In December 2011, President Barack Obama signed the fiscal year 2012 US National Defense Authorization Act. The budget bill also encourages the implementation of procedures to mitigate the possibility of obtaining counterfeit components by making members of all tiers of the defense supply chain accountable. The meaning of the term counterfeit in this context includes fake, substandard, damaged, or mismarked components.
In the fall of 2011, for the first time in history, U.S. Federal Courts prosecuted an individual for trafficking in counterfeit integrated circuits, many of which were targeted for the U.S. military. Others were to be used in brake systems in high-speed trains and instruments used by firefighters to detect nuclear radiation. The administrator of the company that sold the components was sentenced to 38 months in prison and assessed fines of $166,141 for selling almost $16 million worth of semiconductors falsely marked as military, commercial or industrial grade.
As progressive as all of this news seems to be in the fight against counterfeit semiconductors making their way into the US supply chain, it is just the beginning. In fact, it is estimated that 2 percent of all the semiconductors sold last year were counterfeit. That doesn’t sound too threatening until we do the math: the estimated value of counterfeit parts that made their way into the U.S. supply chain in 2011 is over $5 billion.
Not only do counterfeit components threaten lives, they impose a vast negative financial impact to semiconductor manufacturers, distributors, electronics equipment manufacturers and end users. It is estimated that the actual cost of a failed semiconductor that makes its way into production – in any industry – can be more than 100 times the cost of the original component. That makes the bargain price – often 70 percent less than the “real thing” – look less and less attractive. No matter how hard the industry tries, it can’t get around that old adage: you get what you pay for.
"...it is estimated that 2 percent of all the semiconductors sold last year were counterfeit. That doesn’t sound too threatening.." Who with half a brain would think 2% counterfeit parts is not threatening?
If you were in a helicopter that went down owing to a faulty counterfeit part, or the victim of a faulty counterfeit medical appliance, I'm sure you would see things differently. Sure, 2% doesn't sound like a lot to you... but it's a lot to the people it affects. And it's also just an estimate.
It is just simple calculation. Everyone is calculating the risk of failure, in a bicycles, cars and helicopters. Do you believe there are products 100% safe at any possible situation? If I was a victim of such situation, it will be big loss for me, but for manufacturer I am only number on the paper. Busines is cruel, sometime is cheaper to take some bigger risk and buy insurence in case of problems.
No mention of the DLA's Generalized Emulation of Microcircuits (GEM) program which can provide a form, fit, and function replacement for non-available microcircuits using current design and processing technologies? The GEM contract was awarded to SRI International (Cage Code / 03652), perhaps an oversight ?
With regard to "Some companies are working on developing sophisticated identification technologies that can be included in the production of components, but these are also extremely costly." That's hogwash. The DLA recently Mandated the use of botanical DNA marking for high risk chips. It works and it is not "extremely costly".
All in all, a very self-serving article if you ask me.
No oversight at all. Rochester Electronics is well aware of the GEM program and most of the people running it. A part is not "GEM'-ed" and is not available to be "GEM-ed" if it is available by industry. That's a fact. It is only if a part is totally unavailable by an authorized source will the possibility of doing GEM be undertaken.
Regarding your comment about DNA.....Get your facts straight, then come out to talk. Stop quoting the marketing hype and go get a quote for implementation. Figure out how expensive it would be to implement when your part marking is all laser in your current assembly process as it is for many OCM's. Hogwash indeed - ANY part marking strategy mandated onto the OCM's by the 1% market is half-baked at best.
Rochester Electronics is doing something entirely unique here working with the OCM's on our Recreation/Replication strategy. It is an authorized method to get hard-to-get parts like ASICs where the only other choice is to redesign....in a newer technology....and that's not always a good thing to do.
By the way, the GEM program is a good program, but is limited on how far it can go with technology. At some point, having no design archive makes GEMs impossible.
Yippity - your comments show some additional research on your part was needed. Have at it.
Methinks Thou Dost Protest Too Much. It was I who said that the GEM program was an alternative for "non-available"microcircuits, that's "the fact" as you say.
With regard to the Applied DNA Sciences solution, I am not relying on "hype". I know the facts and suggest to you and your competitors call them, or the DLA, and then tell me it's not worth the money to protect our warfighters.
You have magically equated DNA tagging with protecting our warfighters. There is no connection here at all. Tagging known good product penalizes the good guys who can ship pure authorized product. Why didn't DLA insist upon tagging non-Authorized product? Doesn't that make more sense? Tag the product that comes from non-Authorized sources. Tag the material that has the highest odds of being counterfeit. Half the counterfeits reported this year are on active product. HALF! There is a huge procurement problem FIRST. If "protect our warfighters" is getting tossed around, go talk to the CM's and DLA themselves about procurement practices trying to squeeze the last dime out of every system. We should all be focused on the biggest problem first - using the Authorized sources FIRST if available. This alone cuts out half the reported counterfeit product in 2012.
And the lack of mentioning that the min cost will be in the 10's if not 100's of thousands of $$ for the remanufacture !!. There are benefits to both !!.. Military and others need to make contigency for min 25-50% replacement (service) parts adn possibly even 100% relative to the cost of the product !!. Brokers will always have benefits but hte decision to use Broker (Grey Market) need to be made by experienced/educated decision makers not just the bottom of the run purchasing officer who does not understand the consequences of bad decisions !.
You make a very important fact, and it rarely gets implemented well. That is ordering enough spares/repairs capability baked into the funded portion of the program. It's a very difficult program for program managers on long-life programs (20+ years). They are only funded for so many years, but must somehow do all the last time buys to secure enough product. We have seen that the only way for this to change on those types of programs requires a very different mindset at the time of program award.
While the author didn't mention pricing, let's assume your pricing ranges are correct. It can still be significantly cheaper to remanufacture product than redesign and qualify a product.
I am not sure whether I found the article or the comments more interesting. I can say that while at LSI (along with Dan), I was contacted several times by a company, an agency, or DSCC (DLA) needing more parts. When a quote was given, the usual answer was that they would buy salvaged parts because that was cheaper than another wafer run. And yes we were charging a premium ($10K to $100K as most of processes had to be restarted, recreated and re-qualified. Having the assurance of knowing where the part originated can be better than a part with an unknown pedigree? I believe that many of these issues are a result of NOT controlling the supply chain as the article highlights.
Join our online Radio Show on Friday 11th July starting at 2:00pm Eastern, when EETimes editor of all things fun and interesting, Max Maxfield, and embedded systems expert, Jack Ganssle, will debate as to just what is, and is not, and embedded system.