ATX to Lab Bench Power Supply Conversion

In my sophomore year of college at the University of Minnesota, I started into my main electronics classes, and needed a good power supply for working on lab projects at home. My roommate Adam told me about somebody online who had converted an ATX computer power supply into a lab bench power supply, so I decided to do the same thing. You can also check out this link for a very similar guide by their user Abizarl. I have also documented this project on my website at http://www.mbeckler.org/powersupply/ if you are interested.

Warning! There are several large capacitors in ATX power supplies, that will store a dangerous charge for a long time. Please let your power supply discharge, completely unplugged from the wall outlet, for a few days before opening it up. You can probably be seriously hurt, so please be very careful. I am not a lawyer, but I hereby release myself from as much liability as I can, for any sort of injury you sustain, or any trouble you get into.

Step 1: Background

First, a bit of background on a typical ATX power supply:

Computer power supplies are Switch Mode Power Supplies (SMPS), which use high-frequency switching circuit elements to provide a high-quality output voltage, with good energy efficiency. One side effect of this technology is the minimum load requirement that each power supply has. In order to function properly, the power supply needs at least a very small electrical load connected to it. In other words, ATX power supplies will only work if you have something connected to it. We will be using a power resistor to provide this minimum load.

Also, modern power supplies do not simply have an OFF/ON switch, they have what is known as a "soft" power switch. This normally makes no difference to the user, as the computer behaves the same, but when you shutdown your computer, the motherboard can turn off the power supply when it has finished shutting down. This requires us to add our own power switch to the power supply chassis.

To protect our circuit from accidental (and careless!) short circuits, we will install some fuse-holders and fuses, which will disconnect the circuit supply lines if too much current flows. The size of the fuses are up to you, but a 1 amp fuse will work just fine for most circuits. You really should put fuses on all supply lines.

Update: While the diagrams show fuses on all voltage rails and no fuse on the ground line, when I actually built my power supply, I was young and foolish and only put a fuse on the ground wire. It's much safer and a better idea to put fuses on all signal lines and not the ground line. Thanks to many emails and messages on Instructables about this oversight.

Step 2: Planning

Planning is the most important step of any successful project. To plan this project, I created a few images. I am going to be using four binding posts, a power switch, a fuse holder, a power resistor, and two light emitting diodes (LED's) with current-limiting resistors. The first image details the circuit connections inside the power supply, where everything will be connected

When the power supply is connected to the wall socket, but not yet turned on, it provides a +5v standby signal, that can be used by the motherboard for things like wake-on-LAN functionality. We use this signal line to indicate when the power supply is plugged in with a red LED and a 330 ohm resistor. On my power supply, this signal line has a purple wire, and is labeled "+5VSB" on the circuit board.

When the power supply is first turned on, it must go through a start-up sequence, to ensure that everything is working, and that it is able to provide stable power to the computer. When the start-up sequence has completed, it signals the motherboard by providing +5v on the "Power Good/Steady" signal line. We will use another red LED and 330 ohm resistor to indicate when the power supply is running. On my power supply, this signal line has a gray wire, and is labeled "PGS" on the circuit board.

The power resistor is a 10 ohm, 10 Watt resistor, commonly called a "sandbar", because they are usually coated with a material that feels like sand. Most power supplies need a minimum load to keep them running, so this sandbar resistor provides a constant minimum load between the +5V rail and Ground. I've heard that newer power supplies also need a load on the 3.3v rail, your mileage may vary.

In the second image, you can see the diagram for the front of the power supply. Here I have marked where the components will go, including the LED's, the binding posts, the fuse holder, and the switch.

The third image is what the power supply looks like without any modifications. You can see the various voltages I am going to use along the front edge.

Step 3: Drilling Holes

Here, I planned out and drilled the holes in the case. My power supply was a smaller form factor, (It was from a mini-tower case), so there wasn't a lot of space to work with.

Step 4: Connecting Front-panel Items

Here, I am connecting the appropriate wires to the binding posts, power switch, fuse holder, and LED's.

In an ATX power supply, there should be a wire that is used to turn on the power supply. You can see this wire (It's green) in the second picture; it is the green wire in the middle, where it says "ON/OFF" on the PCB. I connected this to the switch, and the other pole of the switch went to ground. The +5, +12, and -12 are connected right to their wires on the PCB. The ground wire is connected through the fuse holder before the binding post.

Initially, I was going to use green LED's, but I realized I had many more red LED's than green LED's, so I switched them over to reds. In the first picture, you can see the holders I installed into the front. I connected the LED's through a common resistor to ground. The LED on the left (from the front view) is a standby LED. It is lit whenever I have the power supply plugged into the wall. It is connected to the +5V standby wire on the PCB. In my PS, it's purple. The other LED is the "Power On" LED, and it is lit when I have the power supply turned on. It's connected to the "Power OK" signal wire, which goes to +5V when the power supply detects that it has stabilized the voltages. In my PS, it's the gray wire.

Step 5: Power Resistor

Most modern ATX power supplies require a small load to stay in the ON mode. I added a 10 ohm, 10 watt resistor between +5V and ground to provide this small load. It is strapped to the back wall of the power supply, where it should get plenty of air flow. It doesn't actually even get warm during normal operation so it's not a big deal.

Step 6: Finished Project

Here you can see the finished project, both with and without the cover. If you have any questions, please leave a comment and I will try to check back often to answer them. Thanks for looking, and good luck!

Keep in mind that while I built my power supply many years ago with only the ground line fused, you should put fuses on all your signal lines and leave the ground line directly connected.

Comments

[ReptilianRascal]

Hi. I'm having trouble with my version of this. I'm trying to use my psu to power several hard drives. I've connected a 10ohm 10W resistor to the 5v and ground wires and the psu powers on ok when nothin else is connected to it (I don't have anything on the 3.3V lines and I'm not using leds on the power and standby circuits - also this is an older psu with molex style power connectors, only has 1 sata power connector). However, when I have 1 or more hdds connected to the psu when I power it on, it turns off again after only 1 second. Any help would be greatly appreciated.

[tschuld]

what is the maximum power you get out of it? my conversion can't supply 1/4th the power it is rated to

[matthewbeckler]

Hello, sorry for the delay. Each voltage "rail" provided by a power supply has an individual max output rating. Together these power ratings should add up to the total PSU power rating. Additionally, some modern power supplies have multiple +12v rails each with their own max output rating, and it's not obvious which wires belong to which rail, so I'm really not sure how to figure out the max in that case.

What are the details of your situation? What does the PSU label say are the max current ratings for +5v and +12v (or whichever voltage you're using)? How are you determining that it can't even supply 1/4th of the power it's rated?

[tschuld]

PS lists the (only) single 12v rail at 45 amps and I am trying to drive a ham radio that has a max power draw of 12.5 amps. did you test yours to see if you was getting max power?

[matthewbeckler]

I did not test mine at such large currents. I typically used under 1 amp. Maybe the radio is actually expecting automotive "12 volts", which is really more like 14 volts?

[JackstandJohnny.]

Hi. Fellow HAM here. These ATX power supplies work up to 12v. HAM radio transceivers need automotive 12v which os actually 13.8v. I think it is possible tomg et this from an ATX, but its not shown here and Im not yet sure how to do it. I believe you will need a step up or buck converter circuit. They are on ebay for super cheap. Around $3. Any updates?

[tschuld]

since then i've done 3 other atx power supplies, and the power hog radio does transmit full power on 12v.

I am betting there is a voltage regulator in the radios that defeats the 13.8, which is only needed if you was using old analog supplies that couldn't supply max current well. more testing in the future for me

note, the one power supply i had appears to have a inrush current cuttof set to high, and that is why it failed on my radio, even though it was supposed to supply 4 times my transmit current

[matthewbeckler]

I don't think I ever tested my PSU's current capacity. I have heard that some PSUs have a "short circuit detection" where a sudden increase in power draw is interpreted as some sort of short, and the PSU will shut off automatically. Is that what you're seeing? Some people here have suggested connecting your load to the PSU, and then turn on the PSU so the entire load is applied at startup, to avoid that short-circuit protection shutdown. Maybe other people here know more?

[pmondal2]

hello sir my atx power supply gets different voltage instead of 3.3v it gets 5.20v,5v it gets 5.70, 12v it gets 13.49v does my atx have fault in it or am i doing something wrong also i'm beginning in electronic so please help me. Another thing is i connect the load 5w 22ohms because i made this few months ago and i find then this load now i see you have used 10w 10ohms do you think my voltages are different because of this. Any help will be appreciated, thanks.

[matthewbeckler]

Some power supply outputs will rise higher than their specification if there is nothing connected, so be sure to attach a load (small light or fan or something) to ensure that the voltage measurement is accurate.

Your 22 ohm resistor will draw about half as much current as the 10 ohm resistor. I doubt that is causing the voltage differences (see above for a possible explanation).

[NickL170]

First off your resistor, 5v @ 22ohm will dissipate about 1.5 watts so you will be fine with that load resistor. You may need to load the 3.3 rail too. But before you do anything make sure your multimeter is properly calibrated but testing on some known voltage sources.

[dalbright1]

Just a word on capacitor discharge. Letting the power supply sit normally won't discharge the circuit. Most modern circuits have a discharge safety circuit in them these days at least in guitar amplifiers I've found. However it is very common for tube amps not have them for fear of sound hindrance. I built a discharge tool from a few alliator clips a large resistor and a led light. In my 100 watt peavy amplifier read 386 volts after sitting unused for 3 months. A handy way to discharge circuits that are switched is to switch the power on without power supplied. Won't work for all circuits but does on some. An ARC supply is either plugged in or not so I would imagine it will discharge itself but it is always imperative to always check the circuit is discharged. 386 dc volts like in my amplifier is extremely dangerous.

[LeanderD4]

Nice project. Easy to follow steps. All was working nicely while I still had the whole thing open. When I put everything inside the case, I guess my aluminum power resistor made contact with 2 heat sinks. When I turned it on again this gave some nice sparks and resulted in a burned out fuse on board of the ATX-pcb and a power resistor reading 0ohm resistance.

So a tip to everyone following these instructions: Make sure your power resistor doesn't connect any other parts, best solution I think is to mount it on the casing and make sure it doesn't touch anything else.

[mspencer8]

if i want to put more binding post do add more specific current for my 5volts and 12 volts (like 1A, 2A, 5A) what should i do? thanks..

[matthewbeckler]

Hello, good question! If you want to limit a specific binding post to only output a certain amount of current (1A, 2A, 5A, etc) then you need to use a fuse that will blow when the current limit is exceeded. Be aware that fuses are not very precise in terms of current limits, and there is usually a range where the fuse may or may not blow.

Hope that helps, let me know if you have further questions, or if I didn't understand what you were asking.

-Matthew

[JackstandJohnny.]

Dimply installing a fuse to the rated amperage you desire will not make that power ouput the amperage you desire. Thats not how fuses work. A fuse is there to kill current when you exceed a certain amperage. Fuses cannot change the amperage of a hot wire. Without using some kind of converter circuit and if you need less amperage than the sum of all the wires tied together, it might be possible to only tie together a certain number of wires, read the amperage with a DMM, then keep trying fewer or more wires until the amperage is close enough to what you require. Bottom line, you cant use a fuse to adjust amperage. Only to limit it by having the fuse blow when the amperage is exceeded.

[JohnathanW15]

Matthew, have you been on ebay recently? CC/CV boards cost 5 bucks for a stated 5 amp boost or buck...reality is 2-2.5amps unless you put some metal fins or odd chunk laying around on them. .One can get a Poly fuse multi value kit for dirt cheap as well.

need more amps? If they have roughly double the budget have them search ebay for ltc3780. 10-12 bucks and a month wait to get one from china using an IC from Linear Tech in Cali.

or you can get a board from an american based ebay shop in a week for about 23 bucks.

worth noting:

its auto regulating boost buck...the chip itself can do just about any form of power converter by re configuring (sepic, cuk etc )

I know all this because im prototyping a product and also keep a bunch of those cheap no name Dc-DC converters on hand for what ever comes up

I got my first 2 3780s for free but I didnt pay attention to the fact that the package was designed in imperial and ebay sop to dips apear to only be made in the 2 common metric pitches. I found one company that wanted like 12 bucks for the damn break out... so i paid the 23 to get the damn premade board asap and I will have 2 chips once I etch the second version at home.

[ksan1]

hi Matthew

currently my ATX power supply have 12v with alot of Amps power...
i wonder it can charge my deep cycle battery 12v 9A or parallel connection with 12v 54Ah.

can you show us a diagrams that 12v have 1A , 2A, 5A , 10A ?

-Kokoro San

[matthewbeckler]

Hello Kokoro San. Be aware that many ATX power supplies have 12v split into 2 or more sub-supplies. Look at the sticker on the side of the ATX power supply for details. I would suggest using a battery charger to manage the charge process.

I am not sure what you're asking about for diagrams?

[ksan1]

Thx for reply Matthew Currently.. my ATX power supply on yellow color wire output is 12v 20A I wonder it able to make it 13v 3A to charge deep cycle batterie