Bridge rectifier on input.

First, you should be aware that the way these supplies operate deal with both dangerous voltages (320 to 400V) and dangerous currents. As long as you just use the supply without opening it, most dangers are contained within the case (although a 350W supply is clearly able to deliver enough power to start a fire, given sufficiently inadvertant stuff connected to it). These supplies are "switch mode" supplies, which means US electricity is rectified and voltage doubled to around 330V DC, while european electricity is just rectified to around 330V DC. In case the supply has active PFC, the voltage is boosted to around 400V DC before entering the main power conversion circuit. This part is called the "hot side", because it is directly connected to mains power.

Your symptom sounds like something is fried in the input circuit before the main transformer. Typical components to be shorted on the hot side are switching transistors and the rectification diodes. They have to be shorted really hard if they trigger the main home circuit breaker ("fuse") before popping the fuse in the supply. The circuit breakers at your home likely react ultra fast if the current exceeds around 5 times the fuse rating, that is around 50 to 80 amps on home circuits. At lower circuits, the time-delayed (also known as slow-acting) 5A-10A fuse in the supply would pop first. This makes it unlikely that the issue is in the part of the circuit used for +5VSB. That part of the circuit will burn if it would handle 50A or more, so the most likely culprits are really big semiconductors. Be aware though, that failing semiconductors may (a) cause collateral damage and (b) have failed as consequence of the circuit controlling them in the wrong way. Also be aware that there may well be more than one shorted component (typically, if the primary switching transistors go bad, they will overload the primary rectifier, and that component may short as well).

zyga64 wrote on 2025-08-21, 10:58:

Bridge rectifier on input.

And where is that?

mkarcher wrote on 2025-08-21, 11:06:

First, you should be aware that the way these supplies operate deal with both dangerous voltages (320 to 400V) and dangerous currents. As long as you just use the supply without opening it, most dangers are contained within the case (although a 350W supply is clearly able to deliver enough power to start a fire, given sufficiently inadvertant stuff connected to it). These supplies are "switch mode" supplies, which means US electricity is rectified and voltage doubled to around 330V DC, while european electricity is just rectified to around 330V DC. In case the supply has active PFC, the voltage is boosted to around 400V DC before entering the main power conversion circuit. This part is called the "hot side", because it is directly connected to mains power.

Your symptom sounds like something is fried in the input circuit before the main transformer. Typical components to be shorted on the hot side are switching transistors and the rectification diodes. They have to be shorted really hard if they trigger the main home circuit breaker ("fuse") before popping the fuse in the supply. The circuit breakers at your home likely react ultra fast if the current exceeds around 5 times the fuse rating, that is around 50 to 80 amps on home circuits. At lower circuits, the time-delayed (also known as slow-acting) 5A-10A fuse in the supply would pop first. This makes it unlikely that the issue is in the part of the circuit used for +5VSB. That part of the circuit will burn if it would handle 50A or more, so the most likely culprits are really big semiconductors. Be aware though, that failing semiconductors may (a) cause collateral damage and (b) have failed as consequence of the circuit controlling them in the wrong way. Also be aware that there may well be more than one shorted component (typically, if the primary switching transistors go bad, they will overload the primary rectifier, and that component may short as well).

I have some basic electricity knowledge, but not it this extend.

My home has some safety built in (by law), so this fuses pop us fast. I am in Europe (Austria).

How this could happen in a first place? PSU was working without problems for few weeks as I have it

boby wrote on 2025-08-21, 11:21:

How this could happen in a first place? PSU was working without problems for few weeks as I have it

Everything dies eventually.

Sometimes it is easier and cheaper to simply buy a working power supply.
However, if you want to repair it yourself, show us some photos of the interior (and the information plate).

Ok, at least I would like to know how to detect where it failed? These PSUs are very cheep and I can get another for 5€ - 10€, but as a hobby I would like to know what happen actually

boby wrote on 2025-08-21, 11:19:

zyga64 wrote on 2025-08-21, 10:58:

Bridge rectifier on input.

And where is that?

A set of four diodes in a (logical, not always physical) square that turn AC into (bumpy) DC. It will usually be followed by a couple of filter stages (capacitor + resistor) to smooth out the DC before it gets fed into the PWM controller that does the switching that gives a switching PSU it's name (followed by more filters)

Here's an explanation:
https://www.electrical4u.com/bridge-rectifiers/

It's near the 230V input of your PSU. Depending on design, it could be four separate diodes, a big block on four legs containing the diodes (with a heatsink - probably the most likely one in a PSU) or a single SMD device (more common in smaller PSUs like say USB chargers).

As you can see here, you'll get warned that PSUs contain lethal charges. While that's true, with a minimum of knowledge and precautions (such as discharging the capacitors before working on the device) you can safely work on them. However if you've not heard of a bridge rectifier, I'd suggest getting a bit more familiar with less powerful electronics before messing around in one of these. Fundamentally a USB charger or 5V-12V PSU for a small consumer electronic device works in exactly the same way as a heavy computer PSU but as they are rated for much lower currents, they store far less charge and are therefore less dangerous. Finding a dead one of those and fixing it would be good practice. Dead bridge rectifiers are definitely a thing (and can explode quite spectacularly, even in small supplies), even if dead capacitors are a more common ailment - but the symptoms are different: electrolytic caps fail open, so no short, just sagging voltage under load until the device fails to work.

dionb wrote on 2025-08-21, 18:35:

A set of four diodes in a (logical, not always physical) square that turn AC into (bumpy) DC. It will usually be followed by a c […]
Show full quote

boby wrote on 2025-08-21, 11:19:

zyga64 wrote on 2025-08-21, 10:58:

Bridge rectifier on input.

And where is that?

A set of four diodes in a (logical, not always physical) square that turn AC into (bumpy) DC. It will usually be followed by a couple of filter stages (capacitor + resistor) to smooth out the DC before it gets fed into the PWM controller that does the switching that gives a switching PSU it's name (followed by more filters)

Here's an explanation:
https://www.electrical4u.com/bridge-rectifiers/

It's near the 230V input of your PSU. Depending on design, it could be four separate diodes, a big block on four legs containing the diodes (with a heatsink - probably the most likely one in a PSU) or a single SMD device (more common in smaller PSUs like say USB chargers).

As you can see here, you'll get warned that PSUs contain lethal charges. While that's true, with a minimum of knowledge and precautions (such as discharging the capacitors before working on the device) you can safely work on them. However if you've not heard of a bridge rectifier, I'd suggest getting a bit more familiar with less powerful electronics before messing around in one of these. Fundamentally a USB charger or 5V-12V PSU for a small consumer electronic device works in exactly the same way as a heavy computer PSU but as they are rated for much lower currents, they store far less charge and are therefore less dangerous. Finding a dead one of those and fixing it would be good practice. Dead bridge rectifiers are definitely a thing (and can explode quite spectacularly, even in small supplies), even if dead capacitors are a more common ailment - but the symptoms are different: electrolytic caps fail open, so no short, just sagging voltage under load until the device fails to work.

Thank you for your answer. My high school was electric engineering and yes I heard of rectifiers, but in my language is called different (hence confusion). However, I am far away from expert so this is why I posted my question here, but I do have basic knowledge of the topic.

Again, this device has no value and I will probably keep it for spare parts, but I am just interested in those things and love if I can fix something.

Device itself:

On the first picture, we clearly see a glass fuse.
It should burn if the diodes are dead. Way before the home fuse pop !
So I don't think the diodes are the problem here.
Do you have a multimeter to check continuity ?
Maybe a cable problem ?
Also remember a home fuse can pop not because there's too much current,but also if there's current leaking to the ground.
I would use a meter and check if there's not a short between ground and the Phase/Neutral at the input of the power supply, and also between Phase and Neutral.

As said above, it can be very dangerous !
Never touch the board just after disconnecting it from the main. Wait a few minutes, it should be enough to discharge the capacitors.
And of course try to measure the voltage on them before touching ...
There's about 340 DC volts if your AC input is 240 ...

Good point about the leaking to ground.

IIRC in German that would trip the Fehlerstrom-Schutzschalter. Was that what tripped, or the regular fuse (Überstromschutzeinrichtung)?

I agree that if the fuse is intact, it shouldn't be the latter - in which case we're looking for a very different failure.

Regarding discharging the caps, any decent remotely modern PSU (which an Astec device from 1998 should be) will have a bleed circuit that will discharge the primary caps within a few minutes. The formal way to properly discharge them is with a heavy resistor (2.2kOhm, 10W or similar) across the leads of the big capacitors - or use a 100W lightbulb if you have retro lighting as well as retro comptuters. Advantage to the latter method is that you get visual feedback. But if you don't have the equipment for that, simply unplug the mains power (ALWAYS unplug the mains power) and turn the PSU on (XT/AT) or keep the power button pressed for ~20sec (ATX) and that will discharge through the attached motherboard.

nali wrote on 2025-08-21, 20:36:

On the first picture, we clearly see a glass fuse. It should burn if the diodes are dead. Way before the home fuse pop ! So I do […]
Show full quote

On the first picture, we clearly see a glass fuse.
It should burn if the diodes are dead. Way before the home fuse pop !
So I don't think the diodes are the problem here.
Do you have a multimeter to check continuity ?
Maybe a cable problem ?
Also remember a home fuse can pop not because there's too much current,but also if there's current leaking to the ground.
I would use a meter and check if there's not a short between ground and the Phase/Neutral at the input of the power supply, and also between Phase and Neutral.

As said above, it can be very dangerous !
Never touch the board just after disconnecting it from the main. Wait a few minutes, it should be enough to discharge the capacitors.
And of course try to measure the voltage on them before touching ...
There's about 340 DC volts if your AC input is 240 ...

I checked the cable and it looks ok. Nothing is shorted and continuity is fine.
I also did check shorts at the input of the power supply (PSU off) and nothing is shorted

dionb wrote on 2025-08-21, 21:39:

Good point about the leaking to ground. […]
Show full quote

Good point about the leaking to ground.

IIRC in German that would trip the Fehlerstrom-Schutzschalter. Was that what tripped, or the regular fuse (Überstromschutzeinrichtung)?

I agree that if the fuse is intact, it shouldn't be the latter - in which case we're looking for a very different failure.

Regarding discharging the caps, any decent remotely modern PSU (which an Astec device from 1998 should be) will have a bleed circuit that will discharge the primary caps within a few minutes. The formal way to properly discharge them is with a heavy resistor (2.2kOhm, 10W or similar) across the leads of the big capacitors - or use a 100W lightbulb if you have retro lighting as well as retro comptuters. Advantage to the latter method is that you get visual feedback. But if you don't have the equipment for that, simply unplug the mains power (ALWAYS unplug the mains power) and turn the PSU on (XT/AT) or keep the power button pressed for ~20sec (ATX) and that will discharge through the attached motherboard.

I have this and the Licht-Schuko 2 pops out

P.S. I am trying the PSU without the board, to turn it on just by shorting green + black wire. This worked before when I first got this PSU

That's the regular fuse, 13A by the look of it; if it trips everything on that circuit combined is drawing more than 13A. The most obvious cause would be a short in the PSU. If there is no short, the other cause would be that the circuit is overloaded already and adding this PSU - which should draw no more than 3A and far less than that if barely loaded - is the straw that breaks the camel's back.

However if that was the case, it would trip when you turn the PSU on, not immediately when you plug it in. So you still seem to have a short somewhere. You say you've checked the cable. Measurements sound good, but have you plugged it in without PSU attached, or connected to a different device? That would completely rule it out.

Assuming the cable is not the cause, one way or another the PSU must be drawing over 13A of current, or expresses differently: must have an equivalent resistance of less than 20 Ohm (as 230V/20Ohm=13A). So first thing would be to measure resistance over the live vs neutral pins on the PSU socket (where you would insert the cable). That will assumedly give you a low value. Then move step by step into the PSU and measure each point vs neutral until the resistance shoots up. Whatever the last thing was you measured before that should be the cause of the problem.

Could be inrush current of the PSU is just enough, with all the other loads, to trip the breaker.

Once again I suggest to use 100W lightbulb in series with the PSU input, and having some very small load on the PSU - up to 25W, even 10W is probably enough to make it work. The bulb should flash once when power is applied and the caps are charging. Then, with such small load and PSU efficiency being at least 50% (which it should be even at such light load), the bulb should be barely lit, if at all.

If the lighbulb is constatly lit at (close to) full brightness then the PSU is shorted on the primary side. Could be the bridge, could be the switching transistor(s), could be the X cap too.

Just an opinion...

As unlikely as it is - i've seen this breakers go bad in a way which makes them trip at way lower than rated load. They also do not like being tripped at all (not turned off by hand - specifically tripped) and usually you want to replace them once it happened a few times.

Also the worst thing you can do when the breaker is tripped is go turn it on and try again. In general it means that it is either faulty, overloaded, or something you just did went really, really wrong. You need to investigate, not just try again. It should never be tripped in any normal conditions, internal fuse in power supply should have popped if something is shorted in the power supply.

If you can find no obvious short/fault in the PSU it makes even more sense to investigate what has happened, perhaps consult with electrician and ask them to check the breakers/wiring just to be sure.

I do it in following way - and this method was working just few days ago:
1. Short the green & black pins with wire
2. Plug the power cable into extension cable which has a switch (as PSU don't have a switch itself) and into PSU
3. Press the switch on extension cable

Home fuse goes down

Ofc, if I don't short green & black, nothing happens, which tells me that the cable and PSU input is ok?

P.S. I don't have bulb to try this out actually and I would not like to mess with wires

boby wrote on Yesterday, 12:33:

I do it in following way - and this method was working just few days ago: 1. Short the green & black pins with wire 2. Plug the […]
Show full quote

I do it in following way - and this method was working just few days ago:
1. Short the green & black pins with wire
2. Plug the power cable into extension cable which has a switch (as PSU don't have a switch itself) and into PSU
3. Press the switch on extension cable

Home fuse goes down

Ofc, if I don't short green & black, nothing happens, which tells me that the cable and PSU input is ok?

OK, that's a lot more than just plug the PSU into the wall outlet...

What happens if you don't short black&green before you plug the PSU in and then short the wires afterwards?

If you don't have a bulb, use a multimeter: all that a bulb is telling you is whether there's current flowing in a highly intuitive way. A multimeter can tell you that too. If you don't want wires, things like current clamps (Zangenstrommesser) exist, but that's a bit next-level unless you intend to do this sort of thing regularly (in which case you'd better get used to wires anyway).