Los Alamos National Laboratory
Abstract-
Existing techniques for determining the product equation-of-state (EOS) of a
condensed-phase explosive observe how detonation products push a metal.
Historically the most common variant has been the cylinder test , in
which a cylindrical charge expands a surrounding copper pipe. The cylinder
test owes its popularity to the fact that its liner motion is readily
measured by rotating mirror streak camera---the experimental method of choice
for three decades. The advent of velocity interferometry in the early
1970's gave the ability to measure liner velocity at a point, making several
additional configurations attractive. Three modern variants are discussed:
the flat plate push test , the big plate test , and the
sandwich test .
Although several EOS forms have been proposed, the JWL (Jones-Wilkins-Lee) equation is used almost exclusively. The reason is largely historical: JWL was developed in conjunction with the cylinder test and, despite known shortcomings, reproduces that data well. From its inception, JWL calibration has involved 1) computing the cylinder test using trial parameter values, 2) observing the outcome, and 3) making adjustments. These steps are repeated until satisfactory agreement with the measured liner trajectory is achieved. Clearly, this methodology yields results that are no better than the chosen analytic form.
Consequently we are developing methods that allow isentropes to emerge naturally---in numeric form. The observed structure then motivates more accurate analytic equations. One strategy (developed at LANL by S. Shaw) is to perform the above iterative process for a tabular EOS, such that basis points rather than equation coefficients are adjusted. This method works well for the one-dimensional plate push test, but will require a super computer to handle axisymmetric tests. A second strategy involves a simplified but direct quasi-analytic method---originally proposed by G.I. Taylor---for calculating the cylinder or sandwich test. Again for historical reasons, Taylor's method has never been carefully evaluated until present. The results are encouraging, yielding isentropes very close existing JWL calibrations but with a more physically realistic sound speed structure.
GALCIT Home Page
|
|