1. INTRODUCTION
In this paper, the bearing capacity of a jointed rock foundation was investigated using the code UDEC. The models include two joint sets with variable orientations and different joint spacing. Three values for orientation and joint spacing were selected and bearing capacity analysis was performed for each case. The results obtained from the numerical analyses were compared with the existing solutions in which both the rock matrix and joint properties are considered and then, the ranges in which the UDEC simulation takes the best agreement with the existing methods are specified. In other part of this paper, the effect of joint normal and shear stiffnesses on the bearing capacity was investigated. Results of this part show that the joint normal and shear stiffnesses don't have very significant effects on the bearing capacity of rock masses. The results obtained in this research show that by increasing the number of joints beneath the foundation, the bearing capacity will be reduced.
The ultimate bearing capacity is an important design consideration for dams, bridges and other engineering structures, particularly when large rock masses are the foundation materials. Because of existence of discontinuities within the matrix material, the mechanical behaviour of a rock mass is governed not only by the intact material but also by the characteristics of the discontinuities. The validity of the subject to the condition that the spacing between two successive joints is small enough when compared with the characteristic length of the problem under consideration. Different suggestions exist for distinguishing the homogenized behaviour in the field of tunneling. Maghous et al [1] proposed that for an underground gallery with diameter D excavated in a jointed rock mass in which the rock matrix is crossed by a set of parallel joints with spacing ?, the homogenization approach can advantageously substitute for direct approaches as soon as e/D<=0.1. Stille et al [2] proposed the continuity factor, CF=tunnel diameter/block diameter, which indicates the number of blocks occurring along the roof of the tunnel. As their proposition, for CF=6-15, the discontinuous (blocky) behaviour and for CF=3-6, the continuous-discontinuous (intact-blocky) behaviour may take place. Very few propositions exist for distinguishing the homogenized behaviour in the field of foundation rock masses. One of the best is that of Serrano et al [3] in which they have restricted their methods of determining the ultimate bearing capacity according to the spacing of joints and dimension of the footing. They proposed a parameter 'the spacing ratio of a foundation (SR)' for quantification of the spacing of joints. As an initial and conservative proposal, a relatively small spacing is suggested when SR is greater than 60. As they proposed, the Hoek and brown criterion is not valid when the SR parameter is so low that the rock can be regarded as intact. It seems that for the range of 2<SR<60 in which the small spacing concept cannot be used, the application of homogenization approach, selecting a suitable domain for numerical modeling is difficult.
