exxos's Atari, Amiga & retro forum

stween wrote: 11 Apr 2022 14:22 I probably missed a step in my own description when testing the DMA chips and boot behaviour. I tried three things, third for fun:

• chip C025914-38A: this is the DMA chip I bought from exxos's store: black screen, nothing more. (This chip will go in my H5 eventually, but I had it

This is not a DMA chip. It is a SHIFTER chip. If you really inserted a SHIFTER chip in the DMA socket, then of course no wonder that it didn't work. And I'm afraid there is a possibility that this provoked permanent damage.

ijor wrote: 11 Apr 2022 21:24This is not a DMA chip. It is a SHIFTER chip. If you really inserted a SHIFTER chip in the DMA socket, then of course no wonder that it didn't work. And I'm afraid there is a possibility that this provoked permanent damage.

hahaha, that'd explain a lot :D

consider this a side-effect of working while jet-lagged!

rubber_jonnie wrote: 11 Apr 2022 15:34On the first item, I wouldn't wedge it in. You have no guarantee that you will have a solid connection and may introduce more problems.

I didn't mean reflowing the sockets, I meant this: https://www.exxosforum.co.uk/forum/viewt ... plcc+tweak

Okay:

• I tweaked the pins on the two PLCCs, and tried a little deoxit on them and on the pins on the chips just in case: same symptoms.
• I dropped in spare glue and MMU chips: same symptoms.
• I cleaned the pins on the DMA and the 1772, since they're both socketed: no change.
• I revisited my assumptions and dropped in a drive from another Atari: no change!

Finally I tried the diagnostics cartridge and it fails fast


The two "invasive" fixes I haven't tried are the DMA pull-up, which I'm hesitating on because I might eventually use that DMA chip in my H5, and the reset fix (but the system boots a-okay every time, so I suspect this is fine).

Think I'm getting to the end of the line on obvious fixes though, without aggressively reflowing a lot of joints (which might introduce more problems than it'd solve!)

@stween

The DMA pullup doesn't need to be invasive or even touch the chip. You could install the resistor network on the rear of the motherboard, or you could solder to the base of the IC socket.

There is also one more, the 1172 pullups: https://www.exxosforum.co.uk/forum/viewt ... =17&t=1395

PS - maybe a silly question, but you've tried different drives too?

Anyhow I'm betting if you do either or both of these fixes, it will solve your issue.

Here is an example where I soldered to the socket for the DMA pullups:

Steve wrote: 16 Apr 2022 15:58 @stween

The DMA pullup doesn't need to be invasive or even touch the chip. You could install the resistor network on the rear of the motherboard, or you could solder to the base of the IC socket.

There is also one more, the 1172 pullups: https://www.exxosforum.co.uk/forum/viewt ... =17&t=1395

Okay yeah, good point. Maybe I'll give these a whirl.

PS - maybe a silly question, but you've tried different drives too?

Yeah. I've tried this drive and this disk in another ST (they work!) and I've tried another drive in this ST (symptoms persist)

stween wrote: 16 Apr 2022 15:27 [*] I dropped in spare glue and MMU chips: same symptoms.
[*] I cleaned the pins on the DMA and the 1772, since they're both socketed: no change.
[*] I revisited my assumptions and dropped in a drive from another Atari: no change!
[/list]
...
The two "invasive" fixes I haven't tried are the DMA pull-up, which I'm hesitating on because I might eventually use that DMA chip in my H5, and the reset fix (but the system boots a-okay every time, so I suspect this is fine).

It does seem to be a DMA issue. Adding the pullups on the DMA bus might help. Have you tried a different DMA chip after knowing you tried a Shifter?

Steve wrote: 16 Apr 2022 15:58 There is also one more, the 1172 pullups: https://www.exxosforum.co.uk/forum/viewt ... =17&t=1395

How this will help on a drive that it is known working good on another machine?

ijor wrote: 16 Apr 2022 18:09It does seem to be a DMA issue. Adding the pullups on the DMA bus might help. Have you tried a different DMA chip after knowing you tried a Shifter?

Hah, no, that was just a brainfart; I was convinced I had a spare DMA out, and that that was it. Turns out I don't have any other DMA chips that I can try easily, not without desoldering them out of their own boards. With my current stock of STs, that's my last resort.

stween wrote: 16 Apr 2022 15:27 Finally I tried the diagnostics cartridge and it fails fast

Yeah, last one I had fail that quick, it was a DMA.

stween wrote: 16 Apr 2022 15:27 The two "invasive" fixes I haven't tried are the DMA pull-up, which I'm hesitating on because I might eventually use that DMA chip in my H5, and the reset fix (but the system boots a-okay every time, so I suspect this is fine).

I'm not entirely sure why you see these as invasive? You can easily remove them, or fix to the back of the board, so I wouldn't worry. I wouldn't suspect anything is fine until it's been proved otherwise, potentially in another machine.

This is one reason I keep a bare H4 with everything socketed, it allows me to test ICs in a known good board.

stween wrote: 16 Apr 2022 15:27 Think I'm getting to the end of the line on obvious fixes though, without aggressively reflowing a lot of joints (which might introduce more problems than it'd solve!)

Have you checked continuity on all of the lines to the FDD? By that I mean every hop to every location following the schematic? One that I looked at needed some bodge wiring since whoever replaced the WD1772 and installed a socket broke a trace and didn't fix it.

ijor wrote: 16 Apr 2022 18:09

Steve wrote: 16 Apr 2022 15:58 There is also one more, the 1172 pullups: https://www.exxosforum.co.uk/forum/viewt ... =17&t=1395

How this will help on a drive that it is known working good on another machine?

Well, it helped Exxos with a 'known working' drive which was having write protect issues. Sometimes these things can be related.

rubber_jonnie wrote: 16 Apr 2022 18:45

stween wrote: 16 Apr 2022 15:27
The two "invasive" fixes I haven't tried are the DMA pull-up, which I'm hesitating on because I might eventually use that DMA chip in my H5, and the reset fix (but the system boots a-okay every time, so I suspect this is fine).

I'm not entirely sure why you see these as invasive? You can easily remove them, or fix to the back of the board, so I wouldn't worry. I wouldn't suspect anything is fine until it's been proved otherwise, potentially in another machine.

Yeah fair. I only say "invasive" because although my soldering isn't outright terrible, I also don't totally trust myself to solder/desolder things repeatedly :lol:

Have you checked continuity on all of the lines to the FDD? By that I mean every hop to every location following the schematic? One that I looked at needed some bodge wiring since whoever replaced the WD1772 and installed a socket broke a trace and didn't fix it.

That's a good idea, I'll find some time to do that.

In case you're wondering: this is 100% spare-time hobby, learn-as-I-go stuff, so I don't always intuit the easy things to check until somebody says them. The day job is all software and networks, hardware is an idle interest, which is why this machine has been on my bench for two months now!