• Please review our updated Terms and Rules here

5150 Power Supply Repair

D

digitalgraveyard

New Member
Joined
Oct 8, 2009
Messages
4
The other day I turned on my 5150 for the first time in about a month or so. It powered on fine ran for about 25 minutes. Then it froze up and I turned it off waited about 30 seconds turned it back on. I was greeted with a crunching sound and an odd smell, no pops or anything like that. It never turned back on. So I removed the power supply from the case toke off the cover. I don\'t see anything really that \"looks\" bad. The power supply will give a short burst of power maybe a second or two then it\'s dead again? Any thing I should be looking at or testing?
 
The crunching sound makes me think "fan"; the smell makes me think "capacitor", but I suppose a dodgy fan could stink a bit too. Does the fan move freely if turned manually? Could it have shorted? IIRC the older 5150s had AC fans which ran off mains (line) voltage, so be careful poking around unless you can guarantee it's safe.
 
If it's like the power supply in my 5150, it won't start up with nothing connected -- it needs a load of some kind.
 
Thanks for the answers, but the fan is fine. The power supply had a load both still installed in the 5150 and when I had it out with a 3.5 IDE hard disk attached.
 
I would start from the 5/12 V output sides and work back. It sounds like the PS "starts" but is driving a short near the output.
 
Yeah, for those of us with PC/XT/AT's it pays to hoard a few Baby AT power supplies while they can still be found.
 
Although a rewired ATX power supply should work just as well. Just connect a switch that grounds the PS_ON pin and wire up an XT/AT connector ... I'd imagine "somewhere out there" exists a direct conversion cable. :)
 
There is a fuse in the original black psu. I remember mine burnt out once during a freeze-shutdown-restart and it made a noise when she blew.

Did you check that?
 
southbird said:
Although a rewired ATX power supply should work just as well. Just connect a switch that grounds the PS_ON pin and wire up an XT/AT connector ... I'd imagine "somewhere out there" exists a direct conversion cable. :)
Click to expand...

Like this or this?

If only all PSU's were that easy. Anyone know how to convert a Compaq WTX460-3505 PSU to normal AT or ATX use? I can't even find a pinout for the connector.
 
digitalgraveyard said:
Thanks for the answers, but the fan is fine. The power supply had a load both still installed in the 5150 and when I had it out with a 3.5 IDE hard disk attached.
Click to expand...
Hi digitalgraveyard.
I just want you to know I'm waiting with interest to see
what will be the resolution of your 5150 PSU problem, as I'm
involved in a 5150 PSU problem situation of my own.
Sounds like the people on this forum may have more than
enough technical knowledge to help me get my PSU situation
cleared up. But lets see what the cure for YOUR 5150 PSU
problem is first.
I have observed, myself with MY PSU situation, that confusing
situation you stated: You said "The power supply will give
a short burst of power maybe a second or two then it's dead again."
I observed that EXACT same thing here, when I had the
multimeter's leads stuck into the connector I'd pulled off
the floppy-drive (the yellow and black wires, for 12 VDC).
Turned on the 5150's power switch several times and the
needle on the meter would bounce up to (12 or 14 volts or so)
for just as YOU stated "maybe a second or two", then needle
back down to 0 VDC. (There was a slight sound, possibly
similar to a circuit breaker blowing, each time too, but I'm
not positive about this.)
I've pulled the PSU out of my other 5150, which had 2 results--
(1)got my main 5150 working again, and (2)make me stop
working on figuring out the actual PROBLEM with that other
PSU (unfortunately).
There was such a FLURRY of helpful suggestions for you on
Thursday, and then on Friday NOTHING! Maybe people will
have time on the weekend for experimentation/research/remembering
and so your situation will be solved soon.
One thing I wanted to mention is that in my experimentation
here, I THINK (but am not positive about this) that I got
no results from the power supply when the only thing plugged
into it was one floppy drive, but it suddenly started
working-BETTER when I had the 2 motherboard connectors
plugged in as well. My theory is that there's some wiring
in the motherboard where the motherboard gives some kind of
signal back to the PSU without which the PSU shuts itself
down. You said your PSU had a load "when I had it out with
a 3.5 IDE hard disk attached". Was wondering if the same
setup with the addition of the 2 PSU motherboard connectors
being plugged in to the motherboard would make any really
noticeable difference. --Like make it work perfectly, for
example.
When I hear what you did to fix your 5150 PSU problem,
maybe I'll know what to do to fix my own 5150 PSU problem,
although my situation is a little different than yours (and
self-inflicted, sorry to say).
 
A 5051 power supply may be different, but I had a similar symptom on an AT recently. The power supply seemed faulty. The fan would move a fraction (or sometimes not at all).

Turned out it wasn't the PSU at all. Rather it was a shorted capacitor on a card. Because of this, the PSU wasn't getting a "power good" signal from the motherboard hence it wasn't firing.

I'm not saying this is the problem in your case, but it's worth considering.

Tez
 
southbird said:
Although a rewired ATX power supply should work just as well.
Click to expand...

The point I was making about using an easily-obtained Baby AT PS is that is it a zero-cost replacement that looks and functions like the original and doesn't require any electrical kludging. Clearly an ATX has the same outputs, but why bother with the effort in installing a momentary switch and adapting the wiring?
 
Most PSUs for the 5150 require a sufficient load to be connected in order for the PSU to start. I've never found that a floppy drive alone was enough of a load. I've found that a motherboard alone provides sufficient loading, and I've found that a hard drive alone provides sufficient loading.

To: digitalgraveyard
The smell means that something has burnt. So the possibility is high that either you are going to see a burnt component or you will see a component that has a crack in it. You may need to use a magnifying glass and take your time. If a component is cracked, you won't always see the crack, such as a crack in a transistor bolted to a heatsink. In your case (a smell), you could also run your nose over the circuit boards (power off obviously) to see where the smell is the strongest, i.e. leading you to the general area of the faulty/smelly component.

To :artpitkin
Welcome to these forums. Have you done a good visual inspection of the insides of your PSU (in case the faulty component/s is also visually damaged)? Note that there can be multiple causes for a particular symptom. And so note that if digitalgraveyard identifies his faulty component/s, that those same components won't necessarily be the cause of your PSU's failure (assuming the board in his PSU is the same as in your PSU).
 
Chuck(G) said:
Like this or this?

If only all PSU's were that easy. Anyone know how to convert a Compaq WTX460-3505 PSU to normal AT or ATX use? I can't even find a pinout for the connector.
Click to expand...

I hadn't actually looked, but it seems like it ought to exist, and apparently it does, pretty much exactly as I imagined it! :) It hasn't come up for me yet so I didn't think too hard about it.

As for your Compaq PSU problem -- figuring out the pinout of a PSU shouldn't be too difficult. If it was powering any IBM Compatible, it likely has the whole -5, +5, -12, +12, PG, and a bunch of grounds. Just pick any ground you want (could even be on a molex drive cable) and start testing each pin of the PSU. You should at least be able to figure out the voltages. Figuring out which one might be PG ... well, maybe you'll be lucky and there's a color coding or label that will hint at it.
 
artpitkin said:
Thanks. That was fascinating reading! And the fact that your logical troubleshooting procedure brought you to a SOLUTION
to the problem makes me feel I can do the same with MY
old-computer-PSU situation. So thanks for the "encouragement"!
Bye. --Art
Click to expand...

You're welcome Art. I'm no techie, and I've found some things impossible to diagnose, but patience, some basic equipment, circuit diagrams and some help from these forums can often result in a fix.

Good luck with it.

Tez
 
southbird said:
As for your Compaq PSU problem -- figuring out the pinout of a PSU shouldn't be too difficult. If it was powering any IBM Compatible, it likely has the whole -5, +5, -12, +12, PG, and a bunch of grounds. Just pick any ground you want (could even be on a molex drive cable) and start testing each pin of the PSU. You should at least be able to figure out the voltages. Figuring out which one might be PG ... well, maybe you'll be lucky and there's a color coding or label that will hint at it.
Click to expand...

It's a 24-pin ATX-style connector. Figuring out he DC voltages isn't tough, but Compaq uses some odd conventions for the fan, power switch and other (unknown) signal lines. I've checked and it isn't ATX or WTX. Seems to be something Compaq-specific.
 
paul said:
The point I was making about using an easily-obtained Baby AT PS is that is it a zero-cost replacement that looks and functions like the original and doesn't require any electrical kludging. Clearly an ATX has the same outputs, but why bother with the effort in installing a momentary switch and adapting the wiring?
Click to expand...
Thanks to all of you for this technical info; until my
power-supply problem is resolved, I'm collecting all info
that'll allow me to UNDERSTAND power-supply-units as well as
step-by-step instructions of what I should do to solve my
particular problem.
I guess the answer to the question "why go to all that
trouble (rewiring an ATX PSU instead of using a readily
available Baby-AT PSU)" is (1)when in the future the Baby-AT
supplies are no longer "readily available", and (2)what if
you don't HAVE a Baby-AT supply (but do have 1 or more ATX
supplies) and don't want to BUY anything right now. --And
also (3)it's interesting to me personally to KNOW how to do
that procedure simply because it helps me understand the PSU
situation in general. (I've been away from computer-tinkering
for quite a few years. I'm familiar with the IBM PC and XT,
but when the 386's came out, although the 386 chip was
supposed to be such an improvement over the 286, I feel the
software that came out then (particularly Windows 3.1) was
such a down to me that I stopped tinkering with computers
at about that time.

So thanks for helping this "old dog" learn "new tricks".
New tricks such as how to keep one's old comp running, even
if it requires adapting the newer readily-purchasable-NEW
computer parts.
Later! --Art
 
You must log in or register to reply here.
Back
Top