Raise switched mode PSU output voltage above nominal voltage

[nour]

Hi,
I have a switched mode PSU (an old laptop PSU) that have a nominal voltage around 20V @ 4.5A, of course this is a fixed output
My question is can i modify it so I can raise the output voltage to 23V?

I Know that if I can increase the switching frequency it would increase the output voltage(correct me if I am wrong), but sometime the switching component's frequency in the circuit is fixed internally!

Is it doable !!
The reason why I want to do it because I want to power up some LEDs in series but the total voltage for them is around 22V  and  I don't want to buy a new PSU just for this specially I want around 3.5 amps to have a enough power to run the LEDs
I will try to reverse engineer the circuit of the PSU and see if I can find the switching component in it.

[joseph nicholas]

I am trying to do the same thing with an atx power supply.  You tube videos say this can be done in a limited range.  On the 12 volt rail you could change the pwm to make the output 14.4 volt to charge a car battery using a potentiometer.  I guess lowering the voltage down woun't be much of a problem.  To double the output voltage might exceed the rating on the output capacitors, and bang.

[nour]

I am trying to do the same thing with an atx power supply.  You tube videos say this can be done in a limited range.  On the 12 volt rail you could change the pwm to make the output 14.4 volt to charge a car battery using a potentiometer.  I guess lowering the voltage down woun't be much of a problem.  To double the output voltage might exceed the rating on the output capacitors, and bang.

I don't want to double it I just want to increase it around 3 volts, also I have checked the output Caps and they be just fine, they are rated up to 25V!
and also I have checked the switching IC and all I have found this

DAP6A
PNLN

I have searched the above symbols and couldn't find Datasheet but found a lot of other Laptop PSU PCB images with the same IC, so presumably it is widely used in this application

[Cliff Matthews]

Just throw in a 99-cent boost pcb and solder in a 10K panel pot. 5v in and ~6v to 37v out. It's adjustable and good for 10watts as outlined in Scullcom Hobby Electronics video #19

**In your application, use a heatsink (he mentions it too) maybe you'll get out 23v at 1amp without making it work too hard.

[nour]

Just throw in a 99-cent boost pcb and solder in a 10K panel pot. 5v in and ~6v to 37v out. It's adjustable and good for 10watts as outlined in Scullcom Hobby Electronics video #19

**In your application, use a heatsink (he mentions it too) maybe you'll get out 23v at 1amp without making it work too hard.

I don't have access for this right now and also that won't work for me at all, I need around 3.5A @22V that's obviously more than 10W 

[nour]

After doing some search I have found someone did the hard work for me, I have found the circuit diagram for the PSU, it seems like someone draw it by hand



Now I need to figure out which resistor should be changed to alter the feedback loop (assuming it has one).
Can someone help over here!

[nour]

According to this diagram the PAD6A is  the NCP1200 with is PWM Current-Mode
Controller.
Searching for the datasheet was very easy
http://www.onsemi.com/pub_link/Collateral/NCP1200-D.PDF

Now after finding the datasheet I think it is much easier to change the values!

Can someone help !

[nour]

According to the datasheet the device has a fixed internal frequency, so is there still a chance to do it?
I have no experience with switching mode power supplies so I think I am missing something 

[nour]

Yes you can modify it, a little anyway safely. The secondary side IC das001 is probably this IC.

http://www.onsemi.com/pub_link/Collateral/NCS1002-D.PDF

Lots of manufactuers make versions of those chips, the pinout seems to match Onsemis version NCS1002 it has an internal 2.5V reference. The chip is used for voltage control and current limiting by the looks of it. One opamp sets current limit pins 6,5 and7 the other one sets the output voltage. The output voltage can be adjusted by increasing R59. I'm assuming there is an error in the value of R59 it seems to be missing a zero, should be 27420. EDITTo check you can measure the voltage across R58 if it's 2.5V under normal operation (correct output voltage) then it likely is that chip. That would mean R58 and R59 set the output.

The output voltage is set by the reference voltage over R58 3831 oHms (these are pretty odd resistor values were they recorded properly?). In this case 2.5V/3831 oHms = 652uA constant current so to get your other 23V -2.5v=  20.5V you need a resistor of 20.5V/625uA =32.8k or nearest standard would be 33k new R59 value.

I just noticied according to your schematic the output caps are only rated to 25V so you might want to increase them to 35V.

I have measured the voltage across R58 and it is 2.5 exactly 
R58 is 3.83 indeed, I have removed it from the board and measured it

Now I don't understand should I increase R59 value or decrease R58 value.
I have tried to decrease R58 value to 3.6K and I have got 21.63V

 because you have mentioned in the first paragraph  increasing R59 but later in the second one you was talking about R58
I am a bit confused.

But really thanks a lot for your reply anyway you have put me on the right spot in the schematic after all.

[nour]

I have decreased R58 down to 3.3K and now I have 23.3V which I consider very good. Also regarding to the caps, the space and the volume of the box wouldn't allow me to add higher rated caps, so I will just leave them like that for now and observe any unusual behavior until I can have another caps with higher ratings and the same volume.

BTW would this affect the max. output current, this unit is capable of 4.5A @20V, and I will be using it around 3A @ 23V?

[nour]

I will use this to power UV box that I have made and in my calculations I wouldn't run it for more than 2 minutes so It will be OK .
also the caps are nichicon rated @ 105C so I think this is also going to be OK.
BTW I only have 2 caps(25V@1000uF) not 3, I don't know why the circuit diagram has 3 instead of 2!!