Slashdot Mirror
Zinc Whiskers Cripple Colorado's Computers
Mr. Christmas Lights writes "While zinc whiskers, small metallic fibers which grow on surfaces that have been electroplated with zinc, aren't a problem for Christmas lights, they can cause serious problems for computers. The Denver Post reports how they caused computer outages for the last three weeks in the Colorado secretary of state's office. This basically halted business and elections document filings. Zinc whiskers are becoming more of a problem as computers electronics get smaller. NASA has a good reference site which includes a interesting PDF summary paper complete with pictures. /.'ers with computer rooms might want to check this out."
They are only blocking slashdot referrers.
IE-ers can drag the link to their address bar.
If you could be told what you can see or read, then it follows that you could be told what to say or think - BoC
The metal frames for the raised floor are where the zinc whiskers are coming from. They get sucked into the power supplies and short them out.
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain
Here's a yummy little google cache for y'all.
(-(friend^2))^(1/2)
Incoming mod-bombing for having a different viewpoint, 2 o'clock! Heads up!
I work in a large midwest hospital, and we've got a constant issue with Zinc whiskers in our network jacks in operating rooms. Supposedly the origin is the cleaning solutions that they use for the floor.
But I could google on and on. Hey, poster (NigritudeUltramarine). Care to explain your nickname? Was it intended to offend? I suspect it's just a trash
It's called oxidation.
bash: rtfm: command not found
I thought almost all quality electronic devices have a conformal coating (non-conducting polymer) applied to PCBs?
Generally, no. Conformal coating is normally only used on circuit boards which will be exposed to extreamly harsh conditions. Unless the designers expect the circuit board to come into contact with liquids or corrosive gasses during typical use, the circuit board will not be conformally coated.
Come test your mettle in the world of Alter Aeon!
Zinc whiskers are quite real, and they can pose a hazard. The reason you don't see data centers around the world having this problem has many aspects, some of which I can think of include:
-Too new. It takes years for these whiskers to grow to any length (1mm/year), and it may take years before it even starts to happen.
-Only happens to electroplated surfaces. Dip galvanized objects (Like electrical boxes and bolts and such) don't seem prone to this effect.
-Newer machines are more vulnerable with their more delicate circuits and smaller, tighter tracings and pins. A data center with older machines might be all but immune to it simply because the equipment is old and robust enough.
High taffic areas they are likely to get stomped on/eroded away long before they pose a problem. So you would have to have a situation where you have a spot like under a table, where you have objects electroplated with zinc sitting undisturbed for a long time, then get disturbed. Then you would have to be unlucky enough to disturb them and get it into the air.
Ironically, there are wood based floors used in data ceneters with steel reinforcing on the back of the tiles that are zinc-electroplated (thus being essentially undisturbed for years). So if a fairly old data center, that happens to have the right (wrong?) type of flooring, undergoes an upgrade or reorganization... well that might do it.
Now if these guys ARE just using it as an excuse, that's another story. But that doesn't make the problem any less real.
=Smidge=
and a narative
Access Floors
I know I speak against my own name, here. But:
Living by the sea, especially with the high humidity that comes with it...and the salt...can give your more than just a few zinc whiskers on the PC board.
Salt oxidation--depending on how long the windows are open--can really eat a PC in two years or less; never mind the quality of the MB.
Sea salt is hydrophilic. If it accumulates on something, episodes of high humidity will attract moisture from the air, and add the basis for typical corrosive effects. I have had containers with dry sea salt, which have pulled moisture out of the air on their own.
(Most acids need water...so does salt to release it's own ions, which can have a corrsive effect similar to an acid on metallic equipment...usually involving the non-metal in the salt. Sea salt has lots of chlorine, a very strong oxidizing agent.)
We already have comp salvage yards piled higher than Iraq's (nonexistant) WMD caches.
Two things are infinite: the universe and human stupidity, though I'm not yet sure about the universe. - A Einstein
You can conformal-coat boards yourself, using Fine-L-Kote spray. We use this stuff on the Overbot.
It's a flammable, toxic chemical mixture until it dries; you need gloves, goggles, a respirator mask, and proper flammable liquid storage. Cover connectors with masking tape before spraying. It's a clear coat, but glows in UV, so you can check for missed spots.
I like Humiseal, myself.
"I'd rather be a lightning rod than a seismometer." -Ken Kesey
and therein lies the problem: you just asserted that this is an item of faith for you, not reason; facts be damned, you cannot be convinced.
Never mind that there are several companies who do raised subfloors who've been addressing this problem for some time. They're all peddling snake oil, and NASA is helping them do it.
Never mind that Zn whiskers grow slowly, Zn-electroplated subfloor panels in data centers aren't that old, and PC board density has been increasing. Or that they only occur on electroplated zinc, and only grow long enough to be problematic in very low traffic areas. Nope, must be bogus.
Never mind that hospitals are affected as well and take this seriously. Or that the condition is well known among electroplaters and materials engineers, and was discussed at least ten years ago in the literature. Or that it's been involved in at least one product liability case. Or that Bell Labs has known about it for over fifty years (since 1948).
And never mind you could have found everything I mentioned above within the first 30 google results for "Zinc Whiskers". Nope. It must all be a myth, because there's no such thing as newly discovered age-related problems.
(Oh, and I hear that automobile corrosion is a myth too ... I went to the new car lot and looked around and didn't see any, so it must not exist)
It's also worth noting that, at the finishing.com link I gave above, you'll find Cisco and well-known companies have known about this problem since at least 1996. Or perhaps finishing.com is involved in the conspiracy, or Cisco power engineers are idiots.
Most of the zinc-plated ducts I've seen are galvanized, not electroplated. It's only a problem with electroplated zinc (not alloy), galvanized doesn't have the same stresses inducing whisker growth.
These whiskers take time to grow and are fragile, you would need to have a situation where the equipment is undisturbed for a period of time, this allows the whiskers to grow, then when there is a disturbance (ie construction vibration)they break off and can cause shorts. I don't think most people leave their computer undisturbed for long periods of time
dendrite growth has a different mechanism can also have similar consequences. This happens in things like old electrical systems, where you can have moisture with high current.
D6 63 0D 70 89 81 BB 8E 7B 7C 5F 5D 54 EA AB 73
No, this is real. Here's why: a tarnished connector can become a non-linear junction. Tarnish such as a thin film of a sulfide, in contact with a 'pure' metal, can be a rectifier. If you put one on a curve tracer, you could actually see this. So there really is merit in a gold, i.e., non-corrodable, connector.
You're thinking of the solder mask-- and the principal purpose of the solder mask is to-- you guessed it-- mask where solder goes. It's not a very good insulator and also there are large areas where a particle of conductive dust would be bad (on component leads, on various places which are left unmasked, probably wouldn't be good in inductors for CPU core power supply, or in the power supply itself..)
Zinc is not attracted by magnets. The only ferromagnetic materials are iron, nickel and cobald.
Magnets will have no effect whatsoever.
What person will donate an airborne act of love?
The EXACT mechanism that creates the metal (tin, zinc, etc.) whisker shaped crystals is unknown but LOTS is known:
What are the Commonly Reported Characteristics of Tin Whiskers?
The vast disparity in the observations reported by different experimenters is evidence of the complications associated with understanding and controlling tin whiskers. The following list is intended to provide a very basic overview of some of the observed characteristics of tin whiskers.
1.
Shapes: Whiskers may be straight, kinked, hooked or forked. Their outer surfaces are often grooved. Some growths may form as nodules or pyramidal structures.
2.
Incubation (Dormancy) Period: Experimenters report the incubation period may range from days to years. This attribute of whisker growth is particularly concerning because meaningful experiments to determine the propensity for a particular process to form whiskers may need to span very long periods of time.
3.
Growth Rate: Growth rates from 0.03 to 0.9 mm/yr have been reported. Growth is highly variable and is likely to be determined by a complex relationship of factors including plating chemistry, plating thickness, substrate materials, grain structure and environmental storage conditions.
4.
Whisker Length: Whiskers as long as a few millimeters are not uncommon with some experimenters observing whiskers as long as10 mm (400 mils) in length.
5.
Whisker Diameter: Typical diameters are a few microns with some reports as large as 10 um
6.
Environmental Factors: There is a great deal of contradictory information regarding environmental factors that might affect whisker formation. Several organizations are attempting to devise accelerated test methods to determine a particular plating process's propensity to form tin whiskers. However, to date, there are no accepted test methods for evaluating whisker propensity. Indeed, much of the experimental data compiled to date has produced somewhat contradictory findings regarding which factors accelerate (or retard) whisker growth.
Temperature: Some experimenters report that ambient temperatures of approximately 50C are optimal for whisker formation, while others observe that room temperatures (22C to 25C) grow whiskers faster. Reportedly, whisker growth ceases at temperatures above 150C
Pressure: Whiskers will grow in vacuum as well as earth based atmospheric pressure.
Moisture: Some observe that whiskers form more readily in high humidity (85% RH) whereas others report moisture is not a contributing factor
Thermal Cycling: Some experimenters report that thermal cycling increases the growth rate of whiskers, but others report no effect due to thermal cycling.
Electric Field: Whiskers grow spontaneously without requiring an applied electric field to encourage their growth. Some recent observations of tin whisker induced field problems in the commercial sector seem to suggest that an electric field could stimulate whisker growth, but more analysis is required to confirm these effects (if any). GSFC has demonstrated that whiskers can bend due to the forces of electrostatic attraction thus increasing the likelihood of tin whisker shorts (ref. to GSFC experiment #4).
7.
Whisker Prone Processes: There is tremendous debate in the industry regarding which plating processes are prone to whisker formation. Most of the literature agrees that "pure tin" electroplated surfaces (especially those that employ brighteners in the plating process) are the most susceptible to whisker formation. There are also reports that tin-lead plating can also grow whiskers; however, such whiskers are generally reported to be less than 50um long.
And before anyone points out that they're the same thing, they're not. Galvanizing involves dipping steel into molten zinc. Despite the name, it has nothing to do with galvanic action, electroplating or Mr Galvani. Galvanized panels don't suffer from zinc whiskers, anyway.
When I am king, you will be first against the wall.