Computer randomly restarts even after changing motherboard, cpu and RAM

Toggle sidebar Toggle sidebar
F

FlyingMandarine

Honorable
Oct 24, 2013
24
0
10,520
Hi everyone! :)

Sorry for the long message! Specs in the attached files.

My partner's computer has been restarting semi-randomly for the past couple of months. There are two ways the computer restarts:

-After a random period of time; the computer could be fine for a couple of days, but sometimes, all it takes is a couple of hours of being on. Usually, once it happens once, it happens again soon after that (sometimes as soon as a couple of minutes later). Leaving the computer off for a few hours and then turning it back on makes it more stable, or should I stay more stable-ish.

-Launching specific games will restart it. Namely, Final Fantasy X (the Steam version), after reaching a certain point in the game after a specific cutscene; and, more easily troubleshoot-able, launching Killing Floor 2, directly after the opening cinematics, right when the main menu should appear.

Because her computer has been displaying weird visual artifacts (and only one of her GPU fans out of two would work), we switched it to another PCI-E slot; same problem. We put her card in my computer and my card in hers -- her card would work fine in mine (at least in terms of both fans running), but there'd be weird stuff happening with my card in her computer, so we didn't push it.

However, we did try her computer without a GPU and it would still reboot, so she thought, since she had an oldish motherboard and CPU, might as well change them. So we changed both of those + the RAM to the following:

http://i.imgur.com/6Y2QT8N.png
http://i.imgur.com/Oco07xM.png

We kept the same GPU, since it seeeemed to work on mine, and we kept the same PSU as it is a recent one (1.5 year old, Corsair RM 750).

However, it looks like the problem hasn't been solved, and the randiom reboots are still there! What, we throw money at the problem and it doesn't get solved?! What world do we live in?! 😉

Launching Killing Floor 2 doesn't cause any problem with these new components though. We also replaced the surge protector just in case it was a house electricity problem.

When the computer restarts, it's like it took a punch to the guts: oftentimes, the Internet just doesn't work anymore for a bit on her computer (on mine, it works fine of course).

So I guess our next step is checking if the PSU is at fault? We checked the voltages on HWMonitor: http://i.imgur.com/nHeUwYP.png
Am I right in checking these values for the motherboard? It does look like the +3, +5 and +12V values are a fair deal lower than they should be.

However, I checked them in the BIOS and they seem more reasonable: http://i.imgur.com/MDhGCHn.jpg

They are quite stable, meaning the 4.860 is also stable. Is that too low to the point where it could explain these problems? I read that the 5V is not necessarily to give power to the more important components in the system... so could it really explain the reboots?

The problem is that it's hard to cause the problem besides just waiting hours (days?) at a time since it doesn't reboot when launching games anymore, so we can't just try something and tell right away if it worked or not.

Thank you so very kindly for reading this and for all the help, assistance and life-saving tips you can give us. :)
 
Sort by date Sort by votes
I would have a look at cpu temps as to me it does sound like it could be a cooling issue but seeing as you have changed cpu and Mobo i'm surprised it hasn't been solved. The next port of call would be to look at the PSU. The Corsair RM are pretty solid PSUs so seems odd that, that would be the outstanding issue.

Sounds a bit odd but have you tried running the PC off a different power outlet in the building? say downstairs instead of upstairs and see if that makes a difference.
 
Upvote 1 Downvote
The stable 4.86v is a bit low and the symptoms can be caused by faulty PSU... I've answered a few posts where the PSU caused random restarts... drivers can also cause random restarts so:

1. I'd suggest interchanging the PSU with your partner's computer for a few days and see what happens.

2. Drivers issues can also cause random restarts and the GPU displaying weird visual artifacts and behaviour should be solved with the last GPU driver updates.

3. Some driver issues with Windows 7 to 10 can be on several components, issues where drivers may not install, or seem to install but not work, and I see possible GPU and Ethernet adapter driver issues.
 
Upvote 0 Downvote
Three answers already! Thank you all. :)

Paul NZ: "Automatically restart" is already unchecked after we read something about it... what, over a month ago? So all good on that front. You are right, though, about the 41(63) errors: https://i.imgur.com/aIHuAU1.png

The last time it happened being today, I think it checks out with when it rebooted, time-wise.

abaday789: You know what? We had a dehumidifier also plugged in on the same outlet (well, "same" as in, there are two plugs on the outlet, one for the dehumidifier, and one with the surge-protected power board with the computer and monitor plugged in), and the dehumidifier just died last month. A couple of other appliances either died or started acting weird in our room, in other outlets, so I'm afraid doing something about that wouldn't help. On the other hand, my computer has been fine so far. Should we try plugging both our computers on the same outlet (mine)? Would it be too much? And if the problem were the outlet indeed, why would it not only crash randomly, but also upon launching certain games -- and why is that problem solved now? I know, too many questions, right? :)

More importantly, why would it restart like this if her computer is plugged into a surge-protected power board? It does say "Surge protector" and "Overload protector" on the power board's box.

As for the CPU temp, they were rather fine and still are, I think? The computer sometimes reboots while idle, and right now it stands at roughly 26 degrees Celsius.

Chicano: We've been thinking of that, but I have to admit it's a bit daunting to interchange PSUs, so I wanted to see if I could try other things before having to do that. But that's a good idea, yes. :)

EDIT: We used DDU (Display Driver Uninstall) to remove the drivers and reinstall them, several times, even, but to no avail. There is, however, a Windows 10 update that doesn't seem to want to install, but it is a recent issue, so I'd say it's unrelated? We'll install the latest GPU drivers again, but I'm not getting my hopes up, I'm afraid.

As for your number 3, what could I do about it? And the fact that the computer shouldn't restart automatically (since we unchecked that box) and it still does makes me think it can't be Windows-related. Or could it?
 
Upvote 0 Downvote
That event id:41 (63) error, if you didnt press the reset button to restart it. Can mean a faulty PSU

If you brush your arm along the case, do you feel anything?? This happened to me once, later found out the earth had come out of the power point, the computer was plugged into. It was earthing on the case.

It didnt damage the computer itself (altho I replaced the CPU under warranty). But in the process, anything that was on the same circuit fried it. Even tho it wasnt plugged into the same power point. ie: The fridge and washing machine I had were dead




 
Upvote 0 Downvote
Paul NZ: That's scary! We don't feel anything brushing our arm along the case, but the house we live in is a bit old and not very well-maintained (unfortunately, we just live in one room, the rest of the house belongs to the landlord and other tenants, so I can't try the computer in another room). However, in another outlet in the same room, her phone fried (!!!), and we figured it would be because it was charging on a faulty outlet which also happens to make a light flicker in a weird way (you know, the types you touch with your hand and it changes the intensity and turns in on/off? well, it would turn on and off and change the light intensity automatically).

Since my computer doesn't have any of these problems, do you think I could try and plug her computer on the same power board as mine, and see if she gets any problems with her computer? Will two computers and two screens overload the power board?
 
Upvote 0 Downvote
It should say on the powerboard what it can handle. I've got a TV, dvd/bluray player, a freeview decoder, an Xbox 1 plugged into this power board lol. And this computer. The board can take up to 8 appliances

But it should be fine. If it's still doing it after this exchange PSU's. if you havent tried yet

Just unplug the ATX12v, 24 pin, and plug them into the other pc. They may have to be side by side




 
Upvote 1 Downvote
So I guess our next step is checking if the PSU is at fault? We checked the voltages on HWMonitor: http://i.imgur.com/nHeUwYP.png
Am I right in checking these values for the motherboard? It does look like the +3, +5 and +12V values are a fair deal lower than they should be.

However, I checked them in the BIOS and they seem more reasonable: http://i.imgur.com/MDhGCHn.jpg

They are quite stable, meaning the 4.860 is also stable. Is that too low to the point where it could explain these problems? I read that the 5V is not necessarily to give power to the more important components in the system...


Hi
First off, the 5v rail is very essential! The PSU usually strives to make that as exact as possible, and the other rails follow along. 4.8V is decidedly borderline if measured with a decent meter. As I understand your description, about all that you've got left of your original PC is the PSU, hard drive and graphics card. Assuming that you updated the drivers when you updated the motherboard, I greatly suspect that either your PSU is running out of power/faulty, or you have an overheating problem. Have you cleaned out the PSU fan and heat Sinks? The cpu fan and heat Sink? Enough airflow? (One of my PCs is erratic if the ambient temperature rises above 22C. It has all new fans, but can't handle it. Runs fine with the cover off! Now I know, it's not worth chasing as it's usually colder than that here.) graphics cards are power hogs: these can really disturb the power rails during heavy processing, not to mention the heat generated.

 
Upvote 1 Downvote


I agree with the PSU issue. The oddity that said power is the two fans. Just not normal. Her PSU(750W) well exceeds her sys reqs but the behavior says PSU. After the cut scene the GPU is displaying things in real time. GPU usage, power consumption, VRAM and other resources get a heavier load. Cut-scenes are normally pre-rendered where the CPU is taking a bigger load and the GPU less of a load. This is why cut-scenes and real-time gameplay can look different. Daunting though it may be your PC is saying HELP ME! And more specifically it sounds as if the cry is coming from your PSU. You really want to address that issue before it gets worse. It also sounds as if the landlord is a bit of a deadbeat. If my place was having that kind of electrical issue it would be immediately addressed. Whether it meant a higher amp circuit breaker or better wiring or just a new light fixture it would be fixed. Now it's changing light levels. Tomorrow it starts a fire. Melodramatic? Sure. Possible? Absolutely.

Lastly. Why are your fail-safes disabled?
 
Upvote 1 Downvote
Paul NZ: Alright then, let's try plugging her computer in my power board and see what happens next! It could take me a while to report back though, if the problem takes time in occurring again (especially since we won't be on our computer for some time in the next few days).

SexySally: Thank you for the reply!

The CPU and CPU fan are new, so it's pretty clean. The PSU seems also pretty clean to me; we have a lot of dust in this room, but we clean our computers often enough that I would thiiink that's not a problem. There's quite a bit of, uh, unremovable-ish dust on the old GPU though, but that wouldn't explain why the computer would reboot back when we tried it without the GPU (before changing all the components).

It doesn't feel like an overheating problem either, for several reasons: first, back when it was quite a fair deal hotter in this room, the problem wouldn't necessarily trigger more often; then, we checked on HWMonitor (and in the BIOS), and the temperatures are quite fine (except for the GPU when running a demanding game; it then easily goes to 70C); finally, if it were an overheating problem, then surely it would be more likely to happen when doing something demanding, and less likely to happen when being idle, right? But it doesn't look like that's a pattern we're seeing here.

Your 4.8V comment is worrying though! If Paul NZ's proposed solution doesn't work, should we go ahead and replace it (warranty is still good to go I think, Corsair 5 years and her PSU is about 1.5 years old).
 
Upvote 0 Downvote
aquielisunari: Thank you for your reply!

"If my place was having that kind of electrical issue it would be immediately addressed. Whether it meant a higher amp circuit breaker or better wiring or just a new light fixture it would be fixed. Now it's changing light levels. Tomorrow it starts a fire. Melodramatic? Sure. Possible? Absolutely."

Unfortunately, I can't do much about the situation. But don't get me wrong: I completely, wholeheartedly agree.

As for your analysis of the cutscene, I'm afraid I should have mentioned it before, but it was an in-game cutscene, so I don't think that would apply?

So if I understand everyone's comments so far, it can be one of two things: the PSU, or the house electricity. So I could try to plug her computer on my (supposedly safe) power board, and if it continues happening, then get a replacement PSU? If it works and there is no more problem but the PSU is still showing 4.8 instead of 5V, then I let it be this way?

I don't know if it matters (for the 4.8V problem, for instance), but my partner has 3 HDDs, 1 SSD and a DVD drive. Could it be too much? If so, well, it wasn't a problem before, so I don't understand why it would become one all of a sudden?
 
Upvote 0 Downvote


I realize it was an in-game cutscene. They are pre-rendered and don't use as much power as real-time gameplay does.
 
Upvote 0 Downvote
We plugged her computer into my power board, but it already restarted twice, while my computer was fine. So it looks like we're entering PSU problem territory here? I've already submitted a ticket to Corsair; we'll see how it goes!

If anyone else has another idea of why that would happen, please feel free to comment. :) And thanks so much for all your answers so far!
 
Upvote 0 Downvote


The PSU is probably one of the easiest HW replacements to do as you can only connect every terminal the right way, so connecting them the wrong way is impossible.. well, unless you really force them but that's not necessary.

One way you can make sure it's a power (PSU) issue is looking for a dump file where the causing error would be described, and if there is no dump file to report the cause, it would mean the sudden shutdown/restart caused by a power issue didn't allow time for a dump file to be created... and if a dump file was created, it would probably mean it's a driver issue or at least help determine the actual cause.

Download BlueScreenView to help you find the dump file
http://www.softpedia.com/get/System/Back-Up-and-Recovery/BlueScreenView.shtml
 
Upvote 0 Downvote
Chicano: I've downloaded BlueScreenView and launched it; without me doing anything, only one dump file appeared, but oldish (mid-October). Nothing regarding today or the other times the computer has restarted. Do I have to tell BlueScreenView where to look for other dumps, or does it mean that the computer did indeed reboot without having enough time to create a dump file?
 
Upvote 0 Downvote


BlueScreenView should have found a recent dumpfile from the last restarts if it had been created so it's probably a power issue... so there's a good chance it's the PSU causing the reboots.


Just in case you still want to avoid replacing the PSU, see here how to configure the system to create a minidump file, a smaller report which you can find easier... you can also make Windows pause with a blue screen displaying error codes, before rebooting (instructions incuded)... and if it still doesn't pause and display a BSOD, then it's no doubt a power issue.

https://www.tenforums.com/tutorials/5560-bsod-minidump-configure-create-windows-10-a.html
 
Upvote 0 Downvote
Okay, my last post was over three months ago, so I think it's high time I post a final update. 😉

My partner received a new PSU which so happens to have slightly off values in the bios too, in much the same way. However, the problem never ever occurred again, and so I think it's now safe to say the PSU might have been the problem all along despite it being rather new and in good condition.

It's often too easy to take for granted the kindness of strangers over the Internet. Thank you so very much for all the help you have given us! :)
 
Upvote 1 Downvote
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom