Military-Aviation Semiconductor Sourcing Remains Problematic

Sustaining often decades-long semiconductor supplies for military and aerospace electronics remains an ever-steeper uphill struggle. Dealing with the daunting challenges for secure continuity of key electronic component supply without compromising quality and/or incurring excessive price premiums needs an entirely fresh business approach.

The reasons for change are clear. There are platforms in service now that have by far exceeded their estimated or scheduled lifetime. Perhaps the most visible case in point is the USAF B-52 strategic bomber. American taxpayers continue receiving a great return on their original 1950s military investment. It may seem hard to believe, but this aircraft has been in service for over half a century and is likely to be in operation for at least another 30 years.

This means that a project that was initially supposed to have a 30- to 50-year life cycle will have exceeded 80 years by the time it is grounded. The B-52 is just one of many similar military and commercial examples of exceptional system service longevity driving the need for continuing electronic systems support.

Looking at more current challenges, many semiconductors designed into the F-35 Joint Strike Fighter will have already generated end-of-life notices even before entering full service. It’s going to get worse. Here’s why:

It‘s estimated that 3 percent of the global pool of electronic components is becoming obsolete each month. On a long-term defense project, typically 30 to 50 percent of the semiconductor products become obsolete before the program is even commissioned. This is surely a sobering thought for any designer preparing to spec a new system, especially if they’re working on a military or aerospace project that may take several years to deliver and then be in service for decades afterward.

Obsolescence mitigation (OM) continues as a reactive and corrective effort by military/aerospace electronics systems contractors to overcome the relatively short-term commercial availability of essential components needed for long-term product support.

The very word “mitigation” quite accurately connotes only a partial fix to the component supply chain problem. Breaking out the oars and rowing like crazy may keep you from going over the falls. However, avoiding the waterfall in the first place is far more pragmatic.

Our industry needs to take a new approach to managing the effects of semiconductor obsolescence. The approach required must address platforms and systems being developed today as well as those already in service. A recent UK Ministry of Defense paper on technology strategy put it quite well:

    Our task is to anticipate, prepare and meet the forthcoming challenges by being highly innovative, agile and flexible in our approach to defense science and technology based R&D. This must be complemented by rapid exploitation to yield military advantage with an ever increasing tempo.

When the semiconductor industry eliminates diminishing manufacturing supply, you eliminate the whiplash effect of obsolescence mitigation. OM is broken. Let’s fix it.

— Joe Bronson is Director of Business Development at e2v Aerospace & Defense, Sunnyvale, Calif. He was previously president of {complink 4773|Sanmina-SCI Corp.}, a leading electronics contract manufacturer.

10 comments on “Military-Aviation Semiconductor Sourcing Remains Problematic

  1. DBertke
    December 10, 2010

    Hi Joe,

    You identified a major problem that is mostly unique to the Aerospace industry.  Few other products look at life cycles measured in decades and it is indeed a serious issue.

    I think the best way to deal with the problem is to look at a planned technology replacement cycle.  The old method of stockpiling components has already proven to be mostly useless, but, if you look at the problem from a Block Cycle view, then a solution presents itself.

    When you design the Avionics or any other electronic device, you incorporate the most advanced Aviation grade component you can find.  As you pointed out, a lot of the chips are only produced for a handfull of years, so the only reasonable solution is to plan for an Avionics replacement cycle.  Set up a group supporting the product and have them prepare a replacement every three or four years with the addition of new components that meet or exceed the now obsolete components in the product they replace.

    Using this approach, you put the Avionics on a similar track to the engines and other aircraft components that get swapped out after every several hundred flight hours or years.  They essentially do this very effectively with engines, so there is no reason why you cannot do this with the Avionics.  It keeps your capability on an increasing track with technology, you continually weed out the obsolete components with each upgrade and it will be less expensive than trying to find or replace a component that no longer exists.

    While electronics can work for years without fail, there is no need to hamstring their function by worrying more about how long they last verses when can we replace it with something better, cheaper or more energy efficient.  Given the tremendous amount of money being spent on aircraft and avionics maintenance, the industry would be better served by a planned product improvement cycle than the traditional World War II era system we live with today.

    One thing that technology has taught us is that it is easier and cheaper to build and toss than it is to keep that old radio around, even if it still works, because some day we may need it and we can no longer find the right batteries.  Planned replacement enables the Avionics to maintain a record of excellence and enhanced capability without the extremely long logistics tail we now attach to every aircraft.




  2. kumar1863
    December 11, 2010

    Whilst semiconductor technology obsolescence is a cause for long term concern in the support of electronic components.

    Today the major thrust of electronics technology development is almost entirely dominated by high volume commercial requirements to satisfy the rapidly expanding market opportunities for video games, personal computers, mobile communication systems and new developments in the automotive industry. The computer and communications industry alone accounts for more than 70% of the market share. Although the military requirement for semiconductor products is now far greater than it has ever been, its share dropped to 0.5% where it is 90% in 1970.

    Proactive obsolescence management will require a culture change in both military and defense contractors. 



  3. Eldredge
    December 11, 2010

    This approach provides another benefit as well. During the upgrade to newer electronics, a stockpile of the previous equipment is available for the equipment not yet scheduled for the upgrade.

  4. Parser
    December 11, 2010

    The problem with any electronics substitute is that it would require very costly design evaluation. It is cheaper to stockpile original replacement parts.

    I am not aware if there is a grading program for parts and subassemblies and treating them as functional module with very precise specification. Therefore one could change part inside but the external functionality would not change.

    Although we can hear time to time that an airplane was modernized with newer systems to perform at higher level. This approach requires design and very costly testing. It is a cheaper approach than designing a new airplane. 

  5. AnalyzeThis
    December 11, 2010

    I agree this is a problem. But it doesn't seem to me like there is one cure-all solution.

    DBertke makes some very good points and having an Avionics replacement cycle is a good idea in theory, but in practice I just don't see how that will work in all situations.

    And that is true that engines and other aircraft components do get swapped out during reasonable time cycles, but those components are designed to be swapped out, unlike electronics. Swapping electronics out is potentially a much more unpredictable and complicated process.

    Obviously, electronics also differ from your standard “traditional” parts in that they cannot be physically repaired or uniquely fabricated by a skilled machinist.

    Anyhow, it is indeed a difficult problem and while I do agree with the idea of implementing planned replacement cycles, I question how realistic that's going be especially considering that part of the reason this problem exists to begin with is the fact that projects has outlived their expected intended use cycle.

    I do agree that it's certainly easier and cheaper to just buy a new radio and throw out the old one even if it still works. But these are some big, expensive, and complicated “radios.”

  6. Anna Young
    December 11, 2010

    This is definitely a problem seeking a prompt resolution but my understanding is that there are companies in the market place that have offerings that can help the military and aviation OEMs resolve the problem. This is underscored in reverse engineering that certain components manufacturers and distributors specialize in.

    These companies take components that have gone out or that are about to go out of production and secure the rights to keep them in the market for as long as the equipment maker wants them. The problem, as Mr. Bronson correctly states, is that the price to get this done can be high because volume production may be so low buyers have to pay a premium.

  7. SP
    December 13, 2010

    Yes its definitely a problem. In defense services, it takes so long first to take a decision to build some design and procurement is a also a lengthy process. And the time taken to build is also quite long. But keeping in mind its for national security dont they get any special preference when it comes to make military components obsolete or end of life. After all whatever are the reasons to obsolete/end of life a component, they can be mostly handled by goernments.

  8. Barbara Jorgensen
    December 13, 2010

    The semiconductor and military businesses are working toward completely opposite goals. Chip companies want to create better, faster, cheaper technologies at an increasing rate while mil-aero wants systems that stand the test of time. Rather than moving closer I think the two are moving farther apart.

  9. elctrnx_lyf
    December 14, 2010

    I’m not really worried about the military services since they are only aimed to protect themselves from other nations and at the same time military is built to show off the dominance. But I’m really worried about the aerospace and medical companies since they have a responsibility and at the same time strong regulations to produce reliable and high life time products. But all the semiconductor companies are actually making products for the markets like communication, networking and consumer electronics to get the short term and high volume income. Only the major semiconductor companies are actually able to produce promising components for the aero and medical products.

  10. prabhakar_deosthali
    December 17, 2010

    This discussion reminds me of the machine tool industry. Those huge press tools and grinders and CNC machining centers  are designed to last for decades and they also use Electronics. Normally the machine outlasts the control electronics. Companies like FANUC who are the leaders in CNC machines, maintain spares inventory for such elctronics and guarantee you a support for the life of the machine. Similar support strategies can be followed by the high profile industries like Aerospace , Medical and Military.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.