Degree Name

Doctor of Philosophy


Department of Civil and Mining Engineering


The shear behaviour of rock joints has been studied in the past, mainly using the conventional direct shear apparatus, where the normal load is kept constant during shearing. This is called the Constant Normal Load (CNL) condition. In this study, the shear behaviour of regular saw-tooth and natural unfilled soft joints was investigated in the laboratory under Constant Normal Stiffness (CNS) condition, using a large-scale shear apparatus. In CNS testing, the normal load varies during shear displacement of a non-planar joint, depending upon the extent of dilation and compression of the joint. Soft joints were prepared in the laboratory using ordinary casting plaster, whereas the natural joints were sampled from a rockslide in Kangaroo Valley, New South Wales. The Coordinate Measuring Machine (CMM) was used to map the joint surfaces precisely before and after the tests. Extensive tests were also conducted on idealised infilled joints under CNS for various infill thickness to asperity height (t/a) ratios. Commercial bentonite was used as the infill material in this study.

A detailed literature review is presented in two separate chapters, one emphasising the shear behaviour of unfilled joints under CNS condition, and the other on the shear behaviour of infilled joints mainly under CNL. This was done to clearly distinguish the differences between CNS and CNL, and to emphasise the relevance of CNS in contrast to the conventional CNL approach.

The effect of rate of shearing on the shear strength of soft joints having an asperity angle of 18.5° (Type II) was studied under CNS condition. Test results show that the rate of shearing has a considerable effect on the shear strength of joints. The peak shear stress of joints increases together with the dilation and normal stress as the shear rate is increased from 0.35 to 1.70 mm/min. A rate of shearing less than 0.50 mm/min has insignificant effect on the strength of joints under CNS.

Test results obtained for natural (tension) joints, and joints having an asperity angle of 9.5 degrees (Type I) under CNS condition are compared with the conventional direct shear tests (Constant Normal Load). It is observed that the peak shear stress obtained under CNL condition is significantly lower than the CNS condition, especially for the natural (tension) joints. The normal stress increases with the shear displacement under the CNS condition whereas it remains unchanged for the CNL condition. The dilation of the joints under CNL condition is much greater in comparison with CNS testing. The strength envelope for CNL testing shows an upper bound for all the tests. Laboratory tests were also extended to include joints having asperity angles of 18.5 degrees (Type II), 26.5 degrees (type III) and natural (field) joints. Plots of shear stress against normal stress show that a non-linear (curved) strength envelope is acceptable for soft rock joints subjected to CNS conditions.

The shear behaviour of soft joints (Type I and II) containing infill materials was investigated in the laboratory under CNS for a given range of initial normal stresses (σno) varying from 0.16 to 1.10 MPa. It was found that the shear strength of joints decreases considerably even with the addition of a thin layer of infill. Results also show that the effect of asperities on shear strength is significant up to a asperity height to infill thickness (t/a) ratio of 1.4-1.8, whereas the shear behaviour is controlled by the infill alone beyond this critical ratio. The shear displacement corresponding to the peak shear stress is considerably reduced once the infill starts to govern the shear behaviour of the joint.

In this study, a new shear strength model for soft joints has been developed for the prediction of unfilled and infilled joint strength based on the Fourier transform method, energy balance principle and the hyperbolic stress-strain simulation. It is verified that the model predictions are in good agreement with the measured data. The applicability of the eNS technique in practice is discussed in detail with regards to the excavation of a mine roadway and in the case of slope instability in jointed media.



Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.