I see. Two TL431's. I'm guessing it uses two feedback loops. One for the 5-volt and one for the 3.3-volt rails since they are powering the most voltage systems.

I've never tried to modify one of those before.

At worst if you can't trace the circuits out you may just have to move your resistor voltage dividers for the TL431's over to their own dedicated places.

 

I see. Two TL431's. I'm guessing it uses two feedback loops. One for the 5-volt and one for the 3.3-volt rails since they are powering the most voltage systems.

I've never tried to modify one of those before.

At worst if you can't trace the circuits out you may just have to move your resistor voltage dividers for the TL431's over to their own dedicated places.

Thanks for your answer. Well the problem is that I'm having a hard time finding those resistors. Since nothing seems to go to the resistors on top of the board only the SMD ones under and there are sooo many of them. Another thing that i found was this: http://www.transmission1.net/viewtopic.php?t=17983

He has the same brand of power supply as i do, so i tried what he did, Found the trace from the 5v and then remove a jumper, connected a pot (1k) like he did, but the psu would not power up no matter how i turned the pot so i guess something is different.

Update: I just checked the there is continuity between dc ground and the middle pin of BOTH TL431, The middle pin is the anode. Here is the datasheet for it (KIA431A) to be exact. http://pdf1.alldatasheet.com/datasheet-pdf/view/69358/KEC/KIA431A.html

 

A small update. I found two SMD resistors, one end that goes from an octocoupler, and the other goes directly to the REF pin of the TL431.

R935 and R933 is the two resistors. The octocouplers is the one in blue.

Edit: Dammit.. i just noticed enother TL431 behind the heatsink covered in that yellow gunk.. So there are 3 of them in total.

 

I'm still looking for a solution to this.. No one else has a tip on how to find the divider network? Getting hard with so much SMD stuff.

 

That's my whole problem with SMD. It's too frigging small for me to accurately trace or steal componentes off so I have very limited use or want for it in general.

The vast majority of the SMD built boards I scrap get the bigger items like capacitors and non SMD components striped and the rest goes in the garbage.

 

Your atx will be using an opto coupler to sense the voltage feedback on the Uc3843 chip pin 2, with a Tl431, if you can retrace the circuit around this pin the Tl431 uses two resistors to set the firing point for the opto.

Looking at your pictures in the first post, there are 3 optos under the heatsink.

http://danyk.cz/s_atx_en.html

 

That's my whole problem with SMD. It's too frigging small for me to accurately trace or steal componentes off so I have very limited use or want for it in general.

The vast majority of the SMD built boards I scrap get the bigger items like capacitors and non SMD components striped and the rest goes in the garbage.

I agree, it's so annoying sometimes..

Your atx will be using an opto coupler to sense the voltage feedback on the Uc3843 chip pin 2, with a Tl431, if you can retrace the circuit around this pin the Tl431 uses two resistors to set the firing point for the opto.

Looking at your pictures in the first post, there are 3 optos under the heatsink.

http://danyk.cz/s_atx_en.html

I will try this again then, Thanks for your answer, About the link, should i look at some particular supply or did you just share the link in general?

 

The link i gave has atx psus using the Uc3843 chip, the circuit maybe similar to yours.

 

The link i gave has atx psus using the Uc3843 chip, the circuit maybe similar to yours.

I see, gonna take a look at thanks and report back when i try tracing it some more, thanks!

Btw if you look at this picture:
http://danyk.cz/s_atx06h.png

From pin 2 of the UC3842, it goes directly between two resistors (R12 and R15) is that the divider network?

 

Yes, its getting a fixed voltage from pin8, that circuit is regulated by the opto pulling the Vsense pin to the Vref supply to shut it down.

 

Not to be nitpicking but those PSU's have 12V, 5V, 3.3V OVP protection on secondary side.
Don't you think those will kick in ?

 

Not to be nitpicking but those PSU's have 12V, 5V, 3.3V OVP protection on secondary side.
Don't you think those will kick in ?

Yes they have OVP protection, but disabling that is not hard at all since it uses a dedicated chip for all the protection, On second hand it can of course be dangerous to disable that, But step one is for me to find the divider network and be able to increase the voltage, then i can look at the protection.

Yes, its getting a fixed voltage from pin8, that circuit is regulated by the opto pulling the Vsense pin to the Vref supply to shut it down.

Alright thanks!

 

Update:

The pin in red is the pin 2 of the UC3842, and the two resistors in blue is the two resistors that pin 2 goes between so my guess is that those two is the divider network for the UC chip (since it's on the primary side i won't mess with that). just wanna report some progress!

 

If that's the case, follow pin 2 it should go to an opto coupler, that will have a Tl431 on the 12v line,,, or you can alter the resistors on pin 2.

 

If that's the case, follow pin 2 it should go to an opto coupler, that will have a Tl431 on the 12v line,,, or you can alter the resistors on pin 2.

Great thanks, Can i just replace the pin 2 resistor with a pot to start with? or will that be dangerous since it is on the primary side?

Also, the layout is very similar to the picture i showed: http://danyk.cz/s_atx06h.png

If that is the case, which is the resistors i should alter? the R36 and R37 after the TL431? Thanks

 

You could lower the resistor on pin 2 to ground, this will alter the feedback loop, and increase the output voltage, or lower the resistors R36/37,, but if there is a shutdown OVP chip that will need to be disabled.

 

You could lower the resistor on pin 2 to ground, this will alter the feedback loop, and increase the output voltage, or lower the resistors R36/37,, but if there is a shutdown OVP chip that will need to be disabled.

Thank you very much! i removed the R36 and 37 and i can finally adjust the voltage!! But on my ATX it was a 115k resistor, i removed that completly and added a 1Mohm pot, i can adjust the voltage betweel 11.8 - 12.99 volt, when i reach 13 volts the PSU shuts down (OVP). Gonna try to remove that ovp protection and replace the caps for higher voltage ones. Also when i adjust the voltage, it changes on the 5v line too, so have to replace them also.

Update: just noticed that my ATX didn't have a single chip for the protection which i thought it would: http://danyk.cz/at_atx_en.html

If you check that sight and then "Forward converter supply (single MOSFET) mod "
He said: " This is done by shorting the emitter and collector of one of optocouplers (there are a total of 3-4). This is the optocouplers, which provides the standby function."

Does it matter which one i short or can i short anyone?

 

You can only short out the led side of the opto, the one that feeds pin 2/8...

 

You can only short out the led side of the opto, the one that feeds pin 2/8...

Thanks, just to be sure, you do mean like this:

right? The guy said "Optocoupler, which turn on the supply from the standby mode. Short the emitter and collector together (they are on primary side). " I tried doing that but the supply wouldn't power up at all..

 

Short out Q2 c/e terminals, this will keep the chip power on and prevent shutdown.

As for the opto led, yes this is the one that regulates the voltage.