Effect of COTS on Connectors
A look at the impact the Perry Memorandum has had on the military industry, particularly connector technology
By Cristi Cristich
 Off-the-shelf commercial equivalent connectors still meet all of the applicable military requirements. |
null
In 1994, Defense Secretary William J. Perry wrote a defining memorandum, "Specifications & Standards – A New Way of Doing Business." Becoming known as the Perry Memorandum, the document stated that military standards would only be used "as a last resort, with an appropriate waiver." It is instructive, eight years later, to look at the impact this document has had on the military industry, particularly connector technology.
The Perry Memorandum was written as a result of the work done by a special team chartered by the Deputy Under Secretary of Defense (Acquisition Reform) to develop a plan to decrease reliance on military specifications and standards. The assumption was that many commercially available standard systems, subsystems and components could perform as well as their "military" counterparts. By specifying such commercial products, the military would decrease cost and development time, while at the same time protect itself from parts obsolescence.
Subsequent to the Perry Memorandum, prime contractors were free to design their complex products to a performance standard as opposed to a litany of specifications that may or may not have been appropriate to the application. These contractors were also able to specify commercial-off-the-shelf (COTS) subsystems that up until that point could not be used in military equipment. Although there have been some bumps in the road, the implementation of these changes certainly produced many success stories of military products developed more rapidly and at reduced costs.
Connector Specifications
As a result of the Perry Memo, the government began to minimize its involvement in the standard, specification and qualification business. Hundreds of military standards are now inactive because of this initiative. When the lower level specifications, such as plating, materials and test procedures were cancelled, there was some initial confusion but much of that has leveled off over the years.
However, significant challenges related to the actual connector specifications themselves remain. Because military specifications became the harbor of "last resort," funding to create new or revised connector military specifications or to correct existing connector military specifications began to dwindle.
The impact of this policy is especially significant for connector specifications. Chips, resistors, capacitors and other components are usually contained inside a box designed, fabricated, controlled and qualified by one design authority. By contrast, connectors are often the interface point between multiple parts of a system or subsystem, and therefore are controlled by multiple design authorities — all with their own preferences. Connectors often perform a variety of mechanical functions as well as their passive function to carry signals with the least possible buildup of circuit resistance. It is imperative that both the electrical and mechanical characteristics of a connector system be well defined and controlled by some authority or standard.
 Off-the-shelf microwave connectors. |
null
Even the most reliable and well-designed connector, by its nature, introduces variability and increased resistance into an electronic circuit. Connectors are almost always used at a point in the system where both mechanical and environmental stresses are at work. Space is usually an issue as well. Design engineers continue to try to put more signals with higher speeds that create more noise into an increasingly smaller package. Smaller sizes are typically at odds with greater durability. Increased densities also mean decreased tolerances to ensure intermateability. Connectors are frequently installed at an environmentally and mechanically critical point in the system and are expected to repeatedly interface with connectors manufactured by multiple suppliers. The expense and risk associated with these circumstances often outweigh the benefits of using COTS connectors. In many cases, the military standard was the best way to define the connectors that customers needed to use, and to ensure intermateability and interchangeability from supplier to supplier.
One important area that needs to be adequately defined is the interface area of the input/ output (I/O) connector. This is the area most vulnerable to process and manufacturer variability. For example, in February, Cristek was shipping sophisticated cable harnesses (about 80 hours of labor in each one) that incorporated a number of its military-grade connectors. However, several parts of the harness were commercial grade. It had to stay that way because they were being plugged into an existing COTS box. All of the harnesses began to come back for repair. The commercial-grade connector was the failure mode. It was being overstressed and used for an application for which it was not intended.
Ironically, the directive to refrain from using military standards to specify connectors has, in some ways, had the opposite intended effect on cost and reliability. When design engineers were pushed to use a military specification part, the selections were limited so they would often go the extra mile to figure out how to make it work. Freed from the limited choices, engineers could now specify exactly what they wanted, and would often customize more than they otherwise might have. This diluted the overall consumption of the standard parts and also limited the ability to easily second source these connectors.
Proactive Steps
Customers may want to buy COTS products, but they still insist on the same level of quality and performance they are used to getting from qualified parts list (QPL) products. They also need to be confident that the products intermate with other supplier's products reliably. Here are some of the proactive steps that the connector community is taking.
In lieu of going to the Defense Supply Center Columbus for a government-coordinated standardization meeting for its nanominiature connector product line, Cristek coordinated a series of meetings among a number of its competitors. It was a bit uncomfortable at first, but everyone emerged knowing that as they went into the marketplace, they could confidently tell customers that if they bought products from certain competitors that not only would they meet the performance requirements but that they would be intermateable.
Many suppliers that focus on the military and aerospace marketplace are taking steps to ensure that the connectors they offer, although they do not bear the QPL designation, are designed and manufactured to the same environmental and reliability standards that customers would expect. In response to many of the COTS subsystems being used that have a commercial standard connector for the I/O, Cristek developed ruggedized versions of connectors that mate with the existing COTS connector. In some cases, a ruggedized system can be retrofitted into both pieces of equipment.
Conclusion
Although Secretary Perry's intention was never to completely eliminate the use of military standards (especially if they were the best way to communicate what is needed), it is clear that some of the early reactions were actually overreactions. It is not uncommon for overreaction to be needed in order to stimulate change and reverse the thinking about an entrenched system. At least that seems to have been the case with military connectors. If a military part is the best selection, customers will not hesitate to use it, but by the same token if a COTS part is better suited for the application they will use that. Nearly a decade after the Perry Memorandum, most military users seem to have adjusted to these changes and apply common sense to their connector specification decisions.
CRISTI CRISTICH is CEO, Cristek Interconnects Inc., 1301 South Lewis St., Anaheim, CA 92805; (714) 618-2002; Fax: (714) 535-4897; Web site: www.cristek.com.
