End of battery-bombed SE/30 restoration?

Snipey

Member
Around April’24 I started a project to restore a battery-bombed SE/30. Having established that all other part of the machine are functional within specs, I obviously focused on the logic board.

I have thoroughly cleaned the logic board (immersed in vinegar, IPA, soapy water, rinsed with water). I replaced several components that came off around the battery area. I recapped the PCB. I measured continuity of the data- and address lines and connectivity between the components and the PCB and finally I spent previous days to apply bodge-wires as necessary.

So this weekend was the big moment: after one more verification of the continuity of the data- and address lines to all involved components; the first ‘live’ test of the logic board!

The good news: when I originally got the machine, it remained completely dark. Only the fan was spinning. Now, I get the infamous horizontal stripes on the screen.
Yet no sound at all; no bong, no chimes of death. I tested this in configurations with and without the presence of ROM, RAM and Video-ROM; all the same result.

Before to take the oscilloscope, I decided once more to test the continuities and doing this I discovered something that I should have tested for much earlier in the process:
  • Shortage between A12 and A13; by removing one of my bodge-wires I was able to isolate the shortage around the CPU and PDC. On the ‘top-side’ of the board the shortage is not present.
  • Shortage between A15 and A16; this proved to be a shortage in the ROM. I guess I may overcome this by obtaining a replacement ROM.
But as for the shortage A12-A13; I seriously doubt now to make any further effort in this project or to accept defeat...

I could de-solder the PDC and/or CPU (non-socketed); perhaps battery residue between the board and the PDC/CPU is the cause the shortage at pin-level.
Even though I have a de-soldering-iron, this seems to be a big job without any guarantee of success; imho. it is quite (more?) likely that the short may be in one of the PCB-layers.

So here are some questions for the experienced Mac-restorers amongst you:
  • Do you agree that mentioned shortages in the address lines is (or could very well be) the cause of the symptoms of the test?
  • Do you agree that there should not be continuity between A15-A16 on the ROM?
  • How likely is it that battery-residue between the board and the PDC and/or CPU causes the shortage A12-A13?
  • Do you see alternative approaches to remedy the shortage?
Any other, generic thoughts on how to proceed from this point on?
 

Kouzui

Well-known member
Do you have some pictures you can share of the problem areas?

Generally, battery acid will eat the traces and break continuity instead of conducting electricity (though capacitor electrolyte can and does do both).
As you said, a possibility is that since the SE/30 board is 6 layers, something may have happened between those layers, which would obviously cause some issues.

I don't have my SE/30 apart to test if the lines you're seeing continuity between should indeed have that, unfortunately.
 

ironborn65

Well-known member
I have a similar issue with a bombed SE/30, in my case the short is across A8 and A9, and I was not able to find where it is.
I guess the acid went down inside the PCB in a via and it shorted two traces. The design of the traces of the original PCB would help to guess where the two traces are close and possibly intercept and fix the issue. There are schematics, but I could not find the PCB design. The one from Bolle is different.

If you have a short in ROM address the Mac is unable to run the self-test code in ROM at boot, this is why you can not hear any chime.

Like you, I'm tempted to go for the Reloaded project.
 

Snipey

Member
Do you have some pictures you can share of the problem areas?

Generally, battery acid will eat the traces and break continuity instead of conducting electricity (though capacitor electrolyte can and does do both).
...
Thanks Kouzi, I was not aware of this difference between battery acid and capacitor electrolyte.

Sincy my board did hardly (not?) suffer from leaking capacitors, this unfortunately would confirm that removing PCD and/or CPU for further cleaning will most likely not provide a solution.

I have added the pictures here.
 

Attachments

  • 20250106_161402.jpg
    20250106_161402.jpg
    390.1 KB · Views: 23
  • 20250106_161426.jpg
    20250106_161426.jpg
    372.1 KB · Views: 23

Snipey

Member
I have a similar issue with a bombed SE/30, in my case the short is across A8 and A9, and I was not able to find where it is.
I guess the acid went down inside the PCB in a via and it shorted two traces. The design of the traces of the original PCB would help to guess where the two traces are close and possibly intercept and fix the issue. There are schematics, but I could not find the PCB design. The one from Bolle is different.

If you have a short in ROM address the Mac is unable to run the self-test code in ROM at boot, this is why you can not hear any chime.

Like you, I'm tempted to go for the Reloaded project.
Yes it seems we have a similar issue,

Like you, and after extensive searches, I have not been able to find the trace layouts of the original PCB in the public domain, set aside if this would really contribute to remedy this particular problem.

I imagine that any short in adress-lines will stop the Mac from running code at all...
 

herd

Well-known member
A common way to find shorts (like a shorted capacitor for example) is to connect a bench supply, slowly turn up the current and look for a hot spot.
 

Snipey

Member
A.o. based on your inputs I decided as a last resort to de-solder the CPU and see if the short can be isolated to the CPU. I am not done desoldering the CPU but I have managed to establish that the short is not in the CPU so it must be in the PCB itself.

As I have another SE/30 project which looks much more promising, I'll put this project on hold. Having now a defective ROM and at least 1 short in the PCB, I may end up using this board as a donor board...
 
Top