VOGONS


First post, by scruit

User metadata
Rank Newbie
Rank
Newbie

I'm trying to replace the original psu in this Mitsubishi MP-3200 386. The original one just puked a month after I recapped it. I will eventually get around to fixing it, but for now I'm adapting a modern PSU.

I have all the wires mapped out except one - it's white.

On the mothreboard side it goes to connector P28, unlabelled. Tracing that line around the mobo gets me to a 14069 inverter, then the matching inverter output goes to the "/CE" chip enable on the RTC. It may go to other places but tracing is a pain.

On PSU end the wire is unlabelled on the PCB inside the PSU. All the other wires connectors are labelled as -12, 5, -5 etc. This one just says "WHT". It goes to the collector of an NPN transistor (emitter is grounded, and the base is +4 through a resistor), which is paired with another NPN transistor, which is connected to the output of two channels of a quad comparitor. The inputs to the comparitor are the 12v signal, 5v signal etc. I labelled it in my notes as ("8"), whereas I labelled the rest of the wires on this connector as "-5", "12", "-12" so I likely stuck a multimeter on it and just read 8v dc.

The two most likely candidates are PS-ON or Power Good.

Looking at schematics of similar power supplies it matches more like a PS-ON signal - but this case does not have a soft power button. It's a big heavy IBM AT-style throw switch on the side of the PSU. No external signal tells the PSU to power to start.

The other candidate, a power-good signal, is also strange.. Why would that go to the chip enable on the RTC?

Right now I'm assuming it's a power good signal. I'm interested to hear if anyone has any other suggestions or advice.

Reply 1 of 1, by MrGatoGrande

User metadata
Rank Newbie
Rank
Newbie

Do you happen to still have your notes on this? I just got my hands on a MP3200 and looks like the PSU is dead. Nothing happens when I flip the switch.

At first I thought it was related to the key on the front (some kind of key-triggered start)... but playing with the switch on both positions still nothing happens... I'm thinking the PSU is a goner.