In 2018 i bought a job lot of Macintosh computers in varying condition. Amongst the lot were two SE/30 computers which had crushed the previous owner’s spirit. They nearly did the same to me.
One machine had a black screen. The other had a stripey “simasimac” display, ie the startup process was not getting to the point where the video ram was initialised.
The previous owner had done some work on the second machine, replacing the surface mount caps with through-hole caps made to fit.
I recapped both boards with surface mount tantalums and thoroughly cleaned the boards. I don’t do a lot of surface mount work, so some of the caps are not as straight as they could be!
There was till no joy with either machine. Some posts on the web suggested that the corrosive juices of the capacitors could penetrate ICs – particularly the row with UE8 which is adjacent to 5 leaky electrolytics. I replaced the entire row on both machines.
After this, The black screen machine progressed to having a classic “simasimac” screen like the other one.
This machine had a broken ROM socket. When i held the ROM firmly in position i was happy to see that the machine would start up. I crafted a kludge to hold the ROM in place, but it really needed a new socket.
The audio on this machine was quiet and crackly. Initially i thought that a transistor had been eaten by the capacitor innards, but i was surprised to find that there was no -12V because a power connector pin had been eaten away. Those capacitors are nasty.
This machine also had a curiosity of a previous repair to the yolk connector. One pin was clearly not making good contact and had overheated. The pin had been bypassed.
The recap and clean had no effect on the other machine: still simasimac. I swapped every component that i could easily swap – no joy.
I did notice that messing with the RAM could cause some changes in behaviour – a curious collection of different chimes at different times – often two chimes. On the scope i could see that the boot behaviour varied. It looked like the RAM test might run for a short time or a long time.
Sometimes it changed just by touching signals with the scope probe.
I spent many hours, coming back several times over about three months, trying to work out what was going on. I Eventually found that it was very sensitive to the probe being on some of the CAS lines. I found that the resistor pack next to the RAM sockets and right next to a leaking capacitor had failed – the internal resistors were not isolated from each other. I replaced the pack and the machine fired up.