Soft Power circuit and /PFW (Q950)

[BadGoldEagle]

Hi all

To keep things short and simple, I am now in a situation that prevents me from devoting as much time to the hobby as I'd want to. Don't worry, I'll come back to it in a few months, my passion for Macs hasn't gone away at all. There's just more important 'life' things going on at the moment. I'll still be lurking on the site from time to time to see what's up but don't expect me to post a lot...

Anyway, before I take that break I'd like to sort this Quadra thing out. I've found a workshop specialized in repairing custom and industrial PSUs. I haven't contacted them yet but given the amount of custom stuff they deal with, this PSU shouldn't cause them that much trouble. I tried to buy the proper schematic but the only source (Bomarc) has apparently disappeared...

I'd like to make sure the PSU can be started without the board (I won't be sending them the whole unit).

I got hold of the second edition of the 'Guide to Macintosh Family Hardware' reference manual. Pages 269 through 272 deal with the Soft Power circuitry used on the Mac IIs.

Could one of you just please confirm that the following statement is correct before I plug it in for a quick test? I don't want to make things worse...

If I want to power the PSU without connecting it to the board, connecting the 5V trickle power pin (#12, blue wire) to /PFW (#1, white wire) powers it on without any load needed (?).

Thanks!

Yours truly,

BGE

 

[BadGoldEagle]

Well, today I got the multimeter out, and noticed that the entire 5V rail appears to be shorted to ground: there's only 26Ohms of resistance between the two and the multimeter in continuity mode beeps so that can't be good, right? 

 

[BadGoldEagle]

Nevermind what I said above. This doesn't actually mean anything as the 'load' isn't purely resistive. If it had been 0 or really close to 0, then it would have been a problem. 

Knowing this, I felt pretty confident about plugging it in for yet another smoke test. 

And it booted right up. I don't usually curse (at least in my posts) but WTAF???

Last time it would simply refuse to do anything but click. The 5V rail used to power the logic board is fine and strong. It's been on for about 25 minutes now and nothing to report. 

I think it won't last long though. I think some component might not particularly like getting hot or one of the joints might still be bad... If only I had my hot air station around! I used it last month to reflow the joints on the 512k Hyperdrive's additional power supply. The joints looked sublime after a few minutes of heat. 

Now I'll try putting a higher load on it (HD, Graphics card and perhaps DOS card...).

 

[BadGoldEagle]

Okay, interesting, so it doesn't like loads. 

 

[BadGoldEagle]

Seems like I'm starting a monologue here...

Anyway, the /PFW wire is indeed getting around 5V (4.4V in my case. FYI, the ATXs work by shorting the green Pwr_ON wire to ground, i.e. the other way round). The soft power circuitry must be using the 5V trickle line but the latter isn't directly connected to the /PFW line when the Mac turns on. It must rather be using the standard 5V line once powered on.

I shorted the blue (5V trickle) and white (/PFW) wires and the power supply was able to start without it being connected to the logic board. So that's at least a good thing. This time, like an ATX, it doesn't need any load to power itself on. Plus when it's powered that way the PSU doesn't reset itself anymore, because all the other rails (-12V, +12V and 5V trickle) are working perfectly fine. However, I am getting absolutely nothing on the 5V rail. ZERO, NADA. Again, that's 'somewhat' good news, as I used to think the PSU was shutting itself off due to a serious problem (shorts etc...). As it turns out, it's just because the /PFW line isn't powered correctly. 

So the problem resides within the 5V rail. I'd say it's more capacitor related as I did wait about 20 hours since yesterday to test it again and it was still in the endless reset loop. But yesterday, after a few of weeks of rest, it worked fine for a while until I started loading it. If heat or a bad joint had been the culprit, it would have had plenty of time to cool down... But again, I only chose Nichicons and all of them were rated way higher than the originals. So IDK.

Problem is nearly off my hands. I'll give the PSU to a professional (again, I don't have the time to troubleshoot more) and report back when it's done. 

 

[LaPorta]

Just reading as you go...

 

[trag]

Just reading as you go...

Yes, silent audience.  You're doing great.  I'm making popcorn...

 

[BadGoldEagle]

I've had to find another workshop as the first one wasn't really interested. The second one I've found is great though. This guy services industrial grade PSUs but he's also a 'Funk' collector (ie radios, TVs, talkie walkies, CB etc.) and an Apple aficionado. He worked on a few Lisas and has a Powerbook 190. TBH, we chatted more about those than about the actual repair...

If he can fix it, he'll even do a 24h stress test. I'll update you guys as soon as I hear from him.