Measurement Uncertainty in Testing of Uniaxial Compressive Strength and Deformability of Rock Samples

Abstract

NE of the basic and most used methods of testing which is performed on rock samples is determination of uniaxial compressive strength and deformability. The testing method is proposed by the International Society for Rock Mechanics [1] in the form of Suggested Method and based on the guidelines in this document, some national standards were derived. Subsequently, additional documents [2] have been published that describe requirements and difficulties that arise from implementation of testing in detail and methods to overcome these difficulties. The latest literature dealing with determination of rock sample properties [3] has replaced [4] and [5]. Reference [3] is extended standard test method which also includes confined testing in various temperatures when compared to [2]. Similar testing is performed on other types of materials like concrete [6] and metals [7] - [9] with some differences in testing procedures and minimum measurement equipment properties. Testing of concrete sample properties has the most similarities with the testing of rock samples. The metals are tested for tensile strength as opposed to rock and concrete. Homogeneity of metallic materials is better than in concrete samples. Also, homogeneity of concrete samples is usually better than in rock samples. Homogeneity is important because it influences repeatability of measurement. According to the method [2], a rock sample of 54 mm in diameter and height from 2.5 to 3 times higher than its diameter is placed in the compression test machine and compressive force is applied to the sample. The force should have constant increase over time until the sample breaks and is disintegrated while simultaneously axial and lateral deformations are measured. The result is uniaxial compressive strength, i.e., stress at which the sample is disintegrated. If axial deformations are measured, Young’s modulus of elasticity can be obtained. If lateral deformations are measured, the sample’s Poisson’s coefficient is also obtained. A large difficulty in implementing the presented measurements is sample nonhomogeneity. Therefore, repeated measurements in equal conditions using different samples will give a variation in measurement results caused by material structure, i.e., nonuniform grain size, arrangement and micro cracks. It is not unusual that this variation is in the range of more than a few percent, i.e., 20 % [10]. Therefore, at least five samples are tested and the results are averaged to represent the properties of the material from which the samples are obtained. It is suggested that test results from samples which substantially deviate from the average are not used for calculating the final result. Rock samples can have uniaxial strength in the range from 5 MPa for soft rock, up to 250 MPa for very brittle rock. Concurrently, Young’s modulus of elasticity is in the range from 5 GPa to 120 GPa. Poisson’s ratio is mostly in the range from 0.1 to 0.45.