2nd attempt to send (message bounced earlier)...
Did somebody say "whiskers"?  Zinc whiskers?

Those who know me from this forum probably realize that the only time I cast in
my two cents is when the topic of "whiskers" rears its ugly head. I help to
maintain the NASA Tin and Other (aka ZINC) Whisker WWW Site at NASA Goddard.

http://nepp.nasa.gov/whisker/other_whisker (for lots about ZINC whiskers)

Regarding zinc and it's potential to grow unwanted whiskers, we have seen it
very recently in a GROUND (non-flight) application.  We believe this particular
application of zinc plating (with possibility to form whiskers) is potentially
so pervasive that we have issued a formal Problem Advisory via the US Govt and
Industry Data Exchange Program (GIDEP) in August 2003. I suspect that many of
you and your organizations may also be exposed to this problem without
knowing it.  I am not authorized to distribute GIDEP documents to a forum such
as TechNET (i.e., no international distribution allowed... GIDEP participants
only,
etc.), but for an overview of the specific situation we have observed, you may
view a presentation I prepared in April 2003:

http://nepp.nasa.gov/whisker/reference/tech_papers/Brusse2003-Zinc-Whisker-Awareness.pdf

The problem is one of zinc-coated structures used in the manufacture of raised
"access" floor structures (commonly found in computer data centers).  These
tiles and the support structures they rest upon (pedestals and rails) are
frequently made with steel that is zinc-electroplated for corrosion protection.
In
the case of the tiles, it is the UNDERSURFACE (the side you DON'T) walk on that
is the potential source of problems.  Over periods that may take months if not
years, the zinc-plated surfaces may grow zinc whiskers. On one tile I retrieved
from a data center, we estimate 10 MILLION whiskers growing on a 4 square foot
area of a single tile.  Many of the whiskers in excess of 2 millimeters. While
they live below floor they pose essentially no
problem.  But when they get dislodged from the tile or rail/pedestal... they
become an airborne NIGHTMARE.  The subfloor space of many raised floor
structures is commonly pressurized for cooling of the electronics in the room.
Tiles with holes or floor vents are strategically located about the center to
distribute the cool air.  The problem is that when someone lifts a tile (they
are called "ACCESS" tiles after all) and slides it across the floor or drops it
back in place or pulls electrical cable under floor, the Zn whiskers become
dislodged and away they go merrily blowing about the room.  The fans on the
equipment in the rooms are quite efficient at pulling in air to keep them cool.
 Along with the air comes a fog of conductive zinc whisker debris... Result...
intermittent glitches... catastrophic shorts... and as Joyce Koo remarked...the
bullet that killed the
beast is so difficult to see due to small size or gets melted during the
event... or gets blown away from the crime scene... that detection of this
problem is HARD!!!

This is not a "rare" situation as I have seen it myself in at least three
different data centers in the last 8 months and in researching the matter I
have
talked to many (at least 7) others who have faced this beast as well.

In my review of this problem, I have spoken with more than one manufacturer of
electronic
equipment who have told me that they too have dealt with zinc whiskers growing
from the Zn plated equipment racks or from the Zn-plated chassis housing their
system.  The fact that this type of problem isn't PUBLICLY reported is often a
matter of each
company's choice to go public or to quietly address the problem internally.  I
think I am echoing Joyce Koo's comments here.

Dave Hillman's suggestion to work closely with your supplier is a good idea,
but
is frequently impractical.  The company I buy a bolt from may not be the one
doing the plating.  The company I buy the relay from may be using Zn-plated
structures INSIDE of the relay that they buy from somewhere else (by the way Zn
whiskers have grown from inside
of such devices leading to field failures).  Dave also mentioned that Zn plating
can be rendered immune to Zn whiskers by "annealing".  I am very interested to
learn more about that this practice and to find out if such practices are
indeed industry standard approaches.  Also as Joyce remarked, I know of no
accepted test methodology by which one could evaluate a particular product's
propensity to whisker in 1 year, 2 years, 10 years, etc.  The floors in many of
our facilities have been in use for 30 years or more in some cases.

Just to cloud the whisker issue a bit further, I've recently seen
pictures of Zn whiskers from Hot Dip Galvanized Steel (albeit rarely).

Just this week during an audit of a manufacturer of electronic components, they
told me they had recently received a survey from a major client asking them
about whether their products
may be prone to "tin whiskers" through the vendor's choice of pure tin
finishes.  The vendor, clueless about what a tin whisker is, responded... "Not
a problem for us" because they interpreted the question to mean are they using
pure tin "solder" to make solder joints.  Of course the vendor does NOT use pure
tin SOLDERs but they do use pure tin-plated terminals and other mechanical
items in the construction of their product.  Buyer Beware!

Cheers,

Jay
--
Jay Brusse
QSS Group, Inc at NASA Goddard
301-286-2019

---------------------------------------------------
Technet Mail List provided as a service by IPC using LISTSERV 1.8e
To unsubscribe, send a message to [log in to unmask] with following text in
the BODY (NOT the subject field): SIGNOFF Technet
To temporarily halt or (re-start) delivery of Technet send e-mail to [log in to unmask]: SET Technet NOMAIL or (MAIL)
To receive ONE mailing per day of all the posts: send e-mail to [log in to unmask]: SET Technet Digest
Search the archives of previous posts at: http://listserv.ipc.org/archives
Please visit IPC web site http://www.ipc.org/contentpage.asp?Pageid=4.3.16 for additional information, or contact Keach Sasamori at [log in to unmask] or 847-509-9700 ext.5315
-----------------------------------------------------