Relating concrete cube, core and cylinder compressive strengths that are cast, cured, prepared and tested in laboratory conditions

In practice, concrete is initially tested for compressive strength by casting a cube/cylinder, which is left to cure in favourable conditions until the date of testing. The results obtained from such tests can give a consultant guidance on the material's properties such as estimated porosity, density and compressive strength. These tests are known as control and conformity tests. Supplementary tests may be needed if the control test fails or further investigation must be done to the concrete. These tests are done by drilling core specimens out of the in-situ concrete and applying the necessary tests. These results are used to verify conformity with specifications set out by the engineer. The outcome of such a test is extremely important as it is often used as the basis to decide the integrity of the structure. Although important, in-situ compressive strength remains as one of the least understood concrete properties due to the difficulty in relating and interpreting the results. Furthermore, there is no reliable universal relationship between compressive strength of cores and; moulded cubes and cylinders. A comprehensive literature and experimental study was attempted to relate standard cube and core compressive strength, as well as, cylinder and core strengths to identify the factors that may affect the analysis and interpretation of results. An experimental program was set out to relate the compressive strength of cubes, cores and cylinders, with a length/diameter ratio of 1.0. All specimens were cast, cured, prepared and tested in the University of Cape Town, New Engineering Building (NEB) laboratory according to South African National Standards. Twelve concrete mixes were designed using two concrete strengths (30 and 50 MPa), three maximum aggregate sizes (9.2, 19.2 and 26.5 mm) and two aggregate types, namely greywacke and quartzitic sandstone. An additional two mixes of high strength concrete were created (60 and 75 MPa) using 19.2mm greywacke aggregate. The compressive tests involved a 100 mm cube, three diameter cylinders (70, 100 and 150 mm) and four core diameter sizes (50, 70, 100 and 150 mm). All core specimens were drilled from beams that were cast. A total of 520 specimens were tested during this study. An analysis of variance (ANOVA) was applied to all the results to identify if the compressive strengths were statistically significantly different. The compressive strength and statistical results indicate that 100 mm cubes and 100 mm diameter cores have statistically similar compressive strengths. The diameter of the core and cylinder influenced the compressive strength. It was found, as the diameter size decreased the strength increased for core specimens and the opposite was found for the cylinder. Both findings were inconsistent with literature. However, as the core and cylinder diameters increased to a size larger than 100 mm, the compressive strengths were statistically similar. With respect to the maximum aggregate size, the strength was influenced in correspondence with the diameter size. As the core diameter decreased and the maximum aggregate size increased, the compressive strength increased. Whereas, the opposite was found with the cylinders. The strength level further determined the influence that the coarse aggregate type had on the compressive strength. At the 30 MPa strength level, the aggregate types produced statistically similar strength. At the 50 MPa strength level, the sandstone produced a statistically higher compressive strength compared to the greywacke aggregate. Finally, as the strength level increased over 50 MPa there was no significant difference between the mean compressive strength of cubes and cores. It was concluded, owing to the controlled environment that the all specimens were cast, cured, prepared and tested; as well as the similarity in the geometric size, statistically comparable compressive strengths were obtained for cubes and cores