Engineering properties of weathered granities for excavation purpose

Abstract

Ripping is a process of breaking harder ground by dragging tines attached to a bulldozer. Rippability of materials in moderately weathered (grade III) to completely weathered (grade V) has always been vague where the material behaviour is not fully understood due to their ‘rock-soil’ characteristics. This research is to examine factors that influence the rippability and establish a comprehensive method that effectively assessed the weathered sedimentary rock masses. The assessment of rock mass properties was mainly based on recording the presence, nature, orientation and occurrence of discontinuities. The information gathered from the monitoring was used for determining the excavatability of rocks. Monitored ripping tests were conducted at four different locations namely; Bukit Indah, Mersing, Kempas and Desa Tebrau which consisted of sandstone, shale and old alluvium. In order to rectify issues in different rock types, investigation was also conducted on granitic area at Masai and Ulu Tiram. Three main factors were identified to affect the rippability performance, these are; rock material, rock mass and the machine properties. Field measurements, in situ and laboratory test results are presented and their relation with the weathering grade was established. Some of the standard strength tests were not suitable to test very weak materials with weathering grade V (completely weathered), due to sampling difficulties. Thus, modifications to the test methods were done in order to get more accurate results. It is also found that there are significant relationships between productivity and the weathering grade. By measuring the ripping process, the relationships between the rock properties and the production were established. The data was analyzed statistically by using the Statistical Package for Social Science (SPSS) software regression analytical techniques to produce sets of equations for different weathering grades from which the machine performance can be predicted. It was revealed that a single parameter is not able to predict significantly the productivity but combinations of different parameters are able to predict satisfactorily the production rate. Different sets of factors were found to influence the productivity of the ripper machine for each weathering grade. By classifying and performing the tests required for that particular weathering grade, a significant correlation between the predicted and actual production rate was established. Identification of weathering grade, joint spacing, ripping direction and some measures of strength in addition to the machine properties are found to be the major factors in predicting the ripper performance. Methodology developed for predicting production rate for different weathering grades of these weathered rock masses is considered to be a significant advance as compared to the previously published methods.