C070523-001 not just layout silly

[Robson]

Hi,

I have problem with the board in title. Put the high res images and a video on the link below. You should be able to access
https://drive.google.com/drive/folders/ ... sp=sharing

Unfortunately this is one of those repairs that I don't like - that you can tell somebody'd already been messed with the board and it's not just a pull and drop back issue, so hoping to get some quality advice here .

Here are the findings and what's done:
- all onboard memory removed and replaced with a 512k expansion in Low bank mode. MAD and CAS resistors are lifted just in case (this ram is working 100%).
- latches socketed but not present for now - there was a corroded track underneath one
- all other large ICs are in sockets now
- CPU is socketed, when I pull all the address busses down to 0, i get something I want to see, 601E (using TOS 1.0 for now) that is about right , however some of the voltages are too low. From D15 - D0 are:
0.11/3.1/3.1/0.13 - 6 (ish)
0.11/0.11/0.132/1.58 - 0 (ish)
.121/.126/.133/4.7 - 1
4.8/4.9/4.8/1.3 - E

When I do this there's the garbage on the screen but static. Sometimes the pattern changes with 1 or 2 other lines juming, then back to this.

- MMU/Glue sockets seem to be fine
- all data and address SILS are changed to 2.2k, resistance on cpu socket measure ok
- psu is ok (not the only one I tried with)
- main elcos are changed on board
- reflown sockets
- All the 3 HIGH TOS eprom sockets are upside down ( definitely not Atari install) but I didn't find problem here.

IMG_7599.jpg (520.26 KiB) Viewed 4509 times

Sometimes (like now when tested again) the CPU is in HALT with the garbage on screen.

So this is not a simple thing (or maybe it is).

Any ideas are welcomed, thanks in advance.

Robson

[Tomswork]

Was there liquid spilled in this machine with sugar in it the discoloration look familiar

Tom

[Robson]

Was there liquid spilled in this machine with sugar in it the discoloration look familiar

Tom

Which area you think is discoloured? Maybe it's just the lighting? I don't see any unusual or sticky areas.
I always cleaned the area (and more) with IPA where I worked. I don't know anything about the history of the machine.

[Tomswork]

It looks like a thin film around some of the resistors leads and the socket have some of the discoloration. I have see this with honey tea spilled in a 1040 really good conductor when it was dried acted like a low ohm resistor.


Tom

[exxos]

It looks like you have pretty much every databit faulty and one floating signal causing the "Random" bits. Datalines would be that ribbon cable..

[Robson]

I see what you mean...I give it a good ipa bath tomorrow and see if there's any improvement. Thanks for the input.

[exxos]

Is that shifter adapter in the right way around ? Datalines are from pins 3-19.. would have thought the connector would be near them not diagonally opposite.. unless thats just how the board was designed

[Robson]

It looks like you have pretty much every databit faulty and one floating signal causing the "Random" bits. Datalines would be that ribbon cable..

Yes, seems like it. I'll double check the mmu and glue sockets tomorrow as well.

[Robson]

Is that shifter adapter in the right way around ? Datalines are from pins 3-19.. would have thought the connector would be near them not diagonally opposite.. unless thats just how the board was designed

It's fitted correctly, ribbon between the boards is as it should be too. Here is tested with another board.

IMG_7601.JPG (270.29 KiB) Viewed 4461 times

[Robson]

After a bit more cleaning it's constant but all lines wrong:

I1 RAM/data line error 0000000000000000 1111111111111111 000008

A bit more tracing around MMU/Glue I guess.