[Elecraft] OT: News Article on "Tin Whiskers" Ruining Electronics

[Don Wilhelm]

Mike and Jack,

My apologies for posting based on my lack of knowledge.

Thanks for the educational updates.  I certainly am not prone to argue 
with the NASA research folks.
I do recall that it was a big concern inside ICs back in the 1980s and 
90s when I was closely associated with the industry and more and more 
density was being crammed inside those multi-legged critters.

I am surprised at the magnitude of inconsistencies and uncertainties 
still present in the research data.
It is very interesting to find that dipping with tin-lead solder appears 
to inhibit the growth, so those of us using tin-lead solders are 
actually helping a bit to inhibit the growth - but then we don't re-tin 
the full length of all the leads on all components either, so there is 
remaining exposure.

So based on reading those articles - my conclusion is that we should 
continue using the tin-lead solder and discourage the use of lead-free 
solder for as long as we are able to do that.  Since the tin-whisker 
induced failures seem to take about a year or more to develop, I might 
consider extended warranty coverage for any expensive electronic devices 
as more and more of them move to RoHS compliance.  This is really a 
two-edged sword.

73,
Don W3FPR

Mike S wrote:
> At 12:28 AM 10/19/2007, Don Wilhelm wrote...
>> The 'tin-whiskers' problem is one that exists mainly at the chip level.
>> The migration of conductive paths between pins that can be soldered 
>> by hand techniques is *not* a problem because the 'whiskers' 
>> typically do not extend that far.  It is a problem inside the chips 
>> where distances are measured in angstroms rather than in fractional 
>> inches or millimeters.
>
> Not according to NASA studies. They state "Whiskers as long as a few 
> millimeters are not uncommon..." - 
> http://nepp.nasa.gov/whisker/background/index.htm
>
> Common through hole IC packages (DIP) have 2.54mm center to center, 
> which means ~1 mm air gap between pins, and that is also a _very_ 
> common hole spacing pattern for discrete components. Here's an example 
> of non-"chip level" failure caused by tin whiskers: 
> http://nepp.nasa.gov/whisker/photos/pom/2004april.htm
>
>
>
>