Increasing capable SMPS from rated 12V ouput to 20V via opto-coupler feedback

Thread Starter

Neddan

Joined Mar 6, 2022
15
Hello everybody.

I am trying to increase my power Supply's default 12V 3.3A output to something like 20V 3A. I Decided to use a potential resistor via the shunt regulator at the opto-coupler which i was sure it had. But the problem is that when i opened the power supply, most of the output side was significantly just SMD components that i am honestly not used to and not confident to work with. Things got very wrong very quickly when i realized i couldn't directly understand what the SMD part of the output was doing.
So at this point i need some help from my fellow colleagues in decrypting the circuit of the power supply and possibly suggesting your opinions on what is the best way to change the output voltage without dramatically over-engineering the entire thing.
Ive attached some camera shots of the board bellow. The PSU itself was a dstv explorer decoder power supply that i salvaged and when i tried to google the schematic, i got nowhere. Thus why i ended up here.
I appreciate any assistance in advance!
 

Attachments

Alec_t

Joined Sep 17, 2013
14,280
Welcome to AAC!
Are you sure the output capacitor(s) would be able to handle 20V plus any ripple safely?
If the supply is rated 12V 3.3A then even if you could do the conversion you'd not get more than 3.3A x 12/20 = 1.98A out.
 

crutschow

Joined Mar 14, 2008
34,281
Without a schematic it's pretty difficult to tell you how the circuit works, or how to modify it to do what you want (or even if it's feasible).
 

Papabravo

Joined Feb 24, 2006
21,159
In terms of time and effort you might be looking at a factor of x20 or x100 as opposed to buying a power supply to meet your needs. In essence you would need to redesign the whole thing to get more power out than it is currently designed for. There was a time, oh three quarters of a century ago when labor and materials were cheap, while finished goods were EXPENSIVE. That is so manifestly not the case anymore.
 

Thread Starter

Neddan

Joined Mar 6, 2022
15
Welcome to AAC!
Are you sure the output capacitor(s) would be able to handle 20V plus any ripple safely?
If the supply is rated 12V 3.3A then even if you could do the conversion you'd not get more than 3.3A x 12/20 = 1.98A out.
Yes the output capacitators are rated for 25V, and I realized that its rated for 45W so the current ratio wont really matter for my purposes, only the voltage.
 

Thread Starter

Neddan

Joined Mar 6, 2022
15
Without a schematic it's pretty difficult to tell you how the circuit works, or how to modify it to do what you want (or even if it's feasible).
Yeah exactly LOL! I am trying to reverse engineer the schematic but if it gets too un-economical then i will likely drop it.
 

Thread Starter

Neddan

Joined Mar 6, 2022
15
In terms of time and effort you might be looking at a factor of x20 or x100 as opposed to buying a power supply to meet your needs. In essence you would need to redesign the whole thing to get more power out than it is currently designed for. There was a time, oh three quarters of a century ago when labor and materials were cheap, while finished goods were EXPENSIVE. That is so manifestly not the case anymore.
Yeah, buying another power supply might be cheaper indeed. But i wanted to at least make use of this old brick if i could. Plus learn a few extra stuff along the way too.
 

Papabravo

Joined Feb 24, 2006
21,159
Yeah, buying another power supply might be cheaper indeed. But i wanted to at least make use of this old brick if i could. Plus learn a few extra stuff along the way too.
Nothing wrong with dissecting it to learn something, and keep it in tip-top shape.
 

Thread Starter

Neddan

Joined Mar 6, 2022
15
So i managed to trace the shunt regulator circuit and located an unidentifiable 3pin SOT-23 chip labeled "4LBJT", or at least I hope thats what it says coz i cant find any references on google. However it seems to be wired to the opto-coupler promisingly. Pin 1 seems to be tapped off a voltage divider network with 30k ohms and 6k ohms approximately to ground. Pin 1 reads 3v and pin 3 11.9v. At this point i am confident its the IC.
But what i want to know is the relationship between varying the 30k resistor up and down and the voltage output i get. If possible i want to use a fixed resistor value rather than a variable resistor.
 

Ya’akov

Joined Jan 27, 2019
9,069
So i managed to trace the shunt regulator circuit and located an unidentifiable 3pin SOT-23 chip labeled "4LBJT", or at least I hope thats what it says coz i cant find any references on google. However it seems to be wired to the opto-coupler promisingly. Pin 1 seems to be tapped off a voltage divider network with 30k ohms and 6k ohms approximately to ground. Pin 1 reads 3v and pin 3 11.9v. At this point i am confident its the IC.
But what i want to know is the relationship between varying the 30k resistor up and down and the voltage output i get. If possible i want to use a fixed resistor value rather than a variable resistor.
It seems to be this LDO but I am not sure how that works according to your pin description.
 

Dodgydave

Joined Jun 22, 2012
11,284
This is what i think you have, it's a TL431 zener with the usual two resistor setting, R45 and R35/R38 , so to increase the output you need to lower the combined resistance of R35/38 so the zener fires later.
You can do a simple test with a volt meter and measure the gate voltage at the junction of the 3 resistors G it will be 2.5V.
I don't think it will give you 20 V maybe 16V,


There may be an extra over voltage protection installed as well. 16467738748132967761402892945720.jpg
 
Last edited:

dendad

Joined Feb 20, 2016
4,451
Just note, if the supply is 12V at 3A, you will not get 20V at 3A out, but maybe 1.5A or so as the max power transfer will remain the same, or most probably drop quite a bit more as you will be running it well outside its design limits.
If you want a 20V supply, old laptop 19V power bricks can usually be picked up for almost nothing and that would be a better one to start from.
Keep your 12V, (or lift it to 13.8V) to run a car radio or something like that.
 

Thread Starter

Neddan

Joined Mar 6, 2022
15
This is what i think you have, it's a TL431 zener with the usual two resistor setting, R45 and R35/R38 , so to increase the output you need to lower the combined resistance of R35/38 so the zener fires later.
You can do a simple test with a volt meter and measure the gate voltage at the junction of the 3 resistors G it will be 2.5V.
I don't think it will give you 20 V maybe 16V,


There may be an extra over voltage protection installed as well. View attachment 262326
Yes, it seemed so to me too. Modifying the resistors increased the voltage to 15V however i was naive because i should have scrapped the rest of the smd components before playing genius with it Lol. Turns out it had some complex overvoltage protection circuitry in conjunction with the regulator circuit i am used to (duh, the thing was an adapter for an expensive PVR decoder. That should have raised my alarm lol). Ended up killing the switching transistor when i pushed it beyond 15V.
So i am still puzzled as to why that happened, could it be that Modifying the voltage didn't reduce the current to maintain the 45W rating and thus pushed the adapter beyond 45W??? I am still clueless. But at least i know now that next time i tweak anything else . I should have all the factors under my control.
Overall, it was a very educational experience for me.
 

Thread Starter

Neddan

Joined Mar 6, 2022
15
This is what i think you have, it's a TL431 zener with the usual two resistor setting, R45 and R35/R38 , so to increase the output you need to lower the combined resistance of R35/38 so the zener fires later.
You can do a simple test with a volt meter and measure the gate voltage at the junction of the 3 resistors G it will be 2.5V.
I don't think it will give you 20 V maybe 16V,


There may be an extra over voltage protection installed as well. View attachment 262326
Also, i forgot to mention that on the 3 resistors junction, the voltage read 3V exactly.
 

Thread Starter

Neddan

Joined Mar 6, 2022
15
Just note, if the supply is 12V at 3A, you will not get 20V at 3A out, but maybe 1.5A or so as the max power transfer will remain the same, or most probably drop quite a bit more as you will be running it well outside its design limits.
If you want a 20V supply, old laptop 19V power bricks can usually be picked up for almost nothing and that would be a better one to start from.
Keep your 12V, (or lift it to 13.8V) to run a car radio or something like that.
Yeah, right now wanna scrap out some old laptop adapters. I did get 15V though from the adapter which i think was as much as it could handle
 
Top