• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

Can this 575 logic board be saved?

Jagmn

Well-known member
Do you want photos of the screen? :)
IMG_6186.jpegIMG_6187.jpeg

I've been (having a go at) identifying the chips to see if there are any likely broken traces between them. I'm using the board, the developer notes and the internet as my points of reference :)
  • U26 should be CLUT/DAC (code-named Antelope), should have:
    • D31-0 that goes to ROM, RAM, VRAM, CPU, MEMCjr and PrimeTime
    • Some pins that go to MEMCjr (presumably for control/timing)
    • Pins that go to the video signals on J13-B
  • U17 seems to be MEMCjr, from what I have found online.
    • A31-0 that go to CPU, ROM (A22-2), PrimeTime, SCSI/53C96 (A7-4), SCC (A2)
    • D11-0 (ROM/RAM/VRAM/Antelope/PrimeTime)
    • VA8-0 (VRAM SIMMS)
    • BDA11-0 (RAM + RAM SIMM)
    • Some Signals that go to Antelope/U26
  • U12 seems to be PrimeTime
I don't have a lot of technical data to go on, though - very open to anyone with schematics or even chip pinouts for these :)
 

jajan547

Well-known member
Do you want photos of the screen? :)
View attachment 41434View attachment 41435

I've been (having a go at) identifying the chips to see if there are any likely broken traces between them. I'm using the board, the developer notes and the internet as my points of reference :)
  • U26 should be CLUT/DAC (code-named Antelope), should have:
    • D31-0 that goes to ROM, RAM, VRAM, CPU, MEMCjr and PrimeTime
    • Some pins that go to MEMCjr (presumably for control/timing)
    • Pins that go to the video signals on J13-B
  • U17 seems to be MEMCjr, from what I have found online.
    • A31-0 that go to CPU, ROM (A22-2), PrimeTime, SCSI/53C96 (A7-4), SCC (A2)
    • D11-0 (ROM/RAM/VRAM/Antelope/PrimeTime)
    • VA8-0 (VRAM SIMMS)
    • BDA11-0 (RAM + RAM SIMM)
    • Some Signals that go to Antelope/U26
  • U12 seems to be PrimeTime
I don't have a lot of technical data to go on, though - very open to anyone with schematics or even chip pinouts for these :)
Ya know strange idea but reflow the ram chips, some times that does it or at least for an LC 475 board and a 550 I had.
 

Jagmn

Well-known member
Ya know strange idea but reflow the ram chips, some times that does it or at least for an LC 475 board and a 550 I had.
That’s an idea. Do you mean the SIMMs? The on-board RAM or both?

And what works best? Hot air? +flux?
 

Jagmn

Well-known member
Unfortunately I've got the chimes of death back after trying a RAM re-flow :( I'll see if I can find anything I've obviously damaged - unless it is the RAM chips themselves I've cooked.
 

jajan547

Well-known member
Unfortunately I've got the chimes of death back after trying a RAM re-flow :( I'll see if I can find anything I've obviously damaged - unless it is the RAM chips themselves I've cooked.
Interesting, any cap goo maybe under them?
 

Jagmn

Well-known member
Well, after an examination with a USB microscope I found that there was a tiny break in the A4 line on pin 14 of U18 - and a discontinuity between the northern DRAM chip's A4 and the southern - funnily enough it is in the same rough latitude as the other broken traces. A solder bridge and a bodge wire has gotten rid of the death chimes and I'm back to the purple/grey screen.

All of the other DRAM chips' pins check out but I haven't verified the data lines yet - I'll see if I can check those next.
 

jajan547

Well-known member
Well, after an examination with a USB microscope I found that there was a tiny break in the A4 line on pin 14 of U18 - and a discontinuity between the northern DRAM chip's A4 and the southern - funnily enough it is in the same rough latitude as the other broken traces. A solder bridge and a bodge wire has gotten rid of the death chimes and I'm back to the purple/grey screen.

All of the other DRAM chips' pins check out but I haven't verified the data lines yet - I'll see if I can check those next.
Do you have any other VRAM?
 

Jagmn

Well-known member
Not yet - I'm hoping some will arrive in the next few days.

RAM data and address lines all seem to check out - from the on-board RAM to the SIMM. I'm going through the VRAM lines now (and learning a bit about how the VRAM setup works in the process).
 

jajan547

Well-known member
Not yet - I'm hoping some will arrive in the next few days.

RAM data and address lines all seem to check out - from the on-board RAM to the SIMM. I'm going through the VRAM lines now (and learning a bit about how the VRAM setup works in the process).
I suspect something might be suspicious near the VRAM slot on the board or it might have a break underneath.
 

Jagmn

Well-known member
Yes; I’m suspicious about the VRAM<->CLUT<->MEMCjr signals. I think I’ve ruled out most of the VRAM SIMM signals (two I need to double check) so I’m going to focus on the CLUT linkage next.

How sensitive is the memory bus, on these boards, to trace length? I’m starting to wonder if my bodge repairs to 4 of the data lines could be causing an issue.
 

jajan547

Well-known member
Yes; I’m suspicious about the VRAM<->CLUT<->MEMCjr signals. I think I’ve ruled out most of the VRAM SIMM signals (two I need to double check) so I’m going to focus on the CLUT linkage next.

How sensitive is the memory bus, on these boards, to trace length? I’m starting to wonder if my bodge repairs to 4 of the data lines could be causing an issue.
Can you supply a picture of the bodges?
 

Jagmn

Well-known member
Yup, here they are:
IMG_6191.jpegIMG_6190.jpeg

The bodged lines are DQ5, DQ4, DQ26 and DQ27 (I believe). I do now have some very fine enamel wire so I could try to repair the traces in-situ, this approach with the vias was just easier for me at my current skill level. :)

I think I've check out most of the CLUT now, with the exception of a few of the bottom pins (same side as the video signals). There are two pins I haven't been able to buzz to anything yet and can just about see their traces going under the chip.
 

Jagmn

Well-known member
So I'm fairly sure I have continuity to every pin on the CLUT now, excepting the bottom-right-most-but-one (see photo). Does anyone know if/what this is supposed to connect to? Thanks
IMG_6192.jpeg
 

Jagmn

Well-known member
Thanks for checking!
The VRAM is due to arrive sometime tomorrow.
I think I'll also see if I can repair the damaged traces 1:1 rather than using the via-to-pin bodges. Looking at the clock and data lines it may make some difference.
 

jajan547

Well-known member
Thanks for checking!
The VRAM is due to arrive sometime tomorrow.
I think I'll also see if I can repair the damaged traces 1:1 rather than using the via-to-pin bodges. Looking at the clock and data lines it may make some difference.
Could be we’ll see when the VRAM arrives.
 

Jagmn

Well-known member
Whelp, the fresh VRAM showed me that it probably isn't (just?) the vram to blame. The symptoms have changed a bit - the screen will now eventually start to flicker with different colours. If I move the mouse cursor to select a menu (blind) you can see obviously correlated changes to the flickering and bands of colour/black on the screen.

IMG_6194.jpeg

I'm pretty sure that I have verified the continuity on the VRAM SIMMs to what look like sensible chips on the board (CLUT & MEMCjr).

The board boots to this state with and without a RAM SIMM in place. I can hit keys and hear audible responses from the OS error chime and even hit the power button followed by return to power the machine off.

Does anyone have suggestions as to what to try next? Another reflow of something and/or removal / cleaning of the VRAM / RAM SIMM sockets?
 
Top