• Please review our updated Terms and Rules here

IBM PS/2 Power Supply - High pitch whine / no voltage under load / super hot 5w resistors

B

BrainStruck

Member
Joined
Jun 5, 2020
Messages
23
Location
Jupiter, FL, USA
Hi,

I recently acquired a dead IBM PS/2 Model 50 Z. Upon initial troubleshooting it appears as if the power supply is the main problem. When turning on the computer, I get no video, no beeps, and a high pitch whine out of the power supply. If I unplug the power supply from the motherboard and power it up there's no whine.

Power Supply Details:
ASTEC AAS13621
IBM FRU P/N: 90X9366

With the motherboard unplugged I checked the voltage at the motherboard connector and I'm getting: -11.31v (-12v rail), 11.79v (+12v rail), and +5.2v (+5v rail).
With the motherboard plugged in, the PSU whines and when I check the voltages I get: -0.01v (-12v rail), +1.63v (+12v rail), +0.77v (+5v rail).

I replaced all of the caps on the low-voltage side. There are 2 power resistors that get very hot. So hot they melted the tube the switch wires run through, faded their labels so bad I can't tell what they're supposed to be and cooked the board enough to leave burn marks on it, and lifted the pads they're soldered to. One tested @ 8 ohms and the other @ 16 ohms.

I've attached a photo of the board calling out some general notes. I'm wondering if someone could point me in a direction. I have a basic concept of how this power supply works, but I can't seem to figure out which component is emitting the high-pitched whine, or if the whine is the cause of the problem or a symptom.

Thanks in advance,

Brian
 

Attachments

  • ps2-psu-notes.jpg
    ps2-psu-notes.jpg
    3.1 MB · Views: 17
The first thing that comes to mind is that the whine is for some reason the PWM is reducing the transistor drive frequency to the point it's audible. The resistors are then used to dump wasted energy as to prevent things from blowing up.
I have seen a similar circuit like that in a Television set where a failure in the feedback circuit for the output went open, the voltage started running away and a sacrificial diode shorts the power supply to ground, putting it in said state.
 
Have you tried to load test it with an external load, like fans or a sacrificial hard drive, WITHOUT it being plugged into the motherboard?

If the power rails are dropping to almost zero when plugged into the motherboard, it sounds like there's a dead short on the motherboard somewhere. Since those older SMPS designs tend to be group regulated, if one rail is dragged down (usually the +5v rail), it will drag the other rails down with it.
 
BrainStruck said:
There are 2 power resistors that get very hot
Click to expand...
that's totally expected, they are here as dummy load. it's thank to them you can run this psu without load.
as GiGaBiTe said, that smell a short on motherboard. you can highly suspect a shorted tantalum, they don't always go boom when shorted.
 
Thanks for your responses. It was a shorted tantalum on the -12v rail on the motherboard. I removed the cap, and it booted. New caps are on the way.

Thanks again!
Brian
 
I had the same problem: whining PSU when connected, green lights when PSU is switched on outside of case, problem was a -12V cap shorted.
But are they tantalums though? How do you know? I think tantalums are usually polarized, these are not. PS/2 have that weird 3 legged caps they don't seem to produce anymore. Both outside legs are usually connected to ground. Mine have markings 22 16 and a 2 in a circle, I've desolders alive one and seems to measure 22uF and I inferred 16(??) volts. Does anyone know for sure?
 
You must log in or register to reply here.
Back
Top