BEARING CAPACITY OF STRING FOOTING ON JOINTED ROCKS

Abstract

Bearing capacity ofjointed rocks is greatly influenced by the presence ofjoints. Orientation of joints, their roughness characteristics and joint spacing are the major controlling factors in the assessment of bearing capacity of jointed rocks. Most of the research carried out over the past four decades has been devoted to isotropic rock condition i.e. for either intact rocks or highly fractured jointed rock mass (Hoek and Brown medium). Very few studies are reported in literature on the bearing capacity of jointed rock mass, where, the mass is anisotropic in nature. Since most of the methods available for finding the ultimate bearing capacity of jointed rocks consider the mass an isotropic medium, their applicability to anisotropic blocky rock mass remains doubtful. Recently two methodologies have been suggested for anisotropic blocky rock mass (Prakoso and Kulhaway, 2004; Singh and Rao, 2005). Singh and Rao (2005) approach is based on the Joint Factor concept. The concept was initially developed from experimental studies on small cylindrical specimens of jointed rocks of artificial and natural origin. The Joint Factor, developed by Ramamurthy and co-workers (Arora, 1987; Ramamurthy, 1993; Roy, 1993; Singh et al., 2002; Ramamurthy, 2004), is a weakness coefficient and depends on frequency, shear strength and orientation of joints. The laboratory tests were performed on small specimens of natural and artificial rocks (Yaji, 1984; Arora 1987). Later, the applicability of the concept was extended to blocky rock mass under uniaxial loading condition by including the effect of failure mode (Singh, 1997; Singh et al., 2002).The concept has been extended to develop a procedure to assess the bearing capacity of shallow foundation in jointed rocks (Singh and Rao, 2005). In its essence, the procedure assesses the anisotropic strength of rock masses to compute the bearing capacity. Prakoso and Kulhawy (2004) presented bearing capacity solution for strip footings on jointed rocks with one and two sets of discontinuities. The solution employs a lower bound bearing capacity model coupled with a simple discontinuity strength model. The model primarily uses the properties of joint and rock material. The bearing capacity computed therefore depends either on the joint property or the intact rock property and not the jointed mass property. As a result, the ultimate bearing capacity predicted by these models are quite differing. In addition, no validation has been reported either in laboratory, or in field, to assess the applicability of available models. It was therefore felt essential to experimentally investigate the behaviour of foundation onjointed rocks containing fewjoint sets which render anisotropy to its engineering response. In the present study, an attempt has been made to investigate the ultimate bearing capacity of jointed blocky rock masses through large size bearing capacity tests under plane strain conditions. Plaster of Paris as a model material has been used to simulate the intact rock material. Specimens of rock mass, having various joint orientations with and without stepping, were assembled using plaster of Paris blocks. These specimens consisted of two joint sets. Joint set-1 was kept continuous and was inclined at various inclinations, 0, with the horizontal direction. The tests were performed for joint inclination, 0, varying from 0° to 90° at an interval of 15°. Joint set- II was orthogonal to set- I and was stepped. Two sets of stepping viz; 0 and 0.5 b, where b is the width of the elemental block, were used. Some additional tests were performed on the specimens with one joint set to verify the applicability of the available bearing capacity theories to the jointed rocks with one joint set. These tests were also performed for joint inclination 0, varying from 0° to 90° at an interval of 15°. All the specimens were tested in plane strain condition in the loading frame designed and fabricated for this purpose. Few tests were also conducted on the intact rock specimens to find out the bearing capacity of intact specimens. During the laboratory experiments, shearing and splitting modes of failure in the rock mass below the footing were observed for all the jointed blocky mass specimens. The joint orientation, 0, was found to be one of the parameters affecting the bearing capacity of blockymass specimens. The ratio of maximum to minimumbearing capacity due to orientation, 0, was found to be 1.61-2.0 for different types of specimens. The bearing capacity of the jointed blocky mass specimens was therefore observed to be low to moderately anisotropic in nature. The bearing capacity of jointed rock mass in present study was found to vary from 12% to 25% of that of intact rocks which indicates that there is pronounced effect of jointing on bearing capacity of rocks. The stepping in jointed blocky mass specimens, which is considered an indirect measure of interlocking conditions, is found to have substantial effect on ultimate bearing capacity for joint u orientations 0= 0° to 30°. However, it is found to have negligible effect on the ultimate bearing capacity in the range of 45°< 0 < 90°. Many empirical strength criteria have been proposed in the past to predict the strength ofjointed rock masses. Some of these include Hoek and Brown (1980, 1988, 1992, 1995, 2002, 2005), Ramamurthy (1993), and parabolic criterion (Singh and Rao, 2005a). The Hoek and brown criterion is applicable for isotropic rock masses only. The parabolic strength criterion is applicable for isotropic as well as anisotropic rock mass and needs least input data for assessing the criterion parameters. The parabolic criterion has been found to predict the ultimate bearing capacity of blocky rock mass with reasonable accuracy. Toassess the applicability of bearing capacity theories available in the literature, the bearing capacity of intact rock specimens tested in laboratory was predicted through these theories. It was observed that most of the theories available in literature could be applied satisfactorily to predict the bearing capacity of intact rocks. For jointed rock masses, however, the applicability of these theories remains doubtful as most of these are applicable only for isotropic rock masses. Probably, the recent approaches those are applicable to anisotropic rock masses are those proposed by Prakoso and Kulhawy (2004) and, Singh and Rao (2005). For these approaches also, the respective authors have only suggested the methodology, however, no validation either in the laboratory or in the field has been reported. When the approaches are applied to predict the bearing capacity ofjointed rock masses tested in the present study, Prakoso and Kulhawy (2004) approach gives zero bearing capacity if the cohesion of thejoint,c,, is considered zero in case of all the specimens having two joint sets and specimens having one joint set with joint orientation, 0 < fy. For other orientations, the bearing capacity predicted is nearly equal to that of intact rock. It should be noted that joint friction and cohesion are not constant values and changes with the level of normal stress. To get reasonable value of joint cohesion cj5 the joints are required to be tested up to fairly high normal stress levels. Using this value of joint cohesion and corresponding joint friction angle, the bearing capacity of the rock mass specimens were predicted. The approach is able to reasonably predict the ultimate bearing capacity for the joint orientation 0 = 15° to 75° but extremely over-predict for the joint orientation, 0= 0°and 90°, for specimens having in two continuous joint sets. Similarly, the approach highly over-predicts the ultimate bearing capacity in case of specimens having one joint set for the joint orientation of 0°, 60°, 75° and 90°. The approach uses shear strength parameters of either rock material or rock joints to assess bearing capacity. As a result, the bearing capacity obtained is either equal to that of intact rock or is completely governed by the shear strength parameters of the joints , and not by the characteristics of the rock mass as a whole. Moreover, the spacing of joints is not accounted for in the approach. It means whether the RQDequals to 0% or 100%, the predicted ultimate bearing capacity will remain same. Singh and Rao (2005) approach, on the other hand, is based on parameters for rock mass as a whole and not on superimposition of intact rock and joint characteristics. The Joint Factor concept which defines the relationship between uniaxial rock mass strength to intact rock strength, takes into account the effect of frequency, orientation and shear strength characteristics of the rock joints. The shear strength parameters of the rockjoints are required at low normal stress level which is easy to assess in laboratoryor field. The approach, in its present form, when employed to estimate bearing capacity, has been found to either over-predict the results for the joint orientation 0 = 0°, 15°, 75°, and 90° or under -predict for the joint orientation 0 = 30°, 45° and 60° respectively. It may be due to the fact that the experimental study conducted by Singh (1997) on blocky mass specimens in uniaxial condition represented a case of unconstrained dilatancy. However, the outcome of the present study indicates that the mass under foundation is constrained to some extent and the blocks are not free to slide or rotate and dilation of the mass is not as prominent as it was in the case of specimen tested by Singh (1997) under uniaxial compression. Based on this outcome, the approach has been modified. The modified approach includes the estimation of the uniaxial compressive strength of rock mass in the desired direction and then determination of the strength enhancement due to confinement. Expression correlating rock mass strength in a given direction with the intact rock strength through Joint Factor has been modified. A non-linear strength criterion has been used to predict the strength of rock mass under confinement. The criterion is based on the critical state concept in rocks. The input parameters of the criterion can be easily obtained in the laboratory or field. The suggested model is able to predict the bearing capacity of rock mass having continuous as well as stepped joints with reasonable accuracy. IV Though the main objective of the study has been to suggest an approach for assessing bearing capacity, an attempt has been made to numerically simulate the experimental findings from the afore-mentioned experimental programme. A simple UDEC model has been developed to assess its applicability to simulate the bearing capacity ofjointed blocky mass specimens tested in the laboratory under plane strain condition. The model uses simple joint and block models. The bearing capacity predicted by the model is found to be comparable with the experimental results for the two continuous joint sets; however the bearing capacity predicted by the model is higher ascompared to laboratory test results in case of stepped joint set and single joint set. The model is also able to capture the failure mode as observed in the laboratory tests