Hi, this one's been driving me mad for a long time so I give up and finally ask for some ideas here…
PSU from Dell Precision 690. Turns on fine when cold. You can stress-test it for a day and it'll keep running fine. But when it's hot it won't turn on.
I think I narrowed down the problem to the PFC area (first & last pics), when I heat there with the hot air station it won't turn on. I'm however not able to narrow down a specific component.
When heated up, the PFC still somewhat works but voltage on the main filter caps drop below 380VDC (even down to 350VDC) and it won't turn on, if you let it cool down, voltage increase and when it's above 380VDC you can turn it on again. It may be a consequence, it may be cause, it may be a coincidence, not sure.
The VREF pin on the PFC controller (ICE1PCS01, in blue) always stays close to 5V, so at least it thinks it's regulating correctly.
To that VREF pin is connected the voltage divider from the output (3x750k top and 28k bottom apparently) as it should be, but also looks to be a pretty complex circuitry with some diodes, an LM393 (in red), a transistor and another IC I don't know what it is (IC4, Q0 11 markings). It's especially when heating up that area that the problem occurs. I'd say it's more sensitive near the LM393 and the diodes. I already replaced the LM393 and tried swapping the diodes so that's not it.
Of course when heated up the diode voltage drop goes a bit lower but that's expected.
That circuit probably allows turning off the PFC when the PSU is off, but there must be something else to it, it's too complex just for a shutdown function.
Even when the transistor (NPN) is removed it doesn't work when warm.
I pulled the diode in yellow and it seems to be working even when warmed up now. Of course the diode is not the problem (tried to replace it with schottky 1N5819 since it measured 0.19V drop) and I have no idea what is actually the problem. I don't really feel confident using this PSU with a random diode removed… Even with this diode removed the PFC seems inactive when the PSU is off (around 315VDC) and active when on (around 385VDC).
Thinking it could be a solder joint issue I reflowed the area, but no luck.
PSU from Dell Precision 690. Turns on fine when cold. You can stress-test it for a day and it'll keep running fine. But when it's hot it won't turn on.
I think I narrowed down the problem to the PFC area (first & last pics), when I heat there with the hot air station it won't turn on. I'm however not able to narrow down a specific component.
When heated up, the PFC still somewhat works but voltage on the main filter caps drop below 380VDC (even down to 350VDC) and it won't turn on, if you let it cool down, voltage increase and when it's above 380VDC you can turn it on again. It may be a consequence, it may be cause, it may be a coincidence, not sure.
The VREF pin on the PFC controller (ICE1PCS01, in blue) always stays close to 5V, so at least it thinks it's regulating correctly.
To that VREF pin is connected the voltage divider from the output (3x750k top and 28k bottom apparently) as it should be, but also looks to be a pretty complex circuitry with some diodes, an LM393 (in red), a transistor and another IC I don't know what it is (IC4, Q0 11 markings). It's especially when heating up that area that the problem occurs. I'd say it's more sensitive near the LM393 and the diodes. I already replaced the LM393 and tried swapping the diodes so that's not it.
Of course when heated up the diode voltage drop goes a bit lower but that's expected.
That circuit probably allows turning off the PFC when the PSU is off, but there must be something else to it, it's too complex just for a shutdown function.
Even when the transistor (NPN) is removed it doesn't work when warm.
I pulled the diode in yellow and it seems to be working even when warmed up now. Of course the diode is not the problem (tried to replace it with schottky 1N5819 since it measured 0.19V drop) and I have no idea what is actually the problem. I don't really feel confident using this PSU with a random diode removed… Even with this diode removed the PFC seems inactive when the PSU is off (around 315VDC) and active when on (around 385VDC).
Thinking it could be a solder joint issue I reflowed the area, but no luck.
Comment