• Updated 2023-07-12: Hello, Guest! Welcome back, and be sure to check out this follow-up post about our outage a week or so ago.

How to turn on a IIci PSU outside the computer?

olePigeon

Well-known member
Are there pins I short to get it to turn on? I don't have a spare PSU, so I'm trying to figure out if it's my PSU or my IIci. :/
 

sfiera

Well-known member
Assuming your IIci PSU is like my Q700 PSU (Delta Electronics), then shorting pin 9 (/PFW, non-square, white) to pin 10 (+5V trickle, non-square, orange) as shown in the above picture should do it. When I did it, the fan started, but I didn’t check the voltages. They might be incorrect while unloaded.
 

re4mat

Well-known member
Yeah, you also have to give it a load, otherwise it trips the over-voltage protection and shuts down. If you've got any 10Ω 5–10W resistors, those should work fine for providing the load.
 

Trash80toHP_Mini

NIGHT STALKER
If you've got a IIsi, you can rig up a test astup with the board out of the case. Running the IIsi or not would be a definitive under load test.
 

olePigeon

Well-known member
Excellent, thank you everyone. I figured it was similar to ATX and just needed a couple pins shorted, but I couldn't find the relevant info.
 

olePigeon

Well-known member
Considering there isn't even fan spin on the power supply when I attempt to turn on the computer, I'm not too worried about the voltages. I just want to see if the PSU will even turn on.

Curiously, on very rare occasions, the computer will turn on and appear to work fine as I'm unplugging the cable. It's weird, and I can't reliably replicate the issue. I'm just hoping it's the PSU and not my Mac. The PSU is a lot easier to diagnose. I've already replaced 90% of the components, so a few more won't hurt anything. But I'm rubbish when it comes to diagnosing computers, even though I have a fancy multimeter.
 

olePigeon

Well-known member
A relay clicks on/off repeatedly with the fan turning on/off. I suppose that's normal with no load.

Sigh.
 

desertrout

Well-known member
If you've got a IIsi, you can rig up a test astup with the board out of the case. Running the IIsi or not would be a definitive under load test.
So a follow up question for me is -- is there any potential risk to the board when doing a smoke test on a newly recapped PSU? I have a IIci PSU I want to recap, but haven't partly because I wasn't sure how to test the voltages before using it on the board...
 

olePigeon

Well-known member
@desertrout I feel like shorting those two pins and letting the load protection trip seems like a good idea to smoke test. It would ramp up, then shut off automatically, which seems safer than plugging it in.
 

olePigeon

Well-known member
Also, I'm going to re-recap my IIci. I suspect I know the problem. I just hope I didn't fry my motherboard somehow. I used solder paste to help me lay down the new capacitors, as there's almost zero room with those surface mount cans with their super tiny leads. I didn't use a heat gun, I just used my soldering iron to melt the paste. When I removed my old tantalums, there was some unmelted solder paste under them. I suspect the same is true for my new caps, and it's short the pads underneath instead of going through the capacitors.

I'm going to put the tantalums back on on and see if that fixes it. If it works, then I'll need to reevaluate how I can put those new caps on. I probably need to buy a heat gun. :/
 

joshc

Well-known member
If it works, then I'll need to reevaluate how I can put those new caps on. I probably need to buy a heat gun. :/
Apply solder to both pads first, heat one pad with your soldering iron, use your other hand to put the capacitor in place while the solder is hot, in effect tacking that side of the cap to the board. Heat the other pad and push down on the cap, tacking the other side.

Or, use hot air to do the above.
 

desertrout

Well-known member
I haven't ever had any issues using an iron and regular solder for SMD cans - just use enough flux (even just a flux pen) on the pads first to ensure the solder gets drawn in.

Edit: Also, a little of pre-solder on the pads helps too as @joshc says
 

olePigeon

Well-known member
@joshc I usually pre-tin one of the pads, then solder one side of a cap to secure it as described above, then apply solder to the other side. It's how I do it with tantalums, anyway. But there's plenty of room with tantalums. These cans ride along the very edge with very little room. The problem I've had with trying to tin both pads simultaneously is that after I solder one side, it's then at an angle as it rests on the cold solder on the other side. It has a tendency to lift the already-soldered-side pad as I press to secure and solder the other side. I've been hesitant to do that again.

@desertrout I'll try a more liberal application of flux. All I have is the thick, brown, sticky stuff. I don't like it because it's hard to clean, but it's all I have.
 

desertrout

Well-known member
I'll try a more liberal application of flux. All I have is the thick, brown, sticky stuff. I don't like it because it's hard to clean, but it's all I have.
Ah, I see. Yeah, and unless it's "no clean" flux you run the risk of causing a short (unless you're liberally cleaning or ultrasonically cleaning the board). And even 'no clean' flux ought to be cleaned off.

Well, you may want to invest in a flux pen or a syringe to assist in this, as it's essential kit imo, and inexpensive. But even so, your description of your issue is interesting, as I can't say that I've encountered it myself. Perhaps you're adding too much solder to the pads? It doesn't take much to create a solid connection, and the cap legs usually accommodate for much of the wiggling (they freely bend). But I certainly I understand your hesitation, as you wouldn't want to lift a pad.

As an alternative approach, you can also just apply solder to one pad to tack it down, then solder the other side down. But flux will be necessary on the unsoldered pad in that case.

This is the 'one-side' method I just described (he doesn't use flux, and you can see it's a bit janky):

This guy is soldering without pre-tinning the pads at all (uses flux):
 
Last edited:

olePigeon

Well-known member
OK, so, some "progress." I re-recapped with tantalums. Nice clean board, nice clean caps. Exact same issue. Won't turn on 95% of the time. So I wasted my nice caps, I had apparently done a good job in the first place. I'll have to order another set.

However, if I press in the auto-power button on the rear, then unplug/plug in the power cable quickly, I am able to get the machine to chime and turn on for about one second, but then it turns off.

So, I still don't know if it's the IIci or the power supply. I'm trying to find a PSU I can borrow. :(

@desertrout That's how I soldered on the tantalums and essentially my cans when I used the paste. I thought you meant tinning both pads, which I had tried previously, and that would often end up with a lifted pad. My flux is a "no clean," but it's a very ugly flux. I got it at Radio Shack, if that's any indication of how old it is.

I should rename and move this thread to the Macintosh II forum as a repair / diagnosis thread.
 

superjer2000

Well-known member
It doesn’t start up when you press the power key on the keyboard, or when you short the PSU pins to turn it on, or both?
I had to replace a few SMD diodes and ICs in my IIci soft power circuit to get it going.
 

olePigeon

Well-known member
It doesn’t start up when you press the power key on the keyboard, or when you short the PSU pins to turn it on, or both?
I had to replace a few SMD diodes and ICs in my IIci soft power circuit to get it going.
When I short the pins, the PSU will start/stop at regular intervals as a relay clicks. I assumed that was the overvoltage protection.

The computer does not turn on via the keyboard except on very rare occasions. That's why I've had the power switch on the rear engaged as I troubleshoot it.
 

superjer2000

Well-known member
When I short the pins, the PSU will start/stop at regular intervals as a relay clicks. I assumed that was the overvoltage protection.

The computer does not turn on via the keyboard except on very rare occasions. That's why I've had the power switch on the rear engaged as I troubleshoot it.
It sounds like your PSU isn’t connected to the logic board when you short the pins though, is that correct? It should be otherwise I’m not sure what you’re testing. I am pretty sure you need to keep the pins shorted for the PSU to stay on (unlike an ATX). If that works then your PSU is likely fine (probably a good idea to still measure voltages at the connector.) I have a jumper cable that lets me connect the PSU without having it sit on the logic board.

I am willing to bet it’s your soft power circuit. As noted, I needed to replace some components plus I had to jumper a few lines that got eaten by cap goo. I posted a thread here on it I believe.
 

olePigeon

Well-known member
@superjer2000 It is not connected to the motherboard when I short the pins, no. I just had it out to see if I could get fan spin, because it wouldn't turn on when installed in the computer (except seemingly randomly.)

Would a bad soft power circuit cause the power supply to turn off?

In any event, I'd love to see your thread. It could point me in the right direction.
 
Top