 250 l/d and a capacity tank of 300 litres. This system is a "normal" dimensioning in the Nice location. For this configuration, the method SOLO gives a solar fraction ranging between 60 and 65 %. Configuration 2: a collector aperture area of 6 m², for a daily load volume of 125 l/d and a capacity tank of 300 litres. This system is an "oversize" dimensioning in the Nice location. For this configuration, the method SOLO gives a solar fraction ranging between 85 and 90 %. The assumptions of the Simulation are : - File weather based on the location of Nice. - Consideration of two desired temperatures of water for each configuration: 60 C: maximum temperature of distribution of hot water without appliance of safety limited temperature of water, and 90 C: a traditional temperature of heating. The system was stopped during August (period of holidays when the constraints of temperatures applied to the system are significant). The digital model doesn t take into account the inertia of the system, we put forth additional assumptions during the treatment of the results resulting from modelling: 1. a stop due to hysteresis does not lead to a setting in stagnation of the collector because the primary circuit will start again when the temperature difference between the collector and the tank is higher than T (5 C in our case), 2. a stop of the primary circuit due to a temperature of the tank higher than the desired temperature leads to a setting in stagnation of the collector. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

189 ANNEX 4 page 4 of 23 pages Solar combined system (SCS) The solar combined system was only modelled in its "summer" configuration (except period of heating) because during this period, constraints for the collector are highest. So the model is based on the normally dimensioned system (configuration 1) of the standard system of production of hot water. The collector aperture area is 12 m² and corresponds to a solar fraction (heating + hot water) calculated of 60 % for a house of 100 m². We used two levels of daily load of hot water : 250 l/d with 50 C (approximately 4 persons) and 125 l/d with 50 C (approximately 2 persons) Simulation results The simulation showed that temperatures higher or equal to 90 C can be reached, in the most critical cases, during 462 to 694 hours per years and, in the cases best dimensioned, during 76 to 436 hours per years. The following tables show the distribution of the number of hours, for which the collector reached a higher temperature or equal to 90 C, according to the two configurations of system of water-heater but also for the combined solar system: SDWH (Reached temperatures) Configuration 2 (T tank = 60 C) (most critical case) Configuration 2 (T tank = 90 C) (most critical case) Total (h) Configuration 1 (T tank = 60 C) Configuration 1 (T tank = 90 C) Table 1 : Number of hours corresponding of reached temperatures by the collector per year without any draw-off in August Solar combined system (Reached temperatures) Daily draw-off = 125 l/d T tank = 60 C (most critical case) Daily draw-off = 250 l/d T tank = 60 C Daily draw-off = 250 l/d T tank = 90 C Total (h) Table 2 : Number of hours corresponding of reached temperatures by the collector per year without any draw-off in August Consequently, we can estimate: - that a water-heater is subjected to constraints of temperatures higher than 90 C between 1 and 8 days per year according to its dimensioning, - that a solar combined system is subjected to constraints of temperatures higher than 90 C between 3 and 14 days per year according to its dimensioning, this type of system is requested more in term of constraint of temperature than the standard system. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

190 ANNEX 4 page 5 of 23 pages Most critical cases The following table presents the distribution of the number of hours of temperature reached (> = 90 C and > = 150 C), according to application (SDWH or SCS), brought back to 15 or 20 years: > = 90 C > = 150 C *configuration 2 SDWH* SCS SDWH* SCS Numbers of hours per year Numbers of hours per 15 years Numbers of hours per 20 years Table 3 : Number of hours of collector stagnation on 15 or 20 years without draw-off in August The following figure presents the distribution of the annual irradiance on collector which led to these numbers of hours of stagnation for one year: 50 Température de stagnation >= 90 C Température de stagnation >= 150 C Nombre d'heures Ensoleillement (W/m²) Figure 1: Irradiance distribution on solar collectors during stagnation periods on one year without any draw-off in August - SDWH - 60 C. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

191 ANNEX 4 page 6 of 23 pages Consequently, if we cross the distribution of the number of hours of stagnation with the irradiance, we obtain the following table: SDWH T Stag >= 90 C T Stag > = 150 C < 850 W/m² > = 850 W/m² Total % 29 % Table 4: Numbers of hours of stagnation according to Irradiance - SDWH. 50 Température de stagnation >= 90 C Température de stagnation >= 150 C Nombre d'heures Ensoleillement (W/m²) Figure 2: Irradiance distribution on solar collectors during stagnation periods on one year without any draw-off in August SCS l - 60 C. Solar combined system T Stag >= 90 C T Stag > = 150 C < 850 W/m² > = 850 W/m² Total % 22 % Table 5: Numbers of hours of stagnation according to Irradiance - SCS. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

192 ANNEX 4 page 7 of 23 pages We notice that to obtain temperatures of stagnation higher or equal to 150 C, the irradiance received on the collector is higher than 850 W/m² during 29% of time for configuration 2 for the SDWH and 22 % of time for the SCS Normal sized cases The following table presents the distribution of the number of hours of temperature reached (> = 90 C and > = 150 C), according to application (SDWH or SCS), brought back to 15 or 20 years: SDWH > = 90 C > = 150 C *configuration 1 SDWH* CSS SDWH* CSS Numbers of hours per year Numbers of hours per 15 years Numbers of hours per 20 years Table 6 : Numbers of hours of stagnation for 15 or 20 years without draw-off in August The following figure presents the distribution of the Irradiance on the collector which led to these numbers of hours of stagnation for one year: 50 Température de stagnation >= 90 C Température de stagnation >= 150 C Nombre d'heures Ensoleillement (W/m²) Figure 3 : Irradiance on collectors during stagnation periods for one year without any draw-off in August - SDWH. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

193 ANNEX 4 page 8 of 23 pages Consequently, if we cross the distribution of the number of hours of stagnation with the irradiance, we obtain the following table: SDWH T Stag > = 90 C T Stag > = 150 C < 850 W/m² > = 850 W/m² 23 2 Total % 15 % Table 7 : Numbers of hours of stagnation according to Irradiance - SDWH. 50 Température de stagnation >= 90 C Température de stagnation >= 150 C Nombre d'heures Ensoleillement (W/m²) Figure 4 : Irradiance distribution on solar collectors during stagnation periods on one year without any draw-off in August SCS l - 90 C. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

194 ANNEX 4 page 9 of 23 pages Consequently, if we cross the distribution of the number of hours of stagnation with the irradiance, we obtain the following table: Solar Combined system T Stag > = 90 C T Stag > = 150 C < 850 W/m² > = 850 W/m² Total % 43 % Table 8 : Numbers of hours of stagnation according to Irradiance - SCS. We notice that to obtain temperatures of stagnation higher or equal to 150 C, the Irradiance received on collector is higher than 850 W/m² during 15% of time for the configuration 2 of SDWH and 43 % of time for SCS Correspondence with the test of one year ageing of the collector We propose to study the correspondence of the preceding results with the test of current ageing carried out on a collector only subjected to one year natural exposure. If we carry out a simulation over the whole year on the same collector tilted to 45 and in stagnation on our site we obtain the following distribution of Irradiance: 120 Température de stagnation >= 90 C Température de stagnation >= 150 C Nombre d'heures Ensoleillement (W/m²) Figure 5 : Irradiance on collector during one year in stagnation and titled to 45 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

195 ANNEX 4 page 10 of 23 pages Consequently, if we cross the distribution of the number of hours of stagnation with the irradiance, we obtain the following table: Ageing exposure test T Stag T Stag > = 90 C > = 150 C < 850 W/m² > = 850 W/m² Total % 14 % Table 9 : Numbers of hours of stagnation according to Irradiance ageing exposure test Summary of results Simulation enabled us to show that the current tests of one year ageing in natural exposure of a solar collector largely cover the equivalent duration of 20 years for the normally dimensioned solar water heaters (113 to 244 %). On the other hand, it is not the same for the solar combined systems even when those are dimensioned normally (20 to 41 %). SDWH SCS Reached Temperatures of the collector 90 C 150 C 90 C 150 C Collector test / numbers of hours per year Normal dimensioned cases Numbers of hours per 15 years Cover of the test Ratio (1 year/15 years) 150 % 244 % 26 % 41 % Numbers of hours per 20 years Cover of the test Ratio (1 year/20 years) 113 % 183 % 20 % 31 % Most critical cases Numbers of hours per 15 years Cover of the test Ratio (1 year/15 years) 25 % 16 % 16 % 10 % Numbers of hours per 20 years Cover of the test Ratio (1 year/20 years) 19 % 12 % 12 % 7 % Table 10 : Comparison of the ageing test of one collector and the collector functioning in its system (SDWH and SCS) NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

196 ANNEX 4 page 11 of 23 pages This summary table enables us to show that for the most critical cases, the cover of the test is only very partial (7 to 25 %). Furthermore, we notice that for the temperatures higher than 150 C, the cover of the test for the normal dimensioned systems is relatively good (31 to 244%). On the basis of these results of simulation, we estimate that the one year natural exposure test is relatively representative of the reality and thus that solicitations are closely the same we could await from a normal functioning during 15 to 20 years. 2.2 TESTS TO REALIZE An often advanced requirement of the user takes 15 years at least use without disorder. To achieve this goal, the CSTB by UEAtc interposed, proposed to distinguish the durability of materials constitutive entering manufacture of the collector, of that of the whole collector. While taking again this procedure, a simplification can be brought in alarming only of the whole collector, with load of the manufacturers whom they select beforehand constitutive materials to obtain a suitable longevity in connection with the functioning conditions reigning in a collector. The exposure tests over 30 days according to the standard appear rather relevant in a procedure of operating requirement of the collector but not easily sufficient with the forecast of durability because the test of the collector even under a high total irradiance of 850 W/m² seems for us too short. Indeed, if this exposure test appears necessary to control the behaviour of the collector in severe conditions, it does not make it possible to consider an extrapolation of the long-term behaviour. Moreover, the constitutive materials undergo during the functioning with the passing of years a chemical evolution which modifies the initial performances. One thirty days duration is too weak to locate a beginning of evolution. We think that a test of accelerated ageing test moderate recommended to obtain an out-of-date state which corresponds to that of the collector at 10 years for example, should be spread out over 6 to 12 months. We thus propose to carry out an aptitude test for a use of long duration which will not give a lifespan but which would be reasonable in duration and cost for the manufacturer. It would consist in laying out the sensor in stagnation outside during one year. The method which seems in first approaches the most adapted, would consist in measuring the evolution of the relevant indicators of performance (average efficiency) in the course of time under requests representative of reality. The acceptable levels of performances for one determined period for the collector could be the measurement of the difference between the initial thermal performances (before ageing) and finales (after ageing) whose threshold could be 10%. In addition to the determination of the variation of the performances of the thermal solar collectors, we propose to carry out a neat visual inspection of the sensor in order to define the deteriorations undergone by the product (possible degradation of the coating of the absorber, degradation of the cover, degradation of the joints...). NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

197 ANNEX 4 page 12 of 23 pages Principal disorders noted on solar collectors The tests carried out in our laboratory enabled us to know the principal disorders caused on the solar collector: concerning the transparent cover: - stains on the external face, - condensation on the internal face, - deposit on the internal face, - ageing in the case of a plastic material, - breakage in the case of glass concerning the absorber: - aspect, - corrosion, - traces of condensation, - deformation, - escapes of liquid. concerning the joints: - cracking of material, - washing away and bad connection cover case (frame). concerning insulator: - deformation, - separation of the absorber concerning the frame: - corrosion for a metal trunk, - degradation of aspect for a plastic trunk. 2.3 CONCLUSION The tests of ageing which we carry out currently must be supplemented by series of real measurements taken on site throughout test i.e. 1 year. These measurements would make it possible to know the real irradiance on the collector, the temperature of stagnation reached in the collector, but also the distribution of these values during the year. Measurements could thus allow the comparison with simulation made previously. A question also arises on the harmonization of the tests and the taking into account of the character of reproducibility of the test, also we propose to make same simulation but for other European climate location. In the second time, it could be interesting to carry out a Round Robin Test on a live test in these same laboratories in order to compare with simulations carried out. An exposure test in a climatic room of solar simulation under strong irradiation 900 W/m² during 4 months at least, collector in stagnation, could be carried out, if it integrates the weather environment bordering the collector (rain, moisture, cold...). Concerning the visual observation carried out on the collector after ageing, we note that a great difficulty of this evaluation resides on the fixing of thresholds of deteriorations. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

198 ANNEX 4 page 13 of 23 pages 3. FAILURE MODE EFFECT AND ANALYSIS 3.1 INTRODUCTION In order to improve the non-quality or poor quality of the building products, and the part of maintenance and operating stages in the cost of a building, CSTB developed an original tool for the capitalization and use of experience and knowledge on building products degradations. The aim of the proposed tool is to modelize the potential evolution of the behaviour of a studied product during his exploitation stage. We have a functional approach of the problem. That is to say that we study the performance of a product as regards of his ability to ensure functions for which it was designed. The originality of our tool is fourfold. Firstly, we capitalize and use the experience and knowledge on building products degradations by developing data bases (for instance impact of climatic factors on materials, incompatibility between materials, list of functions, ) and favouring expert participation (for instance determination of aggressiveness level of climatic factors on materials, classification of failure scenarios by criticality level, ). Secondly, we integrate all the potential degradations which could occur during the process stages (from the design stage to the beginning of the exploitation stage) in the modelization of the product behaviour during his exploitation stage. Thirdly we classify the failure scenarios of the product (chaining of degradations or failures of its components drawing to the failure of the product) by criticality level. Finally, we represent graphically the results of the Failure Modes, Effects and Analysis (FMEA). In the following paragraphs, we will present the proposed methodology, the graph results and finally the potential exploitations of the event-driven graph. In the following paragraphs, each step of the methodology will be presented and illustrated with an example of a solar panel system. 3.2 METHODOLOGY System analysis The system analysis is composed of three steps: a structural analysis, a functional analysis and a process analysis Structural analysis This first analysis allows us to describe the structure of the product being studied. We identify its morphology (geometrical shape, dimensions, etc), its topology of relations with neighbourhood objects, its physico-chemical composition and the nature of its environment in order to know the nature and the aggressiveness level of the environment for materials of components. The Figure 1 describes the schematization of the eight components of the solar panel system. The external components are outside the spatial limit of the studied product. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

199 ANNEX 4 page 14 of 23 pages Secondly, we determine the existence and the nature of links between: each component of the product (one example for the panel solar system is the link between glazing and seals ), the components of the product and the external components of the system (for example: the link between box and external components ). Thirdly, we identify the mediums of the product and their environmental composition (a medium is composed of several environmental agents). For example, we have to take into account two mediums when we study a window: the external environment (outside the building) and the internal environment. To facilitate this third step of the structural analysis, we have created a database including all the principal environmental agents that are generally occurred. We have classified the various data, essentially collected in (EOTA 1999; ISO 1997; Lorusso et al. 1999), on eleven categories: liquids, vapors, gas, electricity, radiations, temperatures, animals, vegetables, noises, mechanical actions, precipitations. Figure 2 regroups the results of the structural analysis of a solar panel system, that is to say the structural representation of components, the schematization of the three mediums that have to be considered (outside environment, inside environment and heat conveying fluid) and the environmental composition of each medium. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

200 ANNEX 4 page 15 of 23 pages Functional analysis We have a functional approach of the problem, indeed we consider that the product had failed when it is not anymore able to ensure the functions for which it was designed. We distinguish two kinds of functions: the need s functions and the technical and constraint functions. The need s functions correspond to the essential functions for which the product is realized and fulfill the user s needs. The satisfaction of technical and constraint functions allows the realization of need s functions. This distinction is useful for the quantitative analysis. (Ex : a solar panel system has to absorb the solar heat source, to transform it into heat fluid and to convey this fluid). The functional analysis is facilitated by the function data base in which the user can select the principal and technical functions of his product and components. At this stage, the functions ensured by each component of the product and the environmental agents in touch with each of those components are known (cf. Structural analysis). Consequently we have identified the reactions of the component solicited by environmental agents. For instance, the glazing permits the ultraviolet radiation and the seals stop the rain. Therefore, we can modelize the evolution of environmental agents through the product. We use a graphical representation (functional diagram) of this evolution. The arrows correspond to the way of the environmental agents flow. The prohibition roadsign schematize the stopped of this flow. We plot a functional diagram for each category of environmental agents (cf. Structural analysis). The Figure 3 displays the functional diagram corresponding to heat flow evolution (geared by solar energy) in the solar panel system. The heat source, solar energy, is symbolized by a sun and the evolution of the heat flow by red arrows. For instance, one of the functions of the glazing is to transmit the infrared radiation, so we plot the heat flow through the glazing by an arrow crossing the element 1- Glazing. The insulation keeps the heat, so the heat flow is held by the element 3- Insulation, what is diagrammed by a closed symbol. 5 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

201 ANNEX 4 page 16 of 23 pages Process analysis The aim of the process analysis is to determine all the errors, defects, damages, degradations that could occur to the product during its construction process (design, manufacturing, packaging, transport, storage, installation ) and could modify the behaviour of the product in service Quantitative analysis Preliminary analysis of hazards The structural, functional and process analysis allows us to describe the structure, the operation and the state of the product at the beginning of its life, once installed in the building. Then, we search all the potential causes and modes of degradations of the product and his components. The principle of the preliminary analysis of hazards is that for each function of the product, we search for each component playing a role to ensure this function, then search for failure modes, the causes, and finally the direct (consequences of degradation on the studied component) and indirect effects (consequence of degradation or failure on neighbourhood components). We distinguish two types of degradation or failure causes: action of climatic factors on a component and unexpected behaviour due to building process. We deduct the first type of causes of the studied product from a data base which defined for each material all the potential climatic factors that present an aggressiveness character. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

202 ANNEX 4 page 17 of 23 pages Nominal behaviour (t=0) From this graph (figure 3), we can deduce the nominal behaviour of the product, i.e. we know the response of a product (and its element) to a given set of climatic and use factors. At t=0 (without any degradation and considering that the product was correctly installed/implemented), we can identify the initial state of stresses. For each element, we know the environment(s) in contact with this element and the environmental agents included in these environments. Knowing the materials of the elements, for each element and each environment, we identify the aggressive environmental agents which could degrade at short or long term the elements and so the product. The stress conditions are summarised in the following table. Figure 4: Environmental stresses Initial conditions Degradation and failure analysis Failure modes (t > 0) Knowing the nominal functioning of the product, we can now analyse potential degradations and failures. We study the behaviour of the product when a material or element deviates from its normal behaviour Failure modes and effects analysis The FMEA is based on an iterative principle: the direct or indirect effects can become the cause of other degradations. With this principle we obtain all the failure scenario of the product (chaining of components degradations leading to the product failure). The failure of the product is obtained when one of its principal functions are no more fulfilled. We consider that the FMEA is finished when all the possible chaining of degradations have led to the components failure or the product failure (when a need s function is at stake). NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

203 ANNEX 4 page 18 of 23 pages Step 1: Preliminary analysis of degradation - Degradation of elements due to climatic or use factors. We first analyse the influence of initial environmental stresses on each element. For instance, we have to study the behaviour of the glazing towards high, low and cyclic temperatures, thermal shock, water, rain, snow, hail, ice, infrared radiation, loads, pressure, wind, shocks, vegetation (including moss, lichens, etc) and vertebrate (bird, small mammals, etc). With this methodology, we are able to identify degradations and failures not only due to these single factors, but also due to a combination of these factors: - combined factors from the same environment, either concomitant factors (water and low temperature from outdoor environment freezing/thawing cycles) or successive factors (sun and high temperature followed by rain thermal shock), - combined factors from different environments (high temperature from outdoor and low temperature from indoor temperature gradient in the element). Knowing the element (and its constituent material) as well as the potential stress factors, we identify the potential behaviours, for instance: - pressure inside the solar panel can lead to the cracking of the glazing, - infrared radiation can lead to the swell of the coating (one part of the absorber), - high temperature can lead to the gassing of the insulation, - vertebrate can provoke the break of the seals, - Step 2: Structural or environmental modifications (degradation or failure identification) The behaviour or the degradation identified during the first step leads to potential modifications of the environment or the structure. As examples to degradation and failure, the following was observed : - the dilatation of coil (one part of the absorber) on effect of pressure due to the coolant liquid draw to the loss of solidity of the absorber, but it still fulfils its main functions ( to absorb infrared radiation, to transmit liquids and to diffuse heat ), - if the glazing has broken, i.e. non ability or partial ability to ensure the to be watertight function, the absorber is no more protected against aggressive environmental agents of the outside environment (temperature, water, rain, snow, hail, ice, vertebrate, loads, wind, ). The initial state is updated and becomes a State 2 condition for which we have to study the effect of temperature, water, rain, snow, hail, ice, vertebrate, loads, wind on the absorber (Figure 5). NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

204 ANNEX 4 page 19 of 23 pages Figure 5: Environmental stresses State 2 This table summarises the new environmental conditions stressing the product when the glazing had failed (State 2). The new environmental agents to take into account are bordered. Steps 3 and following ones: Degradation of elements due to updated climatic or use factors Given the modification of structural or environmental conditions (step 2), the behaviour under new environmental conditions is then studied (action of environmental factors on elements due to structural modifications, mainly loss of protection): - corrosion of the coating due to water stresses, - Again we iteratively identify the modification of both structure and environment (step 2), then step 3, and so on For instance, once the glazing has failed (due to pressure inside stresses) the coating of the absorber is stressed by temperature, water, rain, snow, hail, ice, vertebrate, loads, wind (step 2). It will fail by corrosion, holing or breaking and then will not fulfil or fulfil partially its to absorb infrared radiation function anymore. The infrared radiation will not be absorbed, not be changed into heat source, not be transmitted to the liquid coolant and then to the primary circuit. Consequently the solar panel no fulfil its to absorb and transmit the heat function, it will break down. The second type of causes is stated by experts. They include potential defects, negligence, errors due to materials (quality, chemical reaction between seals and glass frame) mean (unsuitable fixing means), method (surface cleanness ), middle (temperature, organic deposit for absorber), and manpower. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

205 ANNEX 4 page 20 of 23 pages 3.3 RESULTS We thus obtain: - information about the Nominal behaviour of the product in a given environment, - information on the degradation and failures, expressed as a list of degradation and failures (FMEA table) or a failure tree (with scenarios) We capitalize the results of the qualitative analysis in a Failure Mode and Effects Analysis (FMEA) table, presented on Table 1. Functions Components Stage Modes Causes Directs effects Indirect effects Table 1 Propose format of a FMEA table Capitalizing tool of chaining of degradations leading to the product failure: modes, causes and consequences of degradations for each component and each associated function In this table are listed for each element, the modes, causes and effects. In column Stage, we distinguish the single failures (stage = 1) and the complex failures (stage =i, i>1). The single failures are directly due to the aggression of the product by its environment, and the complex failures result to the order of degradations. The below table 2 is an extract of a failure modes and effects analysis of a solar panel system. This example aims to highlight the iterative principle of the failure modes and effects analysis. Indeed the stresses of ultraviolet radiation on the seals create a disintegration of the seals surface (step1) which draws to the holing of the seals (step 2). The step 3 is the failure of seals due to holing and that generates the corrosion of coating (step 4) and its failure (step 5). Table 2 Propose format of a FMEA table Capitalizing tool of chaining of degradations leading to the product failure: modes, causes and consequences of degradations for each component and each associated function NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

206 ANNEX 4 page 21 of 23 pages Graphical representations of the quantitative analysis results The qualitative analysis provides a list of all the potential failure scenario of a product. Those results are identified with difficulty in the FMEA table. That is the reason why we chose to develop clear and synthetic graphical representations of the qualitative analysis results. We develop two types of graph: the event-driven graph, which is the representation of the temporal evolution of degradations of components leading to the failure of the product; the failures tree, which is a deductive method (Hadj-Mabrouk 1997) (from the product failure to the origin causes) Event-driven graph. The Event-driven graph is composed of three parts : the initial state, the degradations states and the failure state. - The initial state (begin of the graph) represents the state of each component of the product at the beginning of its exploitation stage. This state takes into account all the potential degradations (identified with the process analysis) due to errors, mistakes, damages on the product occurred during the building process. - The degradations states regroup all the potential successions or concomitances of degradations of components from the initial state to the failure state. We also schematize the causes of degradations due to environmental agent solicitations. - The failure state (end of the graph) contains the various failures of the product, that is to say that the product is no more able to ensure one of the need s functions for which it was designed. The Figure 6 is an extract of an event-driven graph of a solar collector. This graph represents the failure scenario defined below (cf. Table 2). Figure 6 : extract of an event-driven graph of a solar collector NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

207 ANNEX 4 page 22 of 23 pages Failure tree The failure tree is built from the inverse reading of the FMEA table. We start from the end of the table (which corresponds to the failure state of the event-driven graph) and we search the previous degradations as we find the origin degradation (same as the step 1 of the eventdriven graph). The failure tree is a graphical representation of the deduction of the origin degradations from the failures of the product. It is useful when we search the origin of a specific failure or degradation of the product or one of its components Limits and perspectives Still now, one of the main limits of the method is the non integration of the time scale. Thus we don t know the temporal evolution of degradations of the product. This integration is crucial, as it can be seen in Figure 6. Indeed, on this extract of an event-driven graph of a solar panel system, we didn t quantify the time of a glazing cracking with those environmental solicitations, and the time of a surface degradation of seals. Consequently we represent them at the same step that is not realistic. We are searching to integrate the time within a quantitative analysis and then we have to quantify the kinetic of degradations of product components. Another aspect is that we don t evaluate the phenomenon intensity and the spacial repartition of degradations on the product and its components. The development of a criticality analysis will allow us to take into account those aspects and will permit to classify the failure scenario with a criticality scale and to focus on the more serious failure scenario. For each failure scenario we determinate a criticality level measured by means of three risk indicators: the occurrence probability P (probability to observe the degradation or the failure due to the identified causes), the detection probability D (chance to detect the failure by means of diagnosis, quality control) and the severity of consequences S (consequences of the failure in terms of economic, human aspects). The product of these three risk indicators, for each failure scenario, gives us a criticality level. Consequently we can order the different failure scenarios according this criticality level. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

208 ANNEX 4 page 23 of 23 pages 4. REFERENCES Part 1 : FMEA 1. CHEVALIER, J.L. Durabilité des composants de capteurs solaires, Compte rendu d activité 1978 n 1563, 294 p., avril CHEVALIER, J.L., RUBINSTEIN, M. Fiabilité et durabilité des capteurs solaires thermiques, Cahiers du CSTB 2077, livraison 269, mai CHEVALIER, J.L., Recommandations to prevent internal corrosion damage in solar systems, Rapport ECTS /85/367/CV, novembre Directives UEAtc pour l agrément des capteurs solaires à circulation de liquide, Bulletin des Avis Techniques, cahiers supl , juin CHEVALIER, J.L., DIETZ, R., PERRAD, Y. Assement of a short term ageing process for solar collectors, CSTB ISO/CD Solar Energy - Materials for flat plate collectors - Qualification test procedures for solar absorbers surface durability, 9 janv CHEVALIER, J.L. Revue des matériaux utilisés dans la réalisation des capteurs solaires : caractéristiques et durabilité, Rapport CSTB, Paris, CHEVALIER J.L. Fiabilité, durabilité, comportement en œuvre des capteurs solaires, AFME, Rapport, CSTB, Sophia-Antipolis, CHEVALIER J.L., DIETZ R., FILLOUX A. Compte rendu de la réunion du «Collector and system testing group», Athènes, , CSTB, Ecole des Mines de Paris, Rapport, Sophia- Antipolis, BOURDEAU L., CHEVALIER J.L., GSCHWIND M. Compte rendu de la 2 ème réunion du «Collector and system testing group», CSTB, Ecole Nationale Supérieure des Mines de Paris. Rapport, Sophia-Antipolis, TARDY B. Etude d un processus de vieillissement accéléré des capteurs solaires plans Comparaison avec un processus de vieillissement long, Rapport de stage, CSTB, Sophia- Antipolis, 1986 Part 2 : FMEA 1. Hadj-Mabrouk, H. (1997). L analyse préliminaire de risques. Paris. ED.: HERMES. 127 pages 2. ISO TC59/SC14 (1998) Buildings and constructed assets Service life planning. 3. Lair, J. Chevalier, J.L. (2002). Failure Mode Effect and Criticality Analysis for Risk Analysis (design) and Maintenance Planning (exploitation). 9th Durability of Building Materials and Components, March 2002, Brisbane, Australia. 4. Lair, J. Chevalier, J.L. Rilling, J. (2001). Operational methods for implementing durability in service planning frameworks. CIB World Building Congress, April 2001, Wellington, New- Zealand. 5. Lair, J. Le Téno, J.F. Boissier, D. (1999). Durability assessment of building systems. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

209 WORKING DOCUMENT STAGNATION TEST FOR COLLECTORS AND INTEGRAL COLLECTOR AND CONTAINER (Normative) A.1 SCOPE This Appendix sets out two methods for assessing the ability of a collector or integral collector and container to withstand temperatures close to the maximum temperatures that it will encounter under some or all of the following conditions: (a) (b) (c) NOTE: Either of the two methods set out in this Appendix may be used, but it should be noted that Method 1 is not suitable for use with collectors with heat pipes. When empty during installation; When empty during its service life; and When full of water but not being used in peak summer conditions Such temperatures occur during periods of no useful heat removal from the collector with high solar radiation and ambient temperatures. NOTES: 1 In order to more realistically model extreme conditions, the test conditions have been changed from 1200 W/m 2 radiation and 40 C ambient temperature to 1100 W/m 2 and 50 C. Collectors compliant with the former conditions are deemed to comply with the new conditions. The conditions in ISO for high temperature resistance test Sunny C are also deemed to comply. The effective environmental temperature for New Zealand is 40 C for Method 1. For New Zealand the conditions in the ISO for high temperature resistance test Sunny B are acceptable. However, the collector must be marked in accordance with Clause Error! Reference source not found. Error! Reference source not found. if this test method is used. 2 The Method 2 test has a lower radiation level and ambient temperature to account for the fact that the thermal radiation level received by the collector from the simulator lights is higher than from the natural sky. A.2 APPARATUS C2.1 Method 1 NOTE: This method is not appropriate for use with collectors with heat pipes. The following apparatus is required (see also Figure C1). (a) A pumped heat transfer loop using a suitable heat transfer liquid, with the collector forming part of the loop. A suitable heat transfer fluid is one that will remain in its liquid state at the stagnation temperature and the maximum operating pressure of the collector. (b) (c) A stand on which the collector will face normal to the direct beam solar radiation at solar noon. A suitable stand is described in AS/NZS or ISO Thermometers or temperature-measuring devices to measure the temperature of any critical materials or heat-sensitive components during the test. C2.2 Method 2 The following apparatus is required: (a) A solar simulator of suitable size. (b) (c) A temperature-controlled chamber in which the collector may be placed when exposed to the simulator. Thermometers or temperature-measuring devices to measure the temperature of any critical materials or heat-sensitive components during the test.

210 WORKING DOCUMENT A.3 PROCEDURE C3.1 Method 1 The procedure shall be as follows: NOTE: In this procedure, two approaches are made to determine the stagnation temperature (T s ). The first approach is detailed in Steps (a) and (b), and the second approach (which is a cross-check and correction) in Steps (c) and (d). (a) Test the collector for thermal performance in accordance with AS (b) From the constants supplied from Step (a) calculate, in accordance with the equation below, the stagnation temperature (Ts) when (i) (ii) (iii) the efficiency is zero; the total global radiation is 1100 W/m 2 ; and the effective environmental temperature is 50 C. The stagnation temperature (T s ) is calculated as follows: C3(1) (c) The values for the terms used in the equation are obtained by testing in accordance with AS/NZS As a cross-check for the stagnation temperature, proceed as follows: (i) (ii) Install the collector on the stand in accordance with the manufacturer s instructions and adjust its inclination to provide for maximum (clear sky) solar irradiation. Ensure all the air has been bled from the system. Measure the total global solar radiation (G) on the plane of the collector and the ambient temperature (T a ). Adjust the temperature of the fluid (T f ) entering the collector to C3(2) (iii) Measure the temperature difference (ΔT fluid ) of the fluid flowing through the collector and determine that the collector has reached steady state by plotting the collector outlet temperature versus time. If there is a temperature rise, increase the fluid temperature entering the collector by 5 C, or, if there is a temperature drop, reduce the fluid temperature entering the collector by 5 C. (iv) Measure G, T a, and the temperature difference across the collector. If there is still a rise or drop, repeat Step (iii). If the rise has become a drop or vice versa, then the true stagnation temperature has been saddled. (v) The correction to be applied to the predicted stagnation (T s ) is shown in Figure C2. (d) (e) Adjust the fluid temperature to T s plus the correction obtained in Step (c). Adjust the flow rate through the collector so that the average inlet temperatures and the average outlet temperature over the test period are both greater than the stagnation temperature (measured at least every 5 min). (f) Run the system at this flow rate and temperature for 12 h and then turn the pump off for 12 h. Continue this cycle of operation for 10 days. Ensure that the heat transfer fluid is at the calculated stagnation temperature prior to commencement of each cycle of operation. (g) (h) Visually inspect the collector daily and note any changes in its appearance. Terminate the test after 10 days or when there is evidence of structural or material deterioration that would impair the operation of the collector, whichever is sooner.

211 WORKING DOCUMENT (i) Carry out a test of the thermal performance of the collector in accordance with AS/NZS , at a single test point using a fluid inlet temperature of at least 50 C or a value of (T f T e )/G T, which is half of the stagnation value at G T = 1000 W/m 2 and T a = 25 C; whichever is the lower. Compare the results of this test with those obtained in Paragraph C3.1(a). C3.2 Method 2 The procedure shall be as follows: (a) Test the collector for thermal performance in accordance with AS (b) (c) (d) (e) (f) (g) (h) (i) Install the collector on the stand in a temperature-controlled chamber and adjust its inclination so that it receives normal incident radiation from the solar simulator. Ensure that all the air has been bled from the system. Ensure that the collector is full of heat transfer fluid. It is not necessary to have any flow through the collector for this test method. Ensure that the air temperature adjacent to the collector is greater than 38 C, or, for the New Zealand only test, greater than 30 C, when the lamps are operating. Adjust the solar simulator output so that the average radiation measured at 6 uniformly distributed points on the collector is 1050 W/m 2 or greater, with less than 20% variation across the aperture. Solar spectral lamps as specified in AS/NZS shall be used. NOTE: Alternative arc lamps may be used; however, these lamps have a higher long wave radiation output and will result in a more severe test for some covers and glazing seals. Run the system under these conditions with the simulator being operated 12 h on and 12 h off for 10 days. Visually inspect the collector daily and note any changes in its appearance. Terminate the test after 10 days or when there is evidence of structural or material deterioration that would impair the operation of the collector, whichever is sooner. Carry out a test of the thermal performance of the collector in accordance with AS/NZS , at a single test point using a fluid inlet temperature of at least 50 C or a value of (T f T e )/G T, which is half of the stagnation value at G T = 1000 W/m 2 and T a = 25 C; whichever is the lower. Compare the results of this test with those obtained in Paragraph C3.2(a). A.4 REPORTING OF RESULTS The following results shall be reported: (a) The make and model identification of the system of which the collector forms a part. (b) Full details of the temperature measurements and the dates and duration of the test. (c) Details of the condition of the collector following the test with particular regard to (i) (ii) (iii) (iv) any structural failure; any burning, scorching or heat shrinkage; any effect likely to impair the serviceability of the collector; and any degradation in performance as a result of the test.

212 WORKING DOCUMENT FIGURE C1 SCHEMATIC OF APPARATUS TO MEASURE EFFECTS OF PROLONGED STAGNATION TEMPERATURE (T s )

213 WP 4.1: INVESTIGATION ON TEST METHOD FOR A FLAT PLATE COLLECTOR New Generation of Solar Thermal Systems Reviewer: Peter Kovacs January 2007 CONTENTS 1. INTRODUCTION 2. INVESTIGATED SOLAR COLLECTORS 3. INFLUENCE OF SOLAR COLLECTOR FLUID VOLUME FLOW RATE 4. INFLUENCE OF THE METHOD TO DETERMINE THE MEAN SOLAR COLLECTOR FLUID TEMPERATURE 5. INFLUENCE OF THE METHOD TO DETERMINE THE SPECIFIC HEAT OF THE SOLAR COLLECTOR FLUID 6. INFLUENCE OF THE TEMPERATURE LEVELS USED IN THE TEST 7. INFLUENCE OF THE WEATHER CONDITIONS USED IN THE TEST 8. CONCLUSIONS REFERENCES SUMMARY The test method of the standard EN (European Committee for Standardization, 2004) is used by European test laboratories to determine the efficiency of solar collectors. The aim of this work is to present an evaluation of the test method for a 12.5 m² flat plate solar collector panel, types HT and HTU, from Arcon Solvarme A/S. CFD (Computational Fluid Dynamics) simulations, calculations with a solar collector simulation program SOLEFF and thermal experiments are carried out in the investigation. The influence of flow nonuniformity on the efficiencies of the solar collector is elucidated for different volume flow rates and weather conditions. The influences of the method to determine the mean solar collector fluid temperature, the approximation used to determine the specific heat of the solar collector fluid, the temperature levels used in the tests and the weather conditions on the collector efficiency are investigated. Based on the investigations, it is concluded that with a volume flow rate between 6.0 l/min and 10.0 l/min (corresponding to 0.48 and 0.80 l/min per m 2 solar collector) the collector investigated will have the best efficiency. The investigations indicate that the maximum temperature level used in collector efficiency tests should not be higher than the maximum operation temperature of the collector. Further, if the efficiency of solar collectors is measured for low flow rates in future test methods the investigations of the HT and HTU collectors indicate that it might be needed to change the test method. It is recommended to investigate more solar collectors in order to elucidate if changes of the test method are needed.

214 ANNEX 6 page 2 of 21 pages 1. Introduction The thermal performance of flat-plate collectors is strongly related to the flow distribution through the absorber tubes /Duf91/. The more uniform the flow distribution, the higher the collector efficiency. However, it is shown by numerous investigations that uniform flow distribution is not always present in solar collectors /Jon94/,/Wan90/, /Wei02/, /Fan05/. The flow distribution through the tubes are influenced by the design of the collectors, the collector tilt and operating conditions of the collector such as volume flow rate and properties of solar collector fluid. The aim of this work is to theoretically and experimentally investigate the flow and temperature distribution in a solar collector panel with an absorber consisting of horizontal fins. CFD simulations are used in the determination of the collector efficiency at different collector fluid flow rates. Based on the CFD calculated efficiencies, the collector efficiency expression is obtained by means of regression. The influence of collector fluid flow rate on the flow distribution and on the collector efficiency will be elucidated. In the test method of the standard EN /Eur04/, the mean solar collector fluid temperature in the solar collector, T m is determined by the approximate equation: T m = ( Tin + Tout ) 2, where T in is the inlet temperature to the collector and T out is the outlet temperature from the collector. The specific heat of the solar collector fluid is a function of the temperature of the fluid. In the test method the specific heat of the solar collector fluid is an approximation for each measuring period determined as a constant equal to the specific heat of the solar collector fluid at the temperature T m. The power produced by the solar collector in a steady state test period is determined by the product of the specific heat, the mass flow rate and the temperature increase of the solar collector fluid. The solar collector efficiency is determined by measurements at different temperature levels. Based on these efficiencies, an efficiency equation is determined by regression analysis. In the test method, there are no requirements on the ambient air temperature and the sky temperature. This work will present an evaluation of the test method for a 12.5 m² flat plate solar collector panel from Arcon Solvarme A/S. The investigations will elucidate: How the collector fluid volume flow rate is influencing the flow distribution and the collector efficiency. How the mean solar collector fluid temperature T m is underestimated by the approximate equation in the test standard and how the collector efficiency equation is influenced by the underestimation of T m. The dependence of the volume flow rate is shown. How the use of the approximate specific heat of the solar collector fluid is influencing the collector efficiency. How the temperature levels used in the tests are influencing the collector efficiency expression. How the measured collector efficiency is influenced by the weather conditions such as the ambient air temperature and the sky temperature. 2. Investigated solar collectors The investigated solar collectors are 12.5 m² solar collector panels, type HTU and HT from Arcon Solvarme A/S, designed for medium and large solar heating systems. Fig. 1 shows the design of the HTU solar collector. The HTU solar collector is tested side-by-side with the similar collector HT, which includes a Teflon foil between the absorber and the cover glass. Each collector consists of two manifolds, one dividing and one combining manifold, and 16 parallel connected horizontal fins in a U type configuration, see Fig. 2. The well known Sunstrips are used as fins and the collector is equipped with a low iron antireflection treated glass cover. A propylene glycol/water mixture is used as solar collector fluid. Properties of the NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

215 ANNEX 6 page 3 of 21 pages mixture and their dependences on temperature for a 40% (weight %) glycol solution are as follows /Fur97/: Density, [kg/m 3 ] Dynamic viscosity, [kg/(ms)] Specific heat, [J/(kgK)] Thermal conductivity, [W/(mK)] where T is fluid temperature, [ C]. 2 ρ = * T * T T µ = *( ) C p = * T * T λ = * T 2 Fig. 1. Design and dimensions of the investigated HTU collector Fig. 2. A schematic illustration of the HTU collector configuration. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

216 ANNEX 6 page 4 of 21 pages The solar collector fluid is forced to circulate through the collector panel by a pump. The circulating flow rate, in the range of l/min, is measured using a QEC type Clorius flow meter. The flow distribution through the tubes is difficult to measure directly, therefore it is evaluated by temperature measurement just before the fluid enters the combining manifold. Copper-constantan thermo couples, type TT, attached to the back of the tubes are used for the temperature measurements. The positions of the measurement points are schematically shown in Fig. 2 as circles in the collector panel. Since the absorber tube wall has a higher temperature than the fluid inside, small corrections are made to the temperature measurements to get the fluid temperatures. The corrections are determined based on the difference between the measured solar collector fluid outlet temperature and the mean of the measured absorber tube wall temperatures. Solar irradiance on the collector panel is measured using a calibrated pyranometer type CM11 from Kipp & Zonen. The data collection and control program IMPVIEW is used to measure the volume flow rate and temperature as well as the temperature increase over the collector and the ambient temperature and solar irradiance during the steady state test periods. 3. Influence of solar collector fluid volume flow rate Investigations on how the test conditions and the approximate methods used in the test method to determine the collector efficiency will influence the efficiency and the efficiency expression for the HTU and HT collector will be described in the following sections. A 40% propylene glycol/water mixture is assumed to be the solar collector fluid. A collector tilt of 40 is assumed. 3.1 CFD model description The flow distribution through the absorber tubes is investigated theoretically with CFD calculations. A simplified model is built using the CFD code Fluent 6.1 /Flu03/, where the manifolds and the absorber strips are fully modelled, while the existence of the collector casing is represented by a heat loss from the absorber strips to the surrounding air. Due to the large difference in the dimension of the absorber tube length (5.790 m) and the tube hydraulic diameter ( m), a refined grid distribution is needed in the cross section of the tube. Fig. 3 shows grid distribution for the absorber strips and the manifolds. View A shows grid distribution at the vertical cut-plane of both the absorber strips and the manifold. View B and View C show grid setup at the cross-section of the manifold and the absorber strips respectively. Positions of View plane A, B and C are schematically shown in Fig. 2. As shown in Fig.3, the absorber strips including the tubes and the absorber fins are meshed with hexahedron cells, while the manifolds are meshed with a mixed mesh of tetrahedron and wedge cells. Since flow distribution through the tubes is of interest, the mesh density of the tubes (1.4E-8 m 3 /cell on average) and the manifolds (1.1E-8 m 3 /cell on average) are higher than that of the absorber fins (6.9E-8 m 3 /cell on average), which results in about 0.8 million mesh cells for the whole model. The flow distribution through the absorber tubes is investigated with heat transfer and buoyancy effect considered. The conductive heat transfer of absorbed solar energy along the absorber fin and from the absorber fin to the tube walls as well as the convective heat transfer from the tube walls to the collector fluid are included in the CFD calculation. The absorber strips are considered to be surrounded by air. The heat loss from the absorber strips is strongly related to the air flow conditions inside the panel, which are quite complicated and varied for different operating conditions. This makes determination of the heat loss from the absorber strips difficult. In the CFD calculations, the heat loss from the absorber strips is assumed to take place only by means of convection. The convective heat loss can be determined by means of the heat loss coefficient and the air temperature inside the collector. The heat loss coefficient is determined by thermal measurements and the air temperature in the collector is obtained from a solar collector calculation program, SOLEFF /Ras96/. The influence of the collector fluid volume flow rate on the air temperature distribution inside the collector panel is included in the calculations. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

217 ANNEX 6 page 5 of 21 pages View A View B View C Fig. 3. Grid setup of the solar collector model. Steady state CFD simulations are performed with Boussinesq approximation for buoyancy modelling. The PRESTO and second order upwind method are used for discretization of the pressure and the momentum equations respectively. The SIMPLE algorithm is used to treat the pressure-velocity coupling (Fluent Inc., 2003). One simulation takes approximately 12 hours for a computer with 3 GHz CPU frequency and 1G memory. The fluid flow through the i th tube (counted from the top) is characterized by a parameter β i, defined as follows: Q i i = Q 0 β (1) where Q i is the volume flow rate through the i th tube, while Q 0 is the overall volume flow rate for all the tubes. For an ideal, uniform flow distribution through 16 tubes, β i equals to 1/16. However, this is not always the case in the solar collector. Instead, strong deviations from uniform flow distribution have been observed in this work as well as in the numerous investigations of other collectors in the literature. Φ, a relative flow non-uniformity parameter, is introduced to quantify the flow maldistribution. Φ 16 2 ( β i 1/16) 1 is defined as: i= 1 (2) Φ = 16 ( 1 ) 100% 16 From its definition, it can be seen that the relative flow non-uniformity parameter is similar to the root-mean-square or standard deviation generally used in the statistical treatment of experiment data, except it is divided by the mean value (1/16). NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

218 ANNEX 6 page 6 of 21 pages 3.2 CFD calculated flow and temperature distribution The flow and temperature distribution through the 16 parallel absorber tubes are investigated under different operating conditions. The parameters investigated are volume flow rates: from 2.7 l/min to 25.0 l/min; solar collector fluid inlet temperatures from 20 C to 100 C. Fig. 4 shows CFD calculated flow and temperature distribution in the strips just before the fluid enters the combining manifold. The solar collector has an inlet flow rate of 2.7 l/min and an inlet fluid temperature of 19.4ºC. The solar irradiance is 883 W/m 2 and the ambient temperature is 30.0ºC. It can be seen that the flow distribution through the tubes is not uniform. Fluid flow rate in the strip increases from the top to the bottom. At the top strip 1, the flow rate is 0.11 l/min, while at the bottom strip 16 the flow rate is 0.23 l/min, approx. twice of that at the top strip. The fluid temperature decreases from the top to the bottom. At the top strip 1, the fluid temperature can be as high as 86.6ºC compared with a temperature of 54.6ºC at the bottom strip. The temperature of the tube wall is approx. 5 K higher than the solar collector fluid temperature. It should be pointed out that all the temperatures are calculated by averaging all the computational cells at the cross section with cell masses weighted. The temperature distribution of the collector panel in the middle of the absorber is presented in Fig. 5. The uneven flow distribution through the tubes results in an uneven temperature distribution. At the top of the collector and close to the outlet, the fluid and the absorber fin temperature are high, while the temperature is lower at the bottom of the collector. The outlet fluid temperature is 64.5ºC. As can be seen from Fig. 4, the minimum and the maximum temperature of the fluid before it enters the combining manifold is 54.6ºC and 86.6ºC respectively, correspondingly 9.9 K lower and 22.1 K higher than the outlet temperature. The temperature difference between the absorber fin and the collector fluid can be seen in the temperature profile Temperature, ºC Temperature of tube wall Fluid temperature Volume flow rate Flow rate 2.71 l/min T inlet =19.4ºC T outlet=64.3ºc T ambient=21.7ºc Solar irradiance G 883 W/m Volume flow rate, l/min 10 0 top bottom Strip number, i th Fig. 4. Temperature and volume flow rate distributions in the strips just before the fluid enters the combining manifold NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

219 ANNEX 6 page 7 of 21 pages Fig. 5. CFD calculated temperature distribution ( C) in the middle of the absorber The CFD model was used to calculate heat and fluid flow in the collector panel at different collector fluid flow rates. The investigations are carried out with a solar collector with a slope of 40, a solar irradiance of 1000W/m 2 and an ambient temperature of 30ºC. The investigated parameters include: Collector volume flow rates 3.3 l/min, 4.0 l/min, 6.0 l/min, 10.0 l/min and 25.0 l/min and inlet fluid temperatures 20ºC, 40ºC, 60ºC, 80ºC and 100ºC. The flow and temperature distribution through the absorber tubes just before the fluid enters the combining manifold are shown in Fig. 6 and Fig. 7 respectively. The flow and temperature distribution are calculated with an inlet temperature of 60ºC. Two driving forces control the flow distribution through the absorber tubes: The buoyancy force and the frictional drag force. For low flow rates such as 3.3 l/min and 4.0 l/min, the buoyancy effect is significant, which tends to circulate the collector fluid clockwise in the collector panel, thus decreasing the flow rate in the upper absorber tubes and increasing the flow rate in the lower absorber tubes. For high flow rates such as 25.0 l/min, the frictional drag force tend to be more significant, therefore the flow rate in the top tubes are larger due to less friction loss compared with the fluid flow through the bottom tubes. The temperature distribution of the collector fluid at the end of the absorber tubes is shown in Fig. 7. When the flow rate decreases from 25.0 l/min to 3.3 l/min, the tilt of the temperature distribution curve will increase dramatically. When the volume flow rate is 4.0 l/min, the temperature at the upper part of the collector will have a sharp increase, which will get worse with a decreasing flow rate. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

220 ANNEX 6 page 8 of 21 pages Flow rate 3.3 l/min Flow rate 4.0 l/min Flow rate 6.0 l/min Flow rate 10.0 l/min Flow rate 25.0 l/min Uniform flow distribution Φ = 40.2% Φ = 23.9% Φ = 6.9% β i 0.06 Φ = 2.9% Φ = 7.8% Collector tilt angle 40, solar irradiance 1000 W/m 2, T ambient 30 C Inlet fluid temperature 60 C top bottom Strip number, i th Fig. 6. The flow distribution through the absorber tubes at different collector volume flow rates. 160 Fluid temperature, C Flow rate 3.3 l/min Flow rate 4.0 l/min Flow rate 6.0 l/min Flow rate 10.0 l/min Flow rate 25.0 l/min Collector tilt angle 40, solar irradiance 1000 W/m 2 T ambient 30 C, Inlet fluid temperature 60 C top bottom Strip number, i th Fig. 7. The fluid temperature distribution through the absorber tubes at different collector volume flow rates. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

221 ANNEX 6 page 9 of 21 pages 3.4 Collector efficiency Measurements are carried out with the HTU collector at different collector volume flow rates and fluid inlet temperatures and under different weather conditions. Collector efficiencies are determined according to the test standard /Eur04/. The measured collector efficiencies are compared to the CFD calculations. Fig. 8 shows the comparison between the measured and the calculated efficiencies. The collector mean temperature is calculated by averaging inlet and outlet fluid temperatures. There is a good agreement between the measured and calculated efficiencies with a disagreement of less than 4% point. The calculated temperature increase over solar collector is very close to the measured one with a difference of less than 0.50 K. The uneven flow and temperature distribution will decrease the collector efficiency. Fig. 9 shows the CFD calculated collector efficiency at different collector volume flow rates. Boiling is not considered in the CFD calculations. The Reynolds number of the fluid flow in the tube will be in the range of for a volume flow rate between 3.3 l/min and 25.0 l/min if a solar collector fluid of 40% glycol water mixture and a mean collector fluid temperature of 60ºC are used. Therefore a laminar flow model is used in the CFD calculations. Based on the CFD calculated efficiencies, a regression analysis is carried out to find the efficiency expressions of the collector at an incidence angle of 0º operating at different fluid flow rates (see the following 2 Tm Ta ( Tm T a) table 1): η = η0 a1 * a2 * (3) G G Table 1: collector efficiency expressions of the collector at an incidence angle of 0º and at different fluid flow rates Flow rate 3.3 l/min 4.0 l/min 6.0 l/min 10.0 l/min 25.0 l/min η a 1 W/m 2 K a 2 W/m 2 K It can be seen that collectors at flow rates of 6.0 l/min and 10.0 l/min give the highest efficiency, while collectors at flow rates of 3.3 l/min and 25.0 l/min gives lower efficiency. This is due to the uneven flow distribution when the flow rate is too high or too low. The relative flow nonuniformity for flow rates of 3.3 l/min and 25.0 l/min are 40.2% and 7.8% respectively. If the flow rate is low enough, the flow nonuniformity will dramatically increase resulting in an increased collector mean temperature and a decreased collector efficiency. For high flow rates ( >10.0 l/min), the flow nonuniformity will also increase resulting in a decreased collector efficiency. Besides the flow distribution problems, the air temperature distribution inside the collector panel, which varies for different collector volume flow rates, will also influence collector efficiency. That is the reason why the collector efficiency is relatively low at the flow rate of 25.0 l/min. However, if the flow rate is higher than 25.0 l/min, the fluid flow in the tubes will normally be turbulent. For such high flow rates, the collector efficiency will have a sharp increase because the heat transfer coefficient from the tube wall to the fluid will increase dramatically as the fluid flow transits from laminar to turbulent region. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

222 ANNEX 6 page 10 of 21 pages Case 2 Case 1 CFD Measurement 60 Case 3 Efficiency, % Case 4 Case 6 Case (Tm-Ta)/G, [Km2/W] Fig. 8. Measured and calculated collector efficiencies Case 1: Inlet flow rate 2.7 l/min, T inlet =19.4ºC, T outlet =64.3ºC T ambient =21.7ºC, G=883 W/m 2 Case 2: Inlet flow rate 3.4 l/min, T inlet =20.0ºC, T outlet =57.6ºC T ambient =24.5ºC, G=866 W/m 2 Case 3: Inlet flow rate 5.0 l/min, T inlet =41.2ºC, T outlet =69.1ºC T ambient =27.8ºC, G=1000 W/m 2 Case 4: Inlet flow rate 4.9 l/min, T inlet =68.7ºC, T outlet =89.9ºC T ambient =27.8ºC, G=919 W/m 2 Case 5: Inlet flow rate 10.2 l/min, T inlet =85.9ºC, T outlet =93.4ºC T ambient =25.3ºC, G=822 W/m 2 Case 6: Inlet flow rate 24.3 l/min, T inlet =86.7ºC, T outlet =91.3ºC T ambient =22.7ºC, G=1018 W/m 2 Collector efficiency, [-] CFD calculation - flow rate 3.3 l/min CFD calculation - flow rate 4.0 l/min CFD calculation - flow rate 6.0 l/min CFD calculation - flow rate 10.0 l/min CFD calculation - flow rate 25.0 l/min Regression - flow rate 3.3 l/min Regression - flow rate 4.0 l/min Regression - flow rate 6.0 l/min Regression - flow rate 10.0 l/min Regression - flow rate 25.0 l/min (T m -T a )/G, [Km 2 /W] Fig. 9. the CFD calculated collector efficiency and the efficiency expression at different collector fluid flow rates The flow and temperature distribution through the absorber tubes in a HTU solar collector panel is investigated experimentally and theoretically. Results show that the flow and temperature are NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

223 ANNEX 6 page 11 of 21 pages not uniformly distributed. The uneven distribution is controlled by two driving forces: The buoyancy force and the frictional drag force. For low flow rates (<6.0 l/min, corresponding to 0.48 l/min per m 2 solar collector), the buoyancy effect is significant. The flow and temperature distribution get worse with the decrease of collector volume flow rate, resulting in a decreased collector efficiency. For higher flow rates (> 10.0 l/min, corresponding to 0.80 l/min per m 2 solar collector), the friction drag force tend to be more dominant, the flow distribution get worse with the increase of collector fluid flow rate. This will result in a decreased collector efficiency as long as the flow in the tubes is laminar. It is concluded that with a volume flow rate between 6.0 l/min and 10.0 l/min the collector will have the best efficiency. 4. Influence of the method to determine the mean solar collector fluid temperature CFD calculations have been carried out in order to determine the flow distribution as well as the solar collector fluid temperatures in the HTU collector for different conditions /Fan05/. The CFD model has been validated by temperature measurements. By means of the calculations the mean solar collector fluid temperature in the solar collector can be determined. Figure show results from the calculations. In these calculations, the solar irradiance is 1000 W/m² and the ambient air temperature is 30 C. From Fig. 10 it can be seen that for a low flow rate there are large temperature differences inside the collector and that the fluid temperatures are much higher in the upper strips than in the lower strips of the collector. The temperature profile in the strips from the inlet side to the outlet side is not linear. Fig. 11 shows that the mean solar collector temperature is higher than (T in +T out )/2 which in the test method is used as the mean solar collector fluid temperature, due to the nonlinear temperature profile in the strips. For decreasing flow rate the underestimation of the mean solar collector fluid temperature is increasing T ambient 30 C, solar irradiance 1000 W/m Fluid temperature, C Strip Strip 06 Strip Strip 16 Outlet side Inlet side Distance from the combining manifold, m Fig. 10. Calculated solar collector fluid temperatures inside 4 strips: The top strip, strip no. 6 from the top, strip no. 10 from the top and the bottom strip for an inlet temperature of 40 C and a flow rate of 2.9 l/min. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

224 ANNEX 6 page 12 of 21 pages Temperature difference, K CFD: collector tilt angle 40, solar irradiance 1000 W/m 2 T ambient 30 C, collector fluid 40% glycol/water mixture Collector flow rate 3.3 l/min Collector flow rate 4 l/min Collector flow rate 6 l/min Collector flow rate 10 l/min Collector flow rate 25 l/min Inlet temperature, C Fig. 11. Difference between mean solar collector fluid temperature determined by CFD calculations and (T in +T out )/2 for different flow rates and inlet temperatures. Fig. 12 and Fig. 15 show efficiency data points and efficiency expressions for different collector fluid flow rates. The efficiency expressions are obtained by regression based on the data points. From Fig. 12 it is seen that for a low flow rate of 3.3 l/min the use of (T in +T out )/2 as the mean solar collector fluid temperature results in a too low efficiency, especially at high temperature levels. Yearly thermal performance of the HTU collector is calculated with the two efficiency expressions: one is determined by the approximated mean solar collector fluid temperature (T in +T out )/2; the other is determined by the real solar collector fluid mean temperature. Collector efficiency, [-] Mean fluid temperature T m =(T inlet +T oulet )/2 η = *(T m -T a )/G *(T m -T a ) 2 /G Real mean fluid temperature η = *(T m -T a )/G Flow rate 3.3 l/min, solar irrdiance 1000 W/m 2, T ambient 30 C (T m -T a )/G, [Km 2 /W] Fig. 12. Efficiency expression based on T mean as the real mean solar collector fluid temperature and as (T in +T out )/2 for a flow rate of 3.3 l/min. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

225 ANNEX 6 page 13 of 21 pages Fig. 13 shows yearly thermal performance of the HTU collector as a function of a constant mean solar collector fluid temperature throughout the year. It can be seen that the thermal performance of the collector will be underestimated if the collector efficiency is determined by the approximation (T in +T out )/2, especially for high solar collector mean temperatures. Fig. 14 shows the relative thermal performance of the two HTU solar collectors as presented in Fig. 13. The underestimation of yearly thermal performance by the approximation method to determine collector fluid mean temperature is varying from 1% to 49% for a mean solar collector fluid temperature between 10 and 100 C. Further, if the difference between the real mean solar collector fluid temperature and the approximate mean solar collector fluid temperature, (T in +T out )/2 is high, it is not likely that the normal used equation for the collector efficiency for all weather conditions is suitable for determining the collector efficiency. Therefore the test method is not suitable for low flow rates Mean fluid temperature Real mean fluid temperature T m =(T inlet +T oulet )/2 Yearly thermal performance, [kwh/(year*m 2 )] Mean solar collector fluid temperature, [ C] Fig. 13 Yearly thermal performance of the HTU collector for efficiency expressions determined in different ways. The calculations are based on the weather data from the Danish Design Reference Year Mean fluid temperature Real fluid temperature T m =(T inlet +T oulet )/ Performance ratio, [-] Solar collector fluid mean temperature, [ C] Fig. 14. Relative thermal performance of the HTU collector for efficiency expressions determined in different ways. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

226 ANNEX 6 page 14 of 21 pages From fig. 15, it is seen that the error caused by the approximate equation is of no importance if the flow rate is 6 l/min. It can be concluded that the normally used approximation T ( T + T ) 2 m = only will result in a wrong efficiency expression if the flow rate is lower than 6 l/min, corresponding to 0.48 l/min per m² collector. For low flow rates the approximation will result in too low efficiencies especially at high temperature levels. In the test method a recommended flow rate of 1.2 l/min per m² collector is normally used. Consequently, the approximation does not require any changes of the test method. Changes will only be needed if low flow rates will be used in future test methods. in out Collector efficiency, [-] Flow rate 6.0 l/min, solar irrdiance 1000 W/m 2, T ambient 30 C Mean fluid temperature T m =(T inlet +T oulet )/2 η = 0, *(T m -T a )/G Real mean fluid temperature η = 0, *(T m -T a )/G (T m -T a )/G, [Km 2 /W] Fig. 15. Efficiency expression based on T mean as the real mean solar collector fluid temperature and as (T in +T out )/2 for a flow rate of 6.0 l/min. 5. Influence of the method to determine the specific heat of the solar collector fluid The specific heat of a 40 % propylene glycol/water mixture is determined by /Fur97/: C p ( T ) = * T * T², [J/kg *K] where T is the fluid temperature, [ C] Toutlet The power produced by the solar collector is found by: Q 1 = ρ * v * C p( T ) dt, [W] Tinlet where ρ is the density of the fluid at the temperature of the fluid in the flow meter, [kg/m 3 ] and v is the volume flow rate, [m 3 /s] The test method makes use of the following approximate equation by calculation of the power from the solar collector: Q 2 = ρ * v * C p,, [W] T where C, is the specific heat of the fluid at the temperature (T in +T out )/2, [J/kg *K] p T Table 2 shows measured collector efficiencies found by both methods for different temperature levels and flow rates. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

227 ANNEX 6 page 15 of 21 pages Table 2. Collector efficiencies calculated by two different methods to determine the specific heat of the solar collector fluid for the HTU collector. Flow T inlet T outlet T mean Q 2 η 2 Q 1 η 1 rate l/min C C C W % W % From the table it is concluded that the error introduced by the approximate method used to determine the specific heat of the solar collector fluid is insignificant. 6. Influence of the temperature levels used in the test Thermal measurements have been carried out to investigate how the temperature levels used in the test will influence the efficiency expression. Fig. 16 shows results of the measurements with the HTU collector and a similar collector, type HT which includes a Teflon foil between the absorber and the cover glass. Five temperature levels (mean solar collector fluid temperature) are used in the test: Group 1 (29 C), group 2 (45 C), group 3 (64 C), group 4 (84 C) and group 5 (94 C). The collector fluid volume flow rate is 25.0 l/min, corresponding to 2.0 l/min per m 2 collector. The full curves show efficiency expressions of the HT and HTU solar collectors determined by means of regression analysis based on the temperature levels 1, 2, 3, 4, while the dashed curves show the efficiency expressions determined based on the temperature levels 1, 2, 3, 5. It can be seen that the efficiency expression determined by the high temperature levels has a higher heat loss coefficient than that determined by the temperature levels 1, 2, 3, 4. This is due to the fact that there is a sharp decrease of the measured collector efficiency at the high temperature level 5, caused by boiling in one of the strips. The boiling is only discovered by means of temperature sensors placed inside the solar collector. Efficiency [%] Solar radiation 975 W/m 2 Regression HT based on group 1, 2, 3, 4 Regression HTU based on group 1, 2, 3, 4 Regression HT based on group 1, 2, 3, 5 Regression HTU based on group 1, 2, 3, 5 Measurement HT Measurement HTU 5 Regression based on group 1, 2, 3, 4 Regression based on group 1, 2, 3, 5 HT η = *(T m -T a )/G HTU η = *(T m -T a )/G *(T m -T a ) 2 /G HT η = *(T m -T a )/G *(T m -T a ) 2 /G HTU η = *(T m -T a )/G *(T m -T a ) 2 /G (T m -T a )/G, [Km 2 /W] Fig. 16. Efficiency expressions of the HT and HTU solar collectors determined by tests at different temperature levels. Yearly thermal performance of the HT collector is calculated with the two efficiency expressions: One determined based on the temperature level 1, 2, 3, 4 and one determined based on the NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

228 ANNEX 6 page 16 of 21 pages temperature levels 1, 2, 3, 5. The collector is facing south and the collector tilt is 40. The calculations are based on the weather data from the Danish Design Reference Year. Fig. 17 shows yearly thermal performance of the HT collector as a function of a constant mean solar collector fluid temperature throughout the year. The yearly thermal performance calculated with the efficiency expression determined by 1, 2, 3, 5 is lower than that with the efficiency expression determined by temperature levels 1, 2, 3, 4 except for a mean solar collector fluid temperature less than 10 C. Fig. 18 shows the relative thermal performance of the HT solar collector defined as the ratio of the yearly thermal performance of the solar collector with an efficiency determined at the temperature levels in question during the collector tests and the yearly thermal performance of the solar collector with an efficiency determined at the temperature levels 1, 2, 3 and 4 during the collector tests. It can be seen that the HT collector with efficiency expression determined at the high temperature levels will reduce thermal performance by 5% - 15% comparable to the efficiency expression determined at temperature levels 1, 2, 3 and 4 for a mean solar collector fluid temperature from 45 C to 84 C Yearly thermal performance, [kwh/(year*m 2 )] Based on efficiency determined by group 1, 2, 3, 4 Based on efficiency determined by group 1, 2, 3, Mean solar collector fluid temperature, [ C] Fig. 17. Yearly thermal performance of the HT collector for efficiency expressions determined at different temperature levels Relative performance, [-] Based on efficiency determined by group 1, 2, 3, 4 Based on efficiency determined by group 1, 2, 3, Mean solar collector fluid temperature, [ C] Fig. 18. Relative thermal performance of the HT collector for efficiency expressions determined at different temperature levels. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

229 ANNEX 6 page 17 of 21 pages Fig. 19 shows yearly thermal performance of the HTU collector as a function of a constant mean solar collector fluid temperature throughout the year. The yearly thermal performance calculated with the efficiency expression determined by 1, 2, 3, 5 is lower than that with the efficiency expression determined by temperature levels 1, 2, 3, 4 except for a mean solar collector fluid temperature between 15 and 55 C. Fig. 20 shows the relative thermal performance of the HTU solar collector defined as the ratio of the yearly thermal performance of the solar collector with an efficiency determined at the temperature levels in question during the collector tests and the yearly thermal performance of the solar collector with an efficiency determined at the temperature levels 1, 2, 3 and 4 during the collector tests. It can be seen that the HTU collector with efficiency expression determined at the high temperature levels will reduce thermal performance by up to 7% comparable to the efficiency expression determined at temperature levels 1, 2, 3 and 4 for a mean solar collector fluid temperature from 55 C to 84 C. It can be concluded that for a solar collector operating with a solar collector fluid temperature no larger than 84 C, the collector efficiency and the thermal performance of the collector will be underestimated by up to 15% for HT collector and up to 7% for HTU collector if a high temperature level like 94 C is used in the test to determine the collector efficiency. It is therefore recommended that the maximum temperature level used in the tests is not higher than the maximum operation temperature of the collector Yearly thermal performance, [kwh/(year*m 2 )] Based on efficiency determined by group 1, 2, 3, 4 Based on efficiency determined by group 1, 2, 3, Mean solar collector fluid temperature, [ C] Fig. 19. Yearly thermal performance of the HTU collector for efficiency expressions determined at different temperature levels. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

230 ANNEX 6 page 18 of 21 pages Relative performance, [-] Based on efficiency determined by group 1, 2, 3, 4 Based on efficiency determined by group 1, 2, 3, Mean solar collector fluid temperature, [ C] Fig. 20. Relative thermal performance of the HTU collector for efficiency expressions determined at different temperature levels. 7. Influence of the weather conditions used in the test Calculations with the program SOLEFF /Ras96/ are carried out in order to determine the collector efficiency for the HTU collector and the HT collector for different weather conditions. A collector fluid volume flow rate of 25.0 l/min, corresponding to 2.0 l/min per m 2 collector, is used in the calculations. Fig. 21 shows the calculated efficiencies for the HTU collector for different ambient air temperatures and sky temperatures. The sky temperature is assumed to be 15 K lower than the air temperature. It is seen that the efficiency for high temperature levels is decreased for increased temperature levels of the ambient air and the sky. Fig. 22 shows the calculated yearly thermal performance of the HTU collector without Teflon foil as a function of a constant mean solar collector fluid temperature throughout the year. The calculations are carried out with weather data from the Danish Design Reference Year based on two efficiency expressions: One determined by means of collector tests with an ambient air temperature of 15 C and a sky temperature of 0 C and one determined by means of collector tests with an ambient air temperature of 30 C and a sky temperature of 15 C. Fig. 23 shows the calculated relative performance of the HTU collector as a function of the mean solar collector fluid temperature. The relative performance is the ratio between the yearly thermal performance of the collector determined with an efficiency expression determined at the ambient air/sky temperatures in question and the yearly thermal performance of the collector determined with an efficiency expression tested at an ambient air temperature of 15 C and a sky temperature of 0 C. For solar collector fluid temperature levels between 15 C and 85 C the difference between the yearly thermal performances of the collector determined by means of the tests at the different weather conditions is lower than 3%. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

231 ANNEX 6 page 19 of 21 pages 90 Tambient=15 C, Tsky=0 C η = 0, *(T m -T a )/G *(T m -T a ) 2 /G 80 Tambient=30 C, Tsky=15 C η = 0, *(T m -T a )/G *(T m -T a ) 2 /G 70 Efficiency, [%] HTU collector without Teflon G = 800 W/m (T m -T a )/G, [Km 2 /W] Fig. 21. Calculated efficiency for the HTU collector for different ambient air and sky temperatures for a solar irradiance of 800 W/m² and an incidence angle of Yearly thermal performance, [kwh / (year * m 2 )] HTU collector without Teflon Tambient=15 C, Tsky=0 C Tambient=30 C, Tsky=15 C Mean solar collector fluid temperature, [ C] Fig. 22. Yearly thermal performance of the HTU collector for efficiency expressions determined with two different ambient and sky temperatures. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

232 ANNEX 6 page 20 of 21 pages Relative performance, [-] HTU collector without Teflon Tambient=15 C, Tsky=0 C Tambient=30 C, Tsky=15 C Mean solar collector fluid temperature, [ C] Fig. 23. Relative thermal performance of the HTU collector at different mean solar collector fluid temperatures. The calculations are carried out with the HT solar collector as well. It is shown that for solar collector fluid temperature levels between 15 C and 100 C the difference between the yearly thermal performances of this collector determined by means of the tests at the different weather conditions is lower than 4%. Consequently, taken the measuring accuracy into consideration, the weather conditions used in the tests will not significantly influence the calculated thermal performance of the collector. 8. Conclusion Investigations on the suitability of collector test methods have been carried out for two flat plate collectors from Arcon Solvarme A/S: The HTU collector without a Teflon layer and the HT collector with a Teflon layer. The weather conditions like the ambient air temperature and the sky temperature used in the tests will not significantly influence the calculated thermal performance of the collector. The error introduced by the approximate method used to determine the specific heat of the solar collector fluid is insignificant. The flow and temperature distribution through the absorber tubes in a HTU solar collector panel is investigated experimentally and theoretically. Results show that the flow and temperature are not uniformly distributed. The uneven distribution is controlled by two driving forces: The buoyancy force and the frictional drag force. For low flow rates (<6.0 l/min, corresponding to 0.48 l/min per m 2 solar collector), the buoyancy effect is significant. The flow and temperature distribution get worse with the decrease of collector volume flow rate, resulting in a decreased collector efficiency. For higher flow rates (> 10.0 l/min, corresponding to 0.80 l/min per m 2 solar collector), the friction drag force tend to be more dominant, the flow distribution get worse with the increase of collector fluid flow rate. This will result in a decreased collector efficiency as long as the flow in the tubes is laminar. It is concluded that with a volume flow rate between 6.0 l/min and 10.0 l/min the collector will have the best efficiency. For the HT solar collector operating at a collector fluid volume flow rate of 25.0 l/min (corresponding to 2.0 l/min per m 2 solar collector) and at solar collector fluid temperatures no larger than 84 C, the thermal performance of the collector will be underestimated by up to 15% for HT collector and up to 7% for HTU collector if a high temperature level like 94 C is used in the test to determine the collector efficiency. Consequently the investigations indicate that the NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

233 ANNEX 6 page 21 of 21 pages maximum temperature level used in the tests should not be higher than the maximum operation temperature of the collector. Further, if the collector efficiency is measured for low flow rates in future test method the investigations indicate that there might be needed to change the test method. It is recommended to investigate more solar collectors in order to elucidate if changes of the test method are needed. References /Duf91/ /Jon94/ /Wan90/ /Wei02/ /Fan05/ /Eur04/ Duffie J. A. and Beckman W. A. (1991): Solar Engineering of Thermal Processes, 2nd edition, John Wiley & Sons, New York, pp Jones G. F., Lior N. L. (1994): Flow distribution in manifolded solar collectors with negligible buoyancy effects, Solar Energy 52 (3), Wang X. A., Wu L. G. (1990): Analysis and performance of flat-plate solar collector arrays, Solar Energy 45 (2), Weitbrecht V., Lehmann D., Richter A. (2002): Flow distribution in solar collectors with laminar flow conditions, Solar Energy 73 (6), Fan J., Shah L. J., Furbo S. (2005): Flow distribution in a solar collector panel with horizontal fins, Proceedings of ISES 2005, Orlando, USA, European Committee for Standardization, (2004): Thermal solar systems and components Solar collectors Part 2: Test methods, EN /Fur97/ /Flu03/ /Ras96/ Furbo S. (1997): Varmelagre til solvarmeanlæg, Institut for Bygninger og Energi, Technical University of Denmark, pp Fluent Inc. (2003): Fluent release 6.1, 10 Cavendish Court, Lebanon, NH USA. Rasmussen P. B. Svendsen S. (1996): SolEff Program til beregning af solfangeres effektivitet. Brugervejledning og generel programdokumentation, Thermal Insulation Laboratory, Technical University of Denmark. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

234 WP 4.1 D2, D3 Improved Sky Temperature Measurement and Refined Testing Methods for Unglazed Collectors Author : Dr. Dieter Gottwald, arsenal research February 2006 CONTENTS: SKY TEMPERATURE DIRECT EFFECTS ON SOLAR COLLECTOR TESTS EFFECTS OF THE INCLINATION ANGLE INDOOR MEASUREMENT OF THE SKY TEMPERATURE RATIO ε/α SUMMARY In an effort to increase the accuracy of collector tests according to the European norm EN (from now on referred to as the norm) we propose improvements to the current test methods. The main focus of our investigations was to examine the influence of the sky temperature on the collector tests, especially on the tests of unglazed collectors. We propose several changes to the norm that will lead to more realistic test conditions and thus more accurate results. File: Annex 7 Unglazed collectors.doc NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

235 ANNEX 7 page 2 of 6 pages 1. SKY TEMPERATURE The sky temperature T sky is defined as the temperature of a black body that emits the same long-wave radiation as the atmosphere. The long wave irradiance in the collector plane E L is an important parameter for the calculation of the net irradiance on the collector G (equation 19 in the norm). The sky temperature depends on the atmospheric conditions and can be equivalent to the ambient temperature T amb in hot and humid climates, or can be up to 50K below T amb in cold and dry climates /Duf91/, which our detailed investigations /Bon04/ corroborated. 2. DIRECT EFFECT ON COLLECTOR TESTS 2.1 EFFECTS ON GLAZED COLLECTORS Section in the norm allows a thermal irradiance on a glazed collector during indoor tests that exceeds that of a black body at ambient temperature by up to 5% which corresponds to a sky temperature that is above the ambient temperature. Since outdoor measurements proved that the sky temperature is always lower than the ambient temperature we propose to change the norm to reduce the maximum thermal irradiance on the collector to the one of a black body at ambient temperature to ensure uniform and realistic test conditions for covered collectors. Figure 1: Position of the sensors and the nine measuring points at the artificial sky. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

236 ANNEX 7 page 3 of 6 pages Flow rate Lamp irradiation T m T amb T as σ as T amb - T as η [m³/h] [W/m²] [ C] [ C] [ C] [ C] [ C] [1] 472,60 831,22 30,46 21,44 17,77 0,05 3,67 0, ,38 833,78 30,40 21,07 17,87 0,04 3,20 0, ,68 839,25 30,42 21,20 18,08 0,07 3,12 0, ,74 838,64 30,46 21,35 18,27 0,05 3,08 0, ,75 835,19 30,56 21,09 18,88 0,06 2,21 0, ,88 833,37 30,61 20,04 18,83 0,05 1,21 0, ,02 834,09 30,69 19,94 18,76 0,02 1,18 0, ,93 832,22 30,75 20,04 19,34 0,32 0,70 0, ,02 827,22 30,80 20,08 20,36 0,28-0,28 0, ,62 829,47 30,71 20,08 21,09 0,16-1,01 0, ,45 827,71 30,60 20,09 21,49 0,08-1,40 0, ,35 829,34 30,46 20,18 21,74 0,06-1,55 0, ,54 828,37 30,33 20,19 21,88 0,02-1,69 0, ,41 827,66 30,24 20,18 21,99 0,07-1,82 0, ,28 827,44 30,25 20,19 22,99 0,49-2,80 0, ,56 830,43 30,30 20,24 24,33 0,30-4,10 0, ,52 827,25 30,31 20,35 25,08 0,16-4,73 0, ,43 827,63 30,32 20,45 25,45 0,06-5,00 0, ,38 827,24 30,36 20,54 25,60 0,07-5,06 0, ,09 826,36 30,41 20,56 27,44 0,88-6,89 0, ,86 823,03 30,38 20,57 28,39 0,75-7,82 0, ,79 819,99 30,36 20,64 27,68 0,71-7,04 0, ,80 823,73 30,41 20,77 30,11 0,65-9,34 0, ,68 824,91 30,42 20,81 29,97 0,77-9,16 0, ,57 826,16 30,37 20,78 27,62 0,58-6,83 0, ,57 830,10 30,39 20,72 26,00 0,38-5,28 0, ,91 833,14 30,40 20,68 24,94 0,25-4,26 0, ,79 830,26 30,35 20,70 24,26 0,14-3,56 0,481 Table 1: Effect of the temperature of the artificial sky T as on the efficiency η of an unglazed collector for constant flow rate, irradiation, mean temperature of the heat transfer fluid T m, and no wind /Bon04/. 2.1 EFFECTS ON UNGLAZED COLLECTORS The effect of long wave radiation on the testing results of unglazed collectors is much more pronounced than on glazed collectors (see table 1). Indoor measurements showed that the efficiency of unglazed collectors can increase by 5% absolute if the sky temperature is increased by ~12K. Measurements that are carried out with a too high fraction of long wave radiation will result in unrealistically high efficiencies. To prevent these errors we recommend changing the allowed relative long wave irradiation on unglazed collectors during indoor tests from ±50W.m -2 to no higher then 0 W.m -2 and no less than -100 W.m -2 to achieve testing conditions that are more similar to real conditions. 3. EFFECT OF THE INCLINATION ANGLE If the collector is inclined with an angle β>45 to the horizontal, the norm proposes to add a correction factor (1+cos β)/2 to the long wave irradiance which will decrease the irradiance. Our measurements showed that the effective sky temperature of an inclined collector increases due NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

237 ANNEX 7 page 4 of 6 pages Figure 2: Thermography of the artificial sky to the thermal radiation of surrounding objects that have a temperature at or even above the ambient temperature. Therefore we propose to drop the factor (1+cos β)/2 in equation (25) in the norm as the uncorrected value is a better approximation to the real value than the corrected one. 4. INDOOR MEASUREMENT OF THE SKY TEMPERATURE Section of the norm recommends a pyrgeometer that is mounted in the collector plane to measure the long wave irradiance. We did examinations at a simulator that was equipped with an artificial sky and measured the temperature of that sky with a pyrgeometer in the collector plane, and contact sensors that were radiation shielded and thermally insulated to the ambient air (see figure 1). The results obtained with the contact sensors showed that the temperature distribution on the surface was not constant, which was further substantiated with a thermography of the artificial sky (figure 2). This inhomogeneity caused the pyrgeometer to produce inaccurate results (see table 2). By means of further investigations, it has to be checked, if accurate results can also be obtained by a pyrometer that is oriented to the surface opposite of the collector. E pyrgeo -E contact v [m.s -1 ] 3 1,5 0 Measuring point [W.m - ²] 1 3,09 3,17 2,66 2 1,70 1,83 1,66 3 2,11 2,28 1,63 4 0,23 0,53 0,85 5-1,37-0,74 0,17 6 0,10 1,06 1,70 7 2,10 2,37 3,46 8 0,33 0,49 1,74 9 0,15 0,35 0,55 Table 2: Comparison between E contact, the irradiance corresponding to the temperature measured with the contact sensors, and E pyrgeo, the irradiance measured with the pyrgeometer for three different wind speeds. The measuring points are defined in figure 1 /Bon04/. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

238 ANNEX 7 page 5 of 6 pages The main disadvantages of pyrgeometers are that they are very shock- and touch-sensitive, and that space adjacent to the collector could be scarce, especially for unglazed collectors where a minimum area of 4m 2 has to be measured. Both contact sensors and pyrometers overcome these downsides and can improve the quality of the measurement. 5.0 RATIO ε/α If the ratio ε/α that is required in equation (19) in the norm is not known, the norm proposes to use a value of ε/α=0.95. However, recent measurements for unglazed collectors yielded an absorption ratio α=0.95±0.02 and an emission ratio of ε=0.8±0.05 /Roc01/ which leads to a ratio ε/α We recommend using this value for unglazed collectors if the manufacturer does not provide a measured value. 6.0 CONCLUSIONS We investigated the influence of the sky temperature on tests of solar thermal collectors and propose the following changes to the European norm EN 12975: Section : Change the allowed thermal irradiance on a glazed collector from 105% irradiance of a black body at ambient temperature to 100%. Section : Change the allowed relative long wave irradiation on an unglazed collector from ±50W.m -2 to no higher than +0 W.m -2 and no less than -100W.m -2 Section : Drop the factor (1+cos β)/2 from equation (25). Section : Recommend contact sensors to measure the temperature of the artificial sky. Section : If the value of ε/α is not known, change the suggested value from 0.95 to Furthermore, we would like to encourage tests regarding the usefulness of measuring the long wave irradiation with a pyrometer during indoor tests. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

239 ANNEX 7 page 6 of 6 pages References /Bon04/ /Duf91/ /Roc01/ Bonnet C. (2004): Untersuchung der Himmelstemperatur bei der Prüfung von Sonnenkollektoren, Diploma Thesis, Arsenal Research, Vienna. Duffie J.A. and Beckmann W.A. (1991): Solar Engineering of Thermal Processes, 2 nd ed., Wiley & Sons, New York Rockendorf G., Sillmann R., Bethe T., and Köln H. (2001): Solare Freibadheizung, Absorberprüfung und Testergebnisse, Anlagen Planung und Betrieb, Ist EnergiePlan GmbH, ISFH. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

240 ANNEX 8 page 1 of 41 pages EUROPEAN STANDARD FIRST DRAFT NORME EUROPEENNE EUROPÄISCHE NORM UDC Descriptors: English version Thermal solar systems and components Solar air collectors Test methods Draft standard for testing of solar air collectors based on EN and ASHRAE NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

241 ANNEX 8 page 2 of 41 pages Contents Page 1 SCOPE REFERENCES TERMS AND DEFINITIONS SYMBOLS AND UNITS RELIABILITY TESTING OF LIQUID HEATING COLLECTORS General Internal pressure tests for collectors Objective Apparatus and procedure Test conditions High-temperature resistance test Objective Apparatus and procedure Test conditions Results Exposure test Objective Apparatus and procedure Test conditions Results External thermal shock test Objective Apparatus and procedure Test conditions Results Rain penetration test Objective Apparatus and procedure Test conditions Results Mechanical load test Positive pressure test of the collector cover...17 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

242 ANNEX 8 page 3 of 41 pages Negative pressure test of fixings between the cover and the collector box Negative pressure test of collector mountings Impact resistance test Objective Apparatus and procedure Test conditions Results Final inspection Test report THERMAL PERFORMANCE TESTING OF AIR HEATING COLLECTORS Collector mounting and location General Collector mounting frame Tilt angle Collector orientation outdoors Shading from direct solar irradiance Diffuse and reflected solar irradiance Thermal irradiance Surrounding air speed Instrumentation Solar radiation measurement Thermal radiation measurement Temperature measurements Measurement of collector liquid flow rate Measurement of surrounding air speed Pressure measurements Humidity Measurement Elapsed time Instrumentation/data recorders Collector area Test installation General consideration Heat transfer fluid Test ducts Pump and flow control devices Air-Preconditioning Apparatus...30 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

243 ANNEX 8 page 4 of 41 pages Humidity Ratio Outdoor steady-state efficiency test Test installation Preconditioning of the collector Test conditions Test procedure Measurements Test period (steady-state) Presentation of results Computation and presentation of collector efficiency and thermal performance Steady-state efficiency test using a solar irradiance simulator General The solar irradiance simulator for steady-state efficiency testing Test installation Preconditioning of the collector Test procedure Measurements during tests in solar irradiance simulators Test period Test conditions Computation and presentation of results Determination of the effective thermal capacity and the time constant of a collector Incidence angle modifier Determination of the pressure drop across a collector Determination of the leakage rate Test Apparatus Presentation of results...40 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

244 ANNEX 8 page 5 of 41 pages Foreword This document has been prepared by Josef Buchinger as a resource document for WP4.1 of the NEGST project as recommendations on testing of solar air collectors. It is based on the EN (2002), the ASHRAE and experiences made during various testing of air collectors, mainly within the IEA SHCP Task 19 Solar Air Systems. This draft describes procedures for testing reliability and performance of glazed solar thermal air collectors operated either in open and closed loop systems with either positive or negative pressure applied. Whereby the focus on the work so far has been laid on the adoption of the sections describing the thermal performance tests, especially steady-state outdoor testing. This draft so far does not include procedures for unglazed solar thermal systems and is open for a revision with regards to air collectors made of polymeric or organic materials. Further the complete topic of testing incident angle modifiers is not covered by this document. Annexes as in the basis documents are not included in this draft and have to be adapted or developed newly where required. Introduction This standard specifies test methods for determining the ability of a solar air collector to resist the influence of degrading agents. It defines procedures for testing collectors under well-defined and repeatable conditions. This standard also provides test methods and calculation procedures for determining the steadystate thermal performance of glazed air heating solar collectors. It contains methods for conducting tests outdoors under natural solar irradiance and natural and simulated wind and for conducting tests indoors under simulated solar irradiance and wind. This standard will also provide methods for determining the thermal performance of unglazed air heating solar collectors. For unglazed absorbers, readily fabricated modules with a specific size are seldom used. Therefore, during the test, it is to be checked, that a realistic flow pattern and flow velocity is used. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

245 ANNEX 8 page 6 of 41 pages 1 Scope This European Standard specifies test methods for validating the durability, reliability and safety requirements for air heating collectors as specified in EN xxxx-1. It is not applicable to those collectors in which the thermal storage unit is an integral part of the collector to such an extent that the collection process cannot be separated from the storage process for the purpose of making measurements of these two processes. This standard applies to non concentrating solar air collectors in which a fluid enters the collector through a single inlet and leaves the collector through a single outlet. Collectors containing more than one inlet and more than one outlet may be tested according to this standard provided that the external piping or ducting can be connected so as to provide effectively a single inlet and a single outlet. Collectors that are custom built (built in; e.g. roof integrated collectors that do not compose of factory made modules and are assembled directly on the place of installation) cannot be tested in their actual form for durability, reliability and thermal performance according to this standard. Instead, a module with the same structure as the ready collector may be tested. The module gross area shall be at least 2m 2. The test is valid only for larger collectors, than the tested module. 2 References This European Standard incorporates, by dated or undated reference, provisions from other publications. The normative references are cited at the appropriate places in the text and the publications are listed hereafter. For dated references, subsequent amendments to or revisions of any of these publications apply to this European Standard only when incorporated in it by amendment or revision. For undated references the latest edition of the publication referred to applies. ISO 9060 Solar energy - Specification and classification of instruments for measuring hemispherical solar and direct solar radiation. ISO Test methods for solar collectors Part 1: Thermal performance of glazed liquid heating collectors including pressure drop. ISO Test methods for solar collectors Part 2: Qualification test procedures. ISO : 1995 Test methods for solar collectors - Part 3: Thermal performance of unglazed liquid heating collectors (sensible heat transfer only) including pressure drop. ISO 9846 Solar energy - Calibration of a pyranometer using a pyrheliometer. ISO 9847 Solar energy - Calibration of field pyranometers by comparison to a reference pyranometer. ISO/TR 9901 Solar energy - Field pyranometers - Recommended practice for use. EN ISO 9488 Solar Energy - Vocabulary (ISO 9488:1999) NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

246 ANNEX 8 page 7 of 41 pages EN :2000 Thermal solar systems and components - Solar collectors - Part 1: General requirements EN 12911:2000 Windows and doors : Resistance to wind load Test method 3 Terms and definitions For the purpose of this standard, the Terms and definitions given in EN ISO 9488 apply. 4 Symbols and units a 1 heat loss coefficient at (T m - T a )=0 Wm -2 K -1 a 2 temperature dependence of the heat loss coefficient Wm -2 K -2 A A absorber area of collector m 2 A a aperture area of collector m 2 A G gross area of collector m 2 AM optical air mass b u collector efficiency coefficient (wind dependence) m -1 s b o constant for the calculation of the incident angle modifier b 1 heat loss coefficient at (T m - T a )=0 Wm -2 K -1 b 2 collector efficiency coefficient Wsm -3 K -1 c 1 heat loss coefficient at (T m - T a )=0 Wm -2 K -1 c 2 temperature dependence of the heat loss coefficient Wm -2 K -2 c 3 wind speed dependence of the heat loss coefficient Jm -3 K -1 c 4 sky temperature dependence of the heat loss coefficient Wm -2 K -1 c 5 effective thermal capacity J m -2 K -1 c 6 wind dependence in the zero loss efficiency sm -1 c f specific heat capacity of heat transfer fluid Jkg -1 K -1 C effective thermal capacity of collector JK -1 D date YYMMDD E L longwave irradiance (λ >3µm) Wm -2 E β longwave irradiance on an inclined surface outdoors Wm -2 E s longwave irradiance Wm -2 F radiation view factor F collector efficiency factor G hemispherical solar irradiance Wm -2 G* hemispherical solar irradiance Wm -2 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

247 ANNEX 8 page 8 of 41 pages G'' net irradiance Wm -2 G b direct solar irradiance (beam irradiance) Wm -2 G d diffuse solar irradiance Wm -2 LT local time h K θ incidence angle modifier K θ b incidence angle modifier for direct radiation K θ d incidence angle modifier for diffuse radiation m thermally active mass of the collector kg. m mass flow rate of heat transfer fluid kgs -1 Q. useful power extracted from collector W Q. L power loss of collector W t time s t a ambient or surrounding air temperature C t dp atmospheric dew point temperature C t e collector outlet (exit) temperature C t in collector inlet temperature C t m mean temperature of heat transfer fluid C t s atmospheric or sky temperature C t stg stagnation temperature C T absolute temperature K T a ambient or surrounding air temperature C T * m reduced temperature difference ( = (t m t a )/G*) m 2 KW -1 T s atmospheric or equivalent sky radiation temperature K U measured overall heat loss coefficient of collector, with reference to T * m Wm -2 K -1 U L overall heat loss coefficient of a collector with uniform absorber temperature t m Wm -2 K -1 u surrounding air speed ms -1 V f fluid capacity of the collector m 3 p pressure difference between fluid inlet and outlet Pa t time interval s T temperature difference between fluid outlet and inlet(t e - t in ) K α solar absorptance β tilt angle of a plane with respect to horizontal degrees NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

248 ANNEX 8 page 9 of 41 pages γ azimuth angle degrees ε hemispherical emittance ω solar hour angle degrees θ angle of incidence degrees Φ latitude degrees λ wavelength µm η collector efficiency, with reference to T * m η o zero-loss collector efficiency (η at T * m = 0), reference to T * m σ Stefan-Boltzmann constant Wm -2 K -4 ρ density of heat transfer fluid kgm -3 τ c collector time constant s τ transmittance (τα) e effective transmittance-absorptance product (τα) ed effective transmittance-absorptance product for diffuse solar irradiance (τα) en effective transmittance-absorptance product for direct solar radiation at normal incidence (τα) eθ effective transmittance-absorptance product for direct solar radiation at angle of incidence θ NOTE 1 In the field of solar energy the symbol G is used to denote solar irradiance, rather than the generic symbol E for irradiance. NOTE 2 C is often denoted (mc) e in basic literature (see also Annex H) NOTE 3 For more information about thermal performance coefficients (parameters) c1 to c6, see Annex H. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

249 ANNEX 8 page 10 of 41 pages 5 Reliability testing of liquid heating collectors 5.1 General The detail of numbers of collectors and sequences used to carried out qualifications tests detailed in the list below (table1) shall be given in the report. For some qualification tests, a part of the collector may have to be tampered with in some way, for example a hole may have to be drilled in the back of the collector to attach a temperature sensor to the absorber. In these cases care should be taken to ensure that any damage caused does not affect the results of subsequent qualification tests, for example by allowing water to enter into a previously rain tight collector. Table 1 - Test List Subclause Test 5.2 Internal pressure 5.3 High-temperature resistance 5.4 Exposure 2) 5.5 External thermal shock 3) 5.6 Internal thermal shock 3) 5.7 Rain penetration 4) 5.8 Freeze resistance 5) 5.9 Mechanical load 5.10 Impact resistance Thermal performance 1) For organic absorbers, the high-temperature resistance test shall be performed first in order to determine the collector stagnation temperature needed for the internal pressure test. 2) The high temperature and exposure test shall be carried out on the same collector 3) The external and internal thermal shock tests may be combined with the exposure test or the hightemperature resistance test. 4) The rain penetration test shall be carried out only for glazed collectors. 5) The freeze resistance test shall be carried out only for collectors claimed to be freeze resistant. 6) The Thermal performance test shall be carried out on a collector that had not been used for other tests. 1), 2) 5.2 Internal pressure tests for collectors Objective The absorber shall be pressure-tested (see 5.2.3) to assess the extent to which it can withstand the pressures, which it might meet in service while operating at elevated temperature. The tests shall be carried out at elevated temperatures, because the pressure resistance of an air collector with NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

250 ANNEX 8 page 11 of 41 pages organic parts such as sealings may be adversely affected as its temperature is increased. One of the methods described in may be chosen Apparatus and procedure General The apparatus consists of a pneumatic pressure source, and a means of heating the absorber to the required test temperature. The characteristics of a solar irradiance simulator shall be the same as those of the simulator used for efficiency testing of liquid heating solar collectors. A temperature sensor shall be attached to the absorber to monitor its temperature during the test. The sensor shall be positioned at two-thirds of the absorber height and half the absorber width. It shall be fixed firmly in a position to ensure good thermal contact with the absorber. The sensor shall be shielded from solar radiation. The pressure in the absorber shall be raised in stages as specified in , and the absorber shall be inspected for swelling, distortion or rupture after each increase in pressure. The pressure shall be maintained while the absorber is being inspected. For safety reasons, the collector shall be encased in a transparent box to protect personnel in the event of explosive failure during this test High temperature pneumatic pressure test The absorber may be pressure-tested using compressed air, when heated by either of the following methods: a) heating the whole collector in a solar irradiance simulator (see figure A.5); b) heating the whole collector outdoors under natural solar irradiance (see figure A.5). The compressed air supply to the absorber shall be fitted with a safety valve and a pressure gauge having a standard uncertainty better than 5% Test conditions Temperature For air collectors the test temperature shall be the maximum temperature which the absorber will reach under stagnation conditions. The reference conditions given in table 2 shall be used. The calculations employed to determine the test temperature are included in Annex C and shall either: - use measured collector performance characteristics, or - extrapolate from average values, measured in the high-temperature resistance test (see 5.3.3), of the global solar irradiance (natural or simulated) on the collector plane, the surrounding air temperature and the absorber temperature. Table 2 - Climate reference conditions to determine test temperatures for internal pressure test of organic absorbers Climate parameter Value for all climate classes NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

251 ANNEX 8 page 12 of 41 pages Global solar irradiance on collector plane, G in W/m Surrounding air temperature, ta in C Pressure The test pressure shall be 1.5 times the maximum collector operating pressure specified by the manufacturer. For absorbers made of organic materials, the pressure shall be raised to the test pressure in equal stages of 20 kpa (approximately) and maintained at each intermediate pressure for 5 min. The test pressure shall then be maintained for a least 1 h Results The collector shall be inspected for leakage, swelling and distortion. The results of the inspection shall be reported. Full details of the test procedure used, including the temperature, intermediate pressures and test periods used, shall be reported with the test results. 5.3 High-temperature resistance test Objective This test is intended to assess rapidly whether a collector can withstand high irradiance levels without failures such as glass breakage, collapse of plastic cover, melting of plastic absorber, or significant deposits on the collector cover from out gassing of collector material Apparatus and procedure The collector shall be tested outdoors, or in a solar irradiance simulator. A schema for testing is shown in figure A.6. The characteristics of the solar irradiance simulator to be used for the high-temperature resistance test shall be those of the solar irradiance simulator used for efficiency testing of liquid heating solar collectors. The collector shall be mounted outdoors or in a solar simulator, and shall not be filled with fluid. One of its fluid pipes shall be sealed to prevent cooling by natural circulation of air, but the other shall be left open to permit free expansion of air in the absorber. A temperature sensor shall be attached to the absorber to monitor its temperature during the test. The sensor shall be positioned at two-thirds of the absorber height and half the absorber width. It shall be fixed firmly in a position to ensure good thermal contact with the absorber. The sensor shall be shielded from solar radiation. The test shall be performed for a minimum of 1 h after steady-state conditions have been established, and the collector shall be subsequently inspected for signs of damage as specified in Test conditions The set of reference conditions given in table 3 or conditions resulting in the same collector temperature according to eqn. C.3, shall be used for all climate classes. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

252 ANNEX 8 page 13 of 41 pages Table 3 - Climate reference conditions for high-temperature resistance test Climate parameter Value for all climate classes Global solar irradiance on collector plane, G in W/m 2 >1000 Surrounding air temperature, ta in C Surrounding air speed in m/s < Results The collector shall be inspected for degradation, shrinkage, outgassing and distortion. The results of the inspection shall be recorded together with the 5 min average values of solar irradiance (natural or simulated) on the collector plane, surrounding air temperature and speed, and absorber temperature (and the pressure of the special fluid in the absorber, if that method is used) recorded during the test. 5.4 Exposure test Objective The exposure test provides a low-cost reliability test sequence, indicating (or simulating) operating conditions which are likely to occur during real service and which also allows the collector to "settle", such that subsequent qualification tests are more likely to give repeatable results Apparatus and procedure The collector shall be mounted outdoors (see figure A.7), but not filled with fluid. One of the fluid openings shall be sealed to prevent cooling by natural circulation of air, while the other shall be left open to permit free expansion of air in the absorber. The air temperature shall be recorded to an uncertainty of 1 K and the global irradiance on the plane of the collector recorded using a pyranometer of class I or better in accordance with ISO Irradiation and mean air temperature values shall be recorded every 30 min and rainfall shall be recorded daily. The collector shall be exposed until the test conditions have been met. At the end of the exposure, a visual inspection shall be made for signs of damage as specified in Test conditions The set of reference conditions given in table 4 shall be used. The collector shall be exposed until at least 20 days (which need not be consecutive) have passed with the minimum irradiation H shown in table 4. The irradiation is determined by recording irradiance measurements using a pyranometer. The collector shall also be exposed for at least 30 h to the minimum irradiance level G given in table 4, as recorded by a pyranometer, when the surrounding air temperature is greater than the value shown in table 4 or conditions resulting in the same collector temperature according to eqn. C.3. These hours shall be made up of periods of at least 30 min. NOTE In regions where these conditions cannot be met during certain periods of the year, the 30-h exposure to high irradiance levels (table 4) can be conducted in a solar irradiance simulator having characteristics identical to those of a simulator used for efficiency testing of liquid heating solar collectors. The 30-h exposure test should be conducted after the collector has completed at least 10 days, but no more than 15 days, of NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

253 ANNEX 8 page 14 of 41 pages the exposure to the minimum irradiation level (table 4). If the external and internal thermal shock tests are combined with the exposure test, the first external and internal shocks shall be caused during the first 10 of the 30 h defined above, and the second during the last 10 of the 30 h. Table 4 - Climate reference conditions for exposure test as well as for external and internal thermal shock tests Climate parameter Value for all climate classes Global solar irradiance on collector plane, G in W/m Global daily irradiation on collector plane, H in MJ/m 2 14 Surrounding air temperature, ta in C 10 NOTE Values given are minimum values for testing Results The collector shall be inspected for damage or degradation. The results of the inspection shall be reported together with a record of the climatic conditions during the test, including daily irradiation, surrounding air temperature and rain. 5.5 External thermal shock test Objective Collectors may from time to time be exposed to sudden rainstorms on hot sunny days, causing a severe external thermal shock. This test is intended to assess the capability of a collector to withstand such thermal shocks without a failure Apparatus and procedure The collector shall be mounted either outdoors or in a solar irradiance simulator. One of its fluid pipes shall be sealed to prevent cooling by natural circulation of air, while the other shall be left open to permit free expansion of air in the absorber (see figure A.8). A temperature sensor may be optionally attached to the absorber to monitor its temperature during the test. The sensor shall be positioned at two-thirds of the absorber height and half the absorber width. It shall be fixed firmly in a position to ensure good thermal contact with the absorber. The sensor shall be shielded from solar radiation. An array of water jets shall be arranged to provide a uniform spray of water over the collector. The collector shall be maintained under a high level of solar irradiance for a period of 1 h before the water spray is turned on. It is then cooled by the water spray for 15 min before being inspected. The collector shall be subjected to two external thermal shocks Test conditions The set of reference conditions given in table 4 shall be used. The specified operating conditions shall be: - solar (or simulated solar) irradiance G greater than the value shown in table 4. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

254 ANNEX 8 page 15 of 41 pages - surrounding air temperature ta greater than the value shown in table 4. The water spray shall have a temperature of less than 25 C and a flow rate in the range 0,03 kg/s to 0.05 kg/s per square metre of collector aperture. If the temperature of the water which first cools the collector is likely to be greater than 25 C (for example if the water has been sitting in a pipe in the sun for some time), then the water shall be diverted until it has reached a temperature of less than 25 C before being directed over the collector Results The collector shall be inspected for any cracking, distortion, condensation or water penetration. The results of the inspection shall be reported. The measured values of solar irradiance, surrounding air temperature, absorber temperature (if measured), water temperature and water flow rate shall also be reported. 5.6 Rain penetration test Objective This test is applicable only for glazed collectors and is intended to assess the extent to which glazed collectors are substantially resistant to rain penetration. They shall normally not permit the entry of either free-falling rain or driving rain. Collectors may have ventilation holes and drain holes, but these shall not permit the entry of drifting rain Apparatus and procedure General The collector shall have its fluid inlet and outlet pipes sealed (unless hot air is circulated through the absorber, see ), as shown in figure A.10, and be placed in a test rig at the shallowest angle to the horizontal recommended by the manufacturer. If this angle is not specified, then the collector shall be placed at a tilt of 30 to the horizontal. Collectors designed to be integrated into a roof structure shall be mounted in a simulated roof and have their underside protected. Other collectors shall be mounted in a conventional manner on an open frame or a simulated roof. The collector shall be sprayed on exposed sides, using spray nozzles or showers Detection of ingress of water The collector shall be mounted and sprayed as explained above while the absorber in the collector is kept warm (minimum 50 C). This can be done either by circulating hot air at about 50 C through the absorber or by exposing the collector to solar radiation. The penetration of water into the collector shall be determined by inspection (looking for water droplets, condensation on the cover glass or other visible signs) and by one of the following methods: a) by weighing the collector (standard uncertainty better than 5 gr/m 2 collector area); or b) by means of humidity measurement (standard uncertainty better than 5%) or c) by means of measuring the condensation level. The heating up of the collector should be started before the spraying of the water in order to ensure that the collector box is dry before testing. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

255 ANNEX 8 page 16 of 41 pages In cases of collectors having wood in the backs (or other special cases), the laboratory must take all necessary measures during the conduction of the test so that the final result will not be influenced or altered by the special construction of the collector Test conditions The collector shall be sprayed with water at a temperature lower than 30 C with a flow rate of more than 0.05 kg/s per square metre of sprayed area. The duration of the test shall be 4 h Weighing method If the weighing method is chosen, the collector shall be put on the scale before the start of the test on three consecutive occasions. The weights recorded shall not vary by more than ±5 gr/m 2 collector area Humidity measurement method When measuring the penetration of water into the collector by means of humidity measurement, an absolute humidity sensor is placed in the air gap between absorber and glazing. Collector and sensor are connected to a hot fluid loop for at least five hours before the rain is switched on in order to stabilise. When testing outdoors, in order to minimize disturbances of the measurement, the collector shall be shaded during the whole test. The humidity shall be monitored from five hours before the raining till at least five hours after the raining. Ingress of water might also be detected at a later stage, during the test Final Inspection (Clause 5.11) Condensation level method If the condensation level method is chosen, the penetration of water is determined by measuring the condensation level on the cover glass and by measuring the water that come out of the collector when tipping it. The heating of the collector shall be started at least 30 minutes before the spaying of water and shall continue until it can be ensured that the collector box is dry before testing. This shall be done by circulating hot air above 50 o C through the absorber before but also during the complete test. The water will thereafter condense on the inside of the glazing, which is being cooled by cold water on the outside. After 2 hours an intermediate inspection of condensation on the cover glass shall be done in order to facilitate the reporting of the places where water penetrates. After finishing the spraying the inspection of condensation should be done after a short time for ventilating, in order to distinguish collectors with good ventilation qualifications that are without accumulation of humidity inside the collector. However, the inspection should be done within one minute after finishing the spraying before the collector will make any temperature changes. To ensure that no water has penetrated the collector box without forming condensation on the glazing, the collector shall be tipped on all four sides in turn after the test is terminated. The collector shall not be exposed by solar radiation Results The collector shall be inspected for water penetration. The results of the inspection, i.e. the extension of water penetration and the places where water penetrated shall be reported. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

256 ANNEX 8 page 17 of 41 pages 5.7 Mechanical load test Positive pressure test of the collector cover Objective This test is intended to assess the extent to which the transparent cover of the collector is able to resist the positive pressure load due to the effect of wind and snow Apparatus and procedure The collector shall be placed horizontally on an even ground. On the collector a foil shall be laid and on the collector frame a wooden or metallic frame shall be placed, high enough to contain the required amount of gravel or similar material (see figure A.12). The gravel, preferably type 2-32 mm, shall be weighed in portions and distributed in the frame so that everywhere the same load is created (pay attention to the bending of the glass), until the wanted height is reached. The test can also be carried out installing the collector in accordance with and loading the cover using suction cups, gravel or other suitable means (e.g. water). As a further alternative, the necessary load may be created by applying an air pressure on the collector cover. In this case, apparatus in accordance with EN can be used Test conditions The test pressure shall be increased in steps of 100 Pa to the recommended maximum test pressure, which shall be at least 1000 Pa or optional load test above 1000 Pa up to the value as specified by manufacturer or national requirements Results The pressure at which any failure of the collector cover occurs shall be reported together with details of the failure. If no failure occurs, then the maximum pressure which the collector sustained shall be reported Negative pressure test of fixings between the cover and the collector box Objective This test is intended to assess the extent to which the fixings between the collector cover and collector box are able to resist uplift forces caused by the wind Apparatus and procedure The collector shall be installed horizontally on a stiff frame by means of its mounting fixtures. The frame which secures the cover to the collector box shall not be restricted in any way. A lifting force which is equivalent to the specified negative pressure load, shall be applied evenly over the cover. The load shall be increased in steps up to the final test pressure. If the cover has not been loosened at the final pressure, then the pressure may be stepped up until failure occurs. The time between each pressure step shall be the time needed for the pressure to stabilise. Either of three alternative methods may be used to apply pressure to the cover: - Method (a): The load may be applied to the collector cover by means of a uniformly distributed set of suction cups (see figure A.13). NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

257 ANNEX 8 page 18 of 41 pages - Method (b): For collectors which have an almost airtight collector box, the following procedure may be used to create a negative pressure on the cover (see figure A.14). Two holes are made through the collector box into the airgap between the collector cover and absorber, and an air source and pressure gauge are connected to the collector airgap through these holes. A negative pressure on the cover is created by pressurising the collector box. For safety reasons the collector shall be encased in a transparent box to protect personnel in the event of failure during this test. - Method (c): The load may be created by applying a negative air pressure on the collector cover. In this case, apparatus in accordance with EN can be used. During the test, the collector shall be visually inspected and any deformations of the cover and its fixings reported. The collector shall be examined at the end of the test to see if there are any permanent deformations. Test method (a) is not designed to check the strength of the collector mounting fixtures. If a glass cover fails before the fixings which hold the cover to the collector box, then the pressure at which the failure occurred shall be noted. If the cover fails again then it shall be concluded that the collector is inadequately designed and the test need not be continued Test conditions The test pressure shall be increased in steps of 100 Pa to the recommended maximum test pressure, which shall be at least 1000 Pa or optional load test above 1000 Pa up to the value as specified by the manufacturer or national requirements Results Any deformations observed during the inspection shall be reported together with the pressure at which any failure of the cover or cover fixings was observed. Details of the failures shall also be reported. If no failure occurs, then the maximum pressure which the collector sustained shall be reported Negative pressure test of collector mountings Objective Solar collectors are generally installed on a roof or on the ground by means of mounting brackets and supporting frames. This test is intended to assess the extent to which the mounting brackets, supporting frames and fixing points can withstand the uplift forces caused by the wind Apparatus and procedure The collector shall be installed horizontally on a stiff frame by means of the mounting fixtures supplied by the manufacturer. A lifting force shall be applied to the collector either by applying a negative pressure uniformly to the top of the collector frame and cover using the method (a) described in , or by applying a positive pressure uniformly over the back of the collector by means of air pressure in large air bags (see figure A.14). The pressure shall be increased in steps up to the final test pressure. If the mounting fixtures have not failed at the final test pressure, then the pressure may be stepped up until failure occurs. The time between each pressure step shall be the time needed for the pressure to stabilise. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

258 ANNEX 8 page 19 of 41 pages Test conditions The test pressure shall be increased in steps of 100 Pa to the recommended maximum test pressure, which shall be at least 1000 Pa or optional load test above 1000 Pa up to the value as specified by manufacturer or national requirements Results The pressure at which any failure of the mounting fixtures or fixing points occurs shall be reported together with details of the failure. If no failure occurs, then the maximum pressure which the collector sustained shall be reported. 5.8 Impact resistance test Objective This test is intended to assess the extent to which a collector can withstand the effects of heavy impacts caused by hailstones Apparatus and procedure General The testing of the solar collector to determine its impact resistance can be done by one of two methods, i.e. by using steel balls or ice balls Method 1 The collector shall be mounted either vertically or horizontally on a support (see figure A.15). The support may be stiff enough so that there is negligible distortion or deflection at the time of impact. Steel balls shall be used to simulate a heavy impact. If the collector is mounted horizontally then the steel balls are dropped vertically, or if it is mounted vertically then the impacts are directed horizontally by means of a pendulum. In both cases, the height of the fall is the vertical distance between the point of release and the horizontal plane containing the point of impact. The point of impact shall be no more than 5 cm from the edge of the collector cover, and no more than 10 cm from the corner of the collector cover, but it shall be moved by several millimetres each time the steel ball is dropped. A steel ball shall be dropped onto the collector 10 times from the first test height, then 10 times from the second test height, etc. until the maximum test height is reached (as specified by the manufacturer). The test is to be stopped when the collector sustains some damage or when the collector has survived the impact of 10 steel balls at the maximum test height. NOTE This method does not correspond to the physical effect of hailstones as the deformation energy absorbed by the ice particles is not being considered Method 2 The apparatus consists of the following equipment: a) Moulds of suitable material for casting spherical ice balls of the required diameter (25 mm). b) A freezer, controlled at -10 C ± 5 C. c) A storage container for storing the ice balls at a temperature of -4 C ± 2 C. d) A launcher capable of propelling an ice ball at the specified velocity (as specified by the NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

259 ANNEX 8 page 20 of 41 pages manufacturer), within ± 5 %, so as to hit the collector within the specified impact location. The path of the ice ball from the launcher to the collector may be horizontal, vertical or at any intermediate angle. e) A rigid frame for supporting the collector, with the impact surface normal to the path of the projected ice ball; the support shall be stiff enough so that there is negligible distortion or deflection at the time of impact. f) A balance for determining the mass of an ice ball to a standard uncertainty of ± 2 %. g) An instrument for measuring the velocity of the ice ball to a standard uncertainty of ± 2 ms -1. The velocity sensor shall be no more than 1 m from the surface of the collector. As an example, figure A.16 shows in schematic form a suitable apparatus comprising a horizontal pneumatic launcher, a vertical collector support and a velocity meter which measures electronically the time it takes the ice ball to traverse the distance between two light beams. The testing procedure shall be the following: a) Using the moulds and the freezer, make sufficient ice balls of the required size for the test, including some for the preliminary adjustment of the launcher. b) Examine each one for cracks, size and mass. An acceptable ball shall meet the following criteria: - no cracks visible to the unaided eye; - diameter within ± 5% of the ball (25 mm); - mass within ± 5% of the ball (25 mm). c) Place the balls in the storage container and leave them there for at least 1 h before use. d) Ensure that all surfaces of the launcher likely to be in contact with the ice balls are near room temperature. e) Fire a number of trial shots at a simulated target in accordance with step g) below and adjust the launcher until the velocity of the ice ball, as measured with the velocity sensor in the prescribed position, is within ± 5% of the required hailstone test velocity. f) Install the collector at room temperature in the prescribed mount, with the impact surface normal to the path of the ice ball. g) Take an ice ball from the storage container and place it in the launcher. Take aim at the impact location and fire. The time between the removal of the ice ball from the container and impact on the collector shall not exceed 60 s. The point of impact shall be no more than 5 cm from the edge of the collector cover, and no more than 10 cm from the corner of the collector cover, but it shall be moved by several millimetres each time the ice ball is launched. An ice ball shall be launched onto the collector 10 times; the test shall be stopped when the collector sustains some damage or when the collector has survived the impact of 10 ice balls Test conditions If the test is conducted according to method 1, the steel ball shall have a mass of 150 g ± 10 g and the following series of test heights shall be used: 0.4 m, 0.6 m, 0.8 m, 1,0 m, 1.2 m, 1.4 m, 1.6 m, 1.8 m and 2.0 m. If the test is conducted according to method 2, the ice ball shall have a diameter of 25 mm ± 5%, a mass of 7,53 g ± 5 % and its velocity shall be 23 m/s ± 5 %. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

260 ANNEX 8 page 21 of 41 pages Results The collector shall be inspected for damage. The results of the inspection shall be reported, together with the height from which the steel ball was dropped (if method 1 is used) and the number of impacts which caused the damage. NOTE As test method 2 is closer to reality, this method ( ) is preferable. 5.9 Final inspection When the full test sequence has been completed, the collector shall be dismantled and inspected. All abnormalities shall be reported and accompanied by a photograph Test report The format sheets given in annex B shall be completed for each test, together with the introductory format sheet (B.1) reporting a summary of main results, including the test methods. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

261 ANNEX 8 page 22 of 41 pages 6 Thermal performance testing of air heating collectors 6.1 Collector mounting and location General The way in which a collector is mounted will influence the results of thermal performance tests. Collectors to be tested shall therefore be mounted in accordance with to Full-size collector modules shall be tested, because the edge losses of small collectors may significantly reduce their overall performance Collector mounting frame The collector shall be mounted in the manner specified by the manufacturer. The collector mounting frame shall in no way obstruct the aperture of the collector, and shall not significantly affect the back or side insulation, unless otherwise specified (for example, when the collector is part of an integrated roof array). Collectors designed to be mounted directly on standard roofing material may be mounted over a simulated roof section. In case of roof integrated collectors, a model consisting of a small scale collector placed on an artificial roof should be prepared for the purpose of the tests. If mounting instructions are not specified, the collector shall be mounted on an insulated backing with a quotient of the materials thermal conductivity to its thickness of 1 Wm -2 K -1 ± 0.3 Wm -2 K -1 and the upper surface painted matt white and ventilated at the back. NOTE Example material suited for the insulated backing is 30 mm of polystyrene foam. The collector shall be mounted such that the lower edge is not less than 0.5 m above the local ground surface. Currents of warm air, such as those which rise up the walls of a building, shall not be allowed to pass over the collector. Where collectors are tested on the roof of a building, they shall be located at least 2 m away from the roof edge. The performance of some forms of unglazed solar collectors is a function of module size. If the collector is supplied in fixed units of area greater than 1 m2 then a sufficient number of modules shall be linked together to give a test system aperture of at least 3 m2. If the collector is supplied in the form of strips the minimum built-up module area shall be 3 m2 (gross area) Tilt angle Collectors may be tested at tilt angles, as recommended by manufacturers or specified for actual installations. Otherwise the collector shall be tested at tilt angles such that the incidence angle with direct solar radiation θ is less than 30 or at angles of tilt such that the incidence angle modifier varies by less than ±2 % from normal incidence. Before deciding on a tilt angle it may be necessary to check the incidence angle modifier at two angles prior to commencing the tests. NOTE For most unglazed collectors, the influence of tilt angle and radiation incidence angle on collector efficiency is small and unglazed collectors are commonly installed at low inclinations. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

262 ANNEX 8 page 23 of 41 pages Collector orientation outdoors The collector may be mounted outdoors in a fixed position facing the equator, but this will result in the time available for testing being restricted by the acceptance range of incidence angles. A more versatile approach is to move the collector to follow the sun in azimuth, using manual or automatic tracking Shading from direct solar irradiance The location of the test stand shall be such that no shadow is cast on the collector during the test Diffuse and reflected solar irradiance For the purposes of analysis of outdoor test results, solar irradiance not coming directly from the sun's disc is assumed to come isotropically from the hemispherical field of view of the collector. In order to minimize the errors resulting from this approximation, the collector shall be located where there will be no significant solar radiation reflected onto it from surrounding buildings or surfaces during the tests, and where there will be no significant obstructions in the field of view. Not more than 5 % of the collector's field of view shall be obstructed, and it is particularly important to avoid buildings or large obstructions subtending an angle of greater than approximately 15 to the horizontal in front of the collectors. The reflectance of most rough surfaces such as grass, weathered concrete or chippings is usually low enough so no problem is caused during collector testing. Surfaces to be avoided in the collector's field of view include large expanses of glass, metal or water. In most solar simulators the simulated beam approximates direct solar irradiance only. In order to simplify the measurement of simulated irradiance, it is necessary to minimize reflected irradiance. This can be achieved by painting all surfaces in the test chamber with a dark (low reflectance) paint Thermal irradiance The performance of some collectors is particularly sensitive to the levels of thermal irradiance. The temperature of surfaces adjacent to the collector shall be as close as possible to that of the ambient air in order to minimize the influence of thermal radiation. For example, the outdoor field of view of the collector shall not include chimneys, cooling towers or hot exhausts. For indoor and simulator testing, the collector shall be shielded from hot surfaces such as radiators, airconditioning ducts and machinery, and from cold surfaces such as windows and external walls. Shielding is important both in front of and behind the collector. The major difference between indoor and outdoor testing of unglazed collectors is the long wave thermal irradiance. The relative long wave radiation in a simulator shall not be higher than ±50 Wm -2 (typically -100 Wm -2 for outdoor conditions) Surrounding air speed The performance of many collectors is sensitive to the surrounding air speeds. In order to maximize the reproducibility of results, collectors shall be mounted such that air can freely pass over the aperture, back and sides of the collector. The mean surrounding air speed, parallel to the collector aperture, shall be between the limits specified in section Where necessary, artificial wind generators shall be used to achieve these air speeds. Collectors designed for integration into a roof may have their backs protected from the wind; if so, this shall be reported with the test results. The performance of unglazed collectors is sensitive to air speed adjacent to the collector. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

263 ANNEX 8 page 24 of 41 pages In order to maximize the reproducibility of results, unglazed collectors shall be mounted such that air can freely pass over the front side of the collector, and exposed back and sides of the collector. The average surrounding air speed at a distance of 100 mm above and parallel to the collector aperture shall cover the range 0 ms -1 to 3,5ms -1 subject to the tolerance specified in table B1. If these conditions cannot be achieved under natural conditions then an artificial wind generator shall be used. If a wind generator is used the turbulence level shall be in the range of 20 % to 40 % to simulate natural wind conditions. The turbulence level shall be checked at the leading edge of the collector 100 mm above the collector surface. The turbulence level shall be monitored using a linearised hot wire anemometer with a frequency response of at least 100 Hz. If the absorber is not mounted directly on a roof or a sheet of backing material, the air speed shall be controlled and monitored on the front and back of the absorber. 6.2 Instrumentation Solar radiation measurement The solar radiation measurement shall be handled in accordance with EN Chapter Solar radiation measurement Thermal radiation measurement Measurement of long wave irradiance A pyrgeometer mounted in the plane of the collector shall be used to measure global long wave radiation Precaution for effects of temperature gradient The pyrgeometer used during the tests shall be placed in the same plane as the collector absorber and allowed to equilibrate for at least 30 minutes before measuring Precautions for effects of humidity and moisture The pyrgeometer shall be provided with a means of preventing accumulation of moisture that may condense on surfaces within the instrument and effect its reading. An instrument with a desiccator that can be inspected is required. The condition of the desiccator shall be observed prior to and following each daily measurement sequence Precautions for effect of short wave heating The influence of short wave solar heating effects should be minimised Calibration interval The pyrgeometer shall be calibrated within 12 month preceding the tests, in accordance with ISO A change of more than 5% over a year period shall warrant the use of more frequent calibration or replacement of the instrument. If the instrument is damaged in any significant manner, it shall be recalibrated or replaced. All calibrations shall be performed with respect to the World Radiometric Reference (WRR) Scale Temperature measurements General Three temperature measurements are required for solar collector testing. These are the fluid temperature at the collector inlet, the fluid temperature at the collector outlet, and the ambient air NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

264 ANNEX 8 page 25 of 41 pages temperature. The required accuracy and the environment for these measurements differ, and hence the sensor for temperature measurement and associated equipment may be different Measurement of heat transfer fluid inlet temperature (t in ) Required accuracy The temperature of the heat transfer fluid at the collector inlet shall be measured to an uncertainty of 0.2 K, but in order to check that the temperature is not drifting with time, a very much better resolution of the temperature signal to ±0.04 K is required. NOTE This resolution is needed for all temperatures used for collector testing (i.e. over the range 0 C to 100 C) which is a particularly demanding accuracy for recording by data logger, as it requires a resolution of one part in or a 12-bit digital system Mounting of sensors The determination of the mean temperature in air flows is critical, several layers of different air temperatures are often close adjacent, therefore a specific mixing device optimized according to fluid dynamic experiences - at the outgoing duct just before the sensors and a sophisticated arrangement of temperature-sensors are necessary. If thermocouples are used to measure the temperatures, thermocouple grids shall be fabricated with thermocouples located as shown in Fig. 1 and Fig. 2. There shall be a minimum of eight thermocouples in a grid in the air inlet test duct and in the outlet test duct. Thermocouples in the grid shall be located at the center of equal cross-sectional or concentric areas, as illustrated in Fig. 1Fig. 2 and Fig. 3 Fig. 1: Schematic of the thermophile arrangement used to measure the temperature difference across the solar collector. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

265 ANNEX 8 page 26 of 41 pages Fig. 2: Schematic of equal areas thermocouple grid. Minimum of eight junctions located at the center of equal crosssectional areas are connected in parallel to obtain an average reading. All thermocouple sets must have leads of identical length. Fig. 3: Distribution of thermocouples in round duct for equal cross-sectional areas grid. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

266 ANNEX 8 page 27 of 41 pages When using other temperature sensors than thermocouples a similar arrangement of the sensors within the ducts is required. The sensor for temperature measurement shall be mounted at no more than 200 mm from the collector inlet, and insulation shall be placed around the ducts both upstream and downstream of the sensor. If it is necessary to position the sensor more than 200 mm away from the collector, then a test shall be made to verify that the measurement of fluid temperature is not affected Determination of heat transfer fluid temperature difference ( T) The difference between the collector outlet and inlet temperatures ( T) shall be determined to a standard uncertainty of <0.05 K. Standard uncertainties approaching 0.02 K can be achieved with modern well-matched and calibrated transducers, and hence it is possible to measure heat transfer fluid temperature differences of 1 K or 2 K with a reasonable accuracy. Delta-T sensors shall be calibrated in the relevant flow range and temperature range, using the same fluid Measurement of surrounding air temperature (t a ) Required accuracy The ambient or surrounding air temperature shall be measured to a standard uncertainty of 0.5 K Mounting of sensors For outdoor measurements the sensor shall be shaded from direct and reflected solar radiation by means of a white-painted, well-ventilated shelter, preferably with forced ventilation. The shelter itself shall be shaded and placed at the midheight of the collector but at least 1 m above the local ground surface to ensure that it is removed from the influence of ground heating. The shelter shall be positioned to one side of the collector and not more than 10 m from it. If air is forced over the collector by a wind generator, the air temperature shall be measured in the outlet of the wind generator and checks made to ensure that this temperature does not deviate from the ambient air temperature by more than ±1 K Measurement of collector liquid flow rate The standard uncertainty of the liquid flow rate measurement shall be within ±1.5 % of the measured value, in mass per unit time. The flowmeter shall be calibrated over the range of fluid flowrates and temperatures to be used during collector testing Measurement of surrounding air speed The measurement of surrounding air speed shall be handled in accordance with EN Chapter Measurement of air speed. For unglazed collectors the measurement of surrounding air speed shall be handled in accordance with Chapter of EN Pressure measurements Pressure-measuring stations shall have four externally manifolded pressure taps, as shown in Fig. 4. The pressures in the test circuit and the pressure drop across the solar collector shall be measured using static pressure tap holes and either a manometer or a differential-pressure transducer. The edges of the holes on the inside surface of the duct shall be free of burrs. The hole diameter shall not exceed 40% of the wall thickness or 1.6 mm. Provision shall be made for determining the absolute pressure of the entering transfer fluid. The static pressure drop across an air collector and static pressure upstream or downstream of the collector shall be determined with instruments that have an accuracy of ±2.5 Pa. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

267 ANNEX 8 page 28 of 41 pages Fig. 4: Schematic representation of the measurement of pressure drop across the solar collector. Measuring stations shall be provided upstream and downstream of the collector, as illustrated in Fig. 5. For collectors tested under negative gauge pressure, the collector inlet gauge pressure shall be below the atmospheric pressure by at least 124 Pa minimum or the maximum allowable operating pressure specified by the manufacturer, whichever is smaller. For collectors tested under positive pressure, the gauge pressure at the collector discharge shall be 124 Pa minimum or the maximum allowable operating pressure specified by the manufacturer, whichever is smaller Humidity Measurement When air is used as the heat transfer fluid, its moisture content is needed for the correct determination of the density and specific heat of the air. The humidity ratio W n shall be measured to an accuracy of ±0.005 (kg water/kg dry air). 1 Humidity measurement should be made in accordance with ASHRAE Standard Elapsed time Elapsed time shall be measured to a standard uncertainty of 0.2 % Instrumentation/data recorders In no case shall the smallest scale division of the instrument or instrument system exceed twice the specified standard uncertainty. For example, if the specified standard uncertainty is 0.1 K, the smallest scale division shall not exceed 0.2 C. Digital techniques and electronic integrators shall have an standard uncertainty equal to or better than 1.0 % of the measured value ASHRAE NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

268 ANNEX 8 page 29 of 41 pages Analog and digital recorders shall have an error equal to or better than 0.5 % of the full-scale reading and have a time constant of 1s or less. The peak signal indication shall be between 50 % and 100 % of full scale. The input impedance of recorders shall be greater than 1000 times the impedance of the sensors or 10 MΩ whichever is higher Collector area The collector area (absorber, gross or aperture) shall be measured to a standard uncertainty of 0.3 %. Area measurements shall take place at a collector temperature of (20 ± 10) C and under operating pressure if the absorber is made of organic material. 6.3 Test installation General consideration An example of test configurations for testing solar air collectors are shown in Fig. 5. It is schematic only, and are not drawn to scale. Fig. 5: Example of an open test loop Heat transfer fluid The heat transfer fluid used for collector testing is air. The specific heat capacity and density of the fluid shall be known to within ±1% over the range of fluid temperatures used during the tests. These values are given for air in annex L Test ducts The air ducts between the solar collector and the pressure-measuring station, upstream and downstream of the collector, shall be of the same cross-sectional dimension. The cross-sectional area of these ducts in the pressure-measuring section shall be equal in size to the inlet discharge opening of the collector, whichever is smaller. The air flow pattern inside the collector is very important for a correct assessment of the performance. The air flow pattern inside the collector (especially the partition close to the inlet) mainly depends on the connection between ducting system and collector. In standardised tests only a single collector module is tested, which might not comply with the mode of installation in praxis. To reach an even air flow pattern throughout the collector special distribution ducts at the inlet and outlet should be used for each collector tested. By means of boxes with perforated metal sheets at inlet and outlet a well distributed air flow can be achieved, which means that an even air flow from the centre-line of the collector to the edges from entrance to outlet exists. The air ducts used in the collector loop shall be suitable for operation at temperatures up to 65 C. Duct lengths shall generally be kept short. In particular, the length of duct between the outlet of NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

269 ANNEX 8 page 30 of 41 pages the fluid temperature regulator and the inlet to the collector shall be minimized, to reduce the effects of the environment on the inlet temperature of the fluid. This section of duct shall be insulated to ensure a rate of heat loss of less than 0.2 WK -1, and shall be protected by a reflective weatherproof coating. Ducts between the temperature sensing points and the collector (inlet and outlet) shall be protected with insulation and reflective (for outdoor measurements also weatherproof) covers to beyond the positions of the temperature sensors, such that the calculated temperature gain or loss along either duct portion does not exceed ±0.01 K under test conditions Pump and flow control devices The air blower shall be located in the collector test loop in such a position that the heat from it which is dissipated in the fluid does not affect either the control of the collector inlet temperature or the measurements of the fluid temperature rise through the collector. Where necessary, an appropriate flow control device may be added to stabilize the mass flow rate. The blower and flow controller shall be capable of maintaining the mass or volume flow rate through the collector stable to within ±1.5 % 2 despite temperature variations, at any inlet temperature chosen within the operating range Air-Preconditioning Apparatus The preconditioning apparatus shall control the dry-bulb temperature of the transfer medium entering the solar collector to within ±1.0 C of the desired test values at all times during the test period. Since the rate of energy collection in the collector is deduced by measuring instantaneous values of the fluid inlet and outlet temperatures, it follows that small variations in inlet temperature could lead to errors in the rates of energy collection deduced. It is particularly important to avoid any drift in the collector inlet temperature. Its heating and cooling capacity shall be selected so that the dry-bulb temperature of the air entering the preconditioned apparatus may be raised or lowered to the required amount to meet the applicable test conditions in Section and Humidity Ratio When air is the transfer fluid and the test panel is operated at a negative pressure, the humidity ratio of the test fluid shall be equal to the humidity ratio of the air surrounding the test panel. 6.4 Outdoor steady-state efficiency test Test installation The collector shall be mounted in accordance with the specifications given in 6.1.1, and coupled to a test loop as described in The heat transfer fluid shall flow from the bottom to the top of the collector, or as recommended by the manufacturer. The air flow pattern inside the collector is very important for a correct assessment of the performance. The air flow pattern inside the collector (especially the partition close to the inlet) mainly depends on the connection between ducting system and collector. In standardised tests only a single collector module is tested, which might not comply with the mode of installation in praxis. To reach an even air flow pattern throughout the collector special distribution ducts at the inlet and outlet should be used for each collector tested. By means of boxes with perforated metal sheets at inlet and outlet a well distributed air flow can be achieved, which means that an even air flow from ASHRAE NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

270 ANNEX 8 page 31 of 41 pages the centre-line of the collector to the edges from entrance to outlet exists Preconditioning of the collector The collector shall be visually inspected and any damage recorded. The collector aperture cover shall be thoroughly cleaned. If moisture is formed on the collector components, then the heat transfer fluid shall be circulated at approximately 50 C for as long as is necessary to dry out the insulation and collector enclosure but at least for 30 minutes. If this form of preconditioning is carried out, then it shall be reported with the test results. If the customer wants only a performance test and not the qualification tests, the empty collector shall be exposed to irradiation for 5 hours at the level of more than 700 Wm Test conditions At the time of the test, the total solar irradiance at the plane of the collector aperture shall be greater than 700 Wm -2. NOTE 1 If the manufacturer has limitations on operation with respect to maximum irradiance but not less than 800 Wm -2, this can be requested with the test. That maximum value should be clearly reported. The angle of incidence of direct solar radiation at the collector aperture shall be in the range in which the incident angle modifier for the collector varies by no more than ±2 % from its value at normal incidence. For single glazed flat plate collectors, this condition will usually be satisfied if the angle of incidence of direct solar radiation at the collector aperture is less than 20. However, much lower angles may be required for particular designs. In order to characterize collector performance at other angles, an incident angle modifier may be determined (see 6.7). Where the diffuse fraction of solar irradiance is less than 30 %, its influence may be neglected. The collector shall not be tested at diffuse fractions of irradiance greater than 30 %. The average value of air speed parallel to the collector aperture, taking into account spatial variations over the collector and temporal variations during the test period, shall be 3 ms -1 ± 1 ms -1 Measurements of fluid temperature difference of less than 1 K shall not be included in the test results because of the associated problems of instrument error Test procedure General The collector shall be tested over its operating temperature range and over a range of mass flow rates as specified below. Tests shall be done under clear sky conditions in order to determine its efficiency characteristic. If test conditions permit, an equal number of data points shall be taken before and after solar noon for each fluid inlet temperature. The latter is not required if the collectors are moved to follow the sun in azimuth and altitude using automatic tracking. During a test, measurements shall be made as specified in These may then be used to identify test periods from which satisfactory data points can be derived Fluid inlet temperature range If the fluid inlet temperature range is specified by the manufacturer, data points, shall satisfy the requirements given below and be obtained for at least three fluid inlet temperatures spaced evenly NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

271 ANNEX 8 page 32 of 41 pages over the operating temperature range of the collector. If possible, one inlet temperature shall be selected such that the mean fluid temperature in the collector lies within ±3 K of the ambient air temperature, in order to obtain an accurate determination of η o. If no specification by the manufacturer is provided it is recommended to have at least three evenly spaced fluid inlet temperature within the following ranges, with regards to the system the collector is predominantly designed for: closed loop systems: 0 to 60 C open loop systems: -20 to +40 C Fluid flow rate range Unless the range of fluid flow rate is specified by the manufacturer, the fluid flow rate shall be set to the values equally distributed between 20 to 120 kg/h/m² absorber area. It shall be held stable to within ±2 % of the set value during each test period, and shall not vary by more than ±3 % of the set value from one test period to another. In some collectors the recommended fluid flow rate may be close to the transition region between laminar and turbulent flow. This may cause instability of the internal heat transfer coefficient and hence variations in measurements of collector efficiency. In order to characterize such a collector in a reproducible way, it may be necessary to use a higher flow rate, but this shall be clearly stated with the test results. NOTE: In the transition regime, the flow rate should first be set high (turbulent) and then reduced to the setpoint value. This will prevent transition from laminar to turbulent during the measurements Measurements The following data shall be measured: the gross collector area A G, the absorber area A A and the aperture area A a ; the global solar irradiance at the collector aperture the global long wave radiation at the collector aperture the diffuse solar irradiance at the collector aperture (only outdoors) the angle of incidence of direct solar radiation (alternatively, this angle may be determined by calculation) the surrounding air speed parallel to the collector aperture the surrounding air temperature the dew point temperature of the surrounding air the temperature of the heat transfer fluid at the collector inlet the temperature of the heat transfer fluid at the collector outlet the dew point temperature of the heat transfer fluid at the collector inlet the dew point temperature of the heat transfer fluid at the collector outlet the mass flow rate of the heat transfer fluid at the collector inlet the mass flow rate of the heat transfer fluid at the collector outlet NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

272 ANNEX 8 page 33 of 41 pages Test period (steady-state) The test period for a steady state data point shall include a pre-conditioning period of at least four times the time constant of the collector (if known), or not less than 10 minutes (if time constant is not known), with the correct fluid temperature at the inlet, followed by a steady state measurement period of at least 4 times the time constant of the collector (if known), or not less than 5 minutes (if time constant is not known). A collector is considered to have been operating in steady-state conditions over a given measurement period if none of the experimental parameters deviate from their mean values over the measurement period by more than the limits given in table 5. To establish that a steady state exists, average values of each parameter taken over successive periods of 30 s shall be compared with the mean value over the measurement period. Table 5 - Permitted deviation of measured parameters during a measurement period Parameter Permitted deviation from the mean value (Global)Test solar irradiance ± 50 Wm -2 Surrounding air temperature ± 1 K Fluid mass flow rate ± 2 % Fluid temperature at the collector inlet ± 0.3 K Presentation of results The measurements shall be collated to produce a set of data points which meet the required test conditions (see ), including those for steady-state operation. These shall be presented using the data format sheets given in Annex D Computation and presentation of collector efficiency and thermal performance General The instantaneous efficiency of a solar collector, operating under steady-state conditions, is defined as the ratio of the actual useful extracted power to the solar energy intercepted by the collector. The actual useful power extracted, Q., is calculated from: Q & = mc & T f (3) A value of c f corresponding to the mean fluid temperature shall be used. If m& is obtained from volumetric flow rate measurement, then the density shall be determined for the temperature of the fluid in the flow meter Solar energy intercepted by the collector Provided that the angle of incidence is less than 20, the use of an incident angle modifier, as described in 6.7, is not required for single glazed flat plate collectors. The solar energy intercepted is A G where the area is A A when referred to the absorber area of the collector and A a when referred to the aperture area of the collector, and the collector efficiency is η =. Q mc & = AG ( te t AG f in ) (4) NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

273 ANNEX 8 page 34 of 41 pages Presentation of the thermal performance subject to reference temperature The efficiency η depends on the operation conditions of the collector. It decreases with increasing temperatures, because of the increasing heat losses. Important is, how to define the operation conditions of the collector given by the temperature difference between the overall collector temperature t c and ambient t a. An efficiency curve can be drawn in dependency of a certain reference temperature which corresponds to the collector temperature t c. In a physically correct way one has to take a weighted mean temperature t c of the whole collector box, but in the measuring practice this is not practicable. The instantaneous efficiency shall be presented graphically as a function of either the inlet temperature t in, the outlet temperature t e and a so called mean collector temperature t m which can be calculated as the arithmetic mean value between inlet and outlet temperature. Hence the physical mean temperature t c. of the collector is often much closer to the outlet temperature t e than to the arithmetical mean temperature t m it is recommended that to use the outlet temperature t e for the presentation of the collector efficiency. The value of G to be used for the presentation of second-order fits shall be 800 Wm -2. The test conditions shall be recorded on the data format sheets given in annex D Presentation of thermal performance subject to the mean collector temperature The reduced temperature difference T m *. When the mean temperature of the heat transfer fluid t m is used, where T t m = tin + 2 the reduced temperature difference is calculated as: * tm ta T = m G (6) The efficiency is then calculated as: U t t mc t t L ( m a ) & f ( e in ) η m = FR τ α = G (7) AG Presentation of thermal performance subject to the outlet temperature The outlet temperature t e U t t mc t t L ( e a ) & f ( e in ) η e = Fo τ α = G (8) AG F o is the collector heat removal factor in relation to t e - the collector outlet temperature and η e is the efficiency when the outlet temperature is taken as reference. F 0 accounts for the fact that the absorber temperature is not the same as the outlet air collector temperature neither in the horizontal direction nor vertical Presentation of thermal performance subject to the inlet temperature When using the inlet temperature t in the efficiency is calculated as: (5) NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

274 ANNEX 8 page 35 of 41 pages U t t mc t t L ( in a ) & f ( e in ) η in = FR τ α = G (9) AG Presentation of thermal performance subject to leakage While equations (7-9) characterize the thermal performance of air-heating collectors, significant leakage of the heat transfer medium is likely and different flow rates in the inlet and exit streams must be taken into account in measuring the useful energy collected. Defining the leakage mass flow rate positive for in-leakage, i.e., m & L = m& m& (10) e in the useful energy collected may be determined for both test modes as follows Testing with negative gauge pressure In this configuration with leakage ( m& L > 0), air at ambient temperature is drawn into the heated air stream in the collector. The actual useful energy gain in this situation is Q & = m& h m& h + m& h ) (11) e e ( in in L a where the bracketed term is the total incoming enthalpy flow. Neglecting effects of moisture transfer between different airstreams and assuming constant specific heat, Equation 11 can be expressed in terms of measured quantities as Q& = m& ec f ( te tin) + ( m& in m& in) c f ( tin ta ) For convenience in subsequent expressions for systems with significant leakage, Equations 4 and 12 may be combined to define an effective heat transfer fluid flow rate for air-heating systems, i.e., (12) ( tin ta ) m& = m& e + ( m& e m& in ) (13) ( t t ) e in which can be used for calculations in Sections 7, 8 and 9. Equation 7, 8 and 9 can be used as the basis for plotting measured efficiencies versus (t in t a )/G provided t in is taken as the massweighted mean temperature of the in-leakage and inlet flow rates Testing with positive gauge pressure In this configuration with leakage ( m& L < 0), heated air escapes from the collector to the environment with an attendant loss of useful energy. In addition to the exit enthalpy flow, the collector also introduces an infiltration of ambient air into the load at a flow rate equal to the collector leakage. Therefore, the collector supplies the enthalpy flow m & ehe + ( m& L ) ha ) to the load. With the inlet enthalpy flow rate of m& ihi an enthalpy balance for this case again gives Equation 11 (with due regard for the sign of m& L ). Consequently, Equations 7 to 9 apply also when testing under positive gauge pressure Conversion of thermal performance test characteristics The equations shall be presented in terms of the aperture area A a, as well as in terms of the absorber area A A. The following basic conversions shall be used: NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

275 ANNEX 8 page 36 of 41 pages η 0 A A = η a 1 = a a 2 = a A 1a 0 a 2a A A A A a A A A a A a A (14) (15) (16) 6.5 Steady-state efficiency test using a solar irradiance simulator General The performance of most collectors is better in direct solar radiation than in diffuse and at present there is little experience with diffuse solar simulation. This test method is therefore designed for use only in simulators where a near-normal incidence beam of simulated solar radiation can be directed at the collector. In practice it is difficult to produce a uniform beam of simulated solar radiation and a mean irradiance level has therefore to be measured over the collector aperture The solar irradiance simulator for steady-state efficiency testing A simulator for steady-state efficiency testing shall have the following characteristics: The lamps shall be capable of producing a mean irradiance over the collector aperture of at least 700 Wm -2. Values in the range 300 Wm -2 to 1000 Wm -2 may also be used for specialized tests, provided that the accuracy requirements given in table 5 can be achieved and the irradiance values are noted in the test report. The mean irradiance over the collector aperture shall not vary by more than ±3 % during a test period. At any time the irradiance at a point on the collector aperture shall not differ from the mean irradiance over the aperture by more than ±15 %. The spectral distribution of the simulated solar radiation shall be approximately equivalent to that of the solar spectrum at optical air mass 1.5. Where collectors contain spectrally selective absorbers or covers, a check shall be made to establish the effect of the difference in spectrum on the (τα) product for the collector. If the effective values of (τα) under the simulator and under the optical air mass 1.5 solar radiation spectrum differ by more than ± 1 %, then a correction shall be applied to the test results. 3µ m τ( λ)α( λ)g( λ)dλ 0,3µ m Effective( τα) = 3µ m G( λ)dλ (17) 0,3µ m NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

276 ANNEX 8 page 37 of 41 pages Measurement of the solar simulator's spectral qualities shall be in the plane of the collector over the wavelength range of 0.3 µm to 3 µm and shall be determined in bandwidths of 0.1 µm or smaller. For certain lamp types, i.e. metal halide designs, it is recommended that the initial spectral determination be performed after the lamps have completed their burn-in period. The amount of infrared thermal energy at the collector plane shall be suitably measured (measurements in the wavelength range above about 2.5 µm if possible, but starting not beyond 4 µm) and reported (see 6.x.x). The thermal irradiance at the collector shall not exceed that of a blackbody cavity at ambient air temperature by more than 5 % of total irradiance. The collimation of the simulator shall be such that the angles of incidence of at least 80 % of the simulated solar irradiance lie in the range in which the incident angle modifier of the collector varies by no more than ± 2 % from its value at normal incidence. For typical flat plate collectors, this condition usually will be satisfied if at least 80 % of the simulated solar radiation received at any point on the collector under test shall have emanated from a region of the solar irradiance simulator contained within a subtended angle of 60 or less when viewed from any point. NOTE 1 Additional requirements concerning collimation apply to measurement of the incident angle modifier (see 6.7). The irradiance shall be monitored during the test and shall not vary by more than ± 3 % during the test period. The method used for measuring the irradiance during the test period shall produce values of mean irradiance which agree with those determined by spatial integration to within ±1%. NOTE 2 The spectral distribution of the lamps (indoors) and of the sky (outdoors) can and do lead to very wide discrepancies in spectrally selective absorbers or covers Test installation Collector mounting and location requirements outlined in 6.1 shall be followed. A wind generator shall be used with a solar simulator to produce an air flow in accordance with Preconditioning of the collector The procedure outlined in shall be followed Test procedure The collector shall be tested over its operating temperature range in approximately the same way as specified for outdoor testing (see 6.4.4). However, eight test points shall be adequate for testing in solar simulators provided that at least four different inlet temperatures are used, and adequate time is allowed for temperatures to stabilize. One inlet temperature should lie within ± 3 K of the ambient air temperature, if possible. During a test, measurements shall be made as specified in These may then be used to identify test periods from which satisfactory data points can be derived Measurements during tests in solar irradiance simulators General Measurements shall be made as specified in NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

277 ANNEX 8 page 38 of 41 pages Measurement of simulated solar irradiance NOTE Simulated solar irradiance usually varies spatially over the collector aperture as well as varying with time during a test. It is therefore necessary to employ a procedure for integrating the irradiance over the collector aperture. Time variations in irradiance are usually caused by fluctuations in the electricity supply and changes in lamp output with temperature and running time. Some lamps take more than 30 min to reach a stable working condition when warming up from cold. Pyranometers may be used to measure the irradiance of simulated solar radiation in accordance with Alternatively, other types of radiation detector may be used, provided they have been calibrated for simulated solar radiation. Details of the instruments and the methods used to calibrate them shall be reported with the test results. The distribution of irradiance over the collector aperture shall be measured using a grid of maximum spacing 150 mm, and the spatial mean deduced by simple averaging Measurement of thermal irradiance in simulators The thermal irradiance in a solar simulator is likely to be higher than that which typically occurs outdoors. It shall therefore be measured to ensure that it does not exceed the limit given in The mean thermal irradiance in the collector test plane shall be determined whenever changes are made in the simulator which could affect the thermal irradiance, and at least annually. The mean thermal irradiance in the collector test plane and the date when it was last measured shall be reported with collector test results Ambient air temperature in simulators The ambient air temperature ta in simulators shall be measured, taking the mean of several values if necessary sensors shall be shielded in order to minimize radiation exchange. The air temperature in the outlet of the wind generator shall be used for the calculations of collector performance Test period The test period may be determined in the same way as for outdoor steady-state testing. The more stable environment of an indoor test facility may allow steady-state conditions to be maintained more easily than outdoors, but adequate time shall still be allowed to ensure proper steady-state operation of the collector as specified in Test conditions The test conditions described in for outdoor testing shall be observed with the following additions: - The thermal irradiance in the plane of the collector aperture shall not exceed that from a blackbody cavity at ambient air temperature by more than 5 % of the total irradiance. - The air issuing from the wind generator shall not differ in temperature from ambient air by more than ±1 K Computation and presentation of results The analysis presented in for outdoor testing is also applicable to solar simulator tests, and the results shall be presented on the format sheets shown in annex D. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

278 ANNEX 8 page 39 of 41 pages 6.6 Determination of the effective thermal capacity and the time constant of a collector Work in progress 6.7 Incidence angle modifier Work in progress 6.8 Determination of the pressure drop across a collector Work in progress 6.9 Determination of the leakage rate The testing of the solar collector to determine its air leakage rate can be conducted to generate a leakage curve that can be used to compare the relative leakage of different collectors. Air leakage due to the test loop can have a significant effect on the results when testing a collector. Air leakage of the test loop, excluding the collector, shall be less than one-half of 0.5% of the manufacturer's recommended operating flow rate or m³/min, whichever is greater, at 250 Pa and shall be determined by a static air leakage test. The test for measuring collector leakage rate will utilize equipment that determines volumetric leakage rate with an accuracy to ±3% of reading. Measurement of collector pressure will be made with a manometer that meets the accuracy provisions of Section The collector is allowed to come to thermal equilibrium with the ambient air temperature. Using ambient air, a leakage curve is determined by either evacuating or pressurizing the collector using a calibrated flow-measuring device and by measuring the pressure difference between the solar collector and the ambient. The leakage test will be performed at negative gauge pressure for collectors that will be thermally performance tested at negative pressure and the test performed at positive gauge pressure for collectors that will be performance tested under positive pressure. Sufficient data points of collector pressure and leakage should be measured to allow accurate interpolation of collector leakage at operating pressures between 0 and the manufacturer's maximum published operating pressure, whichever is higher. For those cases where there is no published maximum pressure, the collector shall be tested at an operating pressure between 0 and 250 Pa. At least four data points will be taken at collector pressures between these levels. A leakage curve will be drawn and reported along with the actual data points. An example of a representative leakage curve is shown in Fig. 6. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

279 ANNEX 8 page 40 of 41 pages Fig. 6: Example of a leakage curve for a flat-plate air collector for positive internal pressure with outward leakage Test Apparatus Air leakage shall be minimized by sealing all joints except those that are part of the collector or the manufacturer's collector assembly. The portion of the duct loop that requires testing will include all sections of ductwork that contain temperature-, pressure-, or flow-measuring stations used for measuring collector performance and transition fittings/ducts used for collector hookup. This includes the inlet duct from before the flow-measuring station to the collector and the outlet duct from the collector discharge point to the outlet of the flowmeasuring station. The applicable sections of the test loop may be capped off and tested separately or together to determine air leakage. Fig. 7: Schematic of apparatus used for measuring air leakage in air collectors. The recommended leak check apparatus is shown in Fig. 7. This apparatus utilizes an orifice or nozzle mounted in a section of straight pipe, a motor and blower, and a flow control damper. Pressure taps are installed on either side of the orifice and are connected to a manometer for flow measurement Presentation of results Work in progress graphs for eta t eta t mass flow t-mass flow eta mass flow pressure drop leakage rate NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

280 ANNEX 8 page 41 of 41 pages NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission

281 WP4.1 Resource document Recommendations on testing of solar air collectors Dissemination level: Public Author: Josef Buchinger, arsenal research Reviewer: Peter Kovacs, SP June 2006 CONTENTS INTRODUCTION Introduction, why testing of solar air collectors is important and why standards are lacking so far. EXPERIENCE AND ACTIVITIES OF ARSENAL RESEARCH Description of the experience and activities of one of the main actors in the field to gain an insight and background knowledge of what has happened in testing of solar air collectors so far. ASPECTS OF TESTING SOLAR AIR COLLECTORS Introduction into the various technical problems of testing solar air collectors. RECOMMENDATIONS Recommendations for further activities to achieve a standard for testing. Technical suggestions for testing of solar air collectors. ACKNOWLEDGEMENTS FOR EARLIER REPORTS Hubert Fechner, arsenal research SUMMARY Solar Air systems are a further promising technology in active using solar energy for heating. Differently to solar liquid systems, air systems so far have not entered the market with significant rates. As main obstacle for a wide dissemination of solar air systems appears lacking information as well as lack of confidence on how these systems will perform. Planners and architects as well as end consumers need trustworthy parameters and facts to start applying and investing in this technology. Testing of the respective components is therefore essential but up to now no such widely accepted standard exists for air collectors. Based on the experience of arsenal research the technical problems such as calculation and presentation of the thermal performance are introduced and it is pointed out that testing of solar air systems in general is difficult and many technical aspects still need to be investigated. It is recommended to start serious activities which will ascertain these aspects and to gain a profound base for drafting a standard procedure for testing the performance and quality of solar air collectors. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6 File: ANNEX 9 Recommendations on testing of solar air collectors.doc

282 ANNEX 9 page 2 of 13 pages Table of contents SUMMARY...1 Table of contents Introduction Experience and activities of arsenal research IEA Task 19: Solar air systems Further activities Latest activities Aspects of testing solar air collectors Thermal performance subject to reference temperature Presentation of Efficiency Curves Closed loop system Open loop system Temperature rise Recommendations Need for further investigation Testing Conditions General Testing Features Symbols, Units and Abbreviations References...13 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

283 ANNEX 9 page 3 of 13 pages 1 Introduction Solar air collectors are not wide spread so far. As main obstacle for a wide dissemination appears lacking information as well as lack of confidence on how these systems will perform. Testing of the respective components is therefore essential. Such tests should be reproducible and acknowledged and are therefore commonly standardised in e.g. European Standards. So far no such standard exists for air collectors. Development of a standard could be done by research institutes familiar with testing and standardisation, but require resources. Those resources could be made available from the industry or from public funds. As mentioned in the beginning the market for solar air collectors is weak; hence the current situation is that the air collector industry is not interested in supporting the definition of a standard. Further, to gain national public funds the authorities require contribution from national industry. Unfortunately the situation in Austria is such that a research institute (arsenal research) is willing to start the process of standardisation but is lacking of local support by the industry since there is none in the field of air collectors and therefore lacking financial support. Achieving high standards of air collector measurements is not a simple task; Generally, measuring of air-temperatures and air mass flows requires much higher effort for gaining satisfactory accuracy s. Moreover, leakage, the air flow pattern inside the collector and the much lower heat transfer from the absorber to the heat transfer medium are further complex affects. The assembling of the air system components, the way how the components are connected, how the system is operated are all very decisive factors for the efficiency of the whole air system. Compared to liquid collectors, the measuring procedure for solar air collectors needs even more expenditure; no satisfactory standardised testing procedure exists so far. The Italian Standard UNI only gives an idea of how testing of air collectors can be carried out, but does not touch the specific problems of solar air systems in its 12 pages. Starting a standardisation process for testing solar air collectors has been already discussed in the Technical Committee 180 of the International Standardisation Organisation (ISO), but work is still resting. 1 UNI 8937: Collettori solari piani ad aria Determinazione del rendimento termico, Norma Italiana, NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

284 ANNEX 9 page 4 of 13 pages 2 Experience and activities of arsenal research In order to start solving the problems of solar air collectors arsenal research has had some small and major projects starting from investigating solar air systems up to activities within this project to start a self financed group on defining a standard for testing of solar air collectors. 2.1 IEA Task 19: Solar air systems In order to pool the experience in designing air systems for space heating, the International Energy Agency (IEA) initiated a five year project: Within Task 19 Solar Air Systems of the Solar Heating and Cooling Programme more than twenty experts from nine countries, coordinated by the Operating Agent Arch. Robert S. Hastings worked together. One part of this task was an investigation on series produced solar air collectors, done by arsenal research in 1999 (Fechner 1999). Seven long time proven products as well as prototypes from seven different countries, mainly from Europe but also from Canada and Australia have been tested. The main topics of development, investigation and research during this project have been: Development of a steady state testing procedure for solar air collectors, suited for all types Discussion on physically correct and proper efficiency presentations Development of different performance descriptions adequate for all common operation modes A comparison of available series-produced products Investigation of the technical behaviour of different types of air collectors Recommendations for an optimised utilisation of solar air collectors Recommendations for improvements of tested products Adaptation of the existing solar-laboratory-facilities for testing solar air collectors 2.2 Further activities In recent years arsenal research has tried to gain further development projects in this field. One very promising activity was the thoroughly investigation and testing of prototypes of an air collector totally made of polymeric materials (Selke 2005). 2.3 Latest activities Initiated by the solar air collector producer Grammer Solar GmbH arsenal research has tried to assemble producers and testing institutions to define a draft for a European Standard of testing air collectors. A small survey beyond actual producers of solar air collectors has been undertaken to get an impression on the need for a standard and the willingness to support a project to define such a testing procedure. The findings were that only the initiator had a significant interest where as other producers either did not feel the need for such a standard since they are making their business even without any standards or did not want to invest in such a poor dogs or question mark (regarding the definitions used in SWOT analysis) product. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

285 ANNEX 9 page 5 of 13 pages 3 Aspects of testing solar air collectors The function of solar-air collectors is to efficiently transfer the energy of the sun (radiated to the earth by short-wave rays) into thermal energy. To assess the quality of this conversion process under different operation conditions is the aim of collector-testing. With these results customers and planners should be enabled to compare different products, calculate or simulate the outcome of systems and plan and design systems. The current problem in testing air collectors are missing standards for: the definition of reference temperature for the different types of air collectors presentation of efficiency curves or other performance indicators for best usability and significance testing procedures, instrumentation and sensor arrangements for different types of air collectors procedures to avoid condensation and monitor humidity As stated before, measuring of air temperatures and air mass flows requires higher effort for gaining comparable accuracies. Moreover, leakage, the air flow distribution inside the collector and the much lower heat transfer from the absorber to the heat-transfer-medium are further complex effects. Opposite liquid solar collectors the efficiency of solar air collectors is strongly influenced by the actual mass flow rate inside the collector due to the often rather low heat transfer between absorber and air. This heat transfer is highly dependent on the air speed. It is, therefore, often difficult/impossible to extrapolate from tests of small modules of solar air collectors in test rigs to larger solar air collector arrays as the air flow pattern might be different. In the following some aspects are presented in detail to exemplify the problems of testing solar air collectors. 3.1 Thermal performance subject to reference temperature Optical features (absorption and emittance of the absorber, transmittance of the cover), materials used (absorber material, cover material, frame, insulation) and constructing characteristics (mainly the airflow-principle and the effective heat transfer area) of the collector are of basic importance for the efficiency. However, the respective operation condition of the collector is decisive as well and the efficiency decreases with increasing temperatures within the collector because of the increasing heat losses. The efficiency of a solar-(air)-collector is defined as the ratio of useful gain of the collector to the respective solar performance of the sun G T at the collector reference area A C. Q& u Q& u Efficiency (eta): η = = Q& Q& u = m& c p ( To Ti ) A G sol c T As collector reference area can be considered: aperture area, absorber area or gross area. A general equation for solar collector performance is based on (Hottel 1958) and (Bliss 1959): η = F τ 0 0 α U L ( T Q& o sol T a ) NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

286 ANNEX 9 page 6 of 13 pages where T a is the ambient temperature and F 0 is the collector heat removal factor in relation to T 0 - the collector outlet temperature and η 0 is the efficiency when the outlet temperature is taken as reference. F 0 accounts for the fact that the absorber temperature is not the same as the outlet air collector temperature neither in the horizontal direction nor vertical. It appears that the efficiency η depends on the operation conditions of the collector. It decreases with increasing temperatures, because of the increasing heat losses. Important is, how to define the operation conditions of the collector given by the temperature difference between the overall collector temperature T K and ambient T a. An efficiency curve can be drawn in dependency of a certain reference temperature which corresponds to the collector temperature T K. In a physically correct way one has to take a weighted mean temperature T K of the whole collector box, but in the measuring practice this is not practicable. That is why three temperatures are for choice: The inlet temperature (T i ), the outlet temperature (T O ) and a so called mean collector temperature (T m ) which can be calculated as the arithmetic mean value between inlet and outlet temperature. Efficiency curves of a solar collector corresponding to the three possible reference temperatures for a constant mass flow rate are shown below: Fig. 1: Efficiency related to different reference temperatures (Fechner 1999) For solar liquid collector it is the custom to present the efficiency related to the mean collector temperature (T m ) representative for the heat losses of the collector. For liquid collectors, where the temperature difference between inlet and outlet is very small (normally less than 10 K) and the heat transmission from the absorber to the fluid is high, the arithmetical mean value (T m ) is in fact very close to the physical mean temperature (T K ) of the collector. For air collectors the difference between inlet and outlet can be up to 30K or 40K dependent on the mass flow. Also important is the amount of heat transmission from the absorber to the fluid, which is for air collectors usually not that high. Due to these effects there will be no longer a linear increase of the fluid temperature along the collector plate and the arithmetical mean value (T m ) is often not representative for the heat losses of the collector. Measurements indicate that the physical mean temperature (T K ) of the collector is often much closer to the outlet temperature (T O ) than to the arithmetical mean temperature (T m ). Therefore the presentation of the collector efficiency curve using the outlet temperature (T O ) often seems to be the best solution. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

287 ANNEX 9 page 7 of 13 pages 3.2 Presentation of Efficiency Curves Solar air collectors usually operate in two different ways: 1. closed loop system 2. open loop system It is recommended to use separate ways to present the efficiency curves for the different systems Closed loop system In a closed loop system the inlet temperature can be sometimes much higher than the ambient temperature. For this mode of operation the presentation of efficiency ηversus (T o -T a )/G is adequate. Fig. 2: Efficiency related to Outlet temperature (Fechner 1999) NOTE: Be aware of the fact that the efficiency values close to y-axis would only appear, if the inlet temperature is below ambient, because if T o = T a, T i is always below T a. The maximum efficiency which will therefore appear in practice (if operating with air with at least ambient temperature) can be seen from the second curve (efficiency vs. massflow rate) Open loop system If the collector operates in an open loop, it always sucks in air with ambient temperature (T i = T a ), then a presentation of the efficiency η versus the mass flow rate is better suited for engineering purposes. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

288 ANNEX 9 page 8 of 13 pages Fig. 3: Efficiency versus mass flow rate (Fechner 1999) NOTE: Pay attention to the fact, that both the increased heat transfer and the fact that the higher the mass flow rates the lower the mean temperature of the collector (which is responsible for the heat losses) makes the conversation process more efficient at higher mass flow rates. 3.3 Temperature rise Another important aspect for planners is the temperature-rise. For direct heating you often need a certain level of the outlet-temperature. In such a case, optimisation towards high efficiency is no longer practicable. The diagram below shows the temperature rise for different irradiance levels. Fig. 4: Temperature rise vs. mass flow rate (Fechner 1999) NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

289 ANNEX 9 page 9 of 13 pages 4 Recommendations 4.1 Need for further investigation As it can be seen from the results and experience described above many aspects in testing solar air collectors still need further investigation, research and discussion. It seems that to get out of the vicious circle where the solar air collectors are at the moment, it requires an initiative beyond this project to solve the problems of missing standards and guidelines for planners and architects. It is therefore recommended to initiate an additional activity of producers, planners, architects, research and testing institutes to investigate thoroughly the different aspects of testing. To describe the relevant basics, define procedures for testing and find the best solution of presenting results, so that everybody is enabled to compare solar air collectors and calculate their performance within a solar air system. In cooperation with the German air collector producer Grammer Solar a workshop on solar air collectors will be organised in Austria this autumn. If relevant stakeholders will get together it might be a next step towards working out a standard for air collectors. Still, their further activities will need the support by national and international funding institutions and the industry. 4.2 Testing Conditions As an entry point and until further results are available the following recommendations for testing conditions derived from the experience made under the IEA TASK 19 can be made. For achieving comparable and reproducible test results, the conditions for testing must be defined carefully. Basis, whenever reasonable, are the conditions written in the standards for testing of liquid collectors (EN ). Especially the radiation elements remain the same. In the following the recommendations made earlier in (Fechner 1999) are updated and summarised. From liquid systems we know that judging a solar system is often to much concentrated on the thermal performance of the collector; other features of the system like control strategy, mounting of temperature sensors, connecting the modules, storages, insulation matters and many other questions should also be considered carefully. Further Investigations seems to be necessary in the general issue of presenting the thermal energy output of air collectors. The problem with reference area well known from liquid collectors as well as the problem of the reference temperature is open for further discussions General Testing Features Facilities for indoor measurements Artificial radiation source IR-shielding device simulating the cold sky temperature Surrounding air speed generated by fans (wind simulation) Automatic electro-pneumatic driven facility for measuring the global radiation Temperature precision control for stability of the inlet temperature of about ±0.05K Cooling machines for constant ambient temperature NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

290 ANNEX 9 page 10 of 13 pages Measurement In- and outlet temperature Surrounding air temperature Mass flow rate: orifice plates in combination with high precision pressure gauges Global radiation: Pyranometer Devices for controlling and stabilising the voltage supply Barometric pressure measurement Surrounding air speed measurement Dew point measurement Static pressure measurement at inlet and outlet Tests Efficiency testing (temperature rise, mass flow rate, specific heat) Wind dependence Leakage rate Pressure drop Stagnation temperature Time constant Temperature distribution in the air channel Buoyancy effects Tilt angle Tilt angle for testing differs according to the typical installation and should always be arranged according to the manufacturer Mass flow rates Mass flow rates recommended for testing can be ranging from 20 to 100 kg/h/m² if no special arrangements with the manufacturer are taken Leakage For an accurate measuring process 2 fans are needed, one at the inlet and one at the outlet. In order to minimise the leakage rate the mean static pressure inside the collector should be equal to the atmospheric pressure. Realistic conditions can be simulated by arranging the fan according to the producer s recommendations either before or after the collector Mass flow measurements Due to possible leakages it is recommended to test the mass flow in the inlet ducts as well as in the outlet ducts. Moreover leakages in the duct system will be found easily if the mass flow is measured twice Air-flow pattern The air flow pattern inside the collector is very important for a correct assessment of the performance. The air flow pattern inside the collector (especially the partition close to the inlet) mainly depends on the connection between ducting system and collector. In standardised tests only a single collector module is tested, which might not comply with the mode of installation in praxis. To reach an even air flow pattern throughout the collector special distribution ducts at the inlet and outlet should be used for each collector tested. By means of boxes with perforated metal sheets at inlet and outlet a well distributed air flow can be achieved, which means that an even air flow from the centre-line of the collector to the edges from entrance to outlet exists. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

291 ANNEX 9 page 11 of 13 pages Temperature measurements While correct measurements of the inlet temperature should be of no problem, the determination of the outlet temperature is critical. Several layers of different air temperatures are often close adjacent, therefore a specific mixing device optimized according to fluid dynamic experiences - at the outgoing duct just before the sensors and a sophisticated arrangement of temperaturesensors are necessary Wind simulation For testing collectors it is essential to find out the wind dependency of a collector. Some collectors, especially uncovered, but also collectors with the air flow directly under the cover are strongly dependent on wind. Recommendations for installation of the wind simulation can be found in the existing standard for testing water collectors. Only for uncovered collectors the direction of the wind can be influential as well Humidity The humidity should be monitored during the tests. Care should be taken for effects of condensation, since for some testing points inlet temperatures far below ambient can be necessary. Any eventual condensation inside the collector or the circuit must be removed while heating the collector circuit at 70 C for a period of at least 30 minutes Irradiation level Although irradiation levels used at solar air applications are sometimes low, for testing levels starting with about 700 W/m² are recommended. Testing at lower irradiation levels should be avoided to ensure a significant temperature rise between the inlet and outlet and further accuracy of measurements Conditioning For testing the collector with different air temperatures, the preparation of differently conditioned air is needed. To attain foreseen stationary/stable conditions, the test circuit must be preconditioned for 15 minutes at the test temperature so as to verify that the inlet temperature is well within the conditions. For indoor-testing an enclosed climatic chamber with reasonable size and temperatures between some degrees below zero and up to 60 C is recommended. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

292 ANNEX 9 page 12 of 13 pages 5 Symbols, Units and Abbreviations A C c p Collector reference area (normally: aperture area) [m2] Specific heat [Jkg-1K-1] η Efficiency F 0 collector efficiency factor G T Global solar irradiance on tilted surface (further on only G) [Wm-2] G Global solar irradiance [Wm-2] m Mass flow rate [kgh-1] m i Mass flow rate at collector inlet [kgh-1] m L Leakage Mass flow rate [kgh-1] m o Mass flow rate at collector outlet [kgh-1] Q sol Performance irradiated by the sun at the collector reference area [W] Q u Useful gain of the collector [W] T a Ambient temperature [ C] T am Measured absorber temperature [ C] T i Inlet temperature [ C] T K Collector temperature (physical collector-mean-temperature) [ C] T m Mean Collector temperature (arithmetic mean value between inlet and outlet temperature) [ C] T O Outlet temperature [ C] U L Collector overall heat loss coefficient α Solar absorptance ε Hemispherical emittance τ Transmittance NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

293 ANNEX 9 page 13 of 13 pages 6 References Bliss R. (1959). "The Derivation of Several Plate Efficiency Factors Useful in the Design of Flatplate Solar-heat Collectors." Solar Energy Vol. 3: pp Fechner H. (1999). IEA TASK 19 Solar air systems - Investigation on Series Produced Solar Air Collectors - Final Report. Vienna, arsenal research - Department of Renewable Energy. Hottel H., Whillier, A. (1958). Evaluation of Flat Plate Collector Performance. Trans. Conf. On the use of Solar Energy, Tucson, Arizona, U.S.A. Selke T. (2005). Studie zur Darstellung des technischen und wirtschaftlichen Marktpotentials des Solar-Luftkollektors der BAYER AG. Wien, arsenal research: 73. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

294 ANNEX 10 page 1 of 6 pages WP4.1 Incidence angle modifier measurements and application of EN to tracking and concentrating collectors Dissemination level: Public Author: Stephan Fischer Reviewer: Maria Joao Carvalho, INETI and Peter Kovacs, SP September 2006 NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

295 ANNEX 10 page 2 of 6 pages Incidence angle modifier measurements and application of EN to tracking and concentrating collectors 1 Collector incident angle modifier 1.1 General The angle of incidence θ 1 (fig. 1) is defined as the angle between the line connecting the sun and the collector plane and the perpendicular line from the collector to the collector zenith. sun collector zenith θ collector plane Figure 1: Definition of the angle of incidence The thermal performance of solar thermal collectors changes with the angle of incidence θ. This is taken into account by introducing the incident angle modifier K θ defined by eqn. 1. η0 ( θ ) K θ ( θ ) = (1) η0 ( θ = 0) The European Standard EN /1/ suggests two different methods to do this: 1. multiplying the hemispherical irradiance G with the incident angle modifier K θ (eqn. 2) (steady-state method). This method is only applicable if the diffuse irradiance incident on the collector plane does not exceed 30 % of the hemispherical irradiance. Q& 2 = η GK θ ) a ( t t ) a ( t t (2) 0 θ ( 1 m a 2 m a ) A 2. subdividing the hemispherical irradiance into direct irradiance G b and diffuse irradiance G d, multiplying both fractions with the accordant incident angle modifier (eqn. 3) (quasi-dynamic method). K θb being a function of θ and K θd a constant value independent of θ. Q& 2 4 dt = η0 ( GbKθ b ( θ ) + Gd Kθ d ) c6ug c1 ( tm ta ) c2 ( tm ta ) c3u( tm ta ) + c4 ( EL σta ) c5 A dt (3) m 1 This paper uses symbols according to EN :2006 /1/. In case new values are introduced the corresponding symbols are chosen in close correlation to the one used in EN :2006. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

296 ANNEX 10 page 3 of 6 pages Introducing the diffuse fraction D = G d /G in eqn. 1 the incident angle modifier can be written as η0 ( θ, D) η0kθ b ( θ )(1 D) + η0kθ d D Kθ ( θ, D) = = (4) η0 ( θ = 0, D) η0 (1 D) + η0kθ d D Using eqn. 4 the incident angle modifier can be calculated for any diffuse fraction. 1.2 Collector types Regarding the behaviour in respect to different angles of incidence, basically 3 types of collectors can be distinguished: 1. collectors with isotropic behaviour in respect to the direction of incidence (isotropic collectors). 2. collectors with bi-axial behaviour in respect to the direction of incidence (bi-axial collectors) 3. collectors with multi-axial behaviour in respect to the direction of incidence (multi-axial collectors) To assign a given collector to one of the three types fig. 2 together with the definitions given below are used. collector zenith longidinal axis north sun transversal plane west transversal plane east transversal axis west θl θt θ transversal axis east collector plane longidinal axis south Figure 2: Definition of used planes and axes. Figure 2 shows a 3-dimentional co-ordinate system having its centre in the middle of the collector plane. The longitudinal axis runs from north to south (top to bottom) and the transversal axis from left to right (east to west). Perpendicular to the collector plane and the two axes are the longitudinal and the transversal plane. The intersection line of the two planes is the axis perpendicular to the collector plane. The punching die divides each of the two planes into two halves, the longitudinal plane into north and south and the transversal plane into east and west. At angles of incidence smaller than 90 the connecting line between sun and collector plane will always be between two of the four areas. The longitudinal angle of incidence θ L is the angle between the perpendicular on the collector plane and the projection of the connecting line between sun and collector plane into the longitudinal plane. The transversal angle of incidence θ T is the angle between the perpendicular on the collector plane and the projection of the connecting line between sun and collector plane into the transversal plane. In case a NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

297 ANNEX 10 page 4 of 6 pages distinction is needed between north and south or east and west the indices N, E, S or W are used. The mathematical correlation between θ, θ L, and θ T is given by equation (5) tan θ = tan θ L + tan θ T (5) For isotropic collectors the collector output is independent of the direction of the incoming irradiance. The most common representative of this type is the flat plate collector. The collector output can be described using an incident angle modifier as a function of the angle of incidence only, K θ = f(θ). For bi-axial collectors the collector output is different if the solar irradiance is parallel to the north-south (longitudinal) axis and the east-west (transversal) axis respectively. However symmetry to the longitudinal and the transversal planes is given. Typical representatives of this category are evacuated tubular collectors, collectors using CPCs and parabolic trough collectors. The collector output must be described using an incidence angle modifier as a function of both longitudinal and transversal angle of incidence, K θ = f(θ L, θ T ). For multi-axial collectors the symmetry to either the longitudinal or the transversal plane or to both planes is not given. Representatives of this collector type are evacuated tubular collectors with absorbers not parallel to the collector plane /2/. The collector output must be described using an incidence angle modifier as a function of all relevant angles of incidence, K θ = f(θ LN, θ TE, θ LS, θ TW ). To apply these definitions to a collector the collector top, bottom, left hand and right hand side have to be defined (Fig.3). Top (north) left (west) right (east) bottom (south) Figure 3: Definition of collector orientation. 1.3 Calculation of incidence angle modifiers Isotropic incidence angle modifier According /1/ isotropic incident angle modifiers are calculated using equation 6, 1 K θ ( θ ) = 1 b0 1 (6) cos( θ ) with b 0 determined out of collector test data using a parameter identification method. However this approach does not fit all isotropic collectors. E.g. flat plate collectors with transparent insulation material (TIM) can show a different incident angle modifier behaviour. To overcome this problem and to establish a calculation method that can be used for all three types of incidence angle modifiers a table is introduced to characterise the isotropic incident angle modifier. Table 1 shows the incidence angle modifier of a typical flat plate collector. NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

298 ANNEX 10 page 5 of 6 pages θ K θ (θ) Table 1: Incidence angle modifiers of a typical flat plate collector The value for K θ (θ=0) being 1 is given per definition (eqn. 1) and value for K θ (θ=90) can for all collector types approximated by 0. Incidence angle modifiers for angles in between the supporting points are calculated by linear interpolation between the two adjacent supporting points. The number of necessary supporting points between θ = 0 and θ = 90 depend on the complexity of the incidence angle modifier Bi-axial and multi-axial incidence angle modifier As shown by McIntrie /3/ bi-axial incidence angle modifier can be approximated using equation 7. Kθ b ( θ L, θt ) = Kθ b ( θ L, θt = 0) Kθ b ( θ L = 0, θt ) (7) The values for K θb (θ L,θ T =0) and K θb (θ L =0,θ T ) are taken or calculated from a table in the way described in section This approach may also be used for the approximation of multi-axial incidence angle modifier using the accordant indices. 1.4 Test procedures To determine the incidence angle modifiers the test procedures documented in /1/ shall be used. According to /1/ the measurement for bi-axial and multi-axial collectors has to be performed always for one axis keeping the incident angle modifier for the other axis to ± 2 % of the value at normal incidence. To fulfil this requirement two possibilities are given: 1. Rotating the collector by 90 before the measurement of the incidence angle modifier of the other axis. However this is not for all collectors possible, e.g. collectors with heat pipes. 2. Increasing the collector azimuth to at least 30 east or west. The suitable deviation from the south has to be calculated depending on the time of the year, the tilt angle and the incidence angle modifier already determined. 2 Performance testing of concentrating tracking collectors 1.1 General The EN /1/ is basically applicable for the test of tracking and tracking concentration collectors. However some consideration regarding the tracking system, the concentration ratio, the irradiance measurement and the irradiance used for the calculation of the collector performance have to be made in advance. Tracking system: In order to characterise the thermal performance of a tracking collector (concentrating or not) the test should be performed with the collector mounted on the appropriate tracking system. The collector identification name as well as the identification of the tracking system shall be documented within the test report. In case the collector is delivered without a tracking system this must be stated in the test report. Irradiance used for the calculation of collector performance: Historically DNI (direct normal irradiance) is used for performance testing of tracking concentrating collectors /4/. For high concentrating rations the influence of diffuse irradiance can be neglected. However at NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

299 ANNEX 10 page 6 of 6 pages small concentration ratios this approach might lead to an overestimation of the thermal performance. To avoid this and to be consistent with /1/ both, direct and diffuse irradiance shall be used while testing and calculating the thermal performance. To be able to distinguish between direct and diffuse irradiance the test method according section 6.3 of /1/ (quasidynamic test method) shall be used. Concentration ratio: Due to different definitions of the concentration ratio /4/ a concentration ratio shall only be mentioned/documented together with the accordant definition. Irradiance measurement for tracking concentrating collectors: To be consistent with /1/ the hemispherical irradiance respectively the direct and diffuse irradiance must be measured in the aperture area. For tracking concentrating collectors this shall be done using a pyrheliometer mounted on a suitable tracking device (in case of a two axis tracking of the collector the pyrheliometer may be mounted on the collector tracking device). The hemispherical irradiance shall be measured with a pyranometer in the aperture of the collector. The diffuse irradiance shall be calculated by the difference between hemispherical and direct irradiance in aperture plane. Instrumentation: Pyranometer first class acc. ISO 9060 or better. Pyrheliometer first class acc. ISO 9060 or better. 2.2 Test and calculation procedure The test procedure and conditions according to /1/ section 6.3 under consideration of this paper apply for the test of tracking and concentrating tracking collectors. However K θd and c 2 should be removed from the list of mandatory collector parameters. At larger concentration ratios both parameters are likely not to fulfill the T-ratio criterion lead down in /1/. In this case the parameters should be neglected /5/. For CPC type collectors with concentrations above 1, although used in stationary mode, the test should also be done according to the test method 6.3. Using this method will take into account that this type of collector sees only part of the Diffuse Radiation 1/C. References /1/ EN :2006. Thermal solar systems and components Solar collectors Part 2: Test methods /2/ S. Fischer et al. Test and Simulation of solar thermal collectors with multi-axial incidence angle behaviour. Proceedings of the 2005 Solar World Congress. Orlando USA. /3/ McIntire W. R., Factored approximations for biaxial incident angle modifiers. Solar Energy 29, , 1982 /4/ E. Lüpfert, U. Herrmann, H. Price, E. Zarza, R. Kistner: Towards standard performance analysis for parabolic trough collector fields. draft for SolarPaces Conference Oxaca 2004 /5/ S. Fischer et al., Efficiency testing of parabolic trough collectors using the quasi-dynamic test procedure according to the European Standard EN Proceedings 13 th International Symposium on Concentrating Solar Power and Chemical Energy Technologies, June 20-23, 2006 Seville, Spain, ISBN: NEGST NEW GENERATION OF SOLAR THERMAL SYSTEMS is a project financed by the European Commission DGTREN within FP6

300 WP4-D2.1.k Resource document Definitions and test procedure related to the incidence angle modifier Dissemination level: Public Maria Joao Carvalho, INETI, Peter Kovacs, SP, Stephan Fischer, ITW Reviewer: Jan Erik Nielsen, ESTIF September 2006 CONTENTS COLLECTOR INCIDENT ANGLE MODIFIER General definitions, calculation methods are given. PERFORMANCE TESTING OF CONCENTRATING TRACKING COLLECTORS The application of the European Standard EN for concentrating tracking collectors is briefly described SUMMARY This report aims to clarify some aspects related to the incident angle modifier that are not addressed within detail within the European Standard EN 12975:2006. Coming from some basic information three collector types are defined depending on their behaviour in respect to the direction of incidence of the beam radiation. collectors with isotropic behaviour collectors with bi-axial behaviour and collectors with multi-axial behaviour In addition some suggestions for the calculation of incident angle modifiers are given. In the end the performance testing of tracking concentrating collectors is described.