Lee Davison wrote:
and about half
the ICs in it are leaving some of their pins behind
when removed from sockets as they've corroded right through :-(
This is not uncommon, I've seen it frequently on control systems that
have all socketed ICs. It seems to occur most in the sockets that have
one sprung leaf pressing against a flat side like this \| but also in
other types where the socket contacts the pin surface. All these systems
had been in constant use for more than twenty years.
Hmmm, that's interesting. This particular system does use the single sprung
leaf socket, but then so does a lot of equipment - it just struck me as odd
how the system was otherwise in very good condition except that the corrosion
seemed to have aimed for some of the chips and destroyed them so completely.
None of the soldered chips have problems, and none of the wider socketed ICs
have problems - but about 80% of the socketed 'narrow' (i.e. 14 or 16 pin)
chips are showing severe signs of corrosion.
Makes me wonder if it's a batch-related thing - perhaps some sort of strange
reaction between the metal making up the 14/16 pin sockets and the chips.
I have noticed that the corroded parts are all in the bottom 50% of the boards
as they sat in the original 3U case; whether that implies a
temperature-related issue or just something to do with whatever the machine
sat on in storage, I don't know.
Hard plated pins, such as TI 74xx series seemed to
suffer most
That's certainly what I seem to be seeing - many of the 74xx chips rotted out,
along with virtually all of the DRAM.
Just one of those mysteries, I suppose. It just surprised me - not only have I
never seen corrosion attack chips quite so much, but I've never seen it quite
so selective in the chips that it attacks either...
cheers
Jules