Re: Interesting test tools

On Thu, Apr 29, 2010 at 8:53 AM, Anders Carlsson
<anders.carlsson@sfks.se> wrote:
> I duplicated the five images here:
> http://www.geting.se/viewimage.php?image=238902-P1030084a.jpg
> http://www.geting.se/viewimage.php?image=238903-P1030085a.jpg
> http://www.geting.se/viewimage.php?image=238904-P1030086a.jpg
> http://www.geting.se/viewimage.php?image=238905-P1030087a.jpg
> http://www.geting.se/viewimage.php?image=238906-P1030088a.jpg
>
> Somehow, I'm sure Ruud has seen those before... :-)

I have a couple of those myself.  I've reverse-engineered the PCB
inside to about 95% (I think there are one or two signals I need to
check before I call it complete).

Essentially, these boxes have a local ROM that overrides the
motherboard ROMs (I think the circuit forces the bus to read ROM
addressed at $9xxx rather than $fxxx to capture the normal 6502 reset
vector transaction).  To diagnose your PET, you typically plug
loopback connectors onto the user port and the keyboard connector...

http://zimmers.net/anonftp/pub/cbm/schematics/computers/pet/diagnostics.txt

...and clip the box on the CPU, then power on the machine.

Here's a scan of the manual...

http://www.commodore.ca/manuals/Testing_The_PET/testing_the_pet_computer.htm

(I think I have a copy of the tape mentioned in the manual - I'll
check my PET documentation shelf).

AFAIK, there is no substantial difference between the various "models"
of clip, just different firmware (there might be a 2K vs 4K variety,
though).  I also have a few ROM images for 60Hz machines.  We should
create a new section on zimmers.net to collect these files.
Reproducing these boxes would not be difficult at all - the only
expensive part would be the 40-pin gompers clip.

-ethan

       Message was sent through the cbm-hackers mailing list