Recommendations: Part 7: Transient Creep for service and accident conditions

Transcription

1 Materials an Structures/Matériaux et Constructions, Vol. 31, June 1998, pp RILEM TECHNICAL COMMITTEES RILEM TC 129-MHT: TEST METHODS FOR MECHANICAL PROPERTIES OF CONCRETE AT HIGH TEMPERATURES Recommenations: Part 7: Transient Creep for service an accient conitions The text presente hereuner is a raft for general consieration. Comments shoul be sent to the TC Chairman, Prof. Dr. U. Schneier, Technische Universität Wien, Institut für Baustofflehere un Bauphysik, Karlsplatz 13/206, A-1040 Wien, Austria, Fax: , uschnei@birisc.tuwien.ac.at, before December 1, The raft of this ocument has been prepare by the following 10 Committee full members representing 7 countries. TC Membership: Chairman: U. Schneier, Austria; Secretary: P. Schwesinger, Germany; Members: G. Debicki, France; U. Dieerichs, Germany; J.-M. Franssen, Belgium; F. Furamura, Japan; U.-M. Jumppanen, Finlan; G. A. Khoury, U.K.; A. Millar, France; W. A. Morris, U.K. 1. SCOPE This recommenation is vali for structural applications of concrete uner service an accient conitions. This ocument presents test parameters (material an environmental), an test proceures for etermining the transient creep of concrete cyliners uring first heating at a constant rate R in the range 20 C-750 C or above, uner a constant uniaxial compressive external applie loa prior to heating [1]. In special cases, higher temperatures may be use. For the case when simulating service conitions at constant temperatures, this ocument also presents test parameters an test proceures for etermining the total strain uring the transitional thermal perio when the rate of heating of the specimen reuces from the constant rate R to zero at a constant temperature level T max, at which point steay-state temperature tests will commence [2], Part 8. Note: Transient creep occurs uring first heating uner loa, but not uring subsequent cooling or immeiate re-heating to the same maximum temperature. 2. SERVICE AND ACCIDENT CONDITIONS 2.1 Service conitions Service conitions normally involve long-term exposure to temperatures in the range 20 C-200 C an moisture states between the two bounary conitions: Bounary Conition : Drying (unseale) concrete Bounary Conition n : Moisture saturate (seale) concrete In general, bounary conition applies to rying structures in air with a maximum thickness < 400 mm, or structures with no point which is farther than 200 mm away from a surface expose to air. Bounary conition n is efine for the following wet structures: Seale structures inepenent of their imensions. Zones of structures with a istance > 200 mm from the surface expose to air. Structures uner water. 2.2 Accient conitions Accient conitions normally involve short-term exposure to temperatures in the range 20 C-750 C or above an transient moisture states, i.e. the concrete is allowe to ry uring heating. In this case, the moisture bounary conition is the same as the conition mentione above. 3. DEFINITION 3.1 General Transient creep of concrete is efine as the creep that occurs uring first time heating at a constant rate uner loa. The specific efinitions for non-rying an rying concrete are given in Sections 3.3 an 3.4, respectively. 3.2 List of Symbols an notations ε = strain = ((L - L i )/L i ) σ = stress level (constant) D = thermal iffusivity L = measure length (variable) L i = initial reference length at ambient temperature (constant) r = raius of specimen R = constant heating rate (T s /t) RH = relative humiity t = time (variable) t i = time at initiation of test t Tmax = time when T reaches T max /98 RILEM 290

2 TC 129-MHT T = reference temperature (variable) T ca = temperature at central axis of rotation of specimen (variable) T max = maximum reference test temperature (constant) T s = temperature at the surface of specimen (variable) T s = surface temperature at which T s /t reuces from R TTP = transitional thermal perio T = temperature ifference T s - T ca 0 = superscript inex for zero stress (σ = 0) ca = subscript inex for location at central axis of rotation of specimen co = subscript inex for constant temperature regime cr = subscript inex for creep = superscript inex for rying (unseale concrete) el = subscript inex for elastic an plastic see note below i = subscript inex for initial max = subscript inex for maximum n = superscript inex for non-rying (seale concrete) s = subscript inex for location at surface of specimen sh = subscript inex for shrinkage th = subscript inex for thermal tot = subscript inex for total tr = subscript inex for transient temperature regime 3.3 Non-Drying Concrete Transient creep n ( εtr, cr ) of non-rying concrete uring heating at a constant rate R (Table 2) uner stress level σ is etermine inirectly from three strain components by subtracting the thermal strain 3.4 Drying concrete n ( ε tr, th ) an the n elastic strain εco, el from the total strain uner loa n ( εtr, tot ) i.e.: n n n n εtr, cr = εtr, tot εtr, th εco, el (1) For non-rying concrete heate without loa (i.e. σ = 0) the thermal strain is equal to the total strain, i.e.: n n εtr, th = εtr, tot (2) Transient Creep of non-rying concrete, therefore, becomes: n n n n εtr, cr = εtr, tot εtr, tot εco, el (3) Note: Transient creep of non-rying concrete is sometimes calle transitional thermal creep. Note: The instantaneous strain comprises elastic an plastic strain compenents. For practical purposes, the instantaneous strain is referre to in this ocument simply as the elastic strain. Transient creep ( εtr, cr ) of rying concrete uring heating at a constant rate R (Table 2) uner stress level σ is etermine inirectly from four strain components by subtracting the thermal strain tr, cr, rying shrinkage strain ε tr, sh an the elastic strain εco, el from the total strain uner loa ( εtr, tot ), i.e.: εtr, cr = εtr, tot εtr, th εtr, sh εco, el (4) For rying concrete heate without loa (i.e. σ = 0), the thermal strain plus the rying shrinkage strain equal the total strain: εtr, th + εtr, sh = εtr, tot (5) Transient Creep of rying concrete, therefore, becomes: ε = ε ε ε (6) T, 0 Note: Thermal strain ε tr, th for rying an non-rying concrete is etermine in accorance with the recommenations given in [2], Part 6. T, σ Note: Elastic strain ( ε ) for rying an non-rying concrete is etermine in accorance with the recommenations given co, el in [2], Part 5. Note: The shrinkage strain is influence by temperature in so far as temperature influences moisture loss. Strictly, shrinkage strains are relate to moisture loss. 3.5 Transitional thermal perio When simulating service conitions at a constant temperature T max, the transitional thermal perio is efine as the time between the en of the constant rate of heating R perio an the beginning of the constant temperature T max perio. 4. MATERIAL tr, tot 4.1 Material type This recommenation applies to all types of concrete use in construction incluing high performance concrete. 4.2 Mix proportions tr, tot ( ε tr, th ) co, el Mix proportions shall be etermine accoring to the concrete esign in practice with the following provisos: The maximum aggregate size shoul not be less than 8 mm. Note: The transient creep strain of concrete originates in the cement paste an is sensitive to the aggregate content which normally comprises 60-80% by volume. Varying the aggregate content may result in significant variations in ε T, σ. tr, cr 291

3 Materials an Structures/Matériaux et Constructions, Vol. 31, June SPECIMEN 5.1 Introuction The specimens referre to in this recommenation may be laboratory cast, fiel cast or taken as cores an shoul conform to the recommenations given below. 5.2 Specimen Shape an Size The concrete specimens (Fig. 1) shall be cylinrical with length/iameter ratio between 3 an 5 (slenerness). The specimen s minimum iameter shall be four times the maximum aggregate size for core samples an five times for cast specimens. The recommene iameters of the test specimen are 150 mm, 100 mm, 80 mm, an 60 mm to be taken as stanar. Other iameters, when use, shoul be escribe as non stanar. 5.3 Mouls, casting, an curing Mouls shall be cylinrical an shoul meet the general recommenations of RILEM. The same applies to casting an curing of the specimens. The mouls shoul preferably be constructe from sufficiently stiff, cylinrical or semi-cylinrical shells mae of steel or polymer. The assemble mouls shoul be watertight so as to prevent leakage of the cement paste or water uring casting. If polymer mouls are use, the polymer shoul not be water absorbent. The compaction of the concrete in the moul shoul be one using a vibrating table. Casting shoul be performe in two or three stages. All specimens shall be store uring the first seven ays after casting at a temperature of 20 ± 2 C as follows: in their mouls - uring the first 24 ± 4 hours after casting; uner conitions without moisture exchange - uring the next 6 ays. This can be achieve by several means. The recommene metho is to keep the specimens in their mouls aing a tight cap on top. Other possibilities inclue storage: in a room with a vapour saturate environment (relative humiity > 98%); in a plastic bag containing sufficient water to maintain 100% RH; after wrapping in metal foil to prevent moisture loss; uner water (preferably water saturate with Ca(OH) 2 ). Further storage conitions up to the beginning of testing shall be chosen to simulate the moisture conitions of the concrete in practice. The following storage conitions are propose: Moisture conition (rying concrete) storage in air at 20 ± 2 C an RH of 50 ± 5% Moisture conition n (non-rying concrete) storage within seale bags or mouls or wrappe in water iffusion tight an non-corrosive foil at 20 ± 2 C. In each case, the moisture loss of specimens over the storage perio shoul be etermine by weighing. The weight loss shoul not excee 0.2% of the concrete weight for the case of seale specimens. 5.4 Specimen preparation The length, iameter an weight of the specimen shall be measure before testing. The concrete specimen shall be prepare so that each en is flat an orthogonal to its central axis. This shall be one at an age of at least 28 ays an not later than 2 months before testing. Non-rying concrete specimens shall be seale by polymer resin, metal or polymer foils, or impermeable encasements epening upon the maximum test temperature. The encasement shall not influence the eformation of the specimen or the contact between the specimen an the strain measuring evice. The time for the preparation of seale specimens uner laboratory conitions shoul not excee 4 hours. 5.5 Age at testing The specimen shoul be at least 90 ays ol before testing. 5.6 Stanar an reference strength The stanar cube or cyliner compressive strength at ambient temperatures shall be etermine at 28 ays, an at the time of testing, accoring to national requirements. In aition, the compressive strength of the test specimens shoul be etermine at 28 ays an at the time of testing, using samples of the same type an batch. The latter shall be use as the reference strength of the specimens [2], Part TEST METHOD AND PARAMETERS 6.1 Introuction The following test parameters are recommene as stanar to allow a consistent generation an comparison of test results. However, other test parameters may be substitute when information is require for specific applications. The non-stanar test conitions shoul be carefully etaile in the test report. 6.2 Measurements Length measurement The measure length is etermine in the irection of the central axis of the cylinrical specimen by measuring 292

4 TC 129-MHT the mean istance between two cross-sections at the surface of the specimen with at least two measuring points per cross-section. The cross-sections shall be perpenicular to the central axis an at least one iameter away from each flat en of the specimen (Fig. 1). At the beginning of the test, the length between the two cross-section is efine as the initial reference length, L i. It shall be at least one iameter in length. The initial reference length L i shall be measure at 20 ± 2 C with a precision of at least 0.5%. During the test usually changes in the length are measure. From these measurements strains are erive. For strains up to microstrain, the uncertainty shoul be less than 10 microstrain. For strains exceeing microstrain, the uncertainty shoul be less than 20 microstrain Temperature measurement Thermocouple, e.g. Pt 100 or other types of thermocouples, may be use. In special cases it may be necessary to protect the surface thermocouples against raiation. Surface temperature measurements shall be mae at three points on the surface of the specimen, at the centre an at the level of the two cross-sections (Section 6.2.1), by a temperature measuring system. Temperature measurements at the central axis of rotation shall be mae at two points locate at one thir points between the measuring length cross-sections (Fig. 1). The precision of the temperature measurements shoul be at least 0.5 C or 1% of the measure values whichever is the greater Loa measurement The loa applie shoul be constant with a precision of ± 1%. 6.3 Test proceure The initial moisture content just before testing shall be etermine using control specimens (seale or unseale) from the same batch cure uner the same conitions as 2 r the test specimens. The evaporable moisture content is etermine by rying at 105 C until constant weight is achieve (when moisture loss oes not excee 0.1% of the specimen s weight over a perio of 24 ± 2 hours), an by measuring the maximum weight loss. The specimen shall not be remove from the curing environment more than two hours for unseale specimens an four hours for seale specimens before the commencement of heating. If necessary, a small pre-loa compressive stress not exceeing 0.01 MPa can be applie in the irection of the specimen s central axis. During installation, the specimen shall be subjecte to two loa cycles between the pre-loa level an 20% of the reference strength. The rate of loaing an unloaing shoul be the same as for the moulus of elasticity tests [2], Part 5. During this process, the ifference in length change is recore at the ifferent locations. It shall not excee 20%. If this ifference excees 20%, then the following shoul be checke: strain measuring evice; centering of the specimen; flatness an orthogonality of the flat ens of the specimen. Appropriate ajustments shoul be mae an the test repeate until this criterion is met. If this is not possible within one hour, the specimen shoul be exclue from the test. A uniaxial compressive loa is then applie continuously in the irection of the central axis of the specimen at a rate of the orer of 1.0 MPa/s to the require constant loa level (Section 6.4.3) at 20 C immeiately prior to heating. The loa level must be kept constant accoring to Section The specimen shall then be subjecte to heating at the appropriate constant rate (Section 6.4.1), commencing not later than 1 minute after the application of loa. When simulating accient conitions up to a maximum reference test temperature T max, the test shoul terminate when the temperature at the central axis T ca reaches T max thus ensuring that the reference temperature of the specimen T (Section 8.1) has also reache T max. When simulating service conitions at a constant temperature T max, the transitional thermal perio starts at T s = T s when T s /t reuces from R (Fig. 2). The temperature ifference T max - T s shoul be less than 1 C at 20 C, 2 r crosssection measure length L 2 r 3-5 x 2 r thermocouple 2 r Fig. 1 Cylinrical specimen showing location of five temperature measuring points. Fig. 2 Definition of the Transitional Thermal Perio (TTP). 293

5 Materials an Structures/Matériaux et Constructions, Vol. 31, June 1998 less than 3 C at 100 C, an less than 20 C at 750 C. For intermeiate temperatures, the maximum permitte value for T max - T s shall be calculate by linear interpolation. Changes in length (Section 6.2) are measure in the irection of the central axis of the specimen from at least two points on the surface of the specimen per cross-section. These points shoul be at least one iameter away from each flat en of the specimen. Recoring of length change shall be taken uring initial loa cycling [2], Part 2. Recorings of length change an temperature shall be taken uring heating an uring the transitional thermal perio at the intervals given in Table 1. Table 1 Maximum recommene intervals for recoring length change an temperature for service an accient conitions Specimen iameter Service interval Accient interval (mm) (min) (min) 150, , Note: Specimen failure coul occur uring heating before the maximum test temperature is reache (e.g. for high loa levels). Appropriate safety measures shoul be taken. 6.4 Test Parameters Heating Conitions The recommene constant heating rates R for service an accient conitions are given in Table 2. Table 2 Recommene heating rates R at the surface of the specimen for service an accient conitions Specimen iameter Service Accient (mm) R ( C/min) R ( C/min) 150, , Maximum axial temperature ifferences between any two of the three surface temperature reaings (Section 6.2.2) shall not excee 1 C at 20 C, 5 C at 100 C, an 20 C at 750 C. For intermeiate values, the maximum axial temperature ifferences permitte shall be calculate by linear interpolation between the two ajacent points. Note: The maximum specimen raial temperature ifferences shoul be preferably below 20 C uring heating (Sections an 8.1). Note: Concrete can spall explosively when heate. Precautions shoul therefore be taken to avoi amage or injury Moisture conition The moisture content shall be etermine at t i accoring to Section 6.3. Unseale specimens shall be teste in a heating evice where the moisture can freely escape from the specimen an from the testing evice. Seale an autoclave specimens shall be heate an teste with a total moisture loss uring the test of less than 0.3% by weight of a control specimen rie at 105 C Loa conition The specimen shall be subjecte uring heating to a constant uniaxial compressive loa applie in the irection of the specimen s central axis. For comparison of ata between ifferent laboratories, a constant loa of 30% of the reference strength (Section 5.6) is recommene Number of tests A minimum of two replicate specimens shall be teste for any unique combination of test an material parameters. The relate thermal strain specimens [2], Part 6 an reference strength specimens [2], Part 3, shoul come from the same set of batches an shoul be teste uner the same conitions. 7. APPARATUS The test apparatus normally comprises a heating evice, a loaing evise, an instruments for measuring temperatures, loa, an lengths of the specimen. The test apparatus must be capable of fulfilling the recommenations given in Section 6 for the test parameters an the levels of precision. 8. EVALUATION AND REPORTING OF RESULTS 8.1 Evaluation of the reference temperature The reference temperature of the specimen T is calculate, uring the perio of heating at a constant rate an uring the transitional thermal perio, from the mean surface an central axis temperatures using: T = T s - 2 T / 3 (7) where T represents the temperature ifference between the surface temperature T s an the temperature at the central axis of rotation T ca, i.e.: T = T s - T ca (8) where T ca is the simple average of the two measurements taken on the central axis of rotation of the specimen at one thir points between the measuring length cross-sections (Section 6.2.2), an T s is the mean surface temperature of the specimen calculate as a weighte average of the three temperature reaings (Section 6.2.2) accoring to the formula: T s = (T 1 + 2T 2 + T 3 )/4 (9) where T 2 is the measure centre surface temperature. Note: An approximation to T ca can be mae for the 294

6 TC 129-MHT perio uring heating at a constant rate using the formula T = Rr 2 /4D, where D = Thermal iffusivity of the concrete, r = raius of the specimen, R = rate of heating. The thermal iffusivity D varies significantly with temperature an type of concrete. 8.2 Evaluation of strain results Non-rying concrete n The transient creep strain ( εtr, cr ) of a non-rying seale concrete specimen is evaluate, as the arithmetic mean of two or more of the recore strain values, in accorance with equation (3) from the total strain etermine in accorance with the proceures given in Section 6 an from the thermal strain [2], Part 6 an the elastic strain [2], Part Drying concrete The transient creep strain ( εtr, cr ) of a rying unseale concrete specimens is evaluate, as the arithmetic mean of two or more of the recore strain values, in accorance with equation (6), from the total strain etermine in accorance with the proceures given in Section 6 an from the thermal strain [2], Part 6 an the elastic strain [2], Part 2. Note: Due to the contribution of shrinkage, it is possible for ε tr, tot to have negative values when a concrete of low thermal expansion is heate Average transient creep The transient creep of the concrete is the average transient creep evaluate as the arithmetic mean of the transient creep values of the replicate specimens (Section 6.4.4). 8.3 Test report General The report shall inclue the items highlighte by unerlining below. The other items liste below shoul be reporte when available Mix proportions Cement type an source, cement replacements, aitives, cement content, water/cement ratio, maximum aggregate size, aggregate/cement ratio, aggregate graing, mineralogical type of aggregate, aggregate content by volume of concrete Fresh concrete Air content, bulk ensity, slump (or equivalent) Harene concrete an specimen etails Curing regime, age at testing, initial moisture content of reference specimen, assume thermal iffusivity D, stanar cube strength or cyliner strength, reference compressive strength, iameter an length of specimen, moisture of the teste specimen, weight before an after testing (excluing the weight of items such as thermocouples), moe of preparation of the flat surfaces of the specimen, metho of sealing (if applicable) Test apparatus The apparatus use shall be escribe unless it is in accorance with a publishe stanar, in which case the stanar shoul be reference Test parameters Time between removal of specimen from the curing environment an initiation of heating. Time between en of loaing an start of heating. Initial reference length. Level of the restraining loa (if applicable). The following shoul be reporte as functions of time uring heating: iniviual temperature measurements, mean surface temperature, mean centre temperature, reference temperature, rate of heating, axial an raial temperature ifferences, an changes in the measure length (incluing any ajustments mae for movements of any or all components of the length measuring evice). Any eviation from the recommene test parameters (e.g. heating rate, loaing rate, loa level uring heating) shall be reporte separately as non-stanar. For service conitions: Transitional thermal perio, Maximum constant test temperature T max, T max T s, Deviation from T max with time of both T s an T ca Strain uring initial loa cycling Strains uring initial two loa cycles measure for each location at ambient temperature (Section 6.3) Strain results n The total strain εtr,tot, εtr,tot an transient creep n (Section 8.2) results εtr,cr, εtr,cr of every specimen shall be reporte in tabular an/or graphical form as function of the reference temperature. The elastic strain n εco,el, εco,el use in calculating the transient creep shall also be reporte (Sections 3.2 an 3.3). The average transient creep (Section 8.2.3) shall also be reporte Place, ate, operator Country, city an institution where the experiment was carrie out. The ates of the experiment an report. Name of the operator. 9. REFERENCES [1] Schneier, U., an Schwesinger, P. (E.) Mechanical testing of concrete at high temperatures, RILEM Transaction 1, February 1990, ISBN: X, pp.72. [2] RILEM TC 129-MHT Test methos for mechanical properties of concrete at high temperatures. The Committee is in the process of preparing the following ocuments: Part 1 Introuction; Part 2 Stress-strain relation; Part 3 Compressive strength; Part 4 Tensile strength; Part 5 Moulus of elasticity; Part 6 Thermal strain; Part 7 Transient creep; Part 8 Steay-state creep an creep recovery; Part 9 Shrinkage; Part 10 Restraint; Part 11 Relaxation. 295