• Please review our updated Terms and Rules here

IBM 5160 faulty power good line

A

Amigaz

Experienced Member
Joined
Jan 31, 2008
Messages
426
Location
Stockholm, Sweden
Trying to resurrect an IBM 5160 which doesn't post and I've figured out with the excellent troubleshooting guide at minuszerodegrees.net that my power good line voltage in the PSU is bad .. it is almost zero volts :eek:

I haven't done any PSU repairs before, is there anything in particular that usually goes bad when the power good voltage is bad on the IBM 5160 PSU's?
The PSU brand is "Schrack"

 
Its normally a tantalum cap on the motherboard that prevents power good from going high in these machine.

Are you sure the PSU is faulty ?
 
With nothing connected to the PSU, Do you get any DC output voltages, +12v -12v +5v -5v or is it just the missing PG
 
Fair enough.

So the PG signal is produced by the PSU when the power lines have come up to voltage and stabilised, so the question is, are the lines coming up to the correct voltages or is the detection circuit faulty. An AT supply gives about 0.5 seconds before it gives up.

Do you have a storage scope so you can see the rise on each of the lines and see if they are reaching spec ?

If they are all coming up to normal, you 'can' tie the +5 to the PG line to tell the motherboard all is good (not recommended) but i would then try to fix the fault.
 
I had to attach some load to this PSU to get it working. It wouldn't start with one 5.25" MFM hard drive, but did work with two 5.25" MFM hard drives.
 
As I mentioned earlier, doing so with a small RC network (series R, C to ground) will provide a bit of a delay that might result in more reliable operation.
 
Amigaz said:
Correct
Click to expand...

If all other DC outputs are within spec, I'd trace from the PG wire where it connects to the PSU PCB and test components along that line after looking for obvious signs of damage, Keep in mind the dangers of working on PSU's, Don't power it up and make sure the Caps are drained, The outcome really depends on your experience and what tools you have, What chuck suggested is an option though never gone that route myself.
 
Chuck(G) said:
As I mentioned earlier, doing so with a small RC network (series R, C to ground) will provide a bit of a delay that might result in more reliable operation.
Click to expand...
Example at [here].

Gary C said:
So all the lines are good, you can jump from +5 to PG I believe.
Click to expand...
I brought out one of my IBM 5150 motherboards.

I tried about 10 power-ups of the motherboard, and in about half of those power-ups, I had the POWER GOOD connected to +5V (done by connecting pin 11 of the 8284A chip to +5V).

In the times that POWER GOOD was connected to +5V, the motherboard did not start. That will be because the 8284A (never seeing POWER GOOD in a LOW state) did not generate its RESET signal at motherboard power-up time, and therefore, the CPU was not reset. Related information at [here].

su7tj48ho457i3.jpg
 
Thanks for all your help and suggestions, guys. really appreciate it :)
Haven't had any time yet to tinker with the PSU because life and work came in between ;-)
Think I'll start by cracking the PSU open and look for obvious damage and take it from there with your tips and suggestions ...
 
You must log in or register to reply here.
Back
Top