Issue 78

M. S ł owik et alii, Fracture and Structural Integrity, 78 (2026) 393-408; DOI: 10.3221/IGF-ESIS.78.23

and the specimens are described in [28]. Both cylindrical and cubic specimens can be used in the splitting test. The load is applied centrally with the use of plywood strips, as shown in Fig. 3a. The concrete’s splitting tensile strength can be determined from Eqn. (1):

2 L d     F



f

(1)

, ct sp

where: F – the maximum force, L – the length of the specimen, d – the nominal dimension of a cross section. However, the above formula is correct when the force is applied linearly along a very narrow band and then the tensile stress is theoretically constant in the failure plane (Fig. 3b). When the applied force is transversely distributed by using ply wood strips, an almost constant tensile stress over the most of the failure plane re-mains whereas compressive` stress appears in the upper and lower part of the critical cross section (Fig. 3c). The horizontal stress along the failure plane can be calculated from Eqn. (2) and Eqn. (3):

d

         1 2 F sin   a 

2





(2)

 



L d

2     a   r

2  



arctg

(3)

where: a – the width of the strip, r – the distance of the considered point from the closer strip. The splitting tensile strength depends on the strips’ width. To minimise the influence of the load application type on the tensile splitting strength, the type and dimensions of the bearing strips are codified [2]. They should be made of plywood and of the cross-section 15 mm × 4 mm. The above formulas are used when a standard type of loading configuration with flat platens is applied. When the test is carried out with curved platens of radius of curvature Rp = 1.5R (R=d/2), the different formula (Eqn. (4)) is recommended in the latest version of the ASTM standard [29]:

2 L d     F

, 0.636 ct sp  

f

(4)

The effect of the radius of curvature of the platens has been recently analysed in [30] and it was found that “…The use of flat (Rp = + ∞ ) platens favors the nucleation of fracture near the region of contact with the platens, while the use of curved platens with a radius of curvature Rp that is comparable to that of the specimen, e.g., Rp = 1.1R, favors the nucleation around the center of the specimen…”.

Figure 3: The Brazilian splitting test: loading arrangement of a cylinder (a); stress distribution for linear force application (b) and for force application over the width a (c); σ ct - tensile stress, σ c - compressive stress, P – a chosen point for calculation of stress, r - the distance of the considered point from the closer strip. Two basic specimen geometries are recommended in the European code [28] for determining the tensile strength of concrete: cubes and cylinders. The standard size for cube specimens is 150 mm, whereas cylindrical specimens typically have a diameter equalling 150 mm and a height of 300 mm. Although specimens of other dimensions may be used, appropriate conversion factors should be applied to ensure compatibility of results. Furthermore, the dimensions of the specimen should fulfil the condition of not being less than 3.5 times the maximum aggregate size. This condition applies to both standard

396