[en] The knowledge of the creep behaviour of concrete under multiaxial stress at raised temperatures is necessary for the design and calculation of prestressed-concrete reactor pressure vessels due to the loading conditions prestressing and internal temperature. Therefore, in addition to the research programmes on the behaviour of concrete under multiaxial short-term stress at normal temperature done at the 'Institut fuer Massivbau' in the framework of the prestressed-concrete pressure vessel programme, the influences of raised temperature is experimentally tested. (orig./LH)

[en] Creep and creep recovery tests at stress levels between service stress and long-time strength are described. Two types of cement, viz. - ordinary Portland cement PZ 45F, and - Portland limestone cement PKZ 35F. And two different storage conditions up to the beginning of the creep tests were examined, viz. -at 20 C with full immersion in water, and - at 20 C at 35% relative humidity. During the creep and creep recovery tests the samples were sealed. The concrete mix was identical in all cases (w/c ratio = 0.6, 300 kg cement/m³ concrete). In this study sealing of the pre-dried or pre-moistened samples respectively during the creep and creep recovery tests served to prevent further moisture exchange with the environment. Through sealing, creep and creep recovery behaviour of samples of varying degrees of moisture content at a greatly reduced level of drying shrinkage and drying creep could be examined. (orig.)

[en] The experimental investigations at the 'Institut fuer Massivbau' at the Technical University, Munich, in the field of reactor construction can be essentially grouped into two sections, namely that of the multiaxial concrete stress and that of concrete creep under raised temperatures. The study of these questions was more strongly coupled with the development of suitable test methods and testing equipment than in other fields of reinforced-concrete research. In the present report, a brief survey is given on the machine equipment developed in Munich. (orig./LH)

[en] Long-time tests of three-axially loaded, sealed cylindrical specimens d = 15 cm, h = 40 cm, were carried out. The 20-cm-cube strength of the concrete was app. 45 N/mm². The creep stresses were chosen in the following ranges: 0,3 ^≤ σ_c/β_c^≤ 2,1; 0 ^≤ σ_r/σ_l^≤ 1,0. The creep coefficients obtained were clearly depending on the multi-axial stress conditions. The creep coefficients for a t = 2 years loading were reaching app. 1 for σ_l/β_c = 0,3 and app. 3 for σ_l/β _c = 2,1, when the test evaluation was based on the initial deformations meausred after 1 minute. For σ_l/β_c = 2,1 the creep coefficients obtained were about 4 times as large, proceeding form calculated elastic deformations. Further evaluations concerned the Young's modulus E, Poisson's ratio μ, the bulk modulus K and the shear modulus G. The preceding permanent load leads to an increase in the Young's modulus of the concrete in longitudinal direction of the specimen up to about 4 times the value of not preloaded comparative specimens. (orig.)

[en] To provide detailed information for the calculation of prestressed concrete reactor vessels, investigations of the behaviour of concrete under multiaxial loading and on creep at elevated temperatures were made at the Institut fuer Massivbau of the Technical University of Munich. The strength of concrete under triaxial compression is dependent on the stress ratio. The less the stresses differ from hydrostatic compression the more strength increases. Triaxial compression increases very much the deformability of concrete. Plastic deformations of +-10% and more (all stresses compression, but not equal, strains compression or tension) are possible without large cracks. The creep deformations are considerably dependent on the temperature. Creep at 80⁰C is about three to four times higher than at 20⁰C. The Poisson's ratio of creep at elevated temperature seems to be bigger than at normal temperatures at a rate of loading of 35% and 50% of the ultimate strength. (Auth.)

[en] Within a German PRCV-program at the Institut fuer Massivbau of the Technical University of Munich tests are made in order to provide the information necessary for an analytical modelling of thick concrete cylinders. In 1976 the series was started with 2 tests on large size unreinforced concrete rings (inner diameter 10.9 cm, outer diameter 36.5 cm resp. 62.5 cm, thickness 10 cm) under inside pressure. The loading was electronically controlled under the condition of a linear increase of the mean strain rate (0.0002/hour). This method provides detailed information on the deformation behavior during rupture of the specimen. The main strain was measured at the inner side of the ring by 8 inductive gages. To obtain the information necessary for a finite element analysis the specimen was subdivided by a triangular net for which all tangential strains have been measured by inductive gages, whereas all radial and diagonal strains have been measured using strain gages. Additionally the increase of the ring diameter has been recorded. (Auth.)

[en] Within a German PRCV-program at the Institut fuer Massivbau of the Technical University of Munich tests are made in order to provide those informations necessary for an analytical modelling of thick concrete cylinders. In 1976 the series has been started with 2 tests on large size unreinforced concrete rings (inner diameter 10,9 cm, outer diameter 36,5 cm resp. 62,5 cm, thickness 10 cm) under inside pressure. The loading was electronically controlled under the condition of a linear increase of the mean strain rate, a method well proved in our laboratory since years. This method provides detailed informations on the deformation behavior during rupture of the specimen. According to the information necessary for a finite element analysis the specimen was subdivided by a triangular net, for which all tangential strains have been measured by inductive gages whereas all radial and diagonal strains have been measured using strain gages. Additionally the increase of the ring diameter has been recorded. The strain-controlled loading allowed to observe the behavior of the specimen also beyond the maximum load level at the descending branch of the load down to 0.73 of the maximum load. The strains indicate clearly the formation of two cracks (primary crack and secondary crack at opposite point). Strain reduction at the neighbouring region of the cracks was recorded. At maximum load the tension strain at the primary crack was measured as 0.0002 which corresponds to the ductility of the concrete. It should be noted that due to the small strains at the outer face of the ring the strains at the inner side could be enlarged beyond the maximum load to 0.00054 without

[en] Tests on the triaxial stress-strain-relation of concrete in compression and tension were done utilizing the Munich testing equipment (a rigid prestressed concrete frame with brush bearing platens). The specimens are cubes of 10 cm square. All stress states can be managed. During the test all deformations were measured. The concrete being tested has a similar mixture as concrete used in PCRVs. The strength of concrete under triaxial compression is dependent on the stress ratio. The less the stresses differ from hydrostatic compression the more the strength increases. Triaxial compression increases very much the deformability of concrete. Plastic deformations of +-10 per mille and more (all stresses compression, but not equal, strains compression or tension) are possible without large cracks. Tension in at least one direction which is connected with compression in the other direction(s) causes a sudden failure and a reduction of strength. The deformations measured show a nearly constant modulus of elasticity. The second test series examines creep of concrete under elevated temperatures with multiaxial loading. The specimens (discs of 20/20/5 cm) were loaded in two directions by brush bearing platens under temperatures of 20 deg, 50 deg and 80 deg Celsius. The creep deformations are considerably dependent on the temperature. Creep at 80 deg C is about three to four times higher than at 20 deg C. The Poisson's ratio of creep at elevated temperature seems to be bigger than at normal temperatures at a rate of loading of 35% and 50% of the ultimate strength

[en] In the 10th SMIRT conference, a report was presented on the Munich tests on the creep behavior of sealed concrete cylinders of 15 cm diameter and 30 cm height under triaxial compressive stresses. The method of comprehensive representation of all results for one observed level of the related longitudinal stress is shown. Some further evaluation was done, and the results are reported. In order to get deformation values with very small creep influence, load was applied very quickly. Nevertheless, the measured deformation values at the beginning of creep curves included creep components, and the improvement is necessary. The basic elasticity values K and G used for the multiaxial studies of isotropic materials were derived. The result of the evaluation in accordance with these formulas is shown. With the basic elasticity values K_i and G_i, the initial strains are obtained. An example showed that creep deformation already occurred in a short time despite very fast stress application. The result of the improvement is shown. Creep process is frequently represented by the related values instead of deformation. A creep coefficient is generally used for the creep of concrete. (K.I.)

[en] In this paper a small test program (16 specimens) about creep of concrete is described. The dependance of the final creep on the following parameters was investigated: - level of loading (about 0,3 to 0,8 βsub(c)) - quantity of cement (250 and 360 kg/m³ concrete respectively) - water/cement ratio (0,45 - 0,60 - 0,80) - consistence of the concrete mix (K2 and softer than K3 in the definition of the German code). The influence could be analysed clearly. The final creep increased with - an increased level of loading - the quantity of cement in the conrete - the water/cement ratio - softer consistence of the conrete mix. A comparison with the German code values shows a good conformity with the test results at lower level of loading. (orig.)