It is common engineering practice to estimate the Unconfined Compressive Strength (UCS) of a rock, by multiplying Point Load Index test (IS50) results by a cor-relation factor. The IS50 to UCS correlation is reasonably accurate for crystalline or isotropic rocks but with weak sedimentary rocks, there is the potential for failure to occur along natural planes of weakness. This creates significant scatter in data and questionable accuracy. The strength of a sedimentary rock is a function of inter-particle friction and grain interlocking. Inter-particle friction is affected by degree of consolidation, and inter-locking is affected by grainsize. This paper provides evidence that a system which indirectly measures consolidation (using dry density testing) and incorporates grain-size produces more accurate UCS estimates for sedimentary rocks. This paper uses data from different geological ages and locations to provide corre-lations for typical sedimentary rock types that transcend geological boundaries and locality. Although this paper uses a relatively large data body, the technique remains appli-cable to smaller projects. Small projects have been simulated by drawing correlations from 10, 20 and 40 randomly selected data points to provide an indication of the error envelope and level of confidence when smaller quantities of test results are available.
Rock strength is probably the most important characteristic for engineering design of both structures interacting with rock or earthworks involving excavation through rock. The most common test used to measure the strength of a rock is the unconfined com-pressive strength test (UCS). Once a representative number of UCS tests have been carried out, the confined strength of the rock can be calculated using standard engi-neering calculations. Due to the cost of carrying out numerous UCS tests, it is advantageous from a pro-ject budget perspective to develop a correlation between UCS and values obtained from cheaper laboratory tests.
