Power supply from ATX PSU

Status
Not open for further replies.

Jaggions

Newbie level 3
Joined
Apr 25, 2014
Messages
4
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Visit site
Activity points
43
Hi. After reading some guides on the net, I decided to make my own power supply starting from an ATX PSU. I wrote down a sort of project and I'd like to know if in your opinion it's OK so that I start making it.

Materials:
  • ATX PSU
  • Solder
  • ATX female connector
  • Drill
  • Project box
  • 10x binding posts
  • Potentiometer
  • Green LED
  • Voltmeter
  • Ammeter

Description:
The male ATX connector from the PSU is connected to a female ATX connector. We need to solder to it one +5V red wire, two +12V yellow wires, two +3.3V orange wires, two -12V blue wires and five COM black wires. Furthermore, we have to short the PS.ON pin with a COM one.
One of the +3.3V wires and one of the COM wires go directly to the green LED, indicating that the power supply is powered on. All the other wires go throughout the voltmeter and the ammeter.
Then one +12V wire and one -12V wire are connected to a potentiometer (so that it can give voltage from 0V to 24V) which outputs to a pair of binding posts.
The other wires end with eight binding posts (+3.3V, COM, +5V, COM, +12V, COM, -12V, COM) providing a fixed voltage.
And that's it; all this starting from the female ATX goes into the project box.
To make it easier to understand, I made a drawing of what's inside the project box:



What do you think? Is it ok? Would you change something?



Edit: I just noticed that the voltmeter shouldn't be placed there as it would be useless; it has to be placed between the potentiometer and the binding posts so that it measures the voltage outputted buy the potentiometer.
 

the LED needs a resistor

what is the purpose of the potentiometer?

is the voltmeter in parallel?
 

the LED needs a resistor

what is the purpose of the potentiometer?

is the voltmeter in parallel?
Does the LED need a resistor, even if it's rated exactly 3.3V? The potentiometer is used to vary the voltage of the posts next to it from 0V to 24V.

About the voltmeter and the ammeter: that setting is wrong as the ammeter would make a short circuit in the PSU. I have to put it in series. I made a new scheme:



1) Male ATX connector (from the PSU)
2) Female ATX connector
3) Shorted PS.ON pin
4) Ammeter in series
5) Ammeter alimentation
6) The - wire of the ammeter is connected to the COM wires
7) The + wire of the ammeter is connected to the ground binding posts
8) Binding post
9) 3.3V LED
10) Potentiometer, to regulate tension from 0V to 24V, connected to a -12V wire and a +12V wire
11) Voltmeter, for showing how much tension passes throughout the potentiometer.
 

Potentiometer can not supply enough current. It will burn.

Well it depends on the type of potentiometer, doesn't it? Mine it rated 24V and 4A, so I suppose that if I don't reach 4A it will just work fine (I may be wrong)?
 

What is its value? It will dump current and heat up and still not "regulate". There are other better ways to achieve this goal.
 

Does the LED need a resistor, even if it's rated exactly 3.3V?

either the LED will not light or it will burn out

better to put that LED across the 5V [or 12V] ps lines - see http://ledcalc.com/

- - - Updated - - -

What is its value? It will dump current and heat up and still not "regulate". There are other better ways to achieve this goal.

agreed - hardly the best way to do it
 

never seen a 4 amp potentiometer, please post some pictures here to share..
 

Follow the advice you have been given and also check the rating of the -12V rail of the PSU. Using it in series with the +12V probably won't work, especially if you are using the other supplies as well because the ground reference will cease to be 0V.

The potentiometer rating is probably for the entire track (or winding as for one that size it must be a wire-wound type). As you adjust it the portion of track carrying the current changes and it's rating for part of the track will be lower than the whole. You will of course have zero regulation anyway so don't expect the voltage to be stable as you change the load on it.

Brian.
 

Status
Not open for further replies.

Similar threads

Cookies are required to use this site. You must accept them to continue using the site. Learn more…