Restoring my IIcx

rodders

Active member
Well, I can't believe its been a year but I've finally got round to tackling the IIcx I posted about in My first Apple computer, a Mac IIcx.
I managed to revive the basket case RiscPC I was working on (that was a journey) and then got distracted by an A3020 and an Amiga but the IIcx now has my full attention.
The logic board has suffered from cap leakage and battery damage quite severely but it looks like it was stored on its right side so the alkali from the battery hasn't gone through the RAM.
The startup circuit is toast with several components falling off and I've discovered lots of eaten traces on the back of the board.
I've also taken the PSU apart and lo and behold, one of the Rifas has exploded. Collective wisdom seems to be that its not worth trying to restore the (Astec) PSU so I've ordered a Flex ATX and plan to drop that inside the Astec box.
The floppy drive is full of crud, not surprising as its the main air intake. Not sure if its saveable yet so I'll try to get a spare for parts.
I'll post some pics as the work progresses if anyone is interested.
 

Snial

Well-known member
Well, I can't believe its been a year but I've finally got round to tackling the IIcx I posted about in My first Apple computer, a Mac IIcx.
Wonderful! I had a IIcx in the very late 1990s/early noughties. It had 4MB of RAM, but only a mono video card. I passed it on to another friend who has stuck with Macs since!
I managed to revive the basket case RiscPC I was working on (that was a journey) and then got distracted by an A3020 and an Amiga but the IIcx now has my full attention.
<snip.. brutal amounts of corrosion & damage to multiple parts of the IIcx>
I'll post some pics as the work progresses if anyone is interested.
Of course, a few of us (mostly Brits) also have Archimedes computers too, e.g. I have a 3MB A3020 for which I 'recently' converted a PS/2 mouse to an Archimedes mouse the hard way, by interfacing it directly to the Quadrature encoders!
 

AndyMc1280

Well-known member
Hullo rodders!

Well I'm here because of my Brother's SE30. Having been bitten by the bug, I now have two II ci, plan is to fix one flog one. As you can probably tell, I've a few Acorns - Listed in signature. Including A3010 Risc PC and R260.
Best of luck with the restore.
 

rodders

Active member
Here's a reminder of what I faced:
IMG_20230730_095327987.jpg
I opened up the PSU and found this (notice the blown Rifa):
IMG_20241021_100514804_HDR.jpg
and on the back of the board, this (notice the light blue on the traces where the copper has been eaten away):
IMG_20241020_133124647_MP.jpg
I finally got the battery box removed, the caps look even crustier than before:
IMG_20241019_114014466_HDR.jpg
I've started removing the rotted components, not sure if I should replace the F258s as well, there seems to be crud underneath them. Only managed to lift one pad and that was the dummy on D4:
IMG_20241019_135931827_HDR.jpg
I've started fixing the bad traces on the back. I'll post more pics when that's done.
 

rodders

Active member
Well, I started by scraping back any traces that looked discoloured because something nasty had got under the solder resist.In many cases the trace has not been eaten away, at least not completely, so I tinned the trace to beef it up a bit so they didn't fail later on. In some places the trace looked almost white and when I scraped these away there was nothing underneath. To fix these I scraped back the trace to some good copper and bridged the gap with 33SWG tinned copper wire.
Here's what it looked like (sorry about the quality of the pic but you get the idea).
IMG_20241029_132700185_MP.jpg
The more I looked the more I found. In some places there was a single circular spot with a diameter the width of the trace which had caused a break.
When I felt I'd fixed all of the defects by buzzing all of the repairs, it was time to cover them all with solder resist. I'd bought a kit a while back with some solder resist and a UV torch. I hadn't used it before so I tried a test area only to find the UV torch wasn't and the ink didn't cure so I'm waiting on a new UV torch from China before I can progress. Doh!
Having got some experience with rotted traces on the back I had another look at the top and found several dodgy looking areas so I'm fixing those while I wait for the torch.
 

naruse

Active member
Well, I started by scraping back any traces that looked discoloured because something nasty had got under the solder resist.In many cases the trace has not been eaten away, at least not completely, so I tinned the trace to beef it up a bit so they didn't fail later on. In some places the trace looked almost white and when I scraped these away there was nothing underneath. To fix these I scraped back the trace to some good copper and bridged the gap with 33SWG tinned copper wire.
Here's what it looked like (sorry about the quality of the pic but you get the idea).
View attachment 80049
The more I looked the more I found. In some places there was a single circular spot with a diameter the width of the trace which had caused a break.
When I felt I'd fixed all of the defects by buzzing all of the repairs, it was time to cover them all with solder resist. I'd bought a kit a while back with some solder resist and a UV torch. I hadn't used it before so I tried a test area only to find the UV torch wasn't and the ink didn't cure so I'm waiting on a new UV torch from China before I can progress. Doh!
Having got some experience with rotted traces on the back I had another look at the top and found several dodgy looking areas so I'm fixing those while I wait for the torch.
Kudos to how perfectly well you are getting those traces back! 🙇‍♂️
 

rodders

Active member
Well, the new UV torch arrived but it was pretty pathetic and I had trouble getting the solder resist ink to cure. I bit the bullet and bought a more powerful (10W) one which had good reviews. I still struggled with curing until I applied a little heat with my hot air station at around 270C.
I've now fitted new parts in the startup area:
IMG_20241118_115613102_MP.jpg
I only replaced the first row of F258s as the others seemed fine. I thought the diodes and transistors would be the most difficult but they were pretty easy (gel flux helping to hold them in place). On the other hand, the tantalum caps that I got to replace the electrolytics and the fusible link were tricky as they almost completely obscured the pads. I've also recapped the rest of the board, a couple of the large axial caps were reluctant to come out but I managed it without any damage.
So, the logic board is ready - but I can't test it without a power supply. I intend to fit a modern ATX PSU inside the original case as the general feeling seems to be that its not worth trying to repair the original. However, some adaptation will be needed so I've designed a daughter board (see https://github.com/roddersuk/Apple-IIcx-PSU-ATX-Adapter for details) and I'll post some more pics as I assemble the PSU.
 

rodders

Active member
I have one question. Some people claim this computer won't start without the PRAM battery, is this correct?
 

rodders

Active member
Finally got a usable PCB for my interface board. The Kicad footprint for the 24 pin ATX socket follows the Molex spec. and has a larger pitch between the rows of pins on the PCB side than the plug side whereas the Chinese copies that I have are the same (simpler to manufacture I guess). As a result, the ATX socket wouldn't fit the first PCBs I ordered, so I created a modified footprint and ordered some more boards.
This board handles the switching of the monitor socket and also inverting the soft start signal (with credit to https://tinkerdifferent.com/threads/psu-hack-for-iici-and-other-soft-switch-macs.116/ for the optocoupler design)
Here is the kit of parts with the mains plug and the motherboard plug salvaged from the original PSU:
IMG_20241215_114905879_HDR.resized.jpg
and here it is fully assembled
IMG_20241216_121227325_MP.resized.jpg
On the PSU board I snipped off the drive leads as these aren't needed and replaced the mains leads with longer, more substantial ones as the originals were a bit thin for 220V, too short and had been badly soldered in the factory with the insulation melted. I used crimped terminals to solder into to my PCB.
IMG_20241216_121311789_MP.resized.jpg
Here are all the parts ready to be reassembled with my PCB finally hooked up to the PSU board. Note the 3D printed adapter for the PSU board. This locates on the original posts.
IMG_20241216_122220222_HDR.resized.jpg
Here's a shot showing the adapter in place on top of the original insulating sheet.
IMG_20241216_122642579_HDR.resized.jpg
And here everything except the fan fitted into the box. There is a lot of wire in there, I considered shortening them but will only do so if there is an overheating problem due to restricted air flow. The input module is secured in place by one of the original PCB mounting posts. Because this is raised slightly by my 3D printed adapter I had to tweak the tinware to get it to fit. Note the top insulating sheet preventing shorting of my daughter board.
IMG_20241216_131009521_HDR.resized.jpg
Finally, the fan is fitted. I'd put 12V provision for a fan on my board but this PCB has a temperature controlled socket for the fan so I used that instead. I just needed to change the plug on the fan cable.
IMG_20241216_131758342_MP.resized.jpg
So there it is. Everything is completely reversible so that a purist could restore the original PSU if required.
Now I just need to carefully test it before plugging it into the motherboard. Let's hope for no magic smoke!
 

rodders

Active member
Hmm, when testing I found that the soft start didn't work. It turns out that the optocoupler design I used has pins 3 and 4 labelled the wrong way round. I feel a bodge wire coming on...
 

rodders

Active member
Well, I finally plucked up the courage to switch the thing on. Thankfully no magic smoke but this is all I got:
View attachment VID_20241223_113923229.mp4
Its clearly trying to start but failing and trying again. I'm guessing that its not even getting into the POST.
If anyone has had a similar experience and knows where to start looking I'd be grateful.
The startup circuit is where most of the damage was so its not really a big surprise. I'll take a look at the schematic again and see what I can figure out.
On a slightly better note I also powered up the monitor (after a good look around inside and a bit of de-rusting) and there was no magic smoke either. I won't know if it actually works until I can the the IIcx to start but it seems promising.
 

Arbee

Well-known member
That behavior from Apple power supplies typically means there's an overload or dead short to ground on one of the power rails on the motherboard. Check with an ohmmeter.
 

dougg3

Well-known member
Also I just want to preemptively point out that around 16 ohms between ground and the 5V rail on the motherboard is normal. It's often enough to cause your multimeter to beep in continuity mode, but 16 ohms doesn't count as a dead short. That has confused me and many others in the past :)
 

croissantking

Well-known member
I had similar behaviour from a IIci I restored. I confirmed my PSU was good by trying it with a IIsi. It turned out to be a faulty 74HC132 chip on the logic board startup circuit.
 

rodders

Active member
Thanks guys, I'll have a probe around. I've done lots of repairs so not surprising it didn't work first time.
Still trying to get my head around how the startup circuit works. In particular, does it hold the PS-ON line high all the time it's on or is it a momentary thing? I'm assuming the former.
From the behaviour it seems that the startup part works as the power LED comes on but it then stops and tries to restart so i guess the stop circuit is the place to look first.
I've replaced the 74HC132s so unless the new ones are dodgy they should be OK. I'll get my scope on it when I've worked out what the signals should be. I'm having a go at simulating the circuit in Kicad but I'm a complete novice...
 

rodders

Active member
Another step forward, I buzzed out the stop part of the startup circuit and found an open connection between pins 2 and 8 of the 74HC132 so the SR latch wasn't working. It was a bad solder joint between pin 8 and its pad. Re-soldering it means the power stays on now.
It doesn't get much further though, there is a bit of a thump through the speaker but no chime. Not sure where to look next. Can you pick up the POST anywhere?
 
Top