Volume 14, Issue 3 (September 1991)
On Some Issues in Triaxial Extension Tests
Results of triaxial extension tests on dry sand are presented. Special attention is paid to the factors influencing the results in triaxial extension tests. It is found that factors such as accuracy of the axial force measurement, axial force carried by the rubber membrane surrounding the specimen, gravity, and inhomogeneous deformation have a much larger influence on the results in triaxial extension tests than in triaxial compression tests. Accurate measurement of the axial force is achieved by placing the load cell inside the pressure chamber. The influence of the confining pressure on the axial force measurement is avoided through a proper construction. The force carried by the rubber membrane is corrected with reference to the result of a tension test on a strip of rubber membrane. The effect of gravity is accounted for by adding the axial stress due to gravity to the applied axial stress. The inhomogeneous deformation is traced by three lateral strain collars placed along the specimen height. Results of triaxial extension tests cannot be appreciated if these influential factors are not taken into account.
The failure mode of the specimen is found to be influenced largely by the initial density. Inhomogeneous deformation in the form of necking develops in the upper part of dense specimens and in the middle of loose specimens. For dense specimens, the necked region becomes wider with advanced deformation and develops into one single or two intersecting shear bands. For loose specimens, however, no shear bands can be observed.