Slashdot Mirror
The Tin-Whisker Menace
An anonymous reader writes "Fortune has an article about how the recent environmental push to completely eliminate lead from electronic components and wiring may eventually lead to the next Y2K problem of slowly-growing tin whiskers short-circuiting equipment.""
it's more disposal than consumer protection... tons of lead in a landfill isn't good
Same story as 6 month ago Zinc Whiskers Cripple Colorado's Computers. There's a PDF there that explains it all as well. They are pretty little whiskers, that can only be observed if examined very closely in the right environment.
I quote from a most informative pdf (get it here):
"A tin whisker is a single crystal of tin that grows spontaneously from a surface a pure tin. They are typically only a few microns (?m) in diameter but can grow to lengths of more than 10 mm (though lengths on the order of 1 mm are far more common) [NIST Website, 2002]. Tin whisker growth is spontaneous, not relying on external influences of current or electrolytic action, more commonly associated with mechanisms like "dendritic" growth, conductive filament formation and electromigration. While early studies believed that tin recrystallization (which occurs at 50 deg C) played some role in whisker formation, recent studies have reported as much, if not greater, propensity for whisker formation at temperatures as low as room temperature [NASA Web Site, 2002].(....)Conventional wisdom attributes tin whiskering to internal stresses in the pure tin layer, with a primary source being the compressive stresses caused by electroplating. However, tin whiskers have also been reported from surfaces where tin has been applied by methods other than electroplating. In the presence of compressive stress, whiskers are extruded over time, as a stress release mechanism. Many factors may contribute to the stress in the plating, including intermetallic formation, thermal expansion mismatches, corrosion of the substrate, and externally applied forces such as bending, lead forming and application of pressure. Defects such as scratches and nicks have been reported to magnify the effects by causing local stress concentrations and possibly providing openings in any protective surface oxide layers. In fact, these external factors may cause whiskering in samples that may otherwise be resistant to the phenomenon. For example, tin whiskers have been observed to form on tin finished surfaces that had been exposed to hot oil dip to fuse the tin (a known mitigating process) [Cunningham and Donahue, 1990]. Adding a trace amount of another element (i.e. Pb or Bi) has been shown to reduce the tendency of plating to grow whiskers."
----- One learns to itch where one can scratch.
Yeah, but, now, there's a hell of a lot of other stuff in the landfills.
Not to mention that you don't want this to be happening on your mission-critical server.
Tons more lead free electronics isn't a good thing for a land-fill either. If electronic devices, don't last as long then more will be dumped onto landfills, albeit leadfree.
OK, I'll feed the troll.
Lead is a neurotoxin; children exposed to lead are at risk of developmental delays, reduced IQ, learning disabilities, hearing loss, reduced height and hyperactivity. Levels above 10 mcg/dL (about 100 ppb) are considered to be a health risk to children. Adults are at risk of anemia, nervous system dysfunction, kidney problems, hypertension, decreased fertility, and increased level of miscarriages.
It used to be that the biggest source of environmental lead was automobile exhaust, followed by lead paint. In the late 1970's 14.9 million children in the US had elevated blood lead levels. This figure declined to about 300,000 in the US CDC's 99-00 survey. Banning lead paint in 1978 and leaded gasoline's phase out starting in 1975 removed the sources of exposure.
In adults, levels above 24 mcg/dL are considered elevated and mostly come from workplace exposures, such as demolition, recycling and manufacturing. The number of adults with elevated blood lead levels has also been declining.
The problem with adult exposure is that without proper industrial hygiene, the lead comes home with them, providing a route of exposure to their children. 2-3% of children with elevated blood lead levels are exposed in this way.
Lead in manufacturing is an expense since exposure has to be managed. But, if you don't control it at the source, then you have to try and manage it in the waste stream which is much more expensive and difficult. Given the population density and the emphasis on recycling in the EU, lead exposure through this route is a decreasingly acceptable risk to the population.
P.S. The EU is the largest economy in the world so you tell your directors that you don't need that market anymore!
From the following artical Sealant may be a solution.
t metalwisker.htm
http://physics.about.com/od/condensedmatter/a/sof
Soft Metal Wiskers
from AIP Physics News
AIP Physics News Update #711
Soft-metal whiskers, tiny metallic protrusions that grow like hair from soft metals, are a problem that can cause electronic short circuits leading, in some cases, to the failure of heart pacemakers, avionic relays, and satellites. What to do with the unwanted whiskers---and, in the first place, understanding how they form---is a problem that's been around for fifty years. Now, researchers at Drexel University have arrived at what they think is an explanation for the cause of whiskers and a potential method for alleviating them. Basically, the whiskers form because of reactions between oxygen and the soft metal such as tin or indium. The reaction results in a volume increase that pushes the whiskers out. The whiskers that form do not have to break off in order to be troublesome; sometimes they cause mischief merely by bridging two neighboring electronic pathways that are supposed to be insulated from each other. The Drexel scientists believe that an oxygen-barrier coating on pertinent surfaces should prevent whiskers from developing. (Barsoum et al. Physical Review Letters, 12 November 2004)
Plastic isn't "better", moron. There are applications where plastic is better. If it's going to drop near freezing, copper is much better at resisting occasinoaly freezes. PVC dies inside a year when exposed to sunlight, so if it's outside you better use CPVC or copper. You can also get copper pipe by the roll. It's a tradeoff.
A "leaky capacitor" can be identified by the foreign material seeming stuck to it on the outside of the capacitor, almost like glue, but not sticky. A less extreme sympton of a bad capacitor is a bulge in the sides or top. Really big capacitors (used in high power electrical applications)or older (20+ years,maybe?)capacitors actually have liquid inside, so the sign of leakage might be a stain on the circuit board or chassis.
A capacitor can be identified by the letters "uF" (which stands for "micro-Farads", which capacitors are measured in) after a number. They tend to be either cylendrical, with the leads coming out of the bottom circular base, or "plate shaped", with the leads coming out of the edge.
In automobile batteries? Get real. There's no suitable replacement available.
Contribute to civilization: ari.aynrand.org/donate
It's not really a bath, but you can use conformal coatings to protect PCB's against this and other maladies such as high humidity, salt water spray, etc. I tend to spray down most PCB's I produce myself as they have no solder mask to protect the bare traces otherwise.
Dead board. I've seen a few ibms and no-name boards with these. Most of the affected capacitors seem to have an x on the top of them. They split apart at the seams and ooze all over the board.
I'm not anti-social, I'm anti-idiot.
There are power supplys on silicon now...
Little switch mode chips that drop 240vac to 5vdc with about 800ma.
They don't even get that hot. I don't know how clean they are.
SOC work has already reached this point for many electronics devices - cell phones are about the highest level things that I can think of that can currently be built from a single IC This is also being seen in the processor market, especially on the server side, where chips can be significantly more expensive, with things like memory controllers, network controllers, etc being put on-die on a lot of next generation processors The main problem is that combining all these parts into a single chip raises the cost out of the range that the typical consumer would consider. Since every new process node provides an ~50% reduction in die size, and assuming defect densities are fairly constant, it is conceivable that prices would be low enough to make it to the mainstream market sometime late in the 45nm or early in the 35nm generation (~3-5 years)