Tin Whiskers will cause digital products to fail within 10 years?

[Frank Petronio]

Well I rediscovered Ken Rockwell after years of not reading him, and his mention of the tin whisker problem sounds pretty legit.

With all the engineers here, tell me if this is true or just semi-true? It means my fanciest digital photo toys might well be kaputt around 2022?

Here's some related links and a pull from Kenny boy....

http://www.huffingtonpost.com/2012/0...n_1221076.html

http://www.guardian.co.uk/technology...ch.engineering

From http://www.kenrockwell.com/tech/2012-07.htm:

Tin whiskers are what mysteriously grow out of electronic parts that don't have enough lead in them to stop the problem. When these grow, they start shorting out circuits and causing all sorts of intermittent and permanent weird problems.

Solder is the soft white metal that holds all the parts together on circuit boards. Military and space electronics require at least 3% lead in their solder for precisely this reason. Normal solder is 40% lead, but optional RoHS proposals (that's the little "10" in circles on Nikon lenses and other products) demand no lead.

Consumer products makers love the RoHS/no lead standards, because these whiskers ensure that consumer electronics made without lead will all die in about ten years, so no one will get away with using old products, and have to keep buying new ones.

10 years is a perfect target life: it's long enough that no one blames the makers for deliberately building this in, and long enough that the maker has no warranty or good will problems. "Hey, it was ten years old, what do you expect?" Hey, I have plenty of thirty-year old and older audio and video products working here just perfectly, thank you.

If we all have to replace everything after 10 years, that's a lot of TVs sold. I have a two Visio TVs, a crappy brand, that died on me at 1 year and four years.

Reading the NASA paper, it may be a lot less than 10 years. All AF lenses work with electronics; they have been loaded with circuit boards and microprocessors and memory since the 1980s. Lose a circuit, and even a lens becomes useless.

No one ever died when their lens died, but the news media, sponsored by car ads, certainly doesn't want anyone learning what caused the Toyota unintended acceleration (UA) problems. Toyotas, like most modern cars, are drive-by-wire. Your gas pedal is nothing more than an input to a computer. Cars haven't had gas pedals connected to throttle valves since the 1980s. Toyotas use a variable resistor to encode gas pedal position, and feed that to the computer that does some math, and then controls the fuel injection and air induction systems itself for optimum efficiency.

The unintended accelerations were caused when a tin whisker in the throttle pedal encoder shorted it to make it seem as if the pedal was pushed all the way down!

What do you think, Docs?

[vinny]

We're fuct.

[jp]

More like tinfoil hats rather than tin whiskers. The whiskers are real and do cause problems, but it's highly overrated. This is speculation assuming your electronics will be 24x7 for the next ten years, which is either unlikely, or if your life revolves around it that much, you have backups to prevent data loss due to a failed electronic device. Lenses and AV products don't get that amount of use needed to grow the whiskers, which happen only under specific environmental conditions for energized electronics over a very long period of time. I've had a small datacenter in operation for 11 years now, and don't think anything has failed due to tin whiskers. Blown caps and cheap junk are responsible for most failures.

[Leigh]

I've had a small datacenter in operation for 11 years now, and don't think anything has failed due to tin whiskers.

The RoHS directive only took effect six years ago.
Most equipment sold even after that date used tin/lead solder, I wouldn't expect you to see problems yet.

Wait another ten years.

- Leigh

[jp]

The RoHS directive only took effect six years ago.
Most equipment sold even after that date used tin/lead solder, I wouldn't expect you to see problems yet.

Wait another ten years.

- Leigh

Datacenter environments have been tin whisker risk areas for far longer than RoHS due to a combination of greater operational value, larger amount of electronics, 24x7 operation, and other metal ducting, cases, etc... I doubt RoHS changes my risks much.

Personally, I do like lead solder, as it's much easier to work with.

What I've learned in this business of troubleshooting computer problems is not to hastily jump to conclusions like Ken Rockwell has done, or you'll be a loser at your job. You've got to solve problems with an open mind as to the complex possibilities of why something has stopped working.

[Leigh]

Datacenter environments have been tin whisker risk areas for far longer than RoHS...

Of course. As I mentioned earlier, the whisker problem has been around for 100 years.
The earliest known examples were found in equipment from 1912.

I agree that any troubleshooting effort must proceed methodically, following and analyzing the symptoms.

- Leigh

[Nathan Potter]

As Leigh has stated the silver shorting mentioned is related to electromigration phenomena - mass transport of silver atoms by electrical current. Tin whisker growth is a really complex and difficult problem to explain precisely. Despite hugh investments in research, really good models of tin solid state transport have yet to be formulated.

It has been demonstrated that the principal driver for whisker growth in tin systems is differential stress within the tin body. This is often caused by TCE (Thermal Coefficient of Expansion) mismatch between the tin and its substrate. Recently, in the IEEE Trans. on Advanced Packaging, and related publications the transport of tin has been found to move along grain boundaries which are presumably low energy pathways. The exact mechanism of movement is obscure but there are suggestions that grain boundary slippage due to temperature excursions could play a role in supplying energy sufficient for mass transport.

The industry partial cure for whisker growth has been to be very careful to match the TCE of the substrate to that of the tin alloy (Tin-Silver-Copper) being the choice to replace tin-lead. This can lead to very good reduction in whisker growth but is really an inelegant cure and does not mitigate the general problem of temperature excursions in the bulk of a SnAgCu alloy.

I would tend to agree that the elimination of lead in electronic devices was premature and the use of mostly tin substitutes may eventually become a headache for some consumers. OTOH how many electronic devices are going to be used for say 10+ years.

Nate Potter, Austin TX. / Corea ME.

[E. von Hoegh]

OTOH how many electronic devices are going to be used for say 10+ years.

Nate Potter, Austin TX. / Corea ME.

Grebe MU1 Sychrophase reciever, 1925
RCA Radiola 60, 1928/29
FADA Neutrolette, 1925
Zenith AM/FM table reciever, Feb. 1961
Collins R388 early 1950s

All in regular use. I can go on......

[SergeiR]

Grebe MU1 Sychrophase reciever, 1925
RCA Radiola 60, 1928/29
FADA Neutrolette, 1925
Zenith AM/FM table reciever, Feb. 1961
Collins R388 early 1950s

All in regular use. I can go on......

All the stuff from days when lead was used, right?

[E. von Hoegh]

All the stuff from days when lead was used, right?

Absolutely.