PSU failed?

cehaem · 7 Sep 2019 at 16:58

Guys, I need some input on this one. Since last week I've been experiencing some random reboots (Kernel 41 error). Initially I was maybe getting 1-2 per day but today had at least 30. After doing the usual Windows settings stuff I finally decided to open up the case and noticed the 4 pin connector (my be quiet! straight power 11 comes with a 20 plus 4 pin connector cable) not being really plugged in. I pushed it back in but the computer hasn't come back on again. I tried the paper clip test but the PSU has stayed dead. I am running only SSD so I'm not sure it this is giving it enough load to work. Also no spare case fans here that I could utilise...On some German forum I found a post saying, that someone actually managed to do the paper clip test on that particular model..

Should I just go ahead an get a new power supply? I have no spare one lying around so there's no way for me to test. What's strange: I was monitoring the 12v rail and did not notice any power fluctuations, not even under load. I had steady 12v to 12,2v all the time so it wasn't showing any signs of dying

BriT · 7 Sep 2019 at 17:55

Does sound like the PSU is dead, whats your budget for a new one? PC spec?

keylion · 7 Sep 2019 at 18:00

I'd maybe take everything out of the case and put together a minimal setup just on the desk. Pulling everything apart and reassembling will ensure there are no bad connections plus doing it out of the case will rule out any shorting on the case or motherboard mounting points. It also lets you get close up to the components so you can see what is happening when you power it up.

cehaem · 7 Sep 2019 at 18:02

BriT said:
Does sound like the PSU is dead, whats your budget for a new one? PC spec?

I'll buy the same one - was pretty happy with this one as it was ultra silent. My question is: why did it fail since the reboots were caused by the loose cable which I eventually plugged back in? I wish I had a second system to check the components one by one.

keylion · 7 Sep 2019 at 18:07

Is the PSU still under warranty, I thought they had only been out a year or two.

cehaem · 7 Sep 2019 at 18:10

keylion said:
Is the PSU still under warranty, I thought they had only been out a year or two.

Yes, it is, I will send it back but I don't fancy waiting ages for them to send me a replacement as I need the PC for work as well.

To me, it seems like this is a PSU issue, but there are still questions marks around the symptoms that I experienced...Just before the PC stopped working for good while powering up again I got the message on the screen saying that overclocking had failed and setting were reset. Since I am not overclocking I guess this was related to the XML profile.

I was getting that message a lot just after I built the PC as the XML profiles wasn't working correcty and then, all of a sudden, they disappeared...

Drumroll · 7 Sep 2019 at 19:24

I'm not sure the system would work without the 4 pin bit connected. Not sure. Possibly a problem with the PSU.
And hope is, if need to be some faulty part, and not your Motherboard or other expensive component.

cehaem · 7 Sep 2019 at 19:33

Michael Marini said:
I'm not sure the system would work without the 4 pin bit connected. Not sure. Possibly a problem with the PSU.
And hope is, if need to be some faulty part, and not your Motherboard or other expensive component.

The 4 pin wasn't hanging loose, just not really sitting tight.

an0nym0us · 8 Sep 2019 at 01:31

Normally, the CPU cable's connection to the PSU is 8-pin.

If you did the paper clip test correctly and PSU still can't provide any power, then PSU is definitely dead. What did you do to verify the paper clip test with though? Sometimes PSUs have low power modes where its fan won't turn on unless load exceeds a certain threshold. Don't do the paper clip test whilst the PSU is connected to any precious components though.

cehaem · 8 Sep 2019 at 07:23

an0nym0us said:
Normally, the CPU cable's connection to the PSU is 8-pin.

If you did the paper clip test correctly and PSU still can't provide any power, then PSU is definitely dead. What did you do to verify the paper clip test with though? Sometimes PSUs have low power modes where its fan won't turn on unless load exceeds a certain threshold. Don't do the paper clip test whilst the PSU is connected to any precious components though.

The question still remains: IF the PSU is dead - what killed it? Me, plugging the 24 pin back in? All these restarts happening within the matter of seconds?

an0nym0us · 8 Sep 2019 at 07:39

cehaem said:
The question still remains: IF the PSU is dead - what killed it? Me, plugging the 24 pin back in? All these restarts happening within the matter of seconds?

That's going to be pretty difficult to determine.

Drumroll · 8 Sep 2019 at 10:53

cehaem said:
The question still remains: IF the PSU is dead - what killed it? Me, plugging the 24 pin back in? All these restarts happening within the matter of seconds?

The restarts are clearly the symptoms of the faulty, not the cause, I guess.
A failing part tends to show the symptoms more often until fails completely.
Despite being a proper PSU, there's a chance that you bought one unit which sadly is the margin of bad units. No product is 100% reliable. Manufacturer aim for 100%, but that's estimated, not actual guaranteed lifespan.

cehaem · 8 Sep 2019 at 11:55

Michael Marini said:
The restarts are clearly the symptoms of the faulty, not the cause, I guess.
A failing part tends to show the symptoms more often until fails completely.
Despite being a proper PSU, there's a chance that you bought one unit which sadly is the margin of bad units. No product is 100% reliable. Manufacturer aim for 100%, but that's estimated, not actual guaranteed lifespan.

I had identified the problem for the reboots which was the loose 4 pin connector on the 12v mail rail. Slowly dying PSU drop the voltage well below 12v and mine wasn't doing that as I had HVI log sensor readings while running stress tests and gaming. My PSU was in perfect health.

The question is now - what happened AFTER i plugged it back in? Did the numerious restarts within seconds bust the PSU and this just randomly coincided with me switching off the PC? Also, what's the matter with the BIOS reporting overclocking failure? Did the mobo take a beating, too, or was it just "confused" because of the frequest restart?

These are the questions I am looking for the answers to. If I had a seconds system to test (or a volt meter) I would have found out what the problem is by swapping components. But since I don't have one, I was hoping for someone to point me in the right direction as I can either go ahead and buy a new PSU just to find out afterwards that there's something wrong with the motherboard - again, the paper clip test isn't the ultimate proof that the PSU is toast - or take it to the local PC shop for them to deal with it. If I go with option 1, I may end up with two perfectly fine PSUs one of which I can't return. I'm not comfortable with option 2 either, as there guys are bloodsuckers and famous for overcharging. Plus, it would feel sort of "noobish"....

cehaem · 8 Sep 2019 at 11:57

an0nym0us said:
That's going to be pretty difficult to determine.

I know. That's why I hoping for some hard-boiled builder of even a Technician that's had dozens of systems in his test bench to point me in the right direction.

EsaT · 8 Sep 2019 at 17:39

cehaem said:
I had identified the problem for the reboots which was the loose 4 pin connector on the 12v mail rail. Slowly dying PSU drop the voltage well below 12v and mine wasn't doing that as I had HVI log sensor readings while running stress tests and gaming. My PSU was in perfect health.

You need to measure also ripple with oscilliscope to know if PSU is working properly.
Dumb average voltage meters would still show same 12V even if PSU had enough ripple to be danger even for fans.
Also other voltages are needed.
And without 5V standby PC won't even try to boot.
(though 5Vsb shouldn't have effect once PC is powered)

cehaem · 9 Sep 2019 at 15:16

It was the PSU. Will be getting a new one now.

kmetek · 4 Oct 2019 at 06:16

paper clip check is hard? I have XFX XTR 650W suspecting it's dead.

bulldog147 · 9 Aug 2021 at 06:14

cehaem said:
Guys, I need some input on this one. Since last week I've been experiencing some random reboots (Kernel 41 error). Initially I was maybe getting 1-2 per day but today had at least 30. After doing the usual Windows settings stuff I finally decided to open up the case and noticed the 4 pin connector (my be quiet! straight power 11 comes with a 20 plus 4 pin connector cable) not being really plugged in. I pushed it back in but the computer hasn't come back on again. I tried the paper clip test but the PSU has stayed dead. I am running only SSD so I'm not sure it this is giving it enough load to work. Also no spare case fans here that I could utilise...On some German forum I found a post saying, that someone actually managed to do the paper clip test on that particular model..

Should I just go ahead an get a new power supply? I have no spare one lying around so there's no way for me to test. What's strange: I was monitoring the 12v rail and did not notice any power fluctuations, not even under load. I had steady 12v to 12,2v all the time so it wasn't showing any signs of dying

I have that same issues as I can see Windows Event logs with Kernel 41 error and sometimes it shutdown itself for no reason and now my psu is dead after 5 months. Very poor from bequiet PSU.

https://www.overclockers.co.uk/foru...-power-11-gold-80plus-750w-are-dead.18935065/