Troubleshooting STFM with white screen

[Slartibartfast]

Thanks for the well wishes folks, I wouldn't say I'm over it by any means but a little better today and managed to get into the workshop for an hour to do a little more investigation.

So it looks like the machine is pretty dead. First job was to hook up to my bench PSU to take the possible 5V rail noise out of the equation - still nothing apart from the white screen.

Hooking up to my scope to the motherboard as suggested by badwolf showed that when toggling the Reset button - no activity on anything (AS, ROM CE or DTACK) except the Reset pin on the CPU. So at least I can assume that the Reset circuitry is working.

Using my infrared thermometer, neither the MMU or Glue chips get warm even if the unit has been on a good 10-15 minutes. CPU gets warm but also the RAM & ROMs stay cool.

Biggest problem I have is that I don't know the pinouts of the MMU or Glue chips as I still haven't found the schematic for the C103414 board that I have. I'm also finding it tricky getting info on the MMU & Glue ICs. One has a part number C100109-001 which I think is the Glue and the other is C070714-001 which I think is the MMU. Does anybody have info on the pinouts of these devices (mine are SMD not socketed) and can confirm if I have identified them correctly?

An hour was the best I could manage today - maybe I'll be able to spend a bit more time tomorrow. Thanks for all the help so far.

[exxos]

A lot of the common IC pinouts are in this thread.

https://exxosforum.co.uk/forum/viewtopic.php?f=31&t=149

[Slartibartfast]

That's great information, thanks.

[Slartibartfast]

Last minute update.

I may have stumbled onto something just now as I spent some time looking for broken tracks etc.

I noticed that a couple of the ROM sockets looked like there was corrosion around a couple of the pins.

I removed the ROMs again and this time pin 17 of Hi0 and pin 7 of Lo0 broke. Doh!

However, given this fact I am now wondering if even after reseating the first time I either didn’t notice they were broken or the corrosion is causing a lack of continuity.

So next job will be to replace both the 2 sockets and also the two broken ROM chips.

2 questions following my either my stupidity or bad luck. Firstly, assuming I can read the broken ROM chips in my programmer, can I use a one time programmable 27c256 as a replacement for each (I have a 6 ROM system). Secondly, might this explain my issue or given the results of my earlier test, I’m still not there yet and this is unconnected?

Actually, I have just thought of a 3rd question. Given that I think I have version 1.0 TOS, is it worth moving up to 1.04 and replacing all the ROMs?

[Badwolf]

OK, so if you don't have any AS strobing, not a lot's going to happen. I can't really see how disintegrating ROM sockets could stop AS strobing, so if I were you I'd remove all the ROM chips and concentrate on the CPU first.

You said RESET goes high. What about HALT? They should basically go high at the same time. If HALT is remaining low, that would block AS.

The other thing that might block AS is DTACK being held low somehow. Verify this is high.

Without the ROM chips in place, you should still get a half a dozen cycles or so. the processor will just be getting gibberish on the data lines (0xFFFF) each time.

If all the above checks out, the processor is probably the most likely candiate to swap out.That's a bit of a mission, though. 64 pins and a very fragile motherboard underneath.

In terms of the ROMs, I'm not really familiar with the 6-ROM configuration boards. I think there are maybe jumpers/0-Ohm links on some boards that selects 6 or 2 ROM configuration? Have a read of https://www.exxosforum.co.uk/atari/last/tos/index.htm

BW

[rubber_jonnie]

Last minute update.

I may have stumbled onto something just now as I spent some time looking for broken tracks etc.

I noticed that a couple of the ROM sockets looked like there was corrosion around a couple of the pins.

I removed the ROMs again and this time pin 17 of Hi0 and pin 7 of Lo0 broke. Doh!

However, given this fact I am now wondering if even after reseating the first time I either didn’t notice they were broken or the corrosion is causing a lack of continuity.

So next job will be to replace both the 2 sockets and also the two broken ROM chips.

2 questions following my either my stupidity or bad luck. Firstly, assuming I can read the broken ROM chips in my programmer, can I use a one time programmable 27c256 as a replacement for each (I have a 6 ROM system). Secondly, might this explain my issue or given the results of my earlier test, I’m still not there yet and this is unconnected?

Actually, I have just thought of a 3rd question. Given that I think I have version 1.0 TOS, is it worth moving up to 1.04 and replacing all the ROMs?

OK so if you have TOS 1.0 then I would definitely suggest you go to TOS 1.04. There is no need to read the old ROMs in your programmer in that case. I'd suggest you just make a new 6 chip ROM set, I've added a patched set to this post that (IIRC) contains a patch for IDE booting and also a boot message that shows drives, I've used it in my own machines and it's fine. Remove the .txt after downloading and unzip.

I would say that you need some erasable ROMs to test with first as if you need to byteswap the images and you flash a set of OTP ROMs, you won't be able to correct them and all they will be good for is the bin.

A white screen is often down to ROM issues. Pin 7 is A3 and 17 is D5 so that might be the cause of your white screen. I usually repair what I can see first and then continue investigating if it doesn't work.

[Slartibartfast]

Thanks for the ROM images. I only have one time programmable PROMs to hand at the moment but I think I can get hold of some UV erasable ones quite easily.

Diagnostic has had to go on hold again as just when I thought I was getting over Covid, it has returned and sent me packing off to bed again.

Hopefully I’ll get the chance to do some more work in a couple of days.

[rubber_jonnie]

Thanks for the ROM images. I only have one time programmable PROMs to hand at the moment but I think I can get hold of some UV erasable ones quite easily.

Diagnostic has had to go on hold again as just when I thought I was getting over Covid, it has returned and sent me packing off to bed again.

Hopefully I’ll get the chance to do some more work in a couple of days.

Hope you're feeling better soon.

[Slartibartfast]

Ok, we are winning....

Felt a bit better this morning so fired up the desoldering station and removed the 2 ROM sockets with the broken pins within them. As you can see from the picture, one of the broken pins (7) had corrosion which was also on the board. That's probably why it broke in the first place.

Fitted 2 new sockets and managed to read the broken ROMs enough to be able to program 2 replacement chips and we now have a desktop!

However, although the FD LED comes on for a few seconds, there is no life in the drive and it's not being 'seen' by the computer.

It's an Epson SMD-380. I've taken the lid off and there is nothing obviously wrong with it. I do have a spare Sony drive (MPF520). Would this be a direct swap - even if just for testing?

[exxos]

If the drive ID is set correctly then it should work. If its a 1.44 drive then you have to put the cable in upside down.