IIGS Power Supply Repair

[Rusty Dekema]

Greetings,

None of my original Apple IIGS power supplies will run a IIGS stable-ly anymore. I am using a hacked up AT PSU as a temporary solution to power one machine, but I would like to get back to being able to run them with the cases on.

I assume (I know, I know) that the problem is bad/worn capacitors in the original PSUs. Do I likely need to replace all the caps, or is there a set of "usual suspects" that I should look at first?

Again, the original PSUs power the IIGSes up just fine, but the machines crash frequently. The machines run and do not crash when being run from the newer AT supply.

Cheers,
Rusty

[retrogear]

I'm not aware of any reports of which capacitors are the most likely. Does anyone know of a schematic of a typical IIgs power supply on the 'net ? If the PS is unstable like random shut down or causing reboots, it's more likely a capacitor involved in the feedback loop which would be small value like < 10uF. Problems with larger capacitors would generally be no power at all and more apt to swell physically. Bad caps will swell underneath where the leads are and not necessarily on top. This is from my experience of repairing switchmode regulated power supplies.

Larry G

[Michael Black]

UNless they moved to a supply from a third party, isn't it likely to be
about the same as the Apple II supply (and the schematic for that is in
the original Apple II manual).

One reason I say that is that I traced out the power supply/analog board
on my Mac Plus one time. That gets all the details, but it results in a
schematic that isn't easy to follow. By chance I noticed the Apple II
schematic, and saw a resemblance, so I redrew what I had in the same form
as the Apple II schematic, and it was more or less a match COnsidering
that, I'm not sure why they might have changed things later. Changing the
physical layout I can imagine, but why change the design unless there was
a big need?

I found an Acer LCD monitor, and it nominally worked. But if I left it on
long enough, it would reset and display the Acer logo. A real problem
because then I couldn't see what was going on in the computer I had it
attached to. So I changed the more obvious capacitors (ie the larger
value ones on the output side) and the problem went away. I wouldn't have
thought it would reset because of that. So what may seem to be the
problem may not be.

Change the larger value capacitors on the output, easy to find since
they'll be in the vicinity of the output lines. If that doesn't do it, it
may be more complicated, but replacing those may take care of most
problems.

Michael

[retrogear]

If the IIgs supply is similar to the II or IIe, check out this discussion:

ftp://ftp.apple.asimov.net/pub/apple_II/documentation/hardware/misc/Apple%20II%20power%20supply%20tech%20info%20.txt

and look for references to C7 220uf 16v

I remember that cap being a culprit for dead or intermittent power supply.
The Sams Photofacts are online for the II and IIe which show that cap in the feedback loop. Sharing photos of the component and circuit side of your PS could help identify it.

Larry G

[Antony Mauget]

The IIGS PSU is somehow different from the one in the II and //e.
You can find a schematic on my Picasa at
https://picasaweb.google.com/amauget/AppleIIGSPSU?authkey=Gv1sRgCLvK2a_rtJuGDw
but this is not an official one, just a reverse engineering work.
There is also a tip for the Dynacomp branded PSU (not the Astec one) : 2
capacitors may have to be replaced (see last 2 pictures).
Credits goes to Stephen Buggie for these pictures.


--
________________
Antony
Apple II forever

[retrogear]

Interesting drawing. C5 could be the equivalent feedback path. T2 pin 2 is essentially ground so the signal on pin 1 is derived from the current on T2 which passes thru C5 to control the on time of Q2 the main driver and provide regulation. If C5 were to open, the power supply would be in runaway which would create an overvoltage shutdown via the SCR on the right crowbarring 12v to ground,
reset and then turn back on, repeating rapidly (chirping sound). Or stay latched off and never turn on. A defective C5 would probably be a dead power supply or intermittent without smoke. I would vote for that cap but of course any are suspect. My two cents ...

Larry G

[retrogear]

I just love electronic circuits so I'll comment some more. The goal of a regulated supply is to keep the sample voltage constant, in this case the 5v at J7-4. If the power supply is running, the key voltage to check is the 12v at J7-3 which is unregulated. If that voltage is elevated, say 12.5 to 13v then the problem is with the sample so C11 or C24 would be suspect, or insufficient feedback like C5. If the voltage gets too high it will break over the zener Z2 and shut the supply down. If the voltage is low, say 11v to 11.5v then the problem is insufficient power on the primary like the main cap c4. Generally, the unregulated voltage will tell you whether to look at the primary side of the circuit ie left of T2 or the secondary side.

Larry G

[cordyva...@gmail.com]

My IIGS Power Supply just went down. Can you re-post the schematic? I believe mine is not the Astec version.

[Antony Mauget]

cordyvandenberg wrote:
> My IIGS Power Supply just went down. Can you re-post the schematic? I
> believe mine is not the Astec version.
>

Here it is : https://photos.app.goo.gl/gZ2LLQH6XZYce8Dc8

[cordyva...@gmail.com]

This picture is for the GS power supply designed by Astec. Mine was designed by DynaComp. I have found what my problem was: a failed X2 capacitor. This design has 2, X capacitors (0.22uF) both of which I will replace.

[Michael J. Mahon]

So did a failed X2 capacitor cause the fuse to blow? Generally, X2 caps
fail open (that’s part of the X2 spec), and since they bridge the AC line,
their failure leaves the supply working (but smelling awful).

--
-michael - NadaNet 3.1 and AppleCrate II: http://michaeljmahon.com

[cordyva...@gmail.com]

The fuse did not blow. I turned off the supply as soon as I heard a pop and saw smoke. The X2 cap measured 88Kohms after I removed it. So, in circuit it would still dissipate about 1/6W at 120V. Interestingly, the capacitance reading is still correct: 222nF (component labelled as 0.22uF)
I have installed new X2 and Y2 capacitors and my IIgs is operational again.

[Michael J. Mahon]

Your results are typical for an X2 cap failure.

When the polypropylene dielectric breaks down, usually as a result of a
transient voltage spike on the AC line, a rapid buildup of heat at the
breakdown site melts the dielectric and breaks the (short) circuit. This
leaves the capacitor with higher leakage but essentially unchanged
capacitance.

It’s the ability of polypropylene to “heal” shorts that makes them suitable
for the line-bridging X2/Y2 role. They don’t create persistent shorts, but
they do produce extremely unpleasant odors!

[Brian Patrie]

I wonder if this is why my IIe stinks a bit.

[Polymorph]

It's quite possible.

I've had this happen to 2 separate //e's and in both instances when the caps failed, they exploded and released brownish goo distributed throughout the internals of the PSU. This "goo" seems to retain the stench, so it's best to get in there and clean it up as best you can to remove the odour.

Cheers,
Mike

[James Kruth]

Quick question here regarding IIgs power supplies (I hope that's ok in this thread) - I've noticed occasionally my IIgs seems to have trouble booting - once it's up and running things seem fine. I do have 3 modern cards in it (VidHD, CFFA3000, and Uthernet II). I'm wondering if a ailing power supply could be causing intermittent boot problems? Do these cards draw more power at startup then when they are running?

Thanks!

- James

[Brian Patrie]

Thankfully, nothing has exploded in any of my Apple II PSUs (yet). :)

[Polymorph]

This sure sounds like a marginal PSU. Try taking one of the cards out and see if the problem goes away. If it does, then it is almost certainly a PSU problem. I would be looking at either recapping, or better yet replacing the PSU with a modern equivalent like the one from Reactive Micro.

Cheers,
Mike