ATX to lab bench power supply problem: turn on for a second then turn off

Thread Starter

Phạm Việt Dũng

Joined Aug 29, 2018
13
Hi all,
I just got an old ATX from scraps and wanted to turn it into a decent lab bench power supply. I've read quite many related old topics but none of them work out with my problem so I make this thread:
When turning on, it works for less than 5 sec while introducing more and more "shhhhhhhh" sound then shut off. For those next times I turned on, the working time is less, it turns off faster and faster until it shut down immediately after turning on. (still make some shhhhh sound though).

I've connected my ATX just exactly like following:

Green wire to black wire with a switch in between.
Grey wire I hook with LED and 330ohm resistor in series then go to ground (black wire).
Brown 3.3V sense with Orange 3.3V or course.
When turning on, it works for less than 5 sec while introducing more and more "shhhhhhhh" sound then shut off. For those next times I turned on, the working time is less, it turns off faster and faster until it shut down immediately after turning on. (still make some shhhhh sound though).
Can you point me out what fault and how to fix this ? or it just that my ATX is dead? Then which component should I look for replacement?
Thank in advance!
 

Attachments

DNA Robotics

Joined Jun 13, 2014
580
It Needs a load so put a 10 ohm 10W resistor between any +5 red wire and ground.

An alternative is a 12 volt automotive turn signal bulb. It will be a load and glow just bright enough to be a power indicator.


 
Last edited:

Thread Starter

Phạm Việt Dũng

Joined Aug 29, 2018
13
It Needs a load so put a 10 ohm 10W resistor between any +5 red wire and ground.

An alternative is a 12 volt automotive turn signal bulb. It will be a load and glow just bright enough to be a power indicator.

Yes, sorry I forgot to mention that I've tried to put a dummy load 33-ohm - 10W resistor between the highest power wires (+5V and GND). But, still the same result.
 

AlbertHall

Joined Jun 4, 2014
10,418
As you have the load resistor, and as the time working keeps reducing, I believe the PSU is faulty. They are not very repairable so probably your best bet is to bin it and get a replacement.
 

MrAl

Joined Jun 17, 2014
7,815
Hi,

This could be another case of bad electrolytic capacitors. When they go bad that is what happens.
If that is really it, then replacing the large electrolytic caps could solve the problem.
That happened to one of my computer power supplies and a LCD TV too. After replacing the caps, the units worked fine again.

The only way to know for sure though is to test the electrolytics, or just replace them and see what happens.
 

Thread Starter

Phạm Việt Dũng

Joined Aug 29, 2018
13
Thank you all for the replies,
AlbertHall is correct, those ATX are not very repairable and the fault could be anywhere. Looking for failure component is not easy since that ATX is quite complicated (for me).
So I guess the best choice is just to have it replaced...
 

MrAl

Joined Jun 17, 2014
7,815
Thank you all for the replies,
AlbertHall is correct, those ATX are not very repairable and the fault could be anywhere. Looking for failure component is not easy since that ATX is quite complicated (for me).
So I guess the best choice is just to have it replaced...
Hi again,

Sorry to hear that, but yes, fixing these things is sometimes a hit or miss where you cant be sure unless you have some test equipment or are willing to take a chance on your fix working out the way you hope it will. It also requires some soldering and unsoldering of small parts, and not ruining the PC board at the same time. It takes a little experience.

Power supplies are not that expensive if they are not that high in power either.
 
Top