THE EFFECT OF EXTERNAL SURFACE PROPERTIES ON THE THERMAL BEHAVIOUR OF A TRANSPARENTLY INSULATED WALL. Al. Politechniki 6, Lodz, Poland

Transcription

1 Proceedings: Building Siulation 7 THE EFFECT OF EXTERNAL SURFACE PROPERTIES ON THE THERMAL BEHAVIOUR OF A TRANSPARENTLY INSULATED WALL Dariusz Hei 1, Piotr Kle 1 and Paul Strachan 1 Departent of Building Physics and Building Materials, Technical University of Lodz, Al. Politechniki 6, 9-94 Lodz, Poland Energy Systes Research Unit, Strathclyde University, 75 Montrose Street, Glasgow, Scotland ABSTRACT The properties of transparent plaster covering transparent insulation aterials (TIM) were investigated using a whole building siulation progra (ESP-r). The outer plaster was ade fro glass balls of different diaeter, glued together with synthetic resin. The of the whole transparent covering layer (plaster + TIM) was estiated for different solar incident angles by laboratory easureents. The innovative character of the aterials required refining of ESP-r s optical database in order to take into account these new characteristics. The transparently insulated building facade was proposed as a solar energy storage syste. The results of the initial analysis showed the desirable optical properties, estiated for sun incident angles on the façade at a latitude of 5 degrees north. Then, siulations based on real cliatic data for Central Europe were conducted to predict the theral TIM wall behaviour. The influence of the structure on the diurnal heat storage potential was investigated for selected periods of the year. KEYWORDS transparent insulation; transission; solar INTRODUCTION The ajority of traditional transparent aterials such as glass, polycarbonate, polyethylene, etc. is regarded as having a poor theral resistance. However, increasing the theral resistance leads to a reduction in the direct solar, which consequently decreases solar radiation heat gains. Therefore, using an additional layer of Transparent Insulation Material (TIM) is reasonable in advanced, high efficiency passive solar systes. The general arrangeent of TIM constructions consists of an outer glass pane or plaster (to protect against weather conditions), a shading device (located optionally in the outer gap space) and a translucent capillary structure (in theral contact with the assive, opaque part of the wall), (Braun et al. 199). The shading device is necessary to prevent overheating in suer and to regulate energy gains during the transition periods. However, in the iddle of winter (Central Europe weather conditions), the absorber teperature does not exceed 3 C (Hei and Kle ) and blind/shutter control is not required. For TIM eleents with an air gap between the absorber and the capillary aterial a plastic (polycarbonate) fil or a thin glass pane covers the rear TIM face. In a odular construction, these coponents are held together by a frae. The thero-physical properties of transparent layers exposed to solar radiation play a key role in the energy efficiency of the whole storage syste. Considering the energy balance during the whole year, the aterial properties should provide the axiu heat gains during the winter period and a iniu during the suer period. The part of the solar energy transitted or reflected by the transparent layer strongly depends on solar incident angle. For flat glass panes the highest transparency is obtained for noral incidence of solar radiation. The TIM systes are significantly ore coplex than single glass. First of all, the outer, protective layer can be rough or consist of sall eleents with different orientations. It eans that the transitting/reflecting characteristic of the first layer differs fro a traditional glass covering. The ain part, the capillary transparent insulation, is also nonhoogeneous and the solar energy is transferred through ultiple reflections, refraction and absorption by the eleents of the honeycob structure. THEORETICAL ANALYSIS Traditional passive eleents operate optially during transition periods - the start and end of the heating seasons. Then the solar heat gains during the day equal the heating energy requireents during the night. In these periods the gains entioned above appear alost every day, relatively regularly, and the heat capacity of traditional building aterials is sufficient for diurnal heat storage systes. Nevertheless, in winter, low external air teperature causes excessive heat losses into the external environent. Also the solar heat gains are low and

2 Proceedings: Building Siulation 7 occur periodically. Due to the factors entioned above two kinds of probles are encountered. The first one is due to the unsatisfactory proportions between the solar heat gains and heat losses. The second concerns the and reflection of the outside layer exposed to the solar radiation with changeable incident angle. For these reasons, it is reasonable to eploy particular, additional strategies which iprove solar systes efficiency. The strategies are usually those which ai at axiized during the winter and iniized during the suer period. Hei and Puchala (7) gave an exaple analysis of theoretical liits for incident solar radiation on the south oriented elevation at a latitude of 5 degrees north. The liit of axiu solar elevation is 61 degrees (in June) while the axiu (in Deceber) is only 14 degrees. It eans that only solar elevation angles fro 15 to 6 degrees need to be considered. For sun elevation angles around 15 degrees (winter) the is close to 1 for solar radiation in the central part of the day. For elevation angles close to 6 degrees (suer) the iniu solar is required. In ESP-r the optical properties are defined for angles, 4, 55, 7, 8 degrees fro noral and linear interpolation is undertaken for interediate values. It eans that for purposes of further nuerical analysis only two angles will play the crucial role for Central European latitudes. They are 4 and 55 degrees. The theoretical analysis of the optial solar was done for aterial properties presented in Table 1. To find the optial aterial properties, ore than analyses were done for a south oriented elevation. External weather conditions correspond to a oderate cliate in Central Europe. Based on a ten years real cliate data set, the winter (with the lowest solar radiation) and the suer (with the highest solar radiation) periods were selected. Figure 1 presents the axiu solar gains in the winter and suer periods for total of the transparent layer (south oriented façade). Table 1 Solar at selected angles SOLAR ELEVATION SOLAR TRANSMITTANCE to to solar gains [W/ ] winter 4deg suer 4deg suer 55deg winter 55deg Figure 1 Solar gains through TIM for at 4 and 55 incident angles solar gains [W/ ] deg 55deg Figure The differences in solar gains through TIM in suer and winter at 4 and 55 incident angles The gains in suer at 55 degrees incidence angle are significantly (ore than five ties) higher than in winter. For 4 degrees, solar gains in winter only slightly exceed the gains in suer. Figure shows the gain differences between winter and suer. It clearly shows that the axiu solar (about.9 or higher) is required for angles between 4 and 45 degree. Above 45 deg the total should be close to.1 or less. Theoretically the best, ideal solar is shown in Figures 3 and 4 (dotted lines). This characteristic was prepared only based on the five ain angles in the ESP-r optical database. However, it is difficult to find a real aterial with such optical

3 Proceedings: Building Siulation 7 properties. The ain assuption for the presented analysis was the absence of any additional shading devices. In the presented analysis the ain paraeter was the axiu value of heat gains during the day, but not the total balance of the TIM (average daily heat flux through the TIM). TRANSMITTANCE OF COVERING Beginning in the early 199s, the Departent of Building Physics and Building Materials, Technical University of Lodz began investigating storage potential of solar walls. Many experiental analyses and investigations were done for the so-called intelligent façade including the potential of the storage part with phase change aterials (Hei and Clarke 4, Hei 6) and also the optical properties of transparent coverings (Hei and Kle 3, Grudzinska 4) including transparent insulation aterials. Based on the results fro laboratory easureents two capillary transparent insulation aterials were selected for further analysis: TIM layer thickness of 6 with cells of width (6/), TIM layer thickness of 1 with cells of 1 width (1/1). The external glass plaster differs in total thickness (1, or 3) and structure of coponent eleents (the diaeter of glass balls). The balls were sorted by diaeter in four groups: -315μ, 315-4μ, 4-63μ and 63-8μ. Soe experiental results of solar are presented in figures 3 and 4. The upper lines represent the highest solar s obtained for 1 plaster ade fro 63-8μ granulated glass (sall squares, test- 1). The botto lines show the lowest solar s received for the 3 glass plaster ade fro -315μ diaeter balls (big squares, test-3). The rest of the aterials have the curve between those presented in the graphs. For noral incidence, the of aterial 1/1 is 1% higher than for aterial 6/. The thickness of the plaster and diaeter of glass balls can cause significant differences for low incident angles (for noral incidence, a factor of between two and three). The differences becoe less iportant for higher incidence angles and for 8 and 9 degrees, the is zero for all types of aterials. Based on the experiental results the optical characteristics for five ain angles were calculated and introduced to the ESP-r optical database. The characteristic angles are arked by x. The values for 55deg were approxiated linearly between 5 and 6 deg test-1 test-3 odel-1 odel-3 ideal degrees fro noral Figure 3 The optical characteristics of 6 transparent insulation covered by glass plaster test-1 test-3 odel-1 odel-3 ideal degrees fro noral Figure 4 The optical characteristics of 1 transparent insulation covered by glass plaster Coparing the final results with the required characteristic for a south oriented façade (dotted line), the differences between 4 and 55 degree values is still not sufficient. However,, the excess of solar gains in suer can be eliinated by other strategies such as cutting off (shading), reoving (ventilation) or isotherally storing (application of phase change aterials). NUMERICAL ANALYSIS 6 Theral siulation of dynaic building behaviour under changeable cliatic conditions is perceived as essential to accurately assess the efficiency of

4 Proceedings: Building Siulation 7 eploying TIM. Advanced atheatical odels for calculating the energy transport in TIM systes have been incorporated into several siulation progras.. The ost popular progras are HAUSSIM, TRANSYS and ESP-r. HAUSSIM is based on the atheatical odel proposed by Hollands et al. (1984) which is able to nuerically solve the nonstationary teperature response of a TIM wall. Within TRANSYS a special calculation procedure was developed to odel the different properties of TIM (Platzer 199). The ESP-r progra (Clarke 1) is capable of odelling the energy and fluid flows within cobined building and plant systes. ESP-r uses an advanced nuerical ethod to integrate the various equation types which can be used to represent heat and ass balances within a building. At present, however, ESP-r does not contain a specific odule or specific algoriths to odel TIM explicitly. Considering the solar radiation transission and absorption through the TIM honeycob structure and the rest of the TIM facade, four possible ways of odelling a TIM wall within ESP-r were proposed (Strachan and Johnstone 1994). For the purpose of this work the air gap as an extra zone approach has been chosen (the air gap space was defined between the transparent and opaque part). A proposed odel of the TIM wall within ESP-r was fored as part of the PASSYS progra (Jensen 1993). Experients were conducted in soe of the PASSYS test sites, aong others, in Belgiu, Italy, Gerany and Holland. The objective was to carry out the easureents under different cliatic conditions in order to evaluate the perforance of such solar coponents. RESULTS The nuerical analysis was conducted using the real cliatic data for a city at a latitude of 5 degrees north. The period of the year was chosen based on the lowest and the highest sun altitude angle. For Deceber it is 14.6 and for June The façade with TIM was oriented to the south. The sun aziuth angle changes in the selected winter week fro 4.6 to +4.6 and in the suer week fro 116. to relative to the surface noral. The transparent insulation aterial was defined as hoogenous, covered fro the outside by the glass ball plaster and fro the inside by polyethylene plate. On the inner side of the plate the 5c wide space gap was defined and then the theral storage, a assive concrete wall. Two capillary insulation thicknesses were considered: 6 and 1. The total transissions were assued for the whole syste (three transparent layers) fro the easured data as presented in Figures 3 and 4. PARAMETER INCIDENT ANGLE ABSORPTIVITY plaster 6-1-b plaster 6-3-s plaster 1-1-b plaster 1-3-s insulation polycarbonate REFLECTIVITY solar radiation [W/ ] b 6-3-s 1-1-b 1-3-s 6-ideal Figure 5 Solar energy transitted through different types of TIM panels in winter solar transitted [W/ ] b 6-3-s 1-1-b 1-3-s 6-ideal I dir = 197 I dir = Figure 6 Solar energy transitted through different types of TIM panels in suer Table Solar absorption and reflection

5 Proceedings: Building Siulation 7 The absorptivities of the each eleent (plaster, TIM and polycarbonate) and the reflectivity are presented in Table. The reflectivity is assued to be constant for each type of the plaster as is the absorptivity of the TIM capillary structure and the polycarbonate. The absorptivities of several plaster types are derived fro the overall reflectivity and transissivity data. Solar gains The total solar energy transitted through the TIM syste with outer glass plaster and inner polycarbonate is presented in Figures 5 and 6, for winter and suer respectively. The dotted line represents aterial with ideal transission. Maxiu direct solar radiation for the analyzed winter day is 197 W/ and for the suer day it is 843 W/. In winter the best characteristics were obtained for thin, 1 glass plaster where the diaeters of the balls does not ake a big difference. The energy transitted through 1 of TIM is about 1% greater than through two ties thinner insulation (6). In suer the best characteristics are observed for aterials with 3 of plaster. The aterial with 6 TIM sees to work better than ideal TIM. However, two ties ore energy is transferred by the ideal aterial in winter than in suer, although the direct solar radiation is four ties higher. The other aterials transit siilar aounts of solar energy in both periods. The optical properties of aterials with the thin 1 plaster covering sees to be insufficient for the suer period and there is a danger of overheating. Energy balance The energy balance at the external transparent part of collector wall is presented in Figures 7 and 8. In the zone under consideration no heating and cooling syste was defined. Also casual gains fro equipent, light and occupants were set to zero in order to reove their ipact on the siulations and so ake the results interpretation easier. The positive surface convective flux in winter is noticed fro 9: to 14:. For 1 of TIM with both kind of plaster coverings, the axiu gains at noon are two ties higher than losses at idnight. For 6 of TIM the extree fluxes are siilar, but the balance over the day is negative. For the winter tie the best sees to be 1 of TIM with 1 of transparent plaster. In suer the best characteristics are observed for 6 of TIM, with the sae 1 glass balls plaster. However, the differences in energy balance are sall. The extree positive surface convective flux (at 1:) in suer is unfortunately three ties higher than in winter. surface convection [W] surface convection [W] b 6-3-s 1-1-b 1-3-s Figure 7 TIM surface convection in winter 6-1-b 6-3-s 1-1-b 1-3-s Figure 8 TIM surface convection in suer SUMMARY AND CONCLUSIONS I dir = 197 in o TIext = 4.5 C I dir = 843 in o TIext = 9.3 C This study is a contribution to developing the nuerical odel of radiation processes and refining the optical aterial databases in ESP-r. The additional characteristics added to the progra were obtained by laboratory easureents of existing transparent insulation aterials covered on the outside by glass ball plaster. An analysis was undertaken to deterine desirable TIM/plaster cobinations for Central European cliatic conditions. The obtained results show the effect of angledependent solar on the theral behaviour of the collector wall. This effect causes a considerable difference in solar energy absorbed by

6 Proceedings: Building Siulation 7 the assive part of storage wall as well as in the energy balance of its transparent part. Coparing the solar energy transferred by the TIM structure, the 1 of TIM covered by 1 of plaster ade fro glass balls (-315μ diaeter) turned out to be the best option in winter. However, in the hottest period with significantly higher direct solar radiation it was found that 6 of TIM with 3 of glass ball plaster (63-8μ) was ore suitable. None of the analyzed aterial cobinations were found to be optiu for all seasons. On the other hand, the technical solutions of TIM systes allowed the reduction of suer overheating by the use of additional strategies such as shading and night ventilation. It leads to the conclusion that systes should be designed to axiise solar gains in winter periods. Assuing that transparent insulation is cut off fro solar radiation during the suer to avoid overheating, axiu is required during winter for the low solar elevation. The highest arises fro the use of a very sall thicknesses of the outer layer, in the order of 1. Such a thin covering, with the additional requireent for safety, is possible only with the glass plaster and not with the ordinary glass pane. Looking at the aterials under consideration, the 1 of TIM covering with 1 of glass plaster ade fro sall balls is preferred. Hei, D. and Kle, P. 3. Nuerical Solution of TIM-PCM Solar Theral Storage Syste with ESP-r. Research in Building Physics, Careliet, J., Hens, H. and Vereir G. (ed.), Balkea Publishers: Hei, D. and Puchala, P. 7. Analiza własności optycznych Izolacji Transparentnej i ich wpływ na efektywność energetyczną przgrody. Building Physics in Theory and Practice, vol., pp Hollands, K.G.T., Raithby, G.D.. Russell, F.B. and Wilkinson, R.G Coupled Radiative and Conductive Heat Transfer across Honeycob Panels and through Single Cells. Int. J. Heat and Mass Transfer, 7(11): Jensen, S.O The PASSYS Project: Subgroup Model Validation and Developent Final Report Part I and II (1986-9). Coission of the European Counities, DGXII, EUR EN. Platzer, W.J Calculation Procedure for Collectors with a Honeycob Cover of Rectangular Cross Section. Solar Energy, Vol.48, No 6: Strachan, P.A.and Johnstone, C.M Solar Residences with Transparent Insulation: Predictions fro a Calibrated Model. Proc. North Sun '94, Glasgow: REFERENCES Braun, P.O., Goetzberger, A., Schid, J. and Stahl, W Transparent Insulation of Building Facades Steps fro Research to Coercial Applications. Solar Energy, Vol. 49, No. 5: Clarke, J.A. 1. Energy siulation in building design, nd edition.butterworth-heineann, Oxford. Grudzinska, 4. Warstwa powierzchniowa przegrody budowlanej o szczególnych właściwościach absorpcyjnych i transisyjnych proieniowania. PhD dissertation. Supervised by P. Kle. Hei, D. 6. Phase-Change Material Modeling within Whole Building Dynaic Siulation. ASHRAE Transactions, Vol. 11, Part 1, 6, pp Hei, D. and Clarke, J.A. 4. Nuerical Modelling and Theral Siulation of PCMgypsu Coposites with ESP-r. Energy and Buildings, Vol. 36, No. 8: Hei, D. and Kle, P.. Wplyw zastosowania izolacji transparentnej na dynaiczna wyiane ciepla w budynku. Proceeding of XLVI National Scientific Conference, Opole-Krynica: (in Polish)