Aug 30, 2019
16
1
15
Hello!
Last night I've tried to update my GPU's (RX 560 4GB) driver, when suddenly the computer froze down fully. After a long waiting time I shut down with the power button on the case, then it didn't want to turn back on. When I turn the PSU off and back on, the PC starts (LEDs are on, fans are spinning), but a few seconds later it shuts down, and won't turn the LEDs and fans back on until I turn off and back on the PSU.
About 2 hours earlier from now I removed everything from it, and reapplied thermal paste to the CPU, then I managed to fully turn the PC on, but when I opened the Opera, it shut down itself again, and does the same thing that I mentioned already.
What should cause the problem? A friend of mine said that it's my PSU, but it's worked perfectly until now.

Mobo: ASRock AB350M Pro4 R.20 (new, 1 week old)
CPU: AMD Ryzen 3 1200 (new, 1 week old)
RAM: 2x8 GB 2400MHz Kingston HyperX (new, 1 week old)
GPU: MSI RX560 4GB
HDD: Toshiba DT01ACA050
PSU: nBase nPower 500W
Case: DeepCool Earlkase RGB (new, 1 week old)
Windows 10 1903

Some more things I thought about can cause this to me...:
-When I was installing the new components to the case, I had to forced shut down a lot of times (had to turn off/on with the power button on the case), and maybe it damaged something inside the case? But on a lot of forums and sites people says that it can't damage hardwares, maybe just some files won't be saved.
-The PSU. An another friend said this brand is a garbage, and so the PSU, and it can give only max 350W, which isn't enough for me, he said. But for days was it okay? And now it isn't?


Thanks for any replies!
 
Solution
Should I try with the PSU from my dad's PC? That one is only 420W, so I'm planning to try in my PC without GPU, and with just on module of RAM.
Oh, and one more thing. I tried to turn on without the GPU, and shut down itself that time too. Or if the GPU's driver would be the thing that causes my problem, it would still not turn on?
Well if it is also a poor quality PSU, the same issue could replicate.

It's also not the overall 420W that powers the PSU, only the 12V rail, and poor PSUs tend to have less available on this rail, so it says 420W, but in reality, if it is poor, it could actually be 350W continuous load, or the 12v rail could only supply 300W for example.

So you can try, but only for testing, and verify that the PSU...

PC Tailor

Illustrious
Ambassador
Welcome to the forums my friend!
A friend of mine said that it's my PSU, but it's worked perfectly until now.
Common misconception unfortunately, PSUs fail at any time, and PSUs also degrade over time. Every PSU will stop working at some point, and your particular one looks like one that I would avoid like the plague!

Poor quality PSUs are arguably the biggest risk to your system as point 1 elaborates here: https://forums.tomshardware.com/thr...when-selecting-parts-for-a-custom-pc.3510178/

If this happened when installing GPU drivers, it could well be the GPU or PSU, as the GPU could fault with the new drivers, or the excess power draw from the PSU may cause a big issue that can cause PSUs to shut down.

Do you have a link to the PSU in question? Your friend sounds correct, even if it ISN'T the PSU, I would be replacing it.
 
  • Like
Reactions: DaneeXIII
Aug 30, 2019
16
1
15
First of all, thanks for the answers, @PC Tailor and @Phillip Corcoran !
It's from Hungary, so I don't know how much will it help, but here's the link: https://www.bestmarkt.hu/power-supp...ntrol-low-noise-pfc-protection-retail-p149520

So it's "only" the PSU or in a worse case the GPU? I didn't kill the motherboard with the forced shutdowns when replacing components, or anything else like this? I only got money for a new PSU maybe, and if it would be the motherboard, I really don't want to cut down on that one.
 

PC Tailor

Illustrious
Ambassador
So it's "only" the PSU or in a worse case the GPU? I didn't kill the motherboard with the forced shutdowns when replacing components, or anything else like this?
Well there is no definitive answer, all we're saying, is until you replace that PSU, you'll likely never know, and that regardless as to whether it is or isn't the PSU, you should absolutely replace it.

Also there is no guarantee that if it IS the PSU, that it hasn't damaged other components, poor quality PSUs have a notorious habit of doing this. and based on what I've just seen on that page, I'd avoid the PSU like the plague!

As the guide I linked above states, better quality PSUs always pay off.
 
  • Like
Reactions: DaneeXIII
Aug 30, 2019
16
1
15
Should I try with the PSU from my dad's PC? That one is only 420W, so I'm planning to try in my PC without GPU, and with just on module of RAM.
Oh, and one more thing. I tried to turn on without the GPU, and shut down itself that time too. Or if the GPU's driver would be the thing that causes my problem, it would still not turn on?
 

PC Tailor

Illustrious
Ambassador
Should I try with the PSU from my dad's PC? That one is only 420W, so I'm planning to try in my PC without GPU, and with just on module of RAM.
Oh, and one more thing. I tried to turn on without the GPU, and shut down itself that time too. Or if the GPU's driver would be the thing that causes my problem, it would still not turn on?
Well if it is also a poor quality PSU, the same issue could replicate.

It's also not the overall 420W that powers the PSU, only the 12V rail, and poor PSUs tend to have less available on this rail, so it says 420W, but in reality, if it is poor, it could actually be 350W continuous load, or the 12v rail could only supply 300W for example.

So you can try, but only for testing, and verify that the PSU is actually adequate first which I would suspect it probably won't be, it's better to test with a known good quality working PSU, otherwise the issue could just repeat.

For example if the GPU pulling power is triggering a UVP mechanism to shut down the PSU (Under volt protection = UVP), then this will just repeat on another PSU where the same thing could happen. However if it is adequate, you could always test.
 
  • Like
Reactions: DaneeXIII
Solution
Aug 30, 2019
16
1
15
@PC Tailor That one is a no-name brand too, plus that doesn't have 8 pin, which my new motherboard requires.
Currently I'm trying my 500W PSU, but with the old motherboard (MSI A55M-E33 FM2+) and CPU (X4 860K), without GPU and HDD.
And it runs perfectly since I turned it on (~6-7 mins), it doesn't shut itself down. Then maybe the new motherboard is the faulty component?
 
Last edited:
  • Like
Reactions: PC Tailor

PC Tailor

Illustrious
Ambassador
@PC Tailor That one is a no-name brand too, plus that doesn't have 8 pin, which my new motherboard requires.
Currently I'm trying my 500W PSU, but with the old motherboard (MSI A55M-E33 FM2+) and CPU (X4 860K), without GPU and HDD.
And it runs perfectly since I turned it in (~6-7 mins), it doesn't shut itself down. Then maybe the new motherboard is the faulty component?
Not necessarily it COULD be, but the PSUs will often fail when it is under load, without a GPU, that power draw isn't there.

I would sooner suspect your friend would be correct that the 12v rail could be faulting under the GPU load.

Not saying it isn't the motherboard, but even if it was and you replaced it, I wouldn't run it with that PSU,
 
  • Like
Reactions: DaneeXIII
Aug 30, 2019
16
1
15
Not necessarily it COULD be, but the PSUs will often fail when it is under load, without a GPU, that power draw isn't there.

I would sooner suspect your friend would be correct that the 12v rail could be faulting under the GPU load.

Not saying it isn't the motherboard, but even if it was and you replaced it, I wouldn't run it with that PSU,

Thank you so much for all your help, answers and advices! Will buy a new PSU, especially after your replies, even though I was planning anyway to buy another one.
 

PC Tailor

Illustrious
Ambassador
Part of the issue, is this is a problem I have faced quite a lot of in reality and on the forums, and there is no one answer because there is a multitude of causes, so it's a case of gradually eliminating! :)

And problem is you can't power up the PC without changing major components by the looks of it. I would have been tempted to check your CPU temperatures but you had to change CPU to boot. Being as this is a build that has worked fine then suddenly stopped and isn't software related, it will often be this kind of order:

PSU > GPU > RAM > MB > CPU.

And when we see a poor quality PSU, it's red flag #1. Problem is it's hard to test PSU problems without another PSU that can power everything. So I'd say it's likely to be your PSU or GPU. Problem is even it is the GPU and you replace it, it could be a matter of time before the PSU would go and potentially take other components with it. Which happens all too often.

There are various tests you could run if you could at least boot it for a bit, but if you can't do that, the only sure fire way to guarantee finding of the problem is to replace the components 1 by 1 and eliminate. Were you running any OC or XMP at all?
 
  • Like
Reactions: DaneeXIII
Aug 30, 2019
16
1
15
Were you running any OC or XMP at all?
Yes, I tried CPU OC once with AMD Ryzen Master (from 3.1 to 3.5 then 3.7 and 1.28V if I remember the voltage right), yesterday, or two days ago, I don't know. But worked fine with that too, but I set back everything to default, and worked after that too.
-I don't know what XMP mean, sorry-
 
Aug 30, 2019
16
1
15
Okay, so I bought a new PSU (be quiet! System Power 9, 500W) and an SSD (PNY 120GB). Everything was okay for a day, and now same <Mod Edit> happens. I was watching a YouTube video, when the PC suddenly turned off, and now I can't turn it back on.
 
Last edited by a moderator:

PC Tailor

Illustrious
Ambassador
Okay, so I bought a new PSU (be quiet! System Power 9, 500W) and an SSD (PNY 120GB). Everything was okay for a day, and now same <Mod Edit> happens. I was watching a YouTube video, when the PC suddenly turned off, and now I can't turn it back on.
Don't have to use profanity (mods will remove it).

You haven't lost out, because now you don't have a ticking time bomb in the PC!

So now it would be a case of RMA the GPU if it's still under warranty.
 
Last edited by a moderator: