Saturn 128k Problems

[Steven Hirsch]

Anyone have a schematic for the Saturn 128k ram card? I'm trying to get
RoboCAD going on my 2e and of course the memory expansion it requires doesn't
work :-(.

When I run the RAMTEST128K program it fails immediately with a message that it
cannot write memory after two pokes. The Ram read LED (closest to the rear of
the card) is stuck on. Interestingly, the machine does not respond to a
Ctrl-Reset (!) and must be power-cycled. Never seen that one before.

I suppose I could start changing out LS TTL chips blindly, but wanted to see
if anyone had experience with the card.

Steve

[Steven Hirsch]

Update: Very strange. I took a chance and tried running RoboCAD with the
card installed. It works! I'm guessing the test program is hardware-specific
for Apple ][ and ][+.

Anyone know for sure?

[Hugh Hood]

Steve:

I don't know for _sure_, but I've noticed the very same thing when running
the Saturn 128K test program(s) on my IIe.

I've attributed the failure to my having an accelerator in the test machine,
but it may instead, as you suggest, have been because the program was
designed for and is specific to a ][ or ][+.

I've run the tests with both a genuine Saturn card and also a clone, and
have _never_ passed.

RoboCAD, as you've discovered, still runs fine. Go figure.






Hugh Hood




In article h8GdnbWdpOYxA1bS...@giganews.com, Steven Hirsch at
snhi...@gmail.com wrote on 6/3/12 12:28 PM:

[Steven Hirsch]

On 06/03/2012 07:50 PM, Hugh Hood wrote:
> Steve:
>
> I don't know for _sure_, but I've noticed the very same thing when running
> the Saturn 128K test program(s) on my IIe.
>
> I've attributed the failure to my having an accelerator in the test machine,
> but it may instead, as you suggest, have been because the program was
> designed for and is specific to a ][ or ][+.
>
> I've run the tests with both a genuine Saturn card and also a clone, and
> have _never_ passed.
>
> RoboCAD, as you've discovered, still runs fine. Go figure.

Are you able to interrupt (Ctrl-Reset) or reboot (OA-Ctrl-Reset) the machine
when RoboCAD is active? Neither works for me. It just freezes up with one of
the LEDs on the Saturn card lit.

[Hugh Hood]

Steve:

I booted up RoboCAD on my IIe this morning and tested:

A. Ctrl-Reset -> RoboCAD 'traps' the reset key and presents a message about
'Press Return to Continue'

This keeps you from losing your drawing and keeps (most) folks out of the
monitor for disassembly

B. OA-Ctrl-Reset -> the machine reboots


This is in a machine with a Transwarp (with the Saturn card selection), so
throw that variable into my test results.





Hugh Hood





in article L9Gdnb017NT4lFHS...@giganews.com, Steven Hirsch at
snhi...@gmail.com wrote on 6/3/12 8:03 PM:

[Steven Hirsch]

On 06/04/2012 10:49 AM, Hugh Hood wrote:
> Steve:
>
> I booted up RoboCAD on my IIe this morning and tested:
>
> A. Ctrl-Reset -> RoboCAD 'traps' the reset key and presents a message about
> 'Press Return to Continue'
>
> This keeps you from losing your drawing and keeps (most) folks out of the
> monitor for disassembly
>
> B. OA-Ctrl-Reset -> the machine reboots
>
>
> This is in a machine with a Transwarp (with the Saturn card selection), so
> throw that variable into my test results.

I cannot reproduce that, leading me to believe that I do have a problem with
the Saturn board. I'm going to try swapping out some of the LS TTL logic for
newer parts and see if I can get it to behave.

Would still be great if someone had the schematic.

[Michael Black]

I recall that it used that pair of Motorola ICs designed for controlling
dynamic ram, one is a multiplexer the other is a controller. Large
multipin ICs, and they should jump out when you look at the board. There
should be a datasheet or application note devoted to those ICs, and I
suspect the board can't be much different from that "typical design".
That would make a good start to a schematic, the interface to the bus
can't be too complicated.

Michael

[Steven Hirsch]

Actually, only a single large-package part: MC3242AP. But, that's a good
suggestion. I'll see if I can chase up any data sheets.

[Michael Black]

That's definitely one of the two numbers I'd have expected there. It's
been too long to remember which of the two it is, the multiplexer or the
controller.

Wait, a websearch says it's the multiplexer so it will provide less
information than if it was the controller or both. It just does the work
of 74LS157s or whatever the Apple II uses to multiplex to the address
lines of the dynamic ram. Since the card has to do a lot of work without
the controller, this is less of a start than I'd hoped.

I remember tracing the power supply in the original Macintosh, sort of a
complete schematic but very messy. And then suddenly I realize it looks a
lot like the Apple II power supply, so I dig out the Apple II manual and
yes, the schematic is more ore less the same. With the Apple II power
supply schematic, I then went and recopied what I'd traced in the same
layout, and that worked out the bits I hadn't got.

Which made me realize that sometimes you can make a guess, and then test
the circuitry to see if it matches the guess.

Michael

[Hugh Hood]

Steve:

If you'd like to troubleshoot the Texas 'shade tree' way, I'll be happy to
send you a loaner card.

Just swap parts until you find the culprit, and then order a replacement
chip.

I _think_ every component on the Saturn board is socketed, BTW.

If you're interested, give me a few days to find the loaner board and test
it with Robo. (I scrapped all my II+'s years ago after removing the power
supplies, so I'll bet the Saturn test program will not help me).

I'd better test the card first though, as I purchased a Saturn board a while
back on eBay and it didn't work. You wouldn't want that to pull chips from
that one.

Are you still in New England?





Hugh Hood




in article SrWdnRJ52usQpFPS...@giganews.com, Steven Hirsch at
snhi...@gmail.com wrote on 6/5/12 11:53 AM: