A Technical Review of the Problems Encountered in Leak Testing Small Cavity Ordnance Devices

Abstract

Military Ordnance Devices such as Detonators, Fuzes, Squibs, etc., are usually intended to be hermetic. They are tested for hermeticity, using the same tests we have always used, and then assumed to be sealed. The Military Standards and various “Derived-Procedures” used for these leak tests are widely “Misunderstood” and “Misapplied”, and as a result are not rejecting all of the non-hermetic devices tested. This is largely due to the lack of understanding the seal testing technology used. It is also due to the use of procedures that have not been changed in over twenty years. This paper specifically addresses the issues that are responsible for improper seal testing of ordnance as well as electronic devices. It is also found that many failed ordnance devices are ‘discarded’, ‘replaced’, and go unreported, leaving us with very poor statistical data. Although very commonly used for testing ordnance devices, MIL-STDs 883 (T/M 1014) (1) & 750 (T/M 1071) (2) are for “Microelectronics” & “Semiconductors” respectively. They are not written for the leak testing of ordnance devices. The Defense Supply Center recently investigated their MIL-STD-750 seal testing standards, which are used as one of the criteria to guarantee an “assumed 15 year life” for the devices they purchase. That investigation clearly showed that the leak testing specifications were not guaranteeing internal device environments for even one year for many small cavity devices. As a result of these findings, MIL-STD-750E T/M 1071.8 has just been tightened to 1 x 10 -9 atm cc/sec (air) for small cavity devices, and many of those small-cavity electronic devices have cavities ‘orders of magnitude’ larger than many compressed-charge ordnance devices. This paper addresses the many difficulties encountered in leak testing small cavity ordnance devices. It explains the practices that are allowing those escapes, as well as the stateof-the-art technology and procedures that are now being used to prevent those non-hermetic escapes. I. Introduction We are commonly encountering the conflicts involved in the leak testing data that is produced from the leak testing of common ordnance devices. It has been seen that data from two or more facilities running the same devices, is widely spread, especially when individual test methodologies have been employed. The most difficult problems are due to the wide variation in the procedures applied, the optional modifications in those procedures, and the total lack of specificity by the procuring body in calling out precise test methods to be followed. This, unfortunately, is the result of the ‘chain-of-people’ involved not being adequately informed of the technology involved in leak testing, and the weaknesses of the test procedures that have been allowed. Devices such as semiconductor devices, hybrid-electronic devices, and ordnance devices such as detonators, fuzes, and squibs, are always procured with the assumption they have some degree of hermeticity. Most procurement specifications are very weak in the callouts for hermeticity requirements. They are rarely specific as to the MIL-STD, Test Method, and Procedure to be used, and many do not call out the exact leak rate limit required. The electronics industry has a well developed pattern for the level of hermeticity that they request, (although it has recently been recognized as far inadequate for the life expectancy they procured for). Steps have just been taken to rectify that problem by tightening the MIL-STDs used for the procurement and qualification of electronic devices.