The real problem is that now, a decade after these boards were made,
many of the chips aren't available any more, and even the ones that are
available aren't available in the quantities needed. Hence, I can't just
toss the chips and change the sockets like should have been done at the
outset.
Does anyone have a workable method of removing solder over gold plating,
that would likely work on PGA chip pins? I would think that some circuit
board manufacturer would have made this mistake at sometime and found a
method to deal with it.
Steve Richfield
http://www.americangoldplating.com/
http://www.appearance-plus.com/gold_plating.htm
I would find a few old worthless chips to practice on
first before i tried it on a non replaceable chip.
BOB
As I understand it, the gold dissolves in the tin and forms a nasty
material which conducts badly. So if you remove the tin I don't think
you're likely to find the gold still there, more likely the nickel or
whatever else they plated under the gold.
I know from bitter experience the problems of tinned pins in gold
sockets and vice/versa. Maybe you could find a socket that doesn't
have gold pins in it, which would make a decent contact with the tinned
pins (after you've removed gross solder from them.) Tinned pins in
tinned sockets work well, one tends to get a gas-tight joint I think
as the tinning deforms under the contact pressure.
I've often removed sockets and soldered chips straight in, though
I appreciate you probably don't wish to do that, especially if you
are swapping them around for debug...
Good luck!
Regards,
Mike.
The first link below has a little info on aluminum oxidation and mentions
the term "fretting corrosion". And I found the second link by searching
google.com with the words: fretting corrosion gold solder
http://www.hsb.com/thelocomotive/Story/FullStory/ST-FS-ALUM.html
A few passing thoughts on your project:
1) Gold (even just plated) is very stable as metal elements go;
tin/lead is fairly active. I would think that diping the pins in a
mild acid for a timed period might work. Or perhaps better, an
electroplating techique where the solder would migrate to a
sacraficial metal it likes better then Au. After you talk with a
chemist about specifics, you would need to try this on some junk chips
first.
2) I know for the older DIP line they make carriers ("headers" ?) on
which one can mount components to make simple custom chips. I doubt
the PGA line has such carriers, but could you use an extra socket as a
carrier? Solder the IC into an extra socket and then plug the combo
into the new socket on the PCB.
3) Would a ZIF socket provide a tighter, more reliable grip, on the
fouled pins?
4) Perhaps further down a road you already don't want to go but....
once you ID good vs bad chips, you could just do away with the socket
altogether and solder the chip right into the PCB. It's approaching
something permanent but it would be reliable....
Greg
The only problem with soldering these chips into PGA sockets with gold
plated pins is the difficulty of getting to the tops of the socket pins
to feed solder. One possibility is to solder bare pins onto the chip, or
strips of pins, working in rows to maintain access. I could do the same
for the messed up sockets, leaving brand spanking new socket pins and
gold plated pins on the chip to plug in, without having to remove
anything. It could be a little tricky getting everything perfectly
straight at the same time, but this sounds like a manageable problem.
Hmmm, this sounds like a really good idea.
Thanks.
Steve Richfield
===============
Unfortunately, I missed the question but, just in case it's relevant:
Gold dissolves readily in molten solder.
Gold in solder causes embrittlement which used to be a common cause of bad
solder joints until the problem was recognised.
Jack
Regatrds
Matt Newton-Stott
Retronix
"Steve" <St...@smart-life.net> wrote in message
news:3D41A6A3...@smart-life.net...