The components labeled "VDR" in the last picture are "voltage dependent resistors" (thus the name), which is another term for metal-oxide varistors, used for surge protection. One VDR with skid marks below it, and another VDR obviously missing from the supply looks like the surge suppressors had to handle a massive surge - or the supply was set to 115V and connected to 230V. The supply will work perfectly with missing VDRs, but it will not start with shorted VDRs.

If the VDRs have been killed by some kind of overvoltage, the primary fuse is likely blown, too.

The purple wire is labeled "5VS", likely "5VSB", which is 5V standby supply, the black one on the three-pin connector is extremely likely ground, so the green one is the "remote turn-on" connector.

I just noted the missing component "R14". If that one is on the AC input, maybe it was an inrush current limiting resistor, likely an NTC. In case of a massive surge, that NTC might fail catastrophically. With that component missing, the AC circuit is interrupted, and the supply would be completely dead.

A lot of PC supply circuits (they are mostly similar) is found at https://danyk.cz/s_atx_en.html

I have one of those (or one very similar at least), it's basically an ATX power supply without the 3.3V components fittedand AT-style connectors instead of 20-pin ATX. The soft-off works exactly as in ATX, with the purple wire being +5VSB and green is PS_ON.

By looking at the last picture, both the component that seems blown (VDR1) and the one that was removed (VDR2) are varistors used to protect against voltage spikes. Those two probably failed because someone tried to run the PSU at 220V with the switch set to 110V. They usually short when they die, causing the fuse to blow but they probably saved other components down the line. If you're lucky then only replacing the fuse and varistors should fix the unit (you can even run without the varistors at all, but you lose that protection). Otherwise you may have to replace some other components such as the two big caps or the switching transistors, but you need to troubleshoot in that case.

If I was you I'd try to fix it, Astec units are well made, with quality components. Otherwise you can use an ATX with an adapter for the power and soft-off connectors.

EDIT: damn, mkarcher beat me to it 😛

Rikintosh wrote on 2024-01-20, 17:54:

I have this computer in hand that I'm trying to save, it uses an Astec AS145 3480 PSU, it's an AT standard PSU, but with some differences, the computer works very similarly to an ATX, with a soft button, standby functions, connecting via alarm, remote control, etc... It does not have a traditional on/off switch, but apparently a cable with green, purple and black wires, which turn it on via a soft button.

Unfortunately it is impossible to find a replacement, and someone seems to have already messed with this PSU, can I try to fix it or will I have to adapt an ATX PSU and modify it to work with the needs of this computer... This OR was used on a variety of OEM parts at the time, such as Packard Bell, Olivetti, and possibly Dell as well.

Probably similar to the Astec unit that came with the NEC Proserva V Plus systems a while back (this translated link has more on the PSU in question - https://www-elektroda-pl.translate.goog/rtvfo … en&_x_tr_pto=sc)

As @cpushack bought a number of these systems, maybe he has some spare PSUs

Oh, sorry, I forgot to provide some information: The fuse is perfect (maybe it's new), and when it's connected to the PC, when you try to turn it on, the speaker makes a noise, but it doesn't start if you want the fan itself. When removing it from the socket, I can hear that brief thin noise characteristic of discharging (only when set to 220v)

I believe that the damage to it may have been caused by having connected 220v when it was set to 110v, as in my country, some states use 220v and others use 110v. Here where I live it is 110v.

I suspected the purple wire was 5vsb when I saw the marking on the psu board.

I suspect the problem is in the primary sector.

My motherboard is an Intel Advanced AV/EV, it has a third power connector for 3.3v PCI, but this PSU does not have any plug for that.

I don't know exactly what this 3.3v PCI would be used for, from what I've researched it doesn't seem to power 32-bit PCI slots, but for something else (?)

Edit: I put a VDR in place of the inrush resistor (R14) just to complete the circuit, but it still doesn't turn on. I don't feel very comfortable dealing with PSU since I was electrocuted by a flyback inverter from a CRT monitor
Unfortunately, I think I'll have to adapt an ATX here. I have a seasonic 80 plus which seems to be of good quality. I'll look for the AT to ATX pinout diagram and try to put in a new PSU.

Rikintosh wrote on 2024-01-20, 20:11:

Edit: I put a VDR in place of the inrush resistor (R14) just to complete the circuit, but it still doesn't turn on. I don't feel very comfortable dealing with PSU since I was electrocuted by a flyback inverter from a CRT monitor
Unfortunately, I think I'll have to adapt an ATX here. I have a seasonic 80 plus which seems to be of good quality. I'll look for the AT to ATX pinout diagram and try to put in a new PSU.

It doesn't turn on because you can't use a varistor like that. It's going to be open until it reaches clamping voltage.

You can bridge the inrush resistor with a piece of wire, or better yet another fuse... Just for testing.

But before you do that I'd really like to see the underside of that board to confirm that resistor is what we think it is.