In most numerical codes which simulate shock wave propagation through rocks the stresses which drive the grid are separated into a mean stress and deviatory components. The mean stress (P = -/sup 1///sub 3/ t/sub ii/) is measured as a function of volume on small samples in the laboratory which is normally called the hydrostat. Repeating these measurements on many samples, unloading each sample from a different mean stress, will produce a single loading pressure-volume curve and a series of unloading curves. The difference between the volume of each unloading path at zero mean stress and the original volume of the sample represents the amount of void space that has been irreversibly squeezed out. Ideally the input for the numerical programs should include all of the P-V data measured, however, this is not practical. The present method used in SOC and TENSOR is to input the loading hydrostat and a single unloading path from the maximum mean stress obtained experimentally (usually approximately 40 kbars). Intermediate unloading paths are then interpolated using a weighted average between the slopes of the loading and unloading curves. The model presented is designed to correct deficiencies in the present method.

A general description of the treatment of the Bevatron injection problem is presented, and the programs are described. The program INJECT for the acceptance calculation determines what beam can be accepted into the machine as a coasting beam. The rf trapping calculation, PHASE, tells what beam will be accepted as the rf voltage is turned on. The accepted pulse calculation, HINJ, uses the results of the two previous calculations to determine what fraction of an injected pulse survives the injection, rf trapping process. The appendices contain sample control card checks, input data cases, and selected program output. (WHK)

None

None