2024/04/24 by Nielsen, Stefan, Spagnuolo, Elena, Violay, Marie
paper · doi:10.13127/rpt/215
Rotary shear machines are becoming a popular tool for rock friction experiments. Such machines are of particular interest to study conditions approaching those of natural earthquakes (slip rates of one meter per second or more, normal stress of several tens of MPa, total slip up to several meters). One key issue for a successful rotary shear experiment is the use of a very carefully and precisely shaped sample. Imperfect samples induce misalignment and spurious vibrations which alter the mechanical data or induce sample failure in the early experimental stages. Traditionally, the cylindrical or annular sample is shaped from a block of the rock which is the object of the test and then introduced in the machine sample holder, either with or without a metal jacket slid around its perimeter. A difficult and lengthy preparation and rectification of the bottom, top and sides of the cylinder is necessary. Here we show that it is possible and advantageous to create a composite sample where one portion only is composed of the rocky material to be studied. A synthetic mortar fills the remaining void between a previously machined metal jacket and the rock portion, insuring cohesion of the assembly. After curing the mortar, the rectification needs only to be done on the top part of the sample (the frictional contact surface for the test). The advantages are that the precision in the sample is improved, the labor time is reduced and it is possible to create a water-tight system for fluid pressurized experiments.