IC2 is feedback.
IC3 is the ON/OFF

 

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.

I see, Q2 goes to pin 8, danyk never mentioned that.. he only said "This is done by shorting the emitter and collector of one of optocouplers" not any transistor.. thanks.

 

IC2 is feedback.
IC3 is the ON/OFF

Oh, thank you! but I've tried shorting out every octocouplers emitter/collector one by one. and it does not matter which one i short ut, the supply won't power on at all..

update: my bad, i missed one. now it starts!! wohoo

 

Was it working before you decided to modify it ?
And if after modification did you disable OVP's

 

I would try to trace the circuit out around the Uc3843 chip, and see if it compares to the circuits i posted...that way you will have a better idea how to modify it!!

 

Thanks for all the help, i have now disabled the ovp, and i can increase the voltage up to 15v without shutdown (i didn't go any higher since i haven't replaced the caps yet). But i have a problem, i cannot go lower than 11.8 volts.. i have a 1Mohm pot connected, if i measure the resistance when i have it mounted to the circuit, it shows a maximum of 1.8kohm instead of 1Mohm.. any idea why i can't go lower than 11 volts?

Update: Just replaced all the caps that was rated for 16 volts to 25 volts ones, gonna try to increase the voltage to around 18 volts. Will report back.

 

Update once again: I switched out ALL the capacitors, the 6.3 volt ones to 16v, and the 16v to 25 - 35, I have now stress-tested the supply, 19.5 volt output, with 8.2A (160W). And it is VERY stable (as far as i can tell). the voltage drops with about 0.2 volts under that load which is not a problem for me.. Now i just need to put a crowbar circuit after the output since the ovp is completly off.. Thanks to everyone who have helped me!! Regards.

 

How about modify the OVP to handle your voltage ?
Since you have gone this far, it would be nice.

 

How about modify the OVP to handle your voltage ?
Since you have gone this far, it would be nice.

Hmm i didn't think of that, Definetly going to try it, I'm used with linear power supplies, pretty new with this whole switching thing hehe, Will report back in an hour or so!

 

The Ovp will be using a Tl431 to fire an opto, or it may use another chip to do the same job....

 

The Ovp will be using a Tl431 to fire an opto, or it may use another chip to do the same job....

Thanks then I'll know what to start looking for.

Btw do any of you know if the capacitors is the only thing i have to "worry" about when changing the voltage? For example if i replace all the caps to 50v, could i increase the voltage to like 40v or do i need to replace other stuff that can't handle that high voltage? if so what?

I would also like to decrease the shunt resistor, Since I'm running at a higher voltage, i want to limit the current a bit more, right now the shunt resistor is at 0.15 ohms, i don't really know what the limit is with that.. So does anyone knows how to calculate it? For example if i replace that with 0.27 ohms, what will the limit be? (the shunt is located on the primary side)

 

You cant limit the current by putting a shunt resistor in series with the output, an atx psu monitors the current in the pulse transformer, and then uses that to switch off the oscillator on the I-sense pin.

As for the capacitors, all the Atx psus i have made variable will only go upto 33V at maximum, so i use 63v capacitors for safety.

 

Atleast use caps with voltage that is 10% higher than the nominal voltage

 

You cant limit the current by putting a shunt resistor in series with the output, an atx psu monitors the current in the pulse transformer, and then uses that to switch off the oscillator on the I-sense pin.

As for the capacitors, all the Atx psus i have made variable will only go upto 33V at maximum, so i use 63v capacitors for safety.

I don't mean in series with the output.

Danyk said this:

Was he wrong? Or did i get something wrong? And thanks, then i will replace the caps with even higher voltages to be on the safe side.

 

Current limit on your diagram is provided by R7, increasing it lowers the current output and vice-versa...

http://danyk.cz/s_atx06h.png

 

Current limit on your diagram is provided by R7, increasing it lowers the current output and vice-versa...

http://danyk.cz/s_atx06h.png

Yes and that is the one i meant to change. But in my case it was 0.15 instead of 0.22.

 

Btw, did anyone know how to calculate the shunt resistance? If i want to limit the current to let's say 15 Amps. what does the value need to be? Thanks.

 

According to the datasheet, Isense pin needs 1V to shutdown, so your resistor needs to be 67mOhms (0.067)

 

According to the datasheet, Isense pin needs 1V to shutdown, so your resistor needs to be 67mOhms (0.067)

Awesome thank you! Sounds weird tho, i thought that the resistance had to increase for the current to lower..

 

Resistance increase causes more voltage drop across it causing the output current to decrease.