In this instructable I will show you how to quickly convert one of those many computer power supplies into something that you can use to power your electronics projects, for electroplating, for electroetching, for heating wires for foam cutting, etc.
The voltages that can be output by this unit are 24v (+12, -12), 17v (+5, -12), 12v (+12, 0), 10v (+5, -5), 7v (+12, +5), 5v (+5, 0) which should be sufficient for most electrical testing. Many ATX power supplies with a 24-pin connector for motherboards will not supply the -5V lead. Look for ATX power supplies with a 20-pin connector, a 20+4-pin connector, or an AT power supply if you need -5V.
PLEASE, YOU ARE WORKING WITH AC VOLTAGE!!! IF YOU ARE NOT SURE OF WHAT YOU ARE DOING - DO NOT ATTEMPT THIS.
These instructions were originally posted by me on http://www.wikihow.com/Convert-a-Computer-ATX-Power-Supply-to-a-Lab-Power-Supply
Remove these ads by
Signing UpStep 1Harvesting the ATX power supply
2) Remove the screws (typically 4) that attach the power supply to the computer case and remove the power supply.
3) Discharge the power supply by either letting it sit unconnected for a few days, or by attaching a 10 ohm resistor between a black and red wire (from the power cables on the output side). Using a resistor will only take a few seconds to fully discharge the power supply.
4) Gather the parts you need: binding posts (terminals), a LED with a current-limiting resistor, a switch (optional), a power resistor (10 ohm, 10W or greater wattage, see Tips), and heat shrink tubing.
| « Previous Step | Download PDFView All Steps | Next Step » |












































anyway this project it's very nice and got the chance to reuse many psw that i got at home...cause of my job and passion
Otherwise, your question is not quite clear...
[This is a little old, maybe you have long given up].
In my case, with a 10ohm resistor (pulling 0.5amp) + a small lightbulb (pulling 0.3amp) I'm getting 12.35V. That might be the upper limit of this power supply, but I will check sometime with a bigger load (like 1ohm) to double check. But my 12.3V stays stable under load.
From what I've read,
You might also have a problem if your voltage sense wire isn't done properly. In my case there was an brown wire that went into an orange cable in the biggest connector of the original cables (the 24pin output). I had to make sure that brown wire was again shorted to the orange cables. Whether the brown cable exists an is shorted to the 5v or 3.3v (or in some cases even 12v) line depends on a number of things, so I just had to go through the trash to find the 24pin connection that I cut off and see what the case is.
When you say "it settles at 11.88v" I'm wondering what you mean by settles. Do you mean no-load voltage is 11.88v? The more likely problem tends to be that no-load voltage is "12.xx" V, but under load the line drops to 11v or something like that. That is primarily what the resistor is supposed to help with.
I would also try and see if you can add another 10ohm resistor in parallel, or a small incandescent lightbulb, just to see if increasing the load helps. That will double the amount you pull from the 5v line (from 500ma to 1amp).
So 1) i would try adding another resistor (in parallel, not series)
2) Is there anyway you can work around your 12.6 or 12.4v requirement?
The 12v rail is giving 12.4v (no load), the outputs of these psu's arn't regulated right?
If i use a high current voltage regulator i'l get a fairly large voltage drop, and that wont give me the full 12v i need.. the only v reg i have right now is a sharp pq12rd11, since it has low Vd, the output i get is around 11,9v.
What do you think, should i use this one and some transistors to handle more current?
Need some tips guys =)
anyway, ill powerup the stirp using just the regular 12v output
Otherwise, your question is not quite clear...
[This is a little old, maybe you have long given up].
Brilliant instructable I am new in electronic engineering and wanted a lab power supply will you please tell me that how Can we add a potetiometer(voltage regulator),rheostat, voltmeter and ammeter to calculate and vary voltage and current?
Thanks
You could do the potentiometer (variable resistor) but would need a beefy one if it isfor high current.
I actually constructed a little box (about 1x1x2 inches) with a knob on it to control voltage. This was based on an LM317 chip (check on google). Two input wires from this cube were connected to any voltage source and two output wires went out to whatever I was powering. The knob varied the voltage. Only limitation was that the maximum current needs to be less than 1 ampere and the maximum voltage was input voltage minus about 1.3 V.
So if you connected it to the +5V of the lab power supply you would get any voltage from 1.2 V to 3.6V. If you connected it to +12 V you would get a variable voltage from 1.2 V to about 10V.
Let me know if you are interested and I can post an instructable about this little cube.
I get the desired voltages in the designated ports according to my multi-meter. I can do a small load with an LED light and small computer fan and the power supply operates as it should.
But anything larger (rc motor, motor dyno) the power supply immediately shuts off. I cannot find a reason for this at all. Any ideas?
Current (in Amps) = (Horsepower x 746) / Voltage.
A 1/4th hp motor will require 15 Amps from a 12 V supply. Or, 37 Amps from the 5V supply.
=12.5 W.
So you would burn away about 12.5 W in the resistor which it should be able to handle, though you will waste more power.
Thanks for the help I put in the 10 ohm resistor and we I turn on the power suppy the resistor get kind of warm. Is that normal? But I am now getting +12v and the -12v.
i am trying to convert a pc power box into a power supply as mentioned, however, inside it looks nothing lke yours, it just has a bunch of really fat capacitors.
does this mean its not an "atx" power supply
and, what computers may i find such a powersupply in?
as in, how old must they be , as these capacitor power supplys that i keep finding, never give currents any higher than 5v, and i tried every single pin combination possible with my multimeter.
thanks
Bet the 23 Watt resistor must be incredibly massive.
my PSU doesn't have any brown sense wire, but from reprap wiki (http://reprap.org/wiki/PCPowerSupply) they said that the pin 11 of the ATX connector is the sense wire (mine is orange in color - same as the 3.3v)..I've solderered that wire with all the remaining 3.3v, but, still nothing..
connected the sense wire with +5v (red wire) & still nothing...
for info, I used up 2 working PSU for this conversion & none of it works....
checked with multimeter (using the "buzz" function), it seems, there is a connection between ground & other's output, but, the sound is very slow. I assume it's due to the capacitor connected in the circuit..
can someone highlight to me what am i doing wrong?
thx
what's the rule of the power resistor!