Re: Blank CMS SCSI Card PCBs

[david@uow.edu.au (David Wilson)]

david@uow.edu.au (David Wilson) writes:
>If the PALs are combinational (no feedback) they will be trivial to reverse
>engineer (they have no flip flops). You just apply each possible input
>to the 8 input pins and record the values on the 8 output pins. Then you
>work out the simplest equation to generate each of the 8 outputs from the
>8 inputs. Then you find the nearest equivalent (eg PALCE16V8 or GAL16V8)
>and program it.

Well, I was a little optimistic. I have been reading my NSC "Interface/
Bipolar LSI/Bipolar Memory/Programmable Logic Databook 1983" and the PAL16L8
is a little more complicated than I remembered.

Pins 1..9,11 are inputs
Pins 12,19 are outputs
Pins 13..18 are outputs or inputs

All the outputs can be individually tristated under control of any input.
If the chips are used simply as 10..16 input blocks with 8..2 outputs
and no tristate it will still be relatively easy to decode. I will try the
following:

Attach the 18 i/o pins to a 6821 PIA (8 on port A, 8 on port B and the last
2 to CA2 and CB2). I should be able to identify the output pins as the inputs
are tristate. Anyway, give me a few days and I will see if I can identify
the first of three.

I can buy GAL16V8 parts here for A$3.90+10%GST so that works out at about US$2.
--
David Wilson  School of IT & CS, Uni of Wollongong, Australia