Apple //e randomly shuts off

[DistantStar001]

I have been piecing together my third Apple //e for a few months now. It's mostly junk parts I've accumulated over the years, but as far as I can tell they all work. The motherboard is fine and has been tested at length in another //e. Cosmetically, the case looks good, but I will admit that I'm missing several screws. The keyboard works too, despite the fact that all but seven of the keycaps are actually from a Mac Plus (and the Delete is a c64 f3 key for now). Still, as far as I can tell, everything works and has been tested with my other //e's. The only exception is the power supply, which randomly cuts out somewhere between 20 minutes and an hour of use. 

 

When I got this power supply, it didn't work as I discovered several cracked solder joints on the rectifiers. Also, the switch is a little wonky, but according to my multimeter, it works. So I reflowed the cracked contacts, and it was fine for several months. That is until the current issues began. I've rechecked the joints and their fine. I wish they weren't, because then I'd know what's wrong. It's not modified in any way, and the only capacitor I've replaced is that stupid solid state that I'm always hearing about blowing in a cloud of toxic smoke. I forget the name of it. it was a yellow boxy thing that looked pretty cracked when I got it.  I know that the supply will work just as well without it, but I was told it served as a safety or something, so I put in a proper replacement.

 

I've dismantled the power supply several times, removing the innards, and testing the outputs with a multimeter. All voltages were present, and when I plugged it back into the motherboard the computer booted up without a problem. Then I screwed the power board back into its case, flipped the switch, and nothing. So I tested the lines again, and with the board inside the case, I was getting less than .01v on all six lines. 

 

Eventually, I came upon the idea that the case might somehow be shorting the power supply to ground on the DC side (I figured it wasn't the AC side, since the power supply was in my lap at the time, and I'm not dead). So long story short, I ended up lining the bottom of the case in plastic and coated the bottom of the board in electrical tape. That seemed to work for a while, but then it cut out again. 

 

I switched it on and off a bit with no luck. So I left it off for about an hour, then powered up, and it booted normally. But then, no more than two minutes later it cut out again (this was happening while I was typing). Eventually, I got up to flip off the switch, when it just booted again right in front of me (so far it's still running). 

 

The power supply doesn't seem to be under or overvolting in any way. It just goes from on to off to on again without me doing anything. Right now there are only four of the six screws holding it in the case, so I'm wondering if one of them has a ground contact on it that makes intermittent contact through the electrical tape. 

 

The thing is that this shouldn't be a problem. Since at least one of my other computers has an identical model PSU and has never needed to be shielded from ground with such severe methods (actually, I didn't have to do anything to it. It just works fine). I'm honestly baffled as to why this thing is cutting out, or what's causing the short on the board. 

 

Any ideas?

[The Usotsuki]

RIFA capacitor?

[DistantStar001]

54 minutes ago, The Usotsuki said:

RIFA capacitor?

That's the thing I replaced!  It was actually driving me nuts trying to think of it.  Still not sure what's making it short though.

 

On the other hand it's still running?

Edited November 30, 2019 by DistantStar001

[ChildOfCv]

Does turning it off and back on immediately after a shutoff bring it back?  Or does it have to sit for a while?

[DistantStar001]

22 minutes ago, ChildOfCv said:

Does turning it off and back on immediately after a shutoff bring it back?  Or does it have to sit for a while?

The last time, I had to wait a while.  But the previous times, removing the PSU board from the casing bring it back instantly.  In fact, every time I removed it from the case it's been fine.  Then I would put it back inside the case and it shorted again (sometimes immediately).  

 

This time it's been on solid for a couple of hours now.  But like I said, it shut itself off for a few minutes, and then just as I got up to physically turn the switch off and take it apart again, it just started up in front of me.  It's been on ever since.  But I have no idea for how much longer.  Last night it was fine for hours, then it just wasn't. 

[DistantStar001]

Just tested my //e again.  It booted fine, then shut off two minutes later.  It powered on again about 30 seconds later, then shut off again after a minute, rebooted, stayed on for about three minutes, then shut off.  I haven't touched it since turning it on about 10 minutes ago.

[ChildOfCv]

Well, the question was:  If it turns off, and you immediately flip the switch off then on, does it immediately come back?  Well, I guess you would need a couple of seconds of delay between off and on due to the capacitors.  But anyway, that could indicate that the protection SCR on the +12V line turning on.

 

Otherwise it would seem like an intermittent connection that disconnects due to heat.  For instance, the transformer could have a broken wire.

[DistantStar001]

11 minutes ago, ChildOfCv said:

Well, the question was:  If it turns off, and you immediately flip the switch off then on, does it immediately come back? 

No.  When it cuts out, turning it off and on again does nothing.  When I've left it alone, sometimes it comes back.  Then turning the switch off and on functions normal, even within seconds of turning it off.   But if I leave it on, it either cuts out a few seconds later, or stays on for a while then cuts out again.  Sometimes it reboots on its own, but there's no precise timing on that.  And even when it does reboot, eventually it cuts out again.  It actually just cut out about five minutes ago, and it still hasn't come back.

 

However, all of this only started happening after I covered the bottom of the board in electrical tape.  Before that, the board wouldn't function inside the case no matter what, but if I took it out and propped it on something nonconductive, it was fine.  I'm guessing that there's still some ground contact, but I don't know why that's a problem since the board would only be making contact where it's designed to.

 

15 minutes ago, ChildOfCv said:

that could indicate that the protection SCR on the +12V line turning on.

I'm not sure which component that is.

16 minutes ago, ChildOfCv said:

Otherwise it would seem like an intermittent connection that disconnects due to heat.  For instance, the transformer could have a broken wire.

I'm doubting this since the computer isn't staying on long enough to get warm.  Also it doesn't explain why the power supply works perfectly outside it's case (even on extended tests) and then shorts (when uninsulated) as soon as I put it back in (even when the case is open).  

 

It's almost as if any contact with ground on the board causes it to cut out.  The question is why?

[ChildOfCv]

Do you still have it connected to the Apple when testing it outside of the case?  Well, maybe so since it's hard to tell if it's doing anything otherwise.

 

Anyway, Sams Computerfacts has a diagram of the power supply.  If this looks like yours, the SCR is labeled on page 33 of this PDF:  https://ia803003.us.archive.org/1/items/Sams_Computer_Facts_Apple_IIe/Sams_Computer_Facts_Apple_IIe.pdf

[Aunty Entity]

Flakey cap in there somewhere?

[DistantStar001]

16 hours ago, ChildOfCv said:

Do you still have it connected to the Apple when testing it outside of the case?  Well, maybe so since it's hard to tell if it's doing anything otherwise.

Yes I connected the it to the //e for testing, but I also measured the voltages from the connector as well.

16 hours ago, ChildOfCv said:

Anyway, Sams Computerfacts has a diagram of the power supply.  If this looks like yours, the SCR is labeled on page 33 of this PDF:  https://ia803003.us.archive.org/1/items/Sams_Computer_Facts_Apple_IIe/Sams_Computer_Facts_Apple_IIe.pdf

Thank you for this.  It’s actually been a big help.  
 

I'm guessing that the SCR is the little thing that looks like a 7805 voltage regulator (but is not).  If I replace it, do you think that will solve the problem?  Or is it another component causing it to malfunction?  
 

5 hours ago, Aunty Entity said:

Flakey cap in there somewhere?

Can a flaky capacitor cause these kind of issues when in contact with ground?

 

Like I said, this thing is fine when isolated from ground.  I’m just not sure why, but it is.  I probably could fix it reliably by by removing whatever screws are making contact with ground, thus isolating it completely, but I’d still have a faulty component in my PSU that could cause future problems.

[ChildOfCv]

The SCR might be activating due to either a spike in voltage or maybe because it's worn out.  Either way, an SCR turns on easily but doesn't turn off until you remove all current through it.  So they thought that would make a good protect circuit.

 

C20 is the filter capacitor for that rail.  If it's not good or if it's not making a good connection due to bad traces/pads, then it could allow for voltage spikes.  Other components in the protection circuit include Z1 (a zener diode), R26, R27, Q4, R17, and R18.  Make sure their pads and traces are all good.  If you have the components, you could remove the SCR to see if it stops shutting off.  But you should connect the power supply to a dummy load instead of the computer for that test: a couple of 20-watt, 24-ohm resistors in parallel for the +/-12V rails and a 20-watt, 5-ohm resistor for each of the +/-5V rails.  That would put an amp of load on each rail.