PC Power Supply High Voltages Problem!! Need Help & Suggestions

Thread Starter

OhAffex

Joined Feb 25, 2017
1
Today for an odd reason I decided to go onto my PC Bios just to change some boot settings. I came across the Power Supply Voltage and these were the readings
3.3V: 3.35
5V: 5.775
12V: 24.732

Both the 12V and the 5V readings were in red (which caught my attention) and the 3.3 was in white (which meant it was normal). Do you guys recommend I should keep using my PC? or should I replace the PSU Asap. Also can it Damage any other parts of my PC? I dont know the exact PSU Model (Since I use a Pre-Built PC) but I do know that it is a 300W power supply. The Model of the PC is "Asus M32CD-CA018T"
 

ian field

Joined Oct 27, 2012
6,536
Today for an odd reason I decided to go onto my PC Bios just to change some boot settings. I came across the Power Supply Voltage and these were the readings
3.3V: 3.35
5V: 5.775
12V: 24.732

Both the 12V and the 5V readings were in red (which caught my attention) and the 3.3 was in white (which meant it was normal). Do you guys recommend I should keep using my PC? or should I replace the PSU Asap. Also can it Damage any other parts of my PC? I dont know the exact PSU Model (Since I use a Pre-Built PC) but I do know that it is a 300W power supply. The Model of the PC is "Asus M32CD-CA018T"
AFAICR: 5.775V is out of spec. I'd stop using it immediately - the main filter/smoothing electrolytics can develop high ESR which often interferes with proper regulation. One that most people miss; is a small electrolytic somewhere near the SMPSU chip - it samples the error voltage for the chip - last time that happened to me; an expensive graphics card sacrificed itself to save a relatively cheap motherboard.

That died later....................
 

RichardO

Joined May 4, 2013
2,270
My guess is that the only voltage that is highly regulated is the 3.3 volts. This is because it measures only about 1/2% high. Typically the other voltages not directly regulated. They track the regulated voltage because they are created from the same transformer. If the 5 volt and 12 volt supplies are lightly loaded they could easily be slightly high in voltage.

Wait a minute. After re-reading your post I see that you said the 12 volts was 24 volts. That is a definite problem if true. I suspect it is not. As Dick says, the only way to know for sure is to measure with a meter.
 

ian field

Joined Oct 27, 2012
6,536
My guess is that the only voltage that is highly regulated is the 3.3 volts. This is because it measures only about 1/2% high. Typically the other voltages not directly regulated. They track the regulated voltage because they are created from the same transformer. If the 5 volt and 12 volt supplies are lightly loaded they could easily be slightly high in voltage.

.
On the last ATX PSU I stripped and hand traced; the 5V and 3.3V were summed together by a resistor network into the TL431.

That could well have been one of the least good designs on the market, but they do depend heavily on tight coupling between secondary windings.

24V on the 12V rail with the other rails only slightly over doesn't look like a regulation fault as such - it sounds like dodgy electrolytics, but hard to nail it exactly without having it on the bench.
 

recklessrog

Joined May 23, 2013
985
I recently repaired a computer that would sometimes take a few minutes to start. When I looked inside the psu, the pcb had dark heated areas in several places indicating it had been running hot, the plastic sleeve on a couple of capacitors looked like they had shrunk due to heat. As it was a reasonably good pc, (Dell) I was dubious of it being just a real cheepo psu, so I looked for the cause of the hot running.
First off, the computer was very dusty and a good vacuuming sorted that. I ran it up on the bench and saw that the psu fan was turning slowly, because the bearings were seizing. A drop of oil, and it freed up. Even though it was now running ok, it was obvious that the components, especially the electrolytic's had suffered and all needed replacing. Whilst I could have done that, I didn't really trust it to not fail at a future date. As I had several very good condition power supplies lying around, I selected a slightly higher power one and fitted that in the pc which now runs like new.
On returning it to it's owner, I found out that he had it located under a desk and pushed hard back against a wall in a corner restricting the airflow. Just pulling it away from the wall a few more inches lowered all the temperature readings inside. I use Speccy by Piriform, an excellent free download that tells you a lot about what is going on inside.

Having spent some years in the service industry, the old saying "If in doubt, chuck it out" have proved to be correct many times over.
If you are anyway worried about your power supply, replace it, they are so cheap it's hardly worth the time and effort repairing it. At least doing that will give you peace of mind.
 

ian field

Joined Oct 27, 2012
6,536
I recently repaired a computer that would sometimes take a few minutes to start. When I looked inside the psu, the pcb had dark heated areas in several places indicating it had been running hot, the plastic sleeve on a couple of capacitors looked like they had shrunk due to heat.
.
That's a classic sign of a knackered electrolytic - the ripple current develops voltage across the ESR and produces a lot of heat.

Once I had an ATX PSU come in with ashed coating on the big toroidal secondary choke - on closer inspection, I found the slots punched in a heatsink didn't line up with the slots on the outer casing it was bolted to - it had practically no airflow.

It was hard work filing out the slots so they matched up, then I had to rummage through the "graveyard" for a choke more or less the same physical size - then it took a couple of hours trial and error adjusting the turns to get the output voltages just so.
 
Top