AtNvme

Honorable
Sep 2, 2015
81
4
10,565
Well... (I'm unsure wether this goes into the MOBO or PSU section, I guess both potentially).

I hoped I would never have to come back here again for that specific issues, but it has come back after a whole year, I had nearly forgotten about it.
It all started in June of last year, where I had a few shutdowns followed by a Power Surge Protection message, here is a link to the thread I made last year: https://forums.tomshardware.com/threads/power-surges-detected-shutdowns-asus-z97.3489154/
It had suddenly stopped, and it hadn't happened again since... since yesterday.

Let's talk about my specs first (the ones that matter):
i5-4690K
ASUS Z97-Pro Gamer
750W 80+ Gold EVGA PSU
Take note that every piece of hardware in my PC is upwards of 4 years old, excluding storage.
I have never neglected my PC, as I do clean it somewhat regularly and am fairly conservative with my CPU overclock, mainly because of my underpowered stock cooler.
It's definitely not running out of juice, 750W is more than I would ever need with this computer.

I had pretty bad luck with one of my case fans, granted they are old and broken, one fell out of its socket onto the cpu cooler, which moved it out of place. As my PC started thermal throttling, I shut it off using the power switch on the PSU, since I wasn't able to shut it off properly from the sheer amount of lag.
I took out my cooler and CPU, cleaned them off with alcohol, put new thermal paste and reseated the whole thing, no problem there, in fact It's running cooler than it ever did, thanks to the non pre-applied compound.
I also took the time to clean out the dust out of my system, including PSU, it had been a probably two to three months since I did for the last time.
I boot up, instant shutdown due to Anti-Surge Protection from my ASUS MOBO. Second boot, we're in.
That was the first shutdown in nearly a year, and it has been followed by many today and yesterday.

It has been regularly shutting down since, upwards of 5 times a day, even when I'm away from the computer, it's not necessarily due to me doing things.
I held back the overclock on my CPU to reduce other factors as much as possible, though I don't think it has anything to do with it, since I haven't had any crashes with my OC before, it's at 3.9GHz, it can barely be called an OC.

I know ASUS motherboards have a bad habit of being very sensitive to voltage changes, causing them to shut down systems easily.
I doubt my PSU is ACTUALLY at fault, since it's a good unit from a reputable brand, but I'm not excluding the idea.
I am plugged into a groundless extension cord, which is plugged into a power bar, which my PC is plugged into. I will be changing that to a newer extension cord WITH A GROUND, maybe that'll help.

There is probably extra info on my old post, I'm all ears.
 
Solution
When the PC is next up and running look in Reliability History and Event Viewer for error codes and warnings that correspond to the shutdowns.

Even though the PSU has the necessary wattage, has been well cared for, and not heavily used its' age could be nearing design EOL (End of Life).

In other words the PSU could be at the beginning of the end with minor falters and failures. Could be any single voltage rail or a combination of the voltage rails.
When the PC is next up and running look in Reliability History and Event Viewer for error codes and warnings that correspond to the shutdowns.

Even though the PSU has the necessary wattage, has been well cared for, and not heavily used its' age could be nearing design EOL (End of Life).

In other words the PSU could be at the beginning of the end with minor falters and failures. Could be any single voltage rail or a combination of the voltage rails.
 
Solution
When the PC is next up and running look in Reliability History and Event Viewer for error codes and warnings that correspond to the shutdowns.
I have looked at RH, I do see every crash, but there doesn't seem to be anything more to it than "Previous shutdown wasn't planned". I will be keeping this in mind though, and will, if issues persist, consider a new unit.

Which EVGA PSU? GD, GM, GQ, G3, GA, G+, G2......,
It's a G2, purchased at the end of 2015.

NEVER do this.
Turns out it wasn't that specific extension cord I was thinking, it had a ground, but was old. I changed it for a new one. I'll also be looking for a longer power bar, so I can avoid extension cords altogether.

I have given my PC a thorough PSU dust clean, I believe it's pretty much dust free at this point.
I was suspecting it could be linked to dust accumulation shorting or whatever, will be running my computer as usual now see if those fixed it. I will update again tomorrow.

Thank you for the help so far!
 
After cleaning be sure to check that all cards, cables, RAM, jumpers, etc., are fully and firmly seated.

No cables pinched, kinked, or crooked connectors. Nothing loose or wiggling.
I've done that, also replaced the broken fan.
Haven't had any crashing so far, I've stress tested a bit and had it run for hours without issues, I'm hoping it stays this way.
I'd say it's stable for now, will update if issues persist.

Thanks!
 
So...

It worked for a day. No more.
Ran for over 24 hours after I had cleaned my PSU with an air compressor. No crashes, nothing wrong whatsoever, I thought it was fixed.

Today went differently, it crashed 3 times within 3 hours or so at random times. I'm really confused, why was it fine for so long, and then back to constant crashing?
From what I know, EVGA PSU's are usually designed to last 10 years+, so I would be surprised it's design EOL as they come with 10 year warranty, I've had mine for 4½.

Could BIOS updates fix this? I'm uncomfortable updating it, since it's crashing so much. A mid install crash would brick my board, would rather avoid that, though I would be surprised Anti-Surge protection is enabled in BIOS?

I might try and give it a final dust cleanup, opening it up and blasting some air onto it, taking out every spec of dust visible, though I doubt that'd be the issue at this point.

I will probably be purchasing a new PSU, testing that one in the system, if issues persist, I will disable Anti-Surge protection, as it would most likely be caused by a faulty sensor.

I'm very confused as to how it first happened a year ago, for about a week, then stopped randomly, and now started again after I blew dust out of it with a compressed air can. Then it fixed itself for a day, after a more thorough blast from a compressor, and now it's happening again. I'm trying to make sense of this, but am having a really hard time doing so. It seems to be linked with dust, or the physical PSU, at least, but it started again after a day, I doubt dust wouldve gotten into it in a day. Could it be that the PSU itself has cooling issues and is overheating, causing voltage spikes?

I'm all out of ideas.

EDIT: I'm aware opening up PSU's is dangerous and will be careful, and perhaps opening it up will allow me to see if anything is visibly wrong with it internally.
 
The Asus power surge detective has also been known to giving false positives,never really like that part of their stuff or even Asus for that mater.
I will be changing system soon as I planned to during the summer (though I did plan to use this PSU, perhaps that's not going to happen). I think Gigabyte is where I'm going to put my money next!

I have no idea where you would come to that conclusion.
I based myself on the fact that they offer a 10 year warranty. I don't see the logic in offering that on a product that isn't designed to last at least that long.

Besides, what would be considered acceptable +12V range? +-5% is what I seem to find mostly, so 11.4 to 12.6 roughly?

I will be monitoring voltages, since my +12V does seem somewhat lower relative to % than my other voltages by a good margin.
It seems to be hanging around 11.7-11.9 currently, which isn't far, but it is on the lower end nonetheless. I see people saying below 11.9 is worrying for the PSU, is that true? I'm guessing, following that pattern, it would be undervoltage to the system causing shutdowns?

Should the numbers be fairly stable or depend on system load? I could stress test the system if they are expected to vary, though I don't see any specific pattern in the computer shutting down. I really want to figure this out, as PSU's are quite expensive right now, great timing. Oh and I have to work on finals, even better.

Any extra info about voltages and PSU's in general would be greatly appreciated, software recommendations and whatnot also.

Thanks for the help so far!

EDIT: Here is about 1 hour long idling voltage min/max: https://gyazo.com/dcced8b6e808b500c048d0ad9cabe73b
These numbers look abnormal, but couldn't it also be sensors? Should I test with a mutlimeter to make sure?
 
Last edited:
I will be changing system soon as I planned to during the summer (though I did plan to use this PSU, perhaps that's not going to happen). I think Gigabyte is where I'm going to put my money next!


I based myself on the fact that they offer a 10 year warranty. I don't see the logic in offering that on a product that isn't designed to last at least that long.

Besides, what would be considered acceptable +12V range? +-5% is what I seem to find mostly, so 11.4 to 12.6 roughly?

I will be monitoring voltages, since my +12V does seem somewhat lower relative to % than my other voltages by a good margin.
It seems to be hanging around 11.7-11.9 currently, which isn't far, but it is on the lower end nonetheless. I see people saying below 11.9 is worrying for the PSU, is that true? I'm guessing, following that pattern, it would be undervoltage to the system causing shutdowns?

Should the numbers be fairly stable or depend on system load? I could stress test the system if they are expected to vary, though I don't see any specific pattern in the computer shutting down. I really want to figure this out, as PSU's are quite expensive right now, great timing. Oh and I have to work on finals, even better.

Any extra info about voltages and PSU's in general would be greatly appreciated, software recommendations and whatnot also.

Thanks for the help so far!

EDIT: Here is about 1 hour long idling voltage min/max: https://gyazo.com/dcced8b6e808b500c048d0ad9cabe73b
These numbers look abnormal, but couldn't it also be sensors? Should I test with a mutlimeter to make sure?
https://www.evga.com/support/faq/FAQdetails.aspx?f=59665
 
I will be changing system soon as I planned to during the summer (though I did plan to use this PSU, perhaps that's not going to happen). I think Gigabyte is where I'm going to put my money next!


I based myself on the fact that they offer a 10 year warranty. I don't see the logic in offering that on a product that isn't designed to last at least that long.

They don't give the PSU a 10 year warranty because they're engineered to last 10 years.

They base it on how long the PSU should last and other factors such as how long do people tend to keep receipts or know where to find a link to their receipt and care to actually RMA something 6 or 7 years down the road.

The fan used in a PSU doesn't even tend to last 10 years unless it's an actual FDB fan.
 
Update : My PC died.

One of the shutdowns seems to have claimed either my PSU, motherboard, or maybe even everything in my system.

It shut down as usual (it always seemed to shut down twice, since it'll shut back on then down instantly, then back on again and boot) but that time, when booting, all I heard was my HDD running. No fans, no post, no lights (other than the Power led on my mobo, which is turned on when my PSU is on, not necessarily when my computer is booted).

I'd need someone else's input on this, it seems to me like it's either the PSU who just died and doesn't draw enough power to supply the system other than drives. It's basically as if my PC wasn't turned on minus the PSU running and drives getting power. I tried booting 2-3 times and the last time had the GPU led turn on for about 1/4 of a second before turning back off. Not sure that's a good or a bad thing, but it leads me to believe maybe my motherboard/GPU is still alive? It looks like it might be power buildup, since it was plugged for a bit.

Second option, motherboard fried and wont supply the system anymore. Considering fans aren't turning, it seems likely, though maybe it's due to 12V rail dying? I also checked the heat around my CPU from letting it run for a dew seconds and didn't feel any, so I dont think it's getting any power (or it's dead I have no idea).

Third option, everything is dead and I'll have to spend an extra 800$ for a new PSU and GPU. Though it's somewhat good timing, since I was panning to upgrade this summer.

I didn't hear anything suspect when it happened, nothing different than usual, though I was checking voltages and noticed a min of 11.5 and a max of 11.9 on my +12V, which is definitely not normal. It was definitely happing around voltages often and not only small jumps. Considering what happened, I doubt my motherboard was giving false positives at that point, though it might be too late.

I'll be having someone over with a new PSU, I'll give that a go in my system. If that doesn't work, I'll try and have my GPU tested, as the rest of my hardware is quite old and might have a hard time finding a test bench. If it does work, I'll look into PSU warranty, hopefully I do still have that inbox @jonnyguru 😉

What are your thoughts?

EDIT : I'm not too worried, since my drives are alive and everything (except the GPU/PSU) was meant to be swapped anyway. Perhaps its better it happened now than later in a brand new system.
 
So I tested a different (new) PSU, took out my GPU, the whole system was shutting off instantly on boot, and then pressing power would do anything.

Since I had been unplugging everything, I thought it might have to do with my power button, which didn't seem to work right. There was no way for me to be 100% sure, since the PSU I was using was brand new and unused, so I tried mine again to see if it would replicate what had been happening the day before, where it would sort of "half start". Plug the PSU in and as soon as the power switch turned to on on the PSU, elecrical sounds come out of the motherboard along with some nasty smoke.

At this point, I've come to the conclusion that my PSU 12V rail was defective and killed my board. Now all that is left to know is wether my components took the hit or had to endure electrical hell to their end.

There isn't much more to do, I'll be selling off the parts that lived (if any) and using my GPU again if it's still alive. I'll take this as valuable experience and make use of my warranty next time before it claims my whole system with it.

I'll be waiting patiently for a Ryzen 7 4800X or something along those lines when they release.