Reviving IBM PC 5150

[SpidersWeb]

So after fetching my IBM 5150 and matching 5151 from storage, thought I'd plug it in.
It didn't beep but I thought I'd give it a few seconds. Then things went bad, I let the smoke out.

Inspected the cards and they look as new - even the chips have their original polish, no tarnishing. The floppy controller does seem to have 2-3 dry joints though. Motherboard looks completely clean, 256Kb on board. Then I sniffed the power supply - think I found the fault. Trouble is, not sure if I have another AT supply to test if the motherboard/memory survived - will see what happens. Just about to go pull the PSU apart now.

What I have with/in the IBM, now that it's with me:
- original CPU
- original MDA + Parallel card (looks new)
- original floppy controller card (looks new)
- expect 256Kb on-board memory (its the 64K version 2, and all slots are full, so I suspect 256K)
- two original Tandon double sided full height drives
- original keyboard
- IBM 5151 display (haven't plugged this in yet)
- made in Aus not the US (awww) so the logo is "IBM' with '5150' under it ont he front rather than 'Personal Computer'.
- Guide to Operations + Basic + two IBM handbooks + Protran study book from ChCh Poly.
Build date is Oct 1984.


Also my ESS soundcard died, fitted in my T3200SX. Took 4 attempts to get the machine to even start up (its happy now though).
But I found a huge range of software for my AccentPC and I've worked out what part of the circuity blew on it all those years ago. Just a headphone amp circuit, may make my own and piggy back it at some stage.

[tezza]

Yes, it's common for AC filter caps in a PSU to blow if things of that vintage haven't been on for a while. These are quite easy to replace if you have soldering skills. It's usually very obvious which ones have blown so diagnosis is not a problem.

If it is one of these PSU caps, no motherboard components should have been effected by the blowout.

[SpidersWeb]

yeah i can solder (electronics was an old hobby of mine), my problem at the moment is finding where I put my security bit set - they're all tiny bolts to open it up except ONE torx with the pin in the middle.

Thanks for the info, fingers crossed thats all it is. I don't beleive this is the first time it's happened on this machine, as there is a big mark on the case lid right above where the PSU was (I had the lid off when it popped).

Peered through with a little torch, the big 220V blue caps look fine - no bulging or colouration, but on the other side I could see the top of two big black capacitors - one silver on top, the other one has black gunk on it. Guess I had better just go find my bitset and find out!

[SpidersWeb]

Got distracted but I found the right security bit and a 6.5mm socket.

Everything in there (despite the dust and a little residue on one of the caps) looks fine except for one piece. In the input section that bolts on to the back of the PSU, a rectangular cap right at the top above the first transformer, 0.1uf (was expecting the problem to be a much larger cap to be honest), has burst open and is definately mr stinky. Now the challenge will be removing the rear board - hopefully without having to remove the mainboard on the PSU. (Ok not really a challenge but it's fiddly in there)

[SpidersWeb]

Got the board removed enough to solder/desolder.
My only problem is, what do I replace it with?

Don't usually see 0.01uf caps that are this large, was this for accuracy in 84 or is there something special about this cap I need to take in to consideration? If anyone could provide any hints on what to replace it with it'd be much appreciated.
Model is 2050MC610 and it has 0.01uF stamped on top and two pins only (one at each end).

Looks not tooooo bad here:


But oh god it's titanic:

[SpidersWeb]

Found two similar components at RS Components.

There are contact supressors, 100nF (0.01uf) at varying resistances:
e.g.
http://newzealand.rs-online.com/web/p/r ... k/0209229/

Then there are just caps, made by the same company, that look the same:
e.g.
http://newzealand.rs-online.com/web/p/paper/4148645/

Both made by the same company as the original part, suppressor would make sense, but it looks more similar (in how the company has laid out the wiriting) to the second link.

Thoughts?

The component is connected in parallel with the first transformer (connects to AC via the main power switch).

[tezza]

Yes, in my experience it always seems to be this kind of cap that blows in a PSU.

Jaycar have a good range of these polypropylene caps including the one you're looking for:
http://www.jaycar.co.nz/productResults. ... orm=CAT2#1

[SpidersWeb]

Excellent, cheers for that. Exactly what I needed to know.
Will pick one up on Monday and swap the part out, then try again....

Haven't tried powering up the 5151 yet :/

[SpidersWeb]

Unrelated to the IBM.

Picked up an unopened monitor, 1993. Its genuinely never been used and is now hooked to my T3200. After a couple of goes on the power switch she fired up and displayed the VGA output no worries.
However it has those magnet marks on the display, turned it on and off several times to do a degauss but they're still there, is this normal and does it clear eventually?



Has that "new component" smell that I remember from when I was a kid getting a new screen. Is nice.

Edit: just doing cold start degauss cycles, but may find a magnet and try to reverse it or change it so degauss can get it (worst case)
Edit2: 5160 just cost me $205 but I won. Hoping its MDA because I don't have CGA (well TV I guess) or Hercs. Looks MDA. Also won 286 and 386 auctions. 386 for parts, 286 as a 1.2/360 go-between.
Edit3: picked up the replacement cap for the IBM, local DSE had it, $1.93. 100nF mains supressor. Will use sacrifical HDD as dumby load and test voltages.

[SpidersWeb]

Before replacing the PSU cap I thought I'd plug in the 5151 - which has no light to tell if it's on.

Turned it on, and it didn't go bang, in fact it did nothing. But without an input I can't be sure that it's actually faulty or not.

Can anyone remember if these screens do anything - like flicker - on power up?
80% thinking it's already dead and needs love, 20% thinking that it's just doing what they did. It was working before (around 2-3 years ago).

(and I'm asking because if its faulty I'll just put everything away and save it for Saturday)

[tezza]

From what i've read these can appear dead unless getting a video signal from the machine itself.

[SpidersWeb]

Woo!

Edit: small delay, I seem to have thrown out my capacitor by accident in a cleaning rampage

[SpidersWeb]

Replaced cap, no worries.

PSU starts up and runs a sacrifical hard disc for a few minutes.
Power off, connect a second hard disc (for more load) and ran it again. No worries.

Put it back in the PC case and the fan spins but no action at all from the PC - no beeps, no blips on screen, just silence except the PSU fan.

So flick the power switch back off. Then I wanted to see if a motherboard short was causing the PSU to cut it's output (multimeter probes I had on hand had too short of a probe), so while it was OFF I connected a hard disc - then sssssssssssssssssssssssssss and smoke starts pouring out of the PSU (pretend the air vent is a steam engine funnel). Took about two minutes for the smoke to stop dribbling out.

Figuring its the other A/C cap which is harder to get to. But will be waiting until my other gear arrives in two weeks so that I have more bits to test with.
It is rather fiddly working in there because the two boards and power connectors are all soldered together - and you can't remove the rear connectors on this model because the screws were shaved down at the factory for some reason. So *may* just send it to an electronics reparier.

386 coming down has either a dead PSU or dead motherboard, if it's the motherboard then I'll have a good donor PSU (as another option).

Edit: well tonight's mission is to fix that SVGA I posted earlier. My degauss wand has arrived, and while it looks like an 1400W mains-power-lady-pleaser, I'm sure it'll do the job well

[SpidersWeb]

Update for the DEC monitor (irrelevant of 5150 I know I know).
Used my magic wand, but didn't make a difference, I think the mask is actually bent, and it was put back in the box in the 90's for that reason.

Good backup VGA though, save me lugging the big Dell around! And once I'm done, it can be hooked to the media server and left off just in case I have one of those 'oh noes' moments with it at some stage.

Trying to find a good electronics guy for the IBM PSU not just to replace the filter but to check the rest of it is in good health, but I will resistance test the motherboard power pins in the weekend. If anyone has any recommendations in Wellington let me know!

[SpidersWeb]

Did some more maintenance on the old girl.

Removed PSU and completely removed both boards. The fault from the latest whoopsie was the same as before, the other AC Filter had spilled it's guts out. I've removed it, but need to pop buy dickies to grab another - I figure there is no point rebuilding the PSU only to know that I really should open it up again in the future to install that part.

I also removed the PC mainboard.

Both the PSU boards and PC mainboard got a good wash in isopropyl.

I have an ATX PSU which easily fits inside the old AT power supply and it's easy enough to swap the wires over. I got a plastic food-dish which is thick enough to make a safe 'adaptor' for the different screw sizes and ATX has all the same voltage outputs (even P.G. etc). But first I'm going to set it up on a bench with the motherboard upside down so I can test the voltages and make sure P.G. is grounding correctly. If the PSU is sweet, then its motherboard diagnostics time - piggy backing memory is the only real thing I know can help, otherwise I may need to order a board from the US without further theories.

5150 Rev B board with 256KB memory drying in the sun:

I could not find any visual faults on the MB. The only thing I spotted was on a CMOS chip and the soldered memory - and that was very small amounts of rust building on the pins.

Now the AC filters blowing wont stop the PC from at least beeping at me, and that's what I'm worried about. I did the resistance checks, no shorts on motherboard inputs or PSU outputs.