# ATX PSU to Bench Supply

Discussion in 'The Projects Forum' started by cobhc_420, Jun 17, 2012.

1. ### cobhc_420 Thread Starter New Member

Jun 17, 2012
2
0
I would like to build a bench power supply for doing labs at home. I was wondering what other opinions I could gather on a variable dc supply using voltage divider.

My idea is to run a Vin from the 12V output of a molex connector from the psu into a voltage divider with a 1M ohm variable resistor and a 500k resistor. (see schematic)

Keep in mind that for my lab experiments I do not see myself drawing more than 1A of current MAX.

What I am not sure of is how to make sure that I don't dissipate too much power burning out my resistors, and I'm not sure which power rating resistors to get.

Does anyone think that this would work properly before I blow myself up?

2. ### MrChips Moderator

Oct 2, 2009
12,636
3,454
What you are proposing will not work.

Here is a circuit of what a power source looks like.

Every power source has an internal resistance shown as Rs in the circuit above.

As applied to your circuit, this Rs would be the equivalent combined resistance of R1 and R2 in your circuit.

In practice, it is desirable that Rs is much smaller than the resistance of your load.

As an example, if you wish your output voltage to be 6V while delivering 1A, you would have to set the potentiometer R2 to the 50% position. Your load resistance for 6V @1A would be 6Ω.

Since Rs has to be much smaller than the load, we can use a rule of thumb that Rs has to be at least 10 times lower than that 6Ω, i.e. Rs = 0.6Ω

Hence the value of the potentiometer has to be 1.2Ω

This means that your potentiometer is drawing 10A and will have to dissipate 120W!

In summary, your simple circuit is not a good idea.

Last edited: Jun 17, 2012
3. ### BMorse AAC Fanatic!

Sep 26, 2009
2,675
234
What voltages do you want to work with? An ATX power supply has +5,+12,-12,-5, +3.3, etc, outputs......

4. ### wayneh Expert

Sep 9, 2010
12,389
3,245
Nope.

You should start reading about linear voltage regulators, in particular the LM317 adjustable regulator. Then you might want to learn how to make a variable current supply, or a higher amperage, variable voltage supply (the LM317 is limited to about 1A - plenty for a lot of bench work).

These power supply projects are probably the first ones many hobbyists get started with. There are more modern, "better" solutions that are quite a bit more advanced, but you won't regret learning thoroughly about these basic circuits.

5. ### wmodavis Well-Known Member

Oct 23, 2010
737
150
1 M Ohm is a bad choice!

6. ### wayneh Expert

Sep 9, 2010
12,389
3,245
Not if you just need nanoamps. ANY resistor is a bad choice if you need 1A.

7. ### Sparky49 Well-Known Member

Jul 16, 2011
835
417
Hi there!

As mentioned, it is not such a good idea to use a simple voltage divider as you show as a power supply. If it were good, then we'd all use a couple of resistors, and wouldn't have to pay lots of cash for power supplies.

Try doing a Google search for a variable bench power supply. There are many ICs which provide a smooth DC voltage output, and all you need to control the output is your friend the voltage divider.

Add some smoothing caps either side and you have a much better power supply. It'll cost a couple of pounds/dollars more, but you'll have a much safer and stable power supply.

Best of luck!

8. ### Evil Lurker Member

Aug 25, 2011
117
24
If all you need is a tiny supply hell a \$2 wall wart will get the job done. But if you want a true variable power supply you will need a linear regulator (for simplicity) or if you want a better higher output option a switching regulator.

9. ### cobhc_420 Thread Starter New Member

Jun 17, 2012
2
0
thanks a lot guys, i figured i would get some input from others, however i think in the long run it would just be better to go out and purchase a bench power supply

10. ### MrChips Moderator

Oct 2, 2009
12,636
3,454
While this is an easy solution, you learn more by having to design and build circuits yourself. An adjustable low voltage, low current power supply is a very doable project.

11. ### wayneh Expert

Sep 9, 2010
12,389
3,245
+1
This is a step that shouldn't be skipped by a budding hobbyist. I'll bet 95+% of the senior members here have built at least one power supply for themselves.

12. ### MrChips Moderator

Oct 2, 2009
12,636
3,454
Right. One of my first construction projects was a simple stand alone 5V@1A DC supply from scratch - transformer, rectifiers, regulator, etc..

13. ### wayneh Expert

Sep 9, 2010
12,389
3,245
My last construction was adding an LM317 and a variable resistor to a rectifier board busted out of an old copy machine. Plug in a 1A, 16VAC wall-wart, and voila, instant portable variable voltage supply.

14. ### takao21203 Distinguished Member

Apr 28, 2012
3,578
463
You can build adjustable power supply, using an OpAmp for voltage regulation + one MOSFET + VGA cooler, not too difficult, and not expensive.

But don't expect to get out more than 1 Amp.

Using resistors is good for some 10s mA of current, 1W power resistors are good for this. The disadvantage is voltage goes down under load, so the use of this is limited.

15. ### wayneh Expert

Sep 9, 2010
12,389
3,245
+1 Specifically, it's limited to low power operation for a predictable load. If your needs fit in that box, it's a quick and dirty solution. But that's a bit rare, in my experience. I have been able to use it to power some very low power ICs that needed a lower voltage than my board voltage.