[comp.sys.ibm.pc.hardware] weitek coprocessor

egcuneo@rodan.acs.syr.edu (Ed Cuneo) (02/20/91)

i have a question regarding a weitek 3167 33 mhz math coprocessor
if the chip looks like this

0000000000000
0000000000000
0000000000000
000       000
000       000
000       000
000       000
000       000
000       000
000#      000
0000000000000
0000000000000
0000000000000

a guy i work with has unfortunately broken off the pin denoted with a #
i'm wondering if i have an $800.00 paper weight or if it might still work
can anyone shed light on if there is a way to use this or fix it?
any chance weitek has a repair procedure?
every little bit of info helps

thanks
ed cuneo

egcuneo@rodan.acs.syr.edu

philhowr@unix.cie.rpi.edu (Bob Philhower) (02/21/91)

In article <1991Feb20.013116.4537@rodan.acs.syr.edu> egcuneo@rodan.acs.syr.edu (Ed Cuneo) writes:
>i have a question regarding a weitek 3167 33 mhz math coprocessor
>a guy i work with has unfortunately broken off the pin denoted with a #
>i'm wondering if i have an $800.00 paper weight or if it might still work
>can anyone shed light on if there is a way to use this or fix it?
>any chance weitek has a repair procedure?
>every little bit of info helps

There is practically no chance of a repair procedure.  The whole
ceramic package is hermetically sealed and it would probably cost far
more than $800 to open it up, replace a pin, prepare the surfaces for
sealing again, and reseal all in a clean-room.  I'd advise your friend
to first look at the pin-out for the chip and see what that pin does.
(There's a chance it will be N/C -- no connection -- in which case you
are home free.)  If not, if it is a power supply pin, you may be able
to get away with the degraded supply.  Do you feel lucky?

Robert Philhower		            (philhowr@unix.cie.rpi.edu)
Rensselaer Center for Integrated Electronics                     
CII 6111 / Rensselaer Polytechnic Institute / Troy, NY  12180 / USA