Hello everyone, I thought I'd start a thread on my own SE/30 repair experience both to document what I did for others and to solicit some advice on a few stubborn issues I can't resolve.
I picked up a nice-condition SE/30 from Craigslist that was listed as "something's wrong with it". It turned out to be externally in great condition. It is a later unit (manufactured in 91 I believe) with the Astec power supply, original hard drive, and 4 1MB sticks of RAM. When I got it home I powered it on to a muffled, indecipherable sound and classic somasimac screen. There was no hard disk activity on boot.
I took it apart and found the logic board in decent condition. The original PRAM battery was installed, so I removed that along with the ROM and RAM SIMMs. There was clearly leakage from the capacitors and corrosion of solder, particularly in the typical UE8 area. I removed the electrolytic capacitors with two soldering irons, though I now realize the twist method is probably better. In the process of removing the caps, I lifted the + pad on C9. I used some 3M DP460NS epoxy to glue it back down and verified continuity afterwards.
I next cleaned all of the bare capacitor pads by hitting them with a flux pen and then using solder wick in a gentle scrubbing motion. They all came out nice and shiny. I spot-sprayed flux remover where I had been working and used q-tips. Then the board went into a tray of 99% IPA for about 24 hours. I occasionally scrubbed some of the worst areas (e.g. UE8) with a toothbrush while it was submerged. Once it was dry, I soldered on the new caps (tantalum) and also replaced the axials. I plugged it in and powered it up to the exact same: somasimac and a weird muffled sound.
I figured that there must be more corrosion/residue under some of the ICs so I went for a vinegar bath (5% acidity). I popped the board in the tray and let it sit for about an hour, again scrubbing the UE8 area occasionally. I pulled it out, rinsed it thoroughly in tap water followed by a 1 hour soak in 99% IPA to displace any residual water. Pulled it out, let it dry, turned it on and it worked! Still weird sound, but no somasimac, just a flashing disk.
In the process of scrubbing around C13, I managed to completely tear off Q3 including one of its pads. I ended up replacing the pad by epoxying down a piece of flattened wire and soldering it to the exposed trace. I then cleaned the leads on Q3 and resoldered it to the pads.
The system would start up fine from a floppy, but no sound was produced from the system sounds. It also wouldn't recognize any SCSI devices (either the internal HD, a new SCSI2SD, or a PowerBook connected to the external port in SCSI target disk mode. At this point, the board looks like this:
I decided to tackle the SCSI issue first. I used the schematics here to trace al connections out of JI12 (the 53C80 SCSI chip). I also found this thread immensely helpful. As a result, I discovered that two connections (D4 and D5) between JI12 and the VROM (UK6) were broken. I reflowed the solder on JI12 and the connections were still bad. I checked the continuity through the vias for these pins on the bottom of the board and it was good. I decided to wire jumpers on the bottom side of the board. When I added solder to the via for D4, it actually started showing continuity again. Thus, I only ended up putting in a jumper wire in for D5. It looks like this:
I reinstalled the logic board and it worked perfectly! The system booted instantly from the SCSI2SD and sound was perfect. Unfortunately now that the machine was working better, the power supply decided to stop working. It was outputting nothing and may be a simple switch problem. I had a Sony power supply to swap in so I did so and the system was back to 100%
Sad mac/freezing/crashing/somasimac when "warmed up"
So this brings us to the current state. Now that everything seemed to be working well, I started trying to actually use the computer. At this point, the case was on and it was upright as you'd expect. After maybe 10-15 minutes, the mouse would freeze or I'd get a system bomb and have to restart the system. I took the case off and set the system on its side to do some troubleshooting. The voltages coming into the logic board look fine (5.036V, 12.541V, -11.79V, -4.982V) and there is less than 12mV of noise on the 12V and 30mV on the 5V rails. Thinking maybe that D4 SCSI via had gone open when heated, I checked it (and my D5 connection) when hot and they were good. I decided to let the machine be and come back to it the next day.
The next day I turned it on and it again worked fine for maybe 20 minutes. I decided to try installing the case and it continued working fine for another 10 minutes before freezing. Restarting resulted in the sad mac chimes but no actual sad mac. I pulled the case off and restarted again and it made it to the happy mac but froze. Restarted again and it booted successfully.
Figuring it could be heat-related, I started heating different parts of the logic board with a heat gun set to 200 F. This resulted in no freezing/crashes. I checked it out with a thermal camera and noted that UG12 along with the DIP packages including UC6/7, UE6/7, and UG6/7 were the hottest parts of the board. Nothing else really stood out.
I restarted it several times during this and it would usually work fine, but occasionally freeze after boot or with a happy mac. Thinking it might have resolved itself, I put the case back on and it instantly responded with sad mac chimes but no display of the sad mac screen. It did this in the same orientation it worked without the case (e.g. on its side with the analog board down, propped up by the metal frame). Take the case off, and it worked again. I thought there must be something with the case shorting or causing a grounding issue but as soon as I was thinking that it started freezing without the case. I kept restarting it over and over and it would either give the sad mac chimes (never an actual sad mac), freeze during startup, or freeze shortly after startup. Finally, it got even worse and would consistently produce somasimac along with sad mac chimes on startup.
That's where I am now. In the time it took to write this, the machine sat off and open for an 45 minutes or so. I just went down to turn it on and it started up just fine (to a blinking disk). I plugged the SCSI2SD in and it started up with a somasimac. Unplugged the SCSI cable from the logic board and it started up again to a blinking disk. Plugged just the cable back in. Blinking disk. Plugged the SCSI2SD in, full boot up just fine. Ugh, enough for tonight.
Has anyone experienced anything like this? Any thoughts on what my strategy should be going forward?