Tin Whiskers

[W1RKW]

I'm involved with an investigation at work with an unusual failure of a component used on a product we produce. The subject of  tin whiskers came into discussion. We don't have anything to conclude that the failure is related to such but I found the subject interesting and worth sharing.  Might explain some bizarre failures we might encounter as amateurs especially with modern equipment. see attached.

[Tom WA3KLR]

Hi Bob,

The latest issue of electronic design , November 7, 2013 pages 63 - 66, has an article on this matter. The recent problems are fueled by the RoHS restriction of lead in Europe.  However Bell Labs started to investigate this failure mechanism not long after WWII.  In the article there is a claim that 3 satellites have been lost to this problem.

http://electronicdesign.com/analog/don-t-let-tin-whiskers-destroy-your-design

Enjoy.

[W9BHI]

The company I work for makes products for the government and they specifically request that the solder
or board plating NOT be RoHS.
I guess they don't trust it either.

Don W9BHI

[Rob K2CU]

To determine the possibility of tin whiskers, determine what solder is being used. Is it lead free?  Are the components themselves RoHS qualified?  Tin whiskers can not only form on PCB artwork, but also on the component itself between pins. With a pin pitch as little as .020 Inch, the space between lead exit points on a part can be even less. IS the board artwork tinned with lead free solder? Keep in mind that these questions can help determine if tin whiskers are possible, but not that they are impossible. A high powered scope is needed to examine the boards. Another possibility is that you have copper whiskers in the artwork from incomplete/faulty etching. The former failures show up after time while the latter are from the get go.

Keeping within the confidentiality of the design and your work, can you describe the nature of the failure in general terms?

[w4bfs]

this explains why NASA specs conformal coatings for electronics that go to space

[k7pp]

Bob;
I've been aware of this problem for along time.   I worked at the Motorola Communications Service facility in Seattle for almost 30 years and came across this issue in UHF and 800 mhz transceivers.
Any of the units using an aperture coupled preselector found themselves losing their ability to resonate and pass any kind of signal.   After tracing the problem to the Preselectors, it was found that opening up the top plate and just blowing them out with a high pressure hose would cure the problem.
We found so many that had been in service for less than a year exhibiting this phenomenon that we borrowed a microscope and looked at an undisturbed unit which had the very metal hairs you described.   We took pictures through the scope and sent them to Motorola in Schaumburg Ill as they didn't believe us.  We had engineers crawling all over the plant for weeks looking at each and every unit from a large fleet of 800 mhz units of which 80 percent had detectable metal hairs.
Under the microscope they looked like a metal forest.   
Never did get a conciseness or opinion as to the cause but most of the preselectors were made from pot metal and the top plates which were steel didn't have the problem.
Some of the models with this affliction were 800 mhz maxtracks and UHF Mitrex units.
I'm betting other Techs involved with Motorola at that time might have a few similar stories to tell.
Best recollection time wise was in the early 80's.
Hope this helps.
73,
Pete

[N4zed]

If this failure has anything to do with connectors..look for fretting. Fretting has became a big problem with some of the early equipment we produced before we started using lubricant on the connectors (Tin on Tin). Also look for the problem of Gold on Tin. Gold plating on one connector and Tin plating on the mating connector is also a big problem.

[W1RKW]

we use lead solder in our electronic assemblies we build and we conformal coat them too.  The other issue that came up in discussion was counterfeit semiconductors.

[Jim/WA2MER]

Bob;
I've been aware of this problem for along time.   I worked at the Motorola Communications Service facility in Seattle for almost 30 years and came across this issue in UHF and 800 mhz transceivers.

I was just about to mention the issue with the Motorola radios but you beat me to it.  I was engineering manager, then later FTR manager in Area C back when this issue first surfaced.  It drove everyone nuts until there were so many complaints that the plant could no longer say that "nobody else is complaining." I seem to recall the problem showing up in Mostar radios as well.

[WA2ONK]

When I worked for Verizon we had problems with our "new" GE IMTS receivers. They would loose sensitivity for no apparent reason. We would re-tune the front end filters and sensitivity would be back to spec. This would happen every few months until we found out about the whiskers. We opened up the filters and could see the whiskers with a magnifying glass. We were told by GE that some of those filters weren't plated properly. After a brushing with a brass wire brush and a very light coat of WD40 they were fine. A year or so later, the IMTS system was dumped, cell phones took over. So I don't know how long that "fix" would have lasted.

[Tom WA3KLR]

Yet another tin whiskers article; the October 2013 issue of Microwave Journal has an article titled “Fusion Processing of Surface Mount Components to Mitigate Tin Whiskers”.

http://www.microwavejournal.com/articles/20778-fusion-processing-of-surface-mount-components-to-mitigate-tin-whiskers

Enjoy.