The analytical and experimental methods of determining the compressive yield stress of materials have been reviewed briefly. Using aluminum, EN 24 steel, lead and a tin-lead eutectic alloy, the areas of applicability of the Cook and Larke method (including its modified form for strain-rate sensitive materials) and the ring compression test have been delineated. Under conditions of sliding friction, both the methods give similar values for the coefficient of friction. Under sticking or near-sticking friction conditions, the modified Cook and Larke method is more accurate, because the interpretation of the ring compression test ignores the strain-rate sensitivity of the flow stress. Within the experimental range, the coefficient of friction and the interface friction factor were unaffected by the changes in the strain level, strain rate, grain size and efficacy of lubrication. Under slow loading conditions, the effect of specimen dimensions on the flow stress could be attributed entirely to a change in the frictional contribution. The strain and/or strain-rate hardening behaviour, as well as the grain size dependence of the flow stress of the different materials, were consistent with earlier wellknown results.
Earlier workFor evaluating the true stress-true strain curves of materials, axi-symmetric compression of solid cylinders is often used. In these tests friction is present and this leads to barrelling [1,2]. But the analysis of many metal-forming processes, e.g. extrusion, requires a knowledge of the compressive yield stress of a material, i.e. which is corrected for friction [3,4]. A correction for friction can be effected either analytically or experimentally.For analysing a cold-working process, Coulomb's law 9 = gN, where 9 is the frictional (shear) stress, p is the coefficient of friction and N is the normal stress, is often used. While in most analyses p is treated as constant, in reality it depends on a number of variables in a complicated way, e.g. material properties, efficacy of lubrication, rate of sliding at the interface. In hot working, on the other hand, sticking friction is assumed. Here the frictional stress is taken as equal to the shear yield stress of the material in plane strain, again an idealized picture [5,7].A barrelling correction, which is analogous to necking correction in tension, is often made. But this calculation is influenced by the specimen dimensions and the lubricant used. Additionally, some questionable assumptions are also involved in the analysis [8-113. Slip-line field solutions are often inapplicable for axi-symmetric flow problems [12][13][14][15][16][17].Early experimental methods of Rummel, Meyer and Nehl and Siebel and Pomp are only of historical significance [18][19][20][21][22][23][24]. The methods of Taylor and Quinney [25], Ford [26] and Polakowski [23] are time consuming as they require remachining (which also changes the overall mechanical history of the specimen) and cannot be used when an allotropic transformation is present in the material, e.g. a steel...