Some update on this never ending puzzle. After some time I had the possibility to further investigate the issue. I managed to re-wire the connections of both AC and DC unit to the main board, in order to have access to all pins and run some measurements with logic analyzer and multimeter.
Originally, I though it was the overcurrent safety circuit controlling the fuser unit on the AC PSU to trigger the cut off, but in reality what is happening is that the 24V line goes down just 3 seconds after the power up signal PCONT is fed into the printer.
PCONT is wired at PIN 7 of the DC supply, it’s a 3.7V signal coming from the logic board (in turn, from the CPU through the interface board) and it triggers an opto decoupler, giving way to the high voltage current downstream the rectifier.
The 24 V should then feed the 5V DC PSU unit (the small board linked by wiring).
The PCONT signal last exactly 3”.
After that, I guess the 5 V should be available to power up the logic board and keep the whole thing alive. But there’s no 5 V, the output pin of the regulator is totally dead.
During those 3” the 24V circuit seems to get sufficiently energized (can’t measure the exact voltage, though, it’s too brief), for closing the relais of the fuser circuit (which closes and feeds current to the connector of the fuse) and spinning the cooling fan.
But as the 3” are over, the PCONT signal goes down and the whole thing becomes numb.
You need to unplug and wait for the capacitors to discharge for resetting the unit.
Honestly I cannot tell whether there’s a fault in the switching PSU controller (the IC on the daughter board), in the 5V voltage regulator, in the transistor which modulates the input voltage to the regulator or whatever else. I’m kind of shooting in the dark.
As they are cheap and available components, though, I will try and replace all of them. To be (hopefully) continued...