17 Nov
2005
17 Nov
'05
3 p.m.
So if I have the original code and know what chip it needs to be programmed
on I might be in luck, otherwise there is little hope?
Are PAL's or GALs expensive to get? Do people here have access to
programmers?
----- Original Message -----
From: "Tony Duell" <ard at p850ug1.demon.co.uk>
To: <cctalk at classiccmp.org>
Sent: Thursday, November 17, 2005 5:13 PM
Subject: Re: How exactly do PAL chips get programmed?
>
> As some of you might recall I was looking for drivers for my Tokamac II
FX
> PDS card on my mac IIfx. Well I found the drivers
from the original
company
> founder. I read in Macworld (and had it confirmed
from the company
contact)
> that you need a PAL chip on the IIfx reprogrammed
(or replaced?) because
> Apple screwed up BUS mastering on the IIfx and nobody caught on at the
time
> because those types of cards were rare.
>
> Well I have found another person who has the same card AND the PAL chip
> needed on his unit but is having other problems getting his setup to
work.
>
> What I want to know is are PAL chips somewhat like eprons in getting
> programmed or is there some other process involved? Does anybody know of
a
Well, physically, a true PAL (as opposed to a GAL or similar) is
programmed like a fusible link PROM. There are little 'fuses' on the chip
that you burn out when you program it.
However, there is one important difference between a programmable logic
device (PAL, GAL, CPLD, etc) and a memory (PROM. EPROM, Flash, etc). And
that is that when you read a memory, you are effectively reading out the
programmed pattern directly. When you use a logic device you are not
Therefore, PALs (etc) have a 'programming mode'. You get to it by,
normally, 'overvoltaging' one of the pins. In this mode, the other pins
on the chip select a particular fuse, and enable you to program or read
it. The exact method varies from device to device and manufacturer to
manufacturer, and often was not available without an NDA (although the
methods for the first PALs were published in the data sheets).
This, in general is the only way to read out the contects -- the
programming -- of the chip. In nromal mode, you can attempt to solve it
as an unknown logic circuit, but rememebr there might be feedback terms
or internal D-types (depending on the PAL) which will give sequential as
opposed to combinatiorial behaviour.
And there is some bad news. There is a 'security fuse' in most, if not
all, PALs. If that has been programmed, there is no (official) way to
read out the fusr map. You can only use the device in normal, not
programming, mode. It's a sort of copy-protection.
If you know the device type and manufactuer and have a programmer that
supposts that _exactly_, it won't hurt to try to read it out. But don't
be suprised if you get a file of all 0's or all 1's.
-tony