Tin Whiskers — Fact Or Fiction? 459
bLanark writes "Some time ago, most electronics were soldered with old-fashioned lead solder, which has been tried and tested for decades. In 2006, the EU banned lead in solder, and so most manufacturers switched to a lead-free solder. Most made the switch in advance, I guess due to shelf-life of products and ironing out problems working with the new material. Lead is added to solder as it melts at low temperature, but also, it prevents the solder from growing 'whiskers' — crystalline limbs of metal. The effect of whiskers on soldered equipment would include random short-circuits and strange RF-effects. Whiskers can grow fairly quickly and become quite long. Robert Cringley wrote this up this some time ago, but it seems that the world has not been taking notice. I guess cars (probably around 30 processors in a modern car) and almost every appliance would be liable to fail sooner than expected due to tin whiskers. Note that accelerated life-expectancy tests can't simulate the passing of time for whiskers to grow. I've googled, and there is plenty of research into the effects of tin whiskers. I should point out that the Wikipedia page linked to above states that tin whisker problems 'are negligible in modern alloys,' but can we trust Wikipedia? So: was the tin whisker problem overhyped, was it an initial problem that has been solved in the few years since lead-free solder came into use, or is it affecting anyone already?"
Fact (Score:5, Interesting)
The symptoms were always the same. The radio would be working fine one minute and be stone-deaf the next. Sometimes just opening the cabinet door would be enough to dislodge the whiskers and remove the short. But it always returned a few days or weeks later. We got to the point where whenever we were sent out to fix a deaf base, our first repair technique was to take a large screwdriver and rap the cavities with the handle a couple of times, hard. We got some funny looks from the customers but they were happy to be back on the air.
GE finally admitted that the plating was the problem and shipped us a bunch of cavities with a different alloy to use as replacements. They never would tell us what the difference was. Curious, we disassembled some of the old cavities and shook out tiny metal slivers that were finer than a human hair. Some were up to a centimeter in length.
All of the radios we had problems with were less than five years old at the time.
On a somewhat unrelated note, a friend of mine works for a company building avionics. They're still using Lead/Tin/Silver solder for US military contracts. He thinks they know something the rest of us don't.
Re:Tin Whiskers are fact (Score:4, Interesting)
Is lead truly that dangerous ? (Score:5, Interesting)
Re:Tin Whiskers are fact (Score:3, Interesting)
I wouldn't go that far (Score:5, Interesting)
Also, when a philosophical-minority or fringe group tries to take over a highly-watched article, administrators eventually silence them if they insist on using unreliable sources or not keeping the article in "proper balance," where "proper balance" reflects the real-world opinions on the subject. Pseudoscience, alternative-history, and similar-subject proponents tend to get banned if they aren't careful.
Low-traffic articles nobody cares about are very vulnerable to this kind of abuse though.
Re:Is lead truly that dangerous ? (Score:3, Interesting)
lookit!!! (Score:4, Interesting)
Re:Fact (Score:2, Interesting)
Re:I wouldn't go that far (Score:4, Interesting)
I had an administrator remove factual, documented information from an article because it didn't jive with the rest of the obviously biased article.
On Wikipedia, the truth is what the Admins says is the truth.
Re:This qualifies for a "Are you pulling my leg?" (Score:1, Interesting)
You can have my leaded solder when you take it from my warm tinned iron.
I've worked with both leaded and unleaded solders. For hand work, nothing beats good old 60/40 Pb/Sn, but some lead-free solders are much worse than others. The ones with a little bit of copper or bismuth are somewhat better at wetting components. Unfortunately, most solder manufacturers just say "Lead-free" on the label, so it's hard to tell (at the retail level) whether you're getting "really annoying" lead-free solder or "not too bad" lead-free solder. My advice is to find a brand that works with your iron (and/or preferred temperature range) and stick with it. And to keep a spare set of soldering iron tips around, with a few pounds of leaded solder in the closet.
Reprinted from my blog (Score:5, Interesting)
More recently we got something called "the restriction of the use of certain hazardous substances in electrical and electronic equipment" or RoHS. RoHS prohibits the sale of materials containing more than 1% lead in the EU. (Old-style electronic assemblies use 37% lead solder.) RoHS came into force in 2006 but research into lead-free electronics began decades ago. Initially researchers tried pure tin plating, which lead to tin whiskers. Some products marketed in the late 90's even failed from this problem. But researchers did not throw up their hands in despair. RoHS has led to innovations in metallurgy to the extent that a circuit board designer can now choose from half a dozen different alloys. Today only 2% of printed circuit boards use tin plating.
Some of these new alloys use gold or silver finishes over copper. These are completely immune to tin whiskers. The most popular new system eliminated the plating step, attaching components directly to the bare copper using chemicals called Organic Solderability Preservatives. OSP leads to stronger and more durable assemblies than even the old tin/lead process.
The whining we see today on the subject of RoHS mirrors almost perfectly the doomsaying seen when California began regulating automobile emissions. There was at that time a tremendous amount of yelling about how the catalytic converter spelled the end of civilization as we know it, and only a moron would take the lead out of gasoline. But soon afterwards we saw the introduction of clean, efficient, powerful cars by Honda. Honda was even able to meet California emissions standards without using catalytic converters or even fuel injection. Their brand of engineering eventually trickled down to even the most benighted American car maker, and California emissions standards are now in force in every industrialized nation.
I would expect to see the same thing with RoHS. We have only just entered the initial stage of complaining. The tin/lead dinosaurs with backwards-looking engineering departments face an existential crisis. In other design houses the challenge of lead-free assembly is being embraced as a competitive advantage. Those who can adapt to RoHS will thrive and those who cannot will clearly suffer.
Cringley brags about a 1966 Thunderbird with a 428 cu. in. motor, a car so heavy, so polluting, and so slow by modern standards that it would be impounded by CARB and laughed off a drag strip by a base model minivan. As time goes on I think Cringley's views on the metallurgy of printed circuit boards will seem as antique at that T-Bird.
A question for those who have experience with this (Score:2, Interesting)
Re:Does it matter? (Score:4, Interesting)
Many vintage amps from the 1950's just need a few capacitors replaced, and they will work perfectly, 50-year old vacuum tubes and all.
Tin WhiskersFACT - It's happend in my lab recently (Score:3, Interesting)
Re:Does it matter? (Score:5, Interesting)
Yup...back to a day when people AND companies took pride in their engineering and craftsmanship.
I know now why some of those old McIntosh tube amps from the 60's still sell for $1200 and up. Things built back then were built like tanks.....and meant to last.
Sad you no longer see that in today's disposable society. Strange that in this day or people trying so hard to be "green", that these same people don't demand that companies built products with such quality that they will last and not have to be replaced every other year.
Re:A question for those who have experience with t (Score:3, Interesting)
My understanding was the conformal coating doesn't stop whiskers from forming. Small whiskers can simply grow straight through the coating. Besides, most conformal coating processes are designed to keep hands or screwdrivers or loose wires from shorting the circuit board by touching it. It isn't obvious to me that the conformal coating gets between pins and underneath pins in such a manner that would stop all possible whiskers from shorting to all possible nasty locations. Can conformal coating even get underneath the leads of an LQFP package (with 100% coverage)? A BGA package?
Conformal coatings may mitigate risk, but I don't think they are a "solution". Also, the performance of conformal coatings probably varies widely with type and quality of the coating and quality of application. Conformal coatings are in the category of "any manufacturer that cares enough to get the conformal coating correct, probably also knows enough to get the tin-replacement solder chemistry correct and avoid the problem in the first place."
I think we have to worry about the cheap subcomponents from relatively unknown factories being assembled into larger subsystems that are in turn assembled into larger machines/products/cars/etc. To have a quality problem, we only need one whisker on one circuit board, and there are lots of circuit boards in most machines (and most other devices too.) The "not checking the suppliers" problem is what caused the leaky electrolytic capacitor problem. It only took a few inexpensive capacitors to cause lots of computer problems.
Re:Well here are a few facts... (Score:4, Interesting)
Re:I wouldn't go that far (Score:2, Interesting)
But it was long enough ago to have been included in the musical play "The King And I".
Re:Does it matter? (Score:2, Interesting)
However, with that said there are a lot design techniques that are applied to aerospace electronics but are not applied to commercial electronics. I am not sure that it really matters, does anyone really want their Dell to last for 15 years?
Tin Whiskers Are Real. Very Real. (Score:1, Interesting)
It is an extreme problem in military electronics because both the likelihood and cost of failure rise for several reasons. Military electronics are, indeed, expected to operate for decades, the types of maintenance that can be done do not fix the problem, and the types of repair required on failure are fairly expensive. (Assuming failure does not result in crash and burn.)
The particulars of military make whisker-shorts fairly likely, and the cost effective solution, so far, has been to use the solder that works instead of the solder that doesn't. (As other posters have noted, there are some lead-free solders that avoid the whisker problem, but cause others. One is the identification of a good solder joint. Another is eutecticity. Yet another are the melting and reflow points; understand that complex assemblies undergo multiple solder and temp cycles and using a high temp solder after a low temp solder is a sure recipe for disaster.)
The real devil of a problem is subcontracted designs from commercial-oriented contractors. In practice, there's nothing wrong with leveraging commercial talent, but in practice, they don't always realize or take seriously the no-lead-free dictates under which the military industry operates. This translates directly into migraine headaches for design engineers and the logistics departments of the big four military contractors.
And the gut-grinding aspect of this is that electronics do not use that much lead to begin with. A typical turn-of-the-century PC board used around a gram of solder, of which less than half, typically, is lead by weight. That's a tiny amount-- about a thousandth of a pound. Annually, lead consumption due to solder is much less than one percent. (Compare to storage batteries, though, which is up around 75%.)
These are not problematical amounts of lead.
The only thing that's been overhyped, here, is the dangers of lead-based solder.
Re:Whiskers & Corrosion (Score:2, Interesting)
Parallels with.. processing parallelism (Score:3, Interesting)
In the programming world, I expect things to be resolved the same way: by superior engineering, taking advantage of the mountains of research and practical application of parallel processing designs that have been going on for decades. My favorite language Python is particularly sensitive to this debate due to something called the GIL [wikipedia.org], but solutions abound, including the newly accepted pyprocessing [berlios.de] module or any number of things you can do with Twisted [twistedmatrix.com], both of which stand as examples of better, practical engineering taking advantage of known solutions to the problem.
Engineering fixes most problems that changing standards and regulations cause, eventually.
Re:Exception (Score:1, Interesting)
Commitment to 'lean manufacturing' and the perception of non-compliance means that many manufacturers of "high reliability" assemblies also follow RoHS rules blindly. They avoid inefficient dual standards and are able to withstand the burden of proof of compliance to the new green standard by avoiding 63/37 and RoHS in the same facility.
Remember, NASA was allowed an exemption to CFC ban for the Space Shuttle's external tank foam. They chose instead to switch to a poorer quality insulation to avoid the perception of non-PC behavior... and so we lost Columbia.
Re:Well here are a few facts... (Score:4, Interesting)
What's the implication for hobbyists? (Score:2, Interesting)
Re:NASA Are Worried (Score:3, Interesting)
http://nepp.nasa.gov/WHISKER/photos/pom/2003sept.htm [nasa.gov]
Embedded controllers may need to last a long time (Score:3, Interesting)
The Ford EEC IV engine controller from the 1980s was designed for a 30 year life span. (My 1985 Bronco still has its original unit.) No software updates, either; the program was masked onto the custom CPU chip. Never needed a recall. On the other hand, my 2007 Jeep Wrangler had three separate major software updates in the first year.
Plenty of industrial hardware needs to be able to run for 30-40 years.
Jenkins Valves [cranevalve.com] used to boast of century lifespans. They had pictures of a valve installed around 1900 which had been removed during a water line replacement. It was still working fine, and after a few months of being photographed and shown off, it was re-installed in a new water line.
Xbox 360's RROD is also linked to this (Score:3, Interesting)
Obviously there's a heat / warping problem and the board is rumoured to bend at high heat, none the less the solder has been listed as a problem too.
Oh by the way SOUL-DERR
SOULLL-DER
NOT 'sodder'
Re:Exception (Score:1, Interesting)
On top of that, there is no visual way to distinguish between a lead-tin and a tin-only termination so you never know exactly what you are getting. I do some work with the space industry, and in a recent study they bought parts specified as "not pure tin" from a wide variety of sources and performed metallurgical tests on them. The number was well over 1/3 of those parts were in fact pure tin terminations. It's important enough that they have developed procedures to strip the terminations and replate with known tin-lead (not a cheap alternative).
So the bottom line is this, regardless of exemptions, except for space and high-end military applications willing to pay the big bucks for lot testing and replating, any other electronics made today likely contains pure-tin-coated parts and stands a significant risk of failure due to tin wiskers.
Re:I wouldn't go that far (Score:3, Interesting)
But where I really think wikipedia is great is for pop culture references... Things that don't have a lot of hard-to-understand scientific details, and that are more common knowledge to more people. Things, as you mentioned, like music, movies, TV, etc.
I think wikipedia has really been great for TV shows, especially those with a die-hard, internet-savvy fan base. A single TV series can hundreds of articles on episodes, characters, locations, themes, symbolism, etc., and you could spend all day learning more than you ever wanted to know about a single show.
Re:Well here are a few facts... (Score:3, Interesting)