Atx psu 5Vstb present, no output

[adelage]

Hi guys,
I am currently trying to fix an atx psu which has 5v stb present, but doesn't turn on when shorting the pson wire to ground (I have tried with and without load), the pwm/supervisor is a est7502 (which I have replaced), I have checked every diode, every transistor and every cap and they all look fine (capacity wise, I don't have an esr meter yet), the only anomaly is the second voltage (is it called bias voltage maybe?) coming out from the standby transformer, which reads 13.5v (I believe it should be above 15/16v), any suggestion on what to check next?

[stj]

Re: Atx psu 5Vstb present, no output

check any opto-isolators

[adelage]

Re: Atx psu 5Vstb present, no output

check any opto-isolators

There's only one optocoupler for the 5Vstb line which works fine

[harp]

Re: Atx psu 5Vstb present, no output

Have you few sharp images from both side, may help for suggestions
What is the brief history of psu, suddenly dead or any past simptomps?
How do you test optocoupler?
What is voltege on pson green wire?
Is you check for short anywhere on pcb board or from any output?

[adelage]

Re: Atx psu 5Vstb present, no output

Have you few sharp images from both side, may help for suggestions
What is the brief history of psu, suddenly dead or any past simptomps?
How do you test optocoupler?
What is voltege on pson green wire?
Is you check for short anywhere on pcb board or from any output?

Unfortunately I am at home now so I can’t provide any pictures, will do it tomorrow.
The psu was pretty much new, it suddenly stopped working without any particular symptom,it is a bog standard chinese model (2 big caps in series with 2 bjt), the 5v standby circuit works fine so I didn’t need to test the optocoupler and its feedback circuit (at first I tried replacing the optocoupler and the tl431 just in case, but the readings on both sides remained identical as before), the voltage on the green wire is 5.2v (before shorting it to ground to turn it on).
I couldn’t find any short, I checked all the outputs, diodes, electrolytic caps and transistors (out of circuit) and everything seems fine. I couldn’t check with a scope, but when I try to turn it on I can read a bit of voltage on the outputs for an instant, it’s like the est7502 (which I replaced as well) ic is stopping it for some reason, I have even tried to disable the ovp and uvp shorting the first 4 pins to ground but nothing changed.

[momaka]

Re: Atx psu 5Vstb present, no output

It's not going to be the est7502 IC - those and many of the TL494 / KA7500 clones are pretty robust ICs.

The auxiliary winding on the secondary side (from the 5VSB transformer) at 13.x Volts is OK.

With no images and no further testing, I'd suspect these items first:
1) Bad cap or multiple bad capacitors on the output (low capacitance / high ESR.)
2) Bad solder joint on any of the output inductors
3) Blown BJTs on the primary - a lot of times, one or both can blow open-circuit, along with their Base-drive resistors, so you won't see any short-circuit, but PSU simply won't work.
4) Blown driver BJTs on secondary side (typically small TO-92 or TO-92L NPN BJTs) or bad protection diodes across their output. I had a case like this and the high resistance and relatively high voltage drop (though still a bit low for a diode... but I missed that clue), which made me go in circles several times before I caught this.

Once we get pictures, I might be able to provide more help. Honestly, though, most of the cheap no-name half-bridge PSUs with a ton of undersized or missing components aren't really worth fixing, except for learning purposes.

[adelage]

Re: Atx psu 5Vstb present, no output

It's not going to be the est7502 IC - those and many of the TL494 / KA7500 clones are pretty robust ICs.

The auxiliary winding on the secondary side (from the 5VSB transformer) at 13.x Volts is OK.

With no images and no further testing, I'd suspect these items first:
1) Bad cap or multiple bad capacitors on the output (low capacitance / high ESR.)
2) Bad solder joint on any of the output inductors
3) Blown BJTs on the primary - a lot of times, one or both can blow open-circuit, along with their Base-drive resistors, so you won't see any short-circuit, but PSU simply won't work.
4) Blown driver BJTs on secondary side (typically small TO-92 or TO-92L NPN BJTs) or bad protection diodes across their output. I had a case like this and the high resistance and relatively high voltage drop (though still a bit low for a diode... but I missed that clue), which made me go in circles several times before I caught this.

Once we get pictures, I might be able to provide more help. Honestly, though, most of the cheap no-name half-bridge PSUs with a ton of undersized or missing components aren't really worth fixing, except for learning purposes.

Sorry I had a couple of busy days and couldn't reply before..
Anyway here are some pictures of the board, as per your suggestions: all the electrolytic caps look fine (I only checked capacitance since I don't have an esr meter), I couldn't find any bad solder joints anywhere, and I checked all the bjts and the main diodes on the board and everything seems fine, now I will have a look again in case I missed something.
Btw, I know it's not worth fixing it, but I am just studying and learning how to fix power electronics.
Thanks for your reply.

[harp]

Re: Atx psu 5Vstb present, no output

R14, R18 whats going on there?
R18 150k looks like cutted and lifted but left purely soldered legs
R14 missing...

[adelage]

Re: Atx psu 5Vstb present, no output

R14, R18 whats going on there?
R18 150k looks like cutted and lifted but left purely soldered legs
R14 missing...

R18, believe it or not, was like this when I opened it (and the psu was bought new a few months ago), I tried to solder it but nothing changed, so I put back the way I found it, R14 was taken out to measure it when I took the picture, I am checking pretty much every single component just to find if anything is burned or shorted

[adelage]

Re: Atx psu 5Vstb present, no output

Just a quick update, I put back the resistor on r14, and I managed to test the frequency of the pwm output (pin 7 and 8), and I found out that the ic was actually switching, so I checked the main bjts on the hot side and one was actually giving some strange readings, so I replaced them both, and after that the ic stopped switching (same as the beginning) so I had the amazing idea to cut all the tracks (1 to 4) for the ovp protection pins and short them to ground, and turn it on without any isolation transformer nor incandescent bulb in series to limit the current…
You can see where this is going, not one but two amazing bangs! The second one was the fuse which literally exploded, while the first might have been one of the two bjts (which is now shorted, note that I tested them before and they seemed ok) ��! Now I am not sure if to proceed, since, even if I somehow manage to fix it, I guess that is gonna be kinda hard to understand which one was the original fault