IIsi power supply won't turn off, I'm deep in schematic land

[Addicted]

I wish I had found this thread a few days ago! I don't have the electronics background (or equipment) to delve very deeply, so there's a ton of info here I can use.

Started mid-week to repair a dead IIsi Sony power supply, which I hope to use as a bench PSU for IIsi/cx/ci logic boards under repair.

There was a little white gritty patina on the outside of the PSU case and definite cap ooze within. This is another of my "why not try" projects. If it comes up, I'll buff away the barnacles and primer the case, just for bonus points. (I do wish I had a test box like @dougg3 built! I fear I will smoke a worthy logic board trying to test the PSU.)

Found one broken trace on the vertical card. Waiting for recaps to arrive. I wish I had a replacement 51977 ; I can't get a little ooze that's left out from under there. I'll try to remove-clean-reseat the existing one.

 

[dougg3]

Oh nice! I have a theory that one of the M51977FP chips I bought from China is probably what blew my power supply up to begin with, so my recommendation would definitely be to stick with your original chip!

I don't have any further progress updates on fixing my power supply that has blown up a few times. I just ran out of steam and I'm kind of stuck until I get some motivation to try to build a riser card tester that can find faults in them without blowing up my power supply again.

 

[Addicted]

I wound up with two Sony IIsi PSUs. The second one had a vertical card in slightly better shape than the one I pictured, but a great deal more ooze on the main board.

I recapped both PSUs/verticals and cleaned as much ooze as I could reach out of both. I did not replace the X2/Y2 safety caps on these.

Now, the one pictured above seems to be working.

The PSU powers on with the power switch and with ADB, and powers off with the switch. The host, a recapped and untested IIsi, power-on chimes then immediately sad chimes. ROM serial diags reveal it an MMU failiure code. (details*) So I can't test software power-off yet, although faking that is probably described in this topic. I am seeing +12 and +5 on the HDD power jack, but want to check the 7905 too. I haven't taked leads on to capture the waveforms for /PFW etal., yet.

The other PSU needs more work - it appears to be acting like the ones that caused this topic, so I'll come back to it in a while. Gotta do some other things first.

Aside: I found a Rustoleum spray paint product I can recommend. It's transparent with a satin sheen. I used a rust remover and fine sandpaper to get the white 'barnacles' (ZnO?) and rust off of the ventilated PSU case (as pictured above, bottom center). The metal came out smooth and clean, though a little mottled. This clear spray gives it a rustproof sealing coat and isn't noticeable. It still looks like vintage Mac.

---

(details*)

The host IIsi's MMU failure is likely for real, for if I put my thumb on the `030, it is almost-uncomfortably-hot to the touch. I fear this `030 is literally toast. I checked, the MPU oscillator is the stock 40MHz with no signs of replacement, so this is not a failed-and-tossed overclock. Will get out the microscope and thoroughly review the `030 pads. Some other day.