*Presented at the 45 ^{th} Annual Meeting of the Fuze Section, National
Defense Industrial Association, Long Beach, CA, April 2001.*

*Barry Neyer
PerkinElmer Optoelectronics
1100 Vanguard Blvd
Miamisburg, OH 45342
(937) 865-5586
(937) 865-5170 (Fax)
*Barry.Neyer@PerkinElmer.com

Sensitivity tests are used to determine the amount of energy, current,
charge voltage, etc that is required to reliably initiate detonators, initiators etc.
Various test methods have been developed over the years that guide the experimenter in the
choice of test levels. Some examples of these are the Bruceton test, developed in the
1940s, the Langlie test developed in the 1960s, and the Neyer D-Optimal test developed in
the 1980s. One characteristic of each of these tests is that they are sequential; the
results of test *N *must be known before test
level *N*+1 can be calculated.

For many purposes, the requirement of sequential tests is not an obstacle. The tests take little time to complete, and the result of an individual test is immediately known. However, for tests like VeriComp or VeriDrive tests, the time for construction of the components presents unique difficulties. In many cases, it may take hours or days to complete the component. Thus conducting a purely sequential test is rather difficult. Recent work with a group of toxicologists who have similar constraints, has lead to a new threshold test that overcomes these difficulties. This new design allows a number of items to be tested or constructed at a one time. The new test is a superset of the D-Optimal test. The efficiency of the test is only slightly reduced when compared with the standard D-Optimal test. The new test method will be discussed and practical applications given.