Thanks! Yeah, it's been a while, the last half of 2014 was hectic. Hopefully I'll have more time to get to some of these projects ...
On my IIx board I replaced UB1 and UB2 and it still does not soft start. I probably should have left them on... I noticed that half the circuit seems to be devoted to soft - starting through the rear switch and half through the ADB keyboard power-on button. So I attached a keyboard and confirmed the keyboard portion is working properly. Did you try turning on through the keyboard?
One other thing... Did you see a typo in gamba's schematic? I am wondering if R15 should be labeled as R17.
Hehe, well glad to see you're still around! I tried both the keyboard and the rear switch, and both methods didn't work. Now that I've traced out the whole startup circuit, I'm pretty confident mine will work after I use known-good new batteries though. There were definitely broken traces though, so I still needed to repair it even if my batteries were half of the problem.
I originally thought the top half of Gamba's startup circuit was for the ADB power button and the bottom half was for the rear power button, but I was wrong. Actually, most of the startup portion of the circuit is shared by both the keyboard and the rear power button. The top part of the circuit is for power on, and the bottom part of the circuit is for power off. UB11 (VIA2) pin 13 is the /POWEROFF signal that tells the power supply to shut off. The power off circuit is also hooked up to the second pole of the power button in order to enable pushing the power button to shut off the system.
If your rear power button doesn't work but your keyboard power button does, there are only a few connections that could be causing that problem. I would verify the connections on R9 and C3 (and their connection to the power button). It could also be the connection from R3 to the power button. Everything past R3 is shared by both the keyboard and the rear button.
I didn't notice a typo on Gamba's schematic, but he did note that the components are for the II -- the IIx may be different. My IIx seems to have different component numbering from most, so mine actually matched Gamba's diagram.
Good news: My IIx now chimes and shows video when it's jump-started! I found another broken connection, and it's a pretty important one -- the main clock. The broken connection, at least on my IIx, was between UG8 pin 9 and R22. UG8 is right by a capacitor, so the leaking capacitor had eaten up the trace between that pad and a nearby via. I ran a tiny temporary wire between the pin and the via to repair the trace, and now the IIx is alive. (P.S. Thanks to bbraun for sending me a NuBus video card to use!) I plan on completely removing that chip and cleaning underneath it, but for now it's working!
Thanks for the clarifications on the schematic, I obviously didn't look at it carefully enough and see the note on the different labeling for the IIx. I'm going to start another thread on it so I don't derail this one.
Well, I got the new batteries today and there's still something wrong with my power on circuit. Now what I'm seeing is that with the batteries in, the system immediately powers itself on and the power button won't shut it off.
I think something is screwed up and allowing the battery voltage to get to places it shouldn't get to, or something like that...
OK, now the power circuit is fully working. The culprit is fairly embarrassing...
I had a patch wire on the bottom of the board between R3 and R18. This is another standard fix that has been described in the past, fixing a trace on the top of the board that rots out commonly because it's next to a capacitor. I left it pretty loose and didn't pay attention to it. Well, it turns out this wire was naturally tending to head toward one of the ground holes used for screwing the board into the case. When I screwed the motherboard down, it was tearing into the insulation and shorting that line to ground. This was forcing pin 2 of UB2 (74HC132 NAND gate input) to ground, essentially the same effect as if the power button was held down, except without R9 and C3 involved. It also explains why the power button wouldn't shut off the computer...Q3 was stuck continually allowing current through. This also explains why the batteries seemed to be immediately draining when I stuck them in.
Whenever I pulled the board out to test, that wire was no longer shorted to ground and everything worked.
The moral of this story? There's two of them:
1. Don't be careless with patch wire
2. If you're seeing a problem when the board is in the case, leave the board in the case while you're troubleshooting! In other words, don't change multiple variables at the same time. :-P
Now that I've extensively studied the II/IIx power circuit, I plan on writing up something on how it works and what to test when troubleshooting it. It was really intimidating reading the circuit diagram on Gamba's site at first, but now that I understand it, it's actually fairly simple.
Oh nice you found the issue. Did you take before/after pics of the patch wire? I am looking forward to your explanation since I don't really understand the circuit. It seems complicated for what it does.
As a matter of fact, I did! I attached the before and after to this post. As you can see, in the "after" I made the wire much more tight and wrapped it around various pieces to keep it away from the ground hole. I also went to R18 instead of a via, but that doesn't really matter too much.
Here's what the break on the old wire looks like:
I'm excited to write about the circuit! All the NAND gates look intimidating, but they are really simple. All three gates on UB2 combine to create what amounts to a buffer with enough current capability for controlling Q3, and all four gates on UB1 combine to create an SR latch/flip-flop with enough current capability for controlling Q2. I'll make some diagrams and stuff to describe it all in more detail when I'm ready.
Here's my explanation of the circuit. It ended up being a lot more difficult to write about than I was originally thinking. I think an animation or more pictures would help, but it's a start. It's so difficult to put into easy-to-understand words!
Thanks! That's for sure. I was going to say that it probably helped using a power supply that actually has a +5V standby voltage, but then I remembered the IIcx/IIci have a +5V standby voltage and they still use all the complex NAND gate stuff!
I just got your battery holder board installed on the IIx. Looks great! It was truly impossible to remove the solder from the ground pin on the IIx...I ended up having to heat up the existing solder and push the holder board's pin through.