My IIsi Power Supply is dying. Possible to fix?

[3lectr1c]

I bought a IIsi recapped at the VCF Swap Meet back in June. It worked fine when I got it, but a while later while I was playing some Tetris on it, it shut off. Now, it will start if I unplug the hard drive (known good), but can't get itself started with it installed. It will start, chime, start to spin up the drive, then die. With repeated attempts, it will eventually stay on for a shorter and shorter amount of time, and eventually not start at all until I let it rest.

Clearly, whichever of the voltage rails that runs the HDD is weak.

Are there any parts in the PSU that could be known to cause these sort of issues? I know an ATX conversion would be more reliable, but I'd honestly rather run it off the original if it would be a cheap fix.

I've 100% confirmed that the PSU has been recapped, I opened it up. Looked decent inside but I didn't take too close of a look.

 

[joshc]

Have you looked at the schematics yet?

From memory there is a thermistor on the high voltage side, could be related to what you are seeing. I am trying to think of components that are heat dependent to operate correctly, or ones that will be affected adversely/drifting out of spec as the power supply warms up.

Do you have a component tester so you can remove components and test them?

See the steps here that I had to take to get my IIsi PSU working. Not the same behaviour as yours, so the repair needed will be different but you could use the same methodology to check what's going wrong.

Troubleshooting a IIsi Sony APS-06 Power supply - blown capacitor caused by bad diodes in series with C105

This post is a recreation of a previous troubleshooting thread (lost in the recent forum crash) where I had some problems identifying the cause of a blown capacitor in a recapped Sony APS-06 from a Macintosh IIsi. The exact PSU in question: I used the correct value capacitor and it was...

68kmla.org

 

[3lectr1c]

Thanks for the reply!

I was unaware that schematics existed for this supply. That should be helpful, as long as I can get better at reading those. I’ve never been great at that.

I was thinking that this problem was more load-dependent rather than heat-dependent. I figured that would explain why it starts without the hard drive. I'll certainly check all possibilities though.

To my memory, the time where it died first was the first time I had ran it for an extended period of time, since then it hasn't been able to start off the hard drive properly except for maybe one other time.
The HDD isn't shorted out or anything either by the way, it still works fine on a different system. It's a circa 1996 IBM SCSI drive that I pulled from my 7100.

I do have a proper multimeter, does that work for testing these components? I haven't had to test diodes and the like before with it, just continuity and voltage mostly. I can post a photo of it later. It's made by Klein I believe.

Your thread looks like a good place to start. I'll look into testing those diodes. Another thing I need to do though is just get the supply open and have a closer look at it. I briefly opened it up when I first got it, mostly to verify that it had actually indeed been recapped including the vertical soft power board, and it had been. Could be that it needs an extra good cleaning and a reflow though. I didn't see the bottom of the board, so I don't know the extent of damage that the original ELNA caps caused.

 

[3lectr1c]

I did take these three photos when I opened it earlier.

 

[blturner]

I've also got a flaky IIsi PSU I was considering re-capping, but this looks like it might be a tempting alternative:

https://twitter.com/i/web/status/1703878219244396815

 

[3lectr1c]

Ooh! Specially made. Let me know if that guy starts selling them.

 

[XKSTEVE]

Maybe add a link to the resources section to the psu replacements when they are available to make it eas to find.

 

[3lectr1c]

I finally got the PSU board out of the case now, and yeah, it's had work done...
Nothing looks too bad though. Vertical board looks REALLY iffy, but I don't think an issue there would cause my symptoms. Anything obvious stand out?
I don't think the two HV caps have been changed. Could those be bad?
Pics incoming.
Looks like it might be able to use another clean to the top side.

 

[3lectr1c]

no solder joints look cracked.

 

[carbide]

When it is failing to start up at all, is the PSU making a quiet, rapid clicking noise? That's the failure mode of mine at the moment.

It was recapped before I picked it up at VCF Midwest this past year, and it worked several times after I first got it, but now it only does that. I'm planning to go through and verify polarities, clean up some of the solder on the daughterboard, and give it a general once over when I have a little time.

I'm also watching CayMac's feed for any announcement of a release for their replacement. Ideally I'd like to have the original one working, but I'm not going to say no to a modern replacement either.

 

[3lectr1c]

No ticking I don't think. It does make a high pitched whine when I get it to stay running though. The IIsi only stays on if I disconnect the hard drive, it seems to be an issue that it can't handle load. And then with it disconnected is when I get the whine, it's clearly unhappy even then. I just wonder why.

 

[JeffC]

I had a IIsi PSU I recapped, it had a fair bit of corrosion that I left alone when I did the recap. It worked for a while, then just shut off mid-use and wouldn't stay running after that. I took it back apart, scraped a bunch of corrosion and added solder to the bare copper. It looks a lot like yours now. I also re-flowed all solder joints on the board. I know you said no joints look cracked, but I would strongly recommend re-flowing them anyways, you can't always see cracks with the naked eye. Those two things fixed my PSU and it has been working well since then. Re-flowing the joints may well not fix your problem, but at least it would eliminate one potential point of failure.

 

[3lectr1c]

Will go try that when I get the chance, can't hurt.

Vertical board looks nasty, I'll bet there's problems there, but probably wouldn't cause the symptoms I'm having. Soft power did only work sometimes when the PSU was working though, so I'll bet there are still issues there!

I wish I knew how to troubleshoot switch-mode power supplies... maybe then I'd be able to get one of my PowerBook 500 series supplies to work again. Both recapped, had leaky caps before, and then had the same symptoms they did pre-recap...ugh.

 

[carbide]

An update to my IIsi PSU that could help you.

I finally had some time to look at it some more and noticed that there was some liquid around the large inductor at the bottom-left of the PSU board (with the AC plugs to the right as in your pictures). I de-soldered it (which was a huge pain, needed a lot of flux and fresh solder to finally get it to go) and found a decent amount of cap juice underneath it. I gave it a good scrub with IPA and checked the rest of the board closely, then soldered it back in place. After reassembly, it booted up immediately. It doesn't recognise the 16MB RAM sticks I bought for it, but that's a different problem.

 

[3lectr1c]

Ooooh that does not sound fun. I appreciate the suggestion though, if I ever feel like it, I'll take it out. I should probably take everything off the output section and clean under tbh.
Will I be delaying that for a long time though? Yes