PSI - Issue 42

Monika Středulová et al. / Procedia Structural Integrity 42 (2022) 1537– 1544 / Structural Integrity Procedia 00 (2019) 000–000 M. Strˇedulova´ et al.

1538

2

Fig. 1. From the left: Relative strengths of cylindrical samples with respect to the slenderness ratio (Gonnerman, 1925); specimen divided into zones based on acting stress as described by Kotsovos (1983); failure mechanism Van Mier (1984); confined compression zones in specimens of di ff erent slenderness ratio as described by Van Vliet and Van Mier (1996).

permissible dimension (both diameter and height) is 50 mm. An important parameter of the cylindrical specimen shape is its height to diameter ratio (also called slenderness ratio ), which equals to two in the case of standardized dimensions, while the smallest ratio allowed is one. If specimens of di ff erent dimensions are used, the code suggests the adoption of correction coe ffi cients when evaluating compressive strength, to relate the measured value to the comparable values of standard-sized specimens. Unless experimental results are available, the code provides values of the correction coe ffi cient for certain dimensions. For cylindrical specimens, which are also central to the present numerical analysis, the proposed coe ffi cient is dependent on the slenderness ratio and is to be used only for slenderness between one and two. The formulation suggests that higher slenderness ratio cylinders do not di ff er from the standardized ratio of two or di ff er only to an extent that is admissible to use without correction. Indeed, the trend has been studied during multiple series of experiments which have been done throughout the twen tieth century in an attempt to describe stress state of a concrete specimen in the uniaxial compression. Gonnerman (1925) described the general trend of the slenderness ratio influence (see Fig. 1, left-hand side) without considering other specific conditions during testing. He shows results for ratios from 0.5 up to 4, relatively to the ratio of two. Strengths obtained for higher slenderness ratios (from 2 to 4) slightly decrease. On the other hand, obtained strength soars dramatically for ratios smaller than two. The samples of the smallest ratio of 0.5 show the apparent compressive strength higher by 80 % with respect to strength measured with the ratio of two. The e ff ect of testing techniques has been investigated only later on by Kotsovos (1983) using experiments per formed on cubes and cylinders of constant dimensions. He postulated that the post-peak behavior of a specimen is a reflection of a restraint applied by loading platens at the top and at the bottom of the specimen. According to his observations, the sample may be divided into two types of zones based on developed stress state during testing: (i) an end zone in the vicinity of the loading platen where lateral restraint of the platen creates a zone of triaxial com pression and (ii) a central zone where only uniaxial compressive loading is present. A schematic representation is given in Fig. 1 on center-left. Kotsovos emphasized the role of friction on the interface. The explanation was sup ported by experimental data obtained for high friction (steel platens) and low friction (teflon treated platens) interface, where post-peak behavior represented by the slope of the curve of a stress–strain diagram indeed di ff ered for the two scenarios. The findings were later supported by other experiments performed by Vonk et al. (1989). Van Mier (1984) performed a series of experiments on specimens of varying sizes. He treated the testing in uniaxial compression as essentialy triaxial by measuring both axial and lateral strains. He observed that when specimens of varying height are tested (that is of di ff erent slenderness ratios), there is no di ff erence in terms of the specimen’s strength with a low friction on the interface with the loading platens. On the contrary, considerable increase of up to 90 % was observed for a high friction interface, showing the importance of a specimens slenderness in this setting. Considering the observed influence of slenderness ratio in a high friction conditions, the International Union of Laboratories and Experts in Construction Materials, Systems and Structures (RILEM) established in 1992 a com mittee designated to study strain softening in uniaxial concrete compression. The committee comprised of 11 universi ties where experimental testing was performed. The main aim of the research project was to create a testing procedure recommendation, which would consider the influence of friction for both pre-peak and post-peak regime, when soft-