Interesting failure, Seagate ST41600N
Daniel Lang
recently failed. I replaced the failed drive with a spare and
restored from backups. According to the date codes on the IC's,
the failed drive is 10 years old (date codes of 9204, 9206 etc).
I decided to "Widlarize" the failed drive, first dropping it 4
feet onto the floor (1/2" thick carpet with concrete underneath).
To my surprise, about half the chips popped off the PC board!
I was able to pry off more of the chips and other SMT components
using my fingernails.
Close inspection of the failed joints seems to indicate a failure
of the bond between the solder and the copper of the PC board.
The copper pads have a dull reddish to reddish purple hue while
the solder that was attached to the pads has a dull grayish hue.
The bond between the solder and the SMT components seems to be OK.
Does anybody have any idea what caused the joint between the solder
and the copper of the PC board to fail? I have heard of solder
forming a brittle intermetallic compound with gold but I do not
think a gold plating was used in this case. Are there any other
PC board platings/finishes that might have caused the bonds to fail?
Thanks very much!
Daniel Lang
Sir Charles W. Shults III
failure. Common causes are failure to scrub properly or neutralize the etchant
when the card is removed from tee etch tank, or failure to clean properly after
copper plating if through plated holes are on the card.
Oxide or chloride coatings are very poor substrates to solder to, generally.
Cheers!
Chip Shults
My robotics, space and CGI web page - http://home.cfl.rr.com/aichip
John Woodgate
<db...@xtyrvos.caltech.edu> wrote (in <aagd0n$p...@gap.cco.caltech.edu>)
about 'Interesting failure, Seagate ST41600N', on Sun, 28 Apr 2002:
>Does anybody have any idea what caused the joint between the solder
>and the copper of the PC board to fail? I have heard of solder
>forming a brittle intermetallic compound with gold but I do not
>think a gold plating was used in this case. Are there any other
>PC board platings/finishes that might have caused the bonds to fail?
The purple colour is a very thin layer of copper oxide, which has formed
over the years. There is no plating process AFAIK that would explain
your observation.
The most likely explanation that I can think of is that the board was
not pre-tinned, fluxing was inadequate, the solder was eutectic (63%
tin) and the board was jolted just as the solder set. That's a lot of
assumptions, but the phenomenon is weird enough to require them.
--
Regards, John Woodgate, OOO - Own Opinions Only. http://www.jmwa.demon.co.uk
Interested in professional sound reinforcement and distribution? Then go to
http://www.isce.org.uk
PLEASE do NOT copy news posts to me by E-MAIL!
N. Thornton
> about 'Interesting failure, Seagate ST41600N', on Sun, 28 Apr 2002:
> >Does anybody have any idea what caused the joint between the solder
> >and the copper of the PC board to fail? I have heard of solder
> >forming a brittle intermetallic compound with gold but I do not
> >think a gold plating was used in this case. Are there any other
> >PC board platings/finishes that might have caused the bonds to fail?
>
> The purple colour is a very thin layer of copper oxide, which has formed
> over the years. There is no plating process AFAIK that would explain
> your observation.
I'm not so sure. If you forcibly peel soldered items off boards, with
no heat, just using pliers, you get the same appearance when the
solder joint fails at the copper face: purplish copper, and matt grey
solder. I think it has more to do with the physical structure of the
surafce of the torn apart metals, not oxidation.
Bear in mind also that it worked fine for 10 years: I wouldn't expect
that from oxidised dry joints. I've seen equipemtn dry jointed that
worked for only months before oxidation of the joint caused failure.
Regards, NT
John Woodgate
wrote (in <a7076635.0204...@posting.google.com>) about
'Interesting failure, Seagate ST41600N', on Mon, 29 Apr 2002:
>I'm not so sure. If you forcibly peel soldered items off boards, with
>no heat, just using pliers, you get the same appearance when the
>solder joint fails at the copper face: purplish copper, and matt grey
>solder. I think it has more to do with the physical structure of the
>surafce of the torn apart metals, not oxidation.
If the copper is purple, the solder didn't stick to it.
>
>Bear in mind also that it worked fine for 10 years: I wouldn't expect
>that from oxidised dry joints. I've seen equipemtn dry jointed that
>worked for only months before oxidation of the joint caused failure.
Yes. OTOH, I found a rather critical dry joint in a **28 year old** TV
set - the g2 feed of the mixer had never been soldered to the tag of the
holder. [And that 'g2' isn't the top gate of a dual-gate FET!]
Russell Shaw
Try resoldering everything and see if it works again.
Pctchrisgibson
>
LOLOLOL !
Chris
I'm sure you can work out the reference to xollob in the email address is an
anti spammer and not an insult !
Daniel Lang
Thanks everyone for your replies. It appears that the failure
is due to oxidation of the copper, most likely due to improper
cleaning.
Thanks!
Daniel Lang
"Russell Shaw" <rjs...@iprimus.com.au> wrote in message
news:3CB3C18A...@iprimus.com.au...