RI 7298 Effect Of Increasing End Constraint On The Compressive Strength Of Model Rock Pillars

Model pillars of limestone, marble, sandstone, and granite with length-to-diameter (L/D) ratios of 3, 2, 1, 0.5, and 0.25 (0.286 for granite) were tested to failure in axial compression to determine to what extent an increase in end constraint increased compressive strength. Adjustable steel rings, mechanically tightened about the ends of test specimens to create a radial prestress of 3,000 or 5,000 psi prior to the application of axial load, increased end constraint more than was possible in previous tests using solid steel rings. (Solid rings, epoxy-bonded to the ends of dogbone specimens, increased compressive strength compared with that of straight specimens without ring constraint.) During the loading cycle, additional end constraint was produced by lateral expansion of the specimens against the rings. The effect of the end constraint was to increase the compressive strength 55 to 58 percent for the shorter pillars of dogbone shape compared with the compressive strength of straight pillars with the same L/D ratios without ring constraint. Most pillars with L/D ratios of less than 2 were made stronger by end constraint. Pillars with an L/D ratio of 3 did not benefit significantly from end constraint. When the results from compression testing of straight specimens of rock in the laboratory are used to design mine pillars in situ, the pillars will be larger than necessary if the horizontal compressive stress is several thousand pounds per square inch greater than that resulting from the weight of the overlying rock alone. The L/D ratio can be increased by a factor of 1.76 for ratios in the range 0.25 to 3 and still provide the same compressive strength. The safety factor applied with straight specimen test results should also be acceptable for use with constrained pillar test results.