IBM 5150 - 130W PSU Repair

[Jotschi]

Hello,

I just started my second IBM PC restoration project.

The 5150 has a 130W T2.0A 250V PSU which is not working. I tested it without the FAN and only a dummy HDD attached.
I can hear a relay trip and I get no power on the plugs. I assume this is the over current protection circuit kicking in.
I also tested this without any load attached. When the fan is attached it gets a small amount of power and moves a bit. (no full rotation).



I have also found some documentation on the PSU:
http://worldphaco.com/uploads/The_IBM_5155_POWER_SUPPLY.pdf

On the bottom I noticed a cut in the ground plane. Is this a known bodge?


I checked the PSU pins with no mains power and loads connected and noticed a low resistance between 12V and GND. (24Ohm).
Is this normal or does this indicate a problem? I noticed a 24Ohm resistor on the board. I suspect this to be a bleeder resistor.

Is the board with the relay part of the over current protection circuit?

I know that PSU repairs are hard and I don't have experience in this but I still want to give it a try.

Any hints or tips would be welcome.

 

[modem7]

I just started my second IBM PC restoration project.
The 5150 has a 130W T2.0A 250V PSU which is not working.

Presumably, you tried the PSU in the first IBM 5150 (i.e. known working) and that is how you know that the PSU is faulty. Is that the case ?

I can hear a relay trip and I get no power on the plugs. I assume this is the over current protection circuit kicking in.

Is the board with the relay part of the over current protection circuit?

My recall of earlier discussion is that the PSU has a resistor to limit the 'inrush' current at turn on time. Some PSU's have a relay connected to that resistor, with the relay controlling whether the resistor is bypassed or not. At turn on time, the relay does not bypass the resistor (i.e. limiting inrush current). Shortly after, the relay bypasses the resistor (no longer required), to reduce power loss.

The relay is shown in the partial circuit diagram in one of the documents pointed to at [here].

The fact that you hear the relay clicking indicates that at least very early after power on, there is voltage being produced.

I also tested this without any load attached. When the fan is attached it gets a small amount of power and moves a bit. (no full rotation).

As expected for an unloaded IBM 5150 power supply.

(And from that, we know that the fan is DC powered.)

On the bottom I noticed a cut in the ground plane. Is this a known bodge?

Looks like someone removing a ground loop, or reducing noise. Something that a manufacturer might do.

Any hints or tips would be welcome.

A single hard drive may not be an adequate load. It comes down to the design of the particular PSU.
Because the fan is a DC powered type, if it turns (continuously) during experimentation, it means that the PSU is at least mostly working.
I think that you should reconnect the fan, because maybe it produces some the loading.

A rotating platter type of MFM hard drive draws most of its power from the +12V line. Some PSU's will require the adequate loading to be mostly on the +5V line.

 

[Jotschi]

Thank you for your the feedback.

Presumably, you tried the PSU in the first IBM 5150 (i.e. known working) and that is how you know that the PSU is faulty. Is that the case ?

Since I did not know the condition of the PSU I did not try it in the 5150. I try to test components individually first. I do this to directly narrow down on issues. Especially for the PSU I wanted to check output voltages first. Unregulated output could cause harm.

Looks like someone removing a ground loop, or reducing noise. Something that a manufacturer might do.

Yes, that may be the case. I'll leave the cut untouched.

A single hard drive may not be an adequate load. It comes down to the design of the particular PSU.

…

A rotating platter type of MFM hard drive draws most of its power from the +12V line. Some PSU's will require the adequate loading to be mostly on the +5V line.

That's a good point. I try a larger load on the 12V rail. I also tested a 5150 floppy drive and fan. I check whether I can build a dummy load using some power resistors.

So far I checked every Cap (EL+Ceramics) and all Diodes.

 

[Jotschi]

I made myself a dummy load Molex plug (16W on 12V and 6W on 5V). That was enough for the PSU to start regulation. Output voltages are spot on. A single TM100-2A + PSU fan was not enough. Thanks for help.