• Please review our updated Terms and Rules here

IBM PS/2 Model 25 Not Posting, not sure where to start

Benson86

Experienced Member
Joined
Jun 1, 2022
Messages
61
Location
Largo, FL
Just picked up an IBM Model 25 at an estate. Seems like it’s in pretty nice shape, looks super clean on the inside. But it isn’t doing much. The keyboard lights flash on power up, and the crt lights up but just displays blank. No beeps, no action from the floppy drive, just the fan spinning. Any ideas on what to start looking at first? I reseated all the connectors and the ram with no success.
 

Attachments

  • IMG_2554.jpeg
    IMG_2554.jpeg
    2 MB · Views: 21
  • IMG_2553.jpeg
    IMG_2553.jpeg
    2.3 MB · Views: 21
Generic advise for old PCs:

Does power put out enough voltage on each line? +12, +5, and if it has, -5 and -12. If not, re-cap power supply (replace old capacitors with fresh, of the same type, capacitance and voltage).

Does motherboard have shorts on any of the power rails? If so, re-cap.

Reseat any socketed chips.

If those low hanging fixes don't help, start reading Service Manual.
 
So far it’s testing fine, power supply is good. Also seeing 5v on the 8086 and no short there at least. Tried putting a CGA card into the ISA slot to an external monitor and the screen just flickers. Still no sounds or diagnostics whatsoever. Sometimes get some garbage on the screen. But not much. Feels like there must be something obvious that I’m missing. Reseated the 8086 and both bios chips but no luck.
 

Attachments

  • IMG_2555.jpeg
    IMG_2555.jpeg
    2.1 MB · Views: 4
Still no luck, every rail on the power supply tests perfect. And every rail on the motherboard tests correctly and not shorted. Running out of things I have the ability to test. Everything points to a power fault because it just won’t do anything at all, but all the power is testing perfect.

I need to get one of those parallel port readers off eBay, just wish they didn’t take so long to get here lol
 
Are you sure that none of the tantalums are fried? They tend to short open so this would not help matters ... just looking at your pic the one at C84 looks a bit strange .. unless this is the picture quality?
 
Are you sure that none of the tantalums are fried? They tend to short open so this would not help matters ... just looking at your pic the one at C84 looks a bit strange .. unless this is the picture quality?
I am not, none have exploded, but thats about all I can see visually. Is it possible to test them with a multi-meter?
 
The keyboard lights flash on power up,
On many AT-class keyboards, the LED's flash on momentarily within a second or so of the keyboard receiving power. That is the keyboard's self test flashing the LED's.

On my IBM AT (IBM 5170), with stock IBM keyboard, I see the LED's flash twice at power up:
* Within a second after turning on the 5170. (As soon as the keyboard receives +5V, the keyboard does a self-test, and that includes momentarily turning on the LED's.)
* Shortly after the motherboard's power-on self test completes its memory test. (After the on-screen memory test, the motherboard's POST issues a self-test command to the keyboard.)

Reseated the 8086 and both bios chips but no luck.
Have you considered 'burning' yourself a new set of BIOS chips?
I see images (data) for a replacement set at [here].
Maybe someone or a company at [here] can create a set for you.
 
Does it have 8-bit ISA slots like model 30-8086? If so, try putting a VGA card that works in 8-bit ISA and hook up a VGA monitor. May be the CRT part is broken and PC is fine.
 
Does it have 8-bit ISA slots like model 30-8086? If so, try putting a VGA card that works in 8-bit ISA and hook up a VGA monitor. May be the CRT part is broken and PC is fine.
It does and I did actually try that. It doesn't even activate the card. Im starting to believe there's a power issue somewhere on the board. I wish there was an easy way to test the tants. I may just recap the board at some point that seems like the next logical step, all the normal stuff has been tested.
 
I wish there was an easy way to test the tants.
The subject tantalums are filtering the various voltage rails as they travel about the motherboard.
If one had gone short-circuit, it would be overloading the switch mode power supply, and you would get no voltages out of the power supply at all.
If one has gone open-circuit, it results in less filtering, which is not something that is going to stop the computer from starting.
 
To the right of the isa/mca? riser is a square package PLCC chip. Few of those pins are corroded. I would take a good look there... Also, have you reseated the ram yet?

Actually now that I look closer , also a bigger PLCC to the left of the riser with corrosion as well. That is where I would start. I would use a hot air station, remove the PLCC's and clean up that pcb. Those traces could be toast, and could cause a non boot condition.
 
The subject tantalums are filtering the various voltage rails as they travel about the motherboard.
If one had gone short-circuit, it would be overloading the switch mode power supply, and you would get no voltages out of the power supply at all.
If one has gone open-circuit, it results in less filtering, which is not something that is going to stop the computer from starting.

In that case im not sure where to go from here. None of the rails are shorted and all the voltages are perfect. I assume if it was a ram issue I would at least get some kind of a display or at least a beep. And if it was bad MCGA chips then the VGA card would give me something.
 
To the right of the isa/mca? riser is a square package PLCC chip. Few of those pins are corroded. I would take a good look there... Also, have you reseated the ram yet?

Actually now that I look closer , also a bigger PLCC to the left of the riser with corrosion as well.

Yes have reseated the ram a couple times, both the simm chips and the socketed ram. And im not sure what you mean by corrosion? The board is very clean. The chips you mentioned do have some pins/pads that dont have solder. But they dont have any signs of corrosion, they look like there was never solder there in the first place, I assumed that was just how they came from the factory.
 
So the Right side of the riser in the photo is a NS8250AV, thats a UART chip, so wouldnt prevent the machine from booting.

The left side cant make out all the lettering...
 
Next I would get a hard bristle brush, and hit all those plcc's and anything else SMT, just to rule out tin whiskers. The 5150 board I just repaired a few months ago, after doing that came back to life.
 
Guess it was the lighting, was only thing I could see... You are right its clean.
Its driving me nuts, I've had stuff with rust pouring out of it that works fine. This is one of the cleanest pieces I've found and its just borked lol.
 
Back
Top