PC shuts off randomly/unexpectedly

September 1, 2017 at 13:41:26
Specs: Windows 10, AMD Ryzen 7 1800x
I recently set up a new build and everything is working fine aside from the fact that every few days it will randomly shut off and then will not turn back on for at least 24 hours. I monitor the temperature and haven't seen anything abnormal. I have also done a cardboard build to check all components and checked the PSU. Why could it be doing this?

Here are the specs:

AMD Ryzen 7 1800x
Asus Crosshair VI Hero
Corshair 32GB (2x16)
NVIDIA GeForce GTX 1080 Ti
NZXT Kraken x62
EVGA SuperNOVA 850 Gold

I have been trying for weeks to get this figured out. Please help me out.


See More: PC shuts off randomly/unexpectedly

Reply ↓  Report •

#1
September 1, 2017 at 13:49:29
Over heating of course comes to mind; not necessarily the cpu... The psu could be overheating, and a breakdown of a capacitor (even a diode) could happen as things get warm.. That it all works again after a long cool down has me thinking you might need to consider the psu - even if if it's new... They can and do go flakey at any time; and just because it's a new one doesn't mean it may not be faulty.

I haven't built a box for an age; but one or two of the other regulars here may have, and likely can advise more...

edited per trvlr - for typos

message edited by trvlr


Reply ↓  Report •

#2
September 1, 2017 at 14:09:35
"I monitor the temperature and haven't seen anything abnormal"

How about posting some readings - system & CPU temps, at idle & under load. Also, are you overclocking or running the CPU at it's stock speed? Have you tweaked the BIOS for best performance or did you load defaults?


Reply ↓  Report •

#3
September 1, 2017 at 21:35:35
A full list of components is also important including apparently trivial things like your cooling arrangement (how many fans, where they are, and which direction they are blowing). Sometimes even the case might make a difference depending on space and set up.

You have to be a little bit crazy to keep you from going insane.


Reply ↓  Report •

Related Solutions

#4
September 2, 2017 at 15:16:55
"That it all works again after a long cool down has me thinking you might need to consider the psu - even if if it's new... They can and do go flakey at any time; and just because it's a new one doesn't mean it may not be faulty."

I used a power supply tester on it and all of the readings came back fine.

----------

"How about posting some readings - system & CPU temps, at idle & under load. Also, are you overclocking or running the CPU at it's stock speed? Have you tweaked the BIOS for best performance or did you load defaults?"

Stock speed, default BIOS. The PC turned off again last night and I haven't been able to have it run for more than around 2 minutes since then. Every time this happens I have to wait around 24 hours for it to go back on for long enough to use it. I can post pics then.

----------

"A full list of components is also important including apparently trivial things like your cooling arrangement (how many fans, where they are, and which direction they are blowing). Sometimes even the case might make a difference depending on space and set up."

https://imgur.com/a/JXDEJ Here is a pic of my setup

message edited by _aloy


Reply ↓  Report •

#5
September 2, 2017 at 16:09:16
Did you test (measure volts on) the psu under long load - as in actually working for a while: Also when you tried to power up again immediately after it shuts down?

Reply ↓  Report •

#6
September 2, 2017 at 16:21:34
Specs as in a list of components.
You pic is sideways and confusing. It appears to have two front fans blowing IN over a radiator which is very unusual as it brings in warm air instead of exhausting it. Then I see one exhaust fan (normal enough) in the rear and one apparently top fan that is blowing IN also. This is the most unusual because you have three input fans and one exhaust fan. You should have the same number of exhaust fans or one more than intake. The radiator fans should be blowing air from inside over the radiator and out the case (unless this is absolutely how the case and radiator is designed) and the top fan absolutely needs to be exhaust since the heated air will be rising naturally. I hope there is not also a side case fan because this will make things worse.
PLEASE list all components.
PLEASE list the temperatures at idle, after 10 minutes of normal use, and after 10 minutes of gaming.
PLEASE list the voltages as they appear in BIOS, HWMonitor, or similar program (12V, 5.5V, 3V) under these same conditions.
IF you cannot answer the question as asked then we cannot get a full picture of what is happening and cannot help you. We are not there to see what is there and what is happening so you have to supply what we cannot see.

You have to be a little bit crazy to keep you from going insane.


Reply ↓  Report •

#7
September 4, 2017 at 15:48:11
I have been trying to boot the PC in order to get the info that you need but it will not boot for whatever reason. Sorry about the picture, imgur seems to have distorted it. And the radiator fans are set up as per the manual.

The pc will start for ~20 seconds and then shut off. Not enough time for it to heat up. Any idea what I could do?


Reply ↓  Report •

#8
September 4, 2017 at 16:06:58
"The pc will start for ~20 seconds and then shut off. Not enough time for it to heat up."
Not true IF you did not apply the thermal compound between the CPU and the waterblock, applied too much, incorrectly or there is no water in the system or not flowing. Without the proper application of thermal compound, it will overheat in seconds. Without water flowing immediately (or an air bubble in the system) it will overheat in seconds. If you has an air cooling system and the fan was not working, the heat sink would soak up quite a bit of heat before overheating and take time to shut down but with a water cooled system there is no thermal mass like the copper or aluminum heat sink and the amount of these metals in the water block is only a small fraction of a heat sink.
You should NOT need your system running to answer these questions since you presumably built it and purchased the components, just read the information off the bill of sale or even the boxes and manuals.

Correct application of thermal compound (find your CPU on the list):
http://www.arcticsilver.com/methods...

You have to be a little bit crazy to keep you from going insane.


Reply ↓  Report •

#9
September 9, 2017 at 16:35:49
I've applied thermal paste many, many times and I'm pretty sure it is applied correctly. A few weeks back I removed it and did it over again just to be sure because I was running out of things to try.

I'm not sure what you are asking me to post. If the pc is not operable I can't get any temp readings/etc that was asked for and I already put up specs I am using. Here they are again:

AMD Ryzen 7 1800x
Asus Crosshair VI Hero
Corshair 32GB (2x16)
NVIDIA GeForce GTX 1080 Ti
NZXT Kraken x62
EVGA SuperNOVA 850 Gold

I understand that it definitely can be heat but it just seems unlikely to me because my cooling system seems to be running properly. I have had the Kraken software running and monitoring the pump speed, fan speeds, temps, etc for weeks with no errors.

I also understand why it could shut off after a few seconds because of heat, but sometimes it will shut down immediately and sometimes it will run for days. I don't see why there would be a difference in this if the heat was causing it because wouldn't it shut down immediately every time? Also, some of the times that it has shut down while gaming it has booted right back up immediately which also indicates that it wouldn't be heat.


Reply ↓  Report •

#10
September 9, 2017 at 19:43:33
Possibly air bubble in the water cooling system.
If you have the stock factory heat sink or a suitable air cooler you could try that to eliminate that as a possible source of the problem.
If you can boot partially you can get temps in Bios Set Up.
Have you tried resetting defaults in BIOS and then manually setting the time, date, and hardware specific settings again. Defaults alone is never the best option, CPU, RAM, Graphics, to name a few are not automatically detected properly all of the time so require review.
There might be or have been a setting that raises the CPU core voltage (or other voltage) under load that is causing occasional heating. Resetting defaults might help this. If the system was overclocked this would also bring it back to normal until the issue can be resolved, then you can proceed to try it again (with more research), step by step with proper monitoring and stress testing.

You have to be a little bit crazy to keep you from going insane.


Reply ↓  Report •

#11
September 10, 2017 at 04:59:03
I also thought that it could be an air bubble back when this issue started and it was within the return window on Amazon so I replaced the cooling system and the problem stayed.

I've been running on default settings and if I can get it back up again I'll change that. It seemed in the past that resetting the bios would prompt it to boot but that isn't working this time. This is the first time that the problem has lasted this long without me being able to boot it. It has always stayed off for ~24 hours and then came back working normally. I haven't overclocked.

I did try a stress test on the CPU at 100% two weeks back and it ran for a few hours before shutting off but it's hard to tell if that's what caused it or not because it shuts off randomly all the time.

Thanks for your help on this.


Reply ↓  Report •

#12
September 10, 2017 at 05:09:30
This is not my area of expertise - but... Has the CPU overheated at any time; certainly prior to the problem arriving or when it arrived?

If so then I'd be very suspicious of the cpu as the culprit.


Reply ↓  Report •

#13
September 10, 2017 at 05:11:05
Reset CMOS and then see if you can boot into BIOS ans reset defaults. This is what you would have had to do if you had set an unusable overclock and it will rest everything and clear every setting so maybe you can get a clean start. Unplug the power supply or turn off manual power switch on power supply if it has one and then move CMOS jumper (or press clear CMOS button if there is one) for a moment, return jumper and power it on.

You have to be a little bit crazy to keep you from going insane.


Reply ↓  Report •

#14
September 10, 2017 at 11:16:42
"This is not my area of expertise - but... Has the CPU overheated at any time; certainly prior to the problem arriving or when it arrived?
If so then I'd be very suspicious of the cpu as the culprit."

This problem has been going on for a while and at some point I decided it must just be a problem with a component so I returned everything on amazon for a new one. The problem came back immediately after installing new components. So unless they both overheated and the second one overheated immediately after installing it then I don't think it is that.

"Reset CMOS and then see if you can boot into BIOS ans reset defaults...."

I've been resetting the CMOS every time I've tried to boot it (it has a button) and it hasn't worked but for whatever reason it just started up. I was able to run it for nearly ten minutes and as a result I only have idling data:

http://imgur.com/a/YxfPp


Upon boot CAM told me that my GPU exceeded 81C on 9/1 which was the same day that it last worked. So it does seem like it could heat related but what I don't understand is why it will not boot again even after cooling down. That makes it sound like it isn't heat.


Reply ↓  Report •

#15
September 10, 2017 at 20:18:05
Do you have an onboard GPU?
Have you booted with GTX GPU removed??

Reply ↓  Report •

#16
September 10, 2017 at 20:53:17
Your CPU, HD, SSD, GPU temps are good enough but your motherboard temp (presumable the chipset) peaks at idle at 59C which is a bit warm. If you had an air cooled CPU then the overflow of air would help cool the motherboard. Since you have a liquid cooler for the CPU the chipset is not getting much air flow. I also see that the one fan speed presumable a case fan has been pushed way high to try to cool it. You will need to see if your case has the option of an additional rear or top rear exhaust fan and if not then swap out the current exhaust fan for a ball bearing model with a much higher CFM rating (look for a low Db rating to avoid a noisy system) and the largest diameter that will fit.

The suggestion of trying the onboard graphics or if necessary a more basic graphics card (one proven to be working well) to see if you can isolate the problem or at least eliminate more possibilities.

You have to be a little bit crazy to keep you from going insane.


Reply ↓  Report •

#17
September 12, 2017 at 05:07:24
I have had my pc crash by overclocked ram before, no blue screen nothing.
try a memory test if you havnt already

Simple solutions are often the best


Reply ↓  Report •

#18
September 15, 2017 at 15:09:43
"Do you have an onboard GPU?
Have you booted with GTX GPU removed??"
No, and yes but it has behaved more or less the same.

"Your CPU, HD, SSD, GPU temps are good enough but your motherboard temp (presumable the chipset) peaks at idle at 59C which is a bit warm...."
Do you suggest trying out a different cooler? I have no more case room for another fan. I am up for buying a different cooling system if that may help. I also don't have an on board GPU. Do you think that a new cooling system or a new GPU would be more beneficial to try first?

"I have had my pc crash by overclocked ram before, no blue screen nothing..."
I've checked memory and it is fine. Completely ruled out RAM at this point.


Reply ↓  Report •

#19
September 15, 2017 at 18:11:59
If your rear exhaust fan is the stock one that came with the case then you can change it out for one rated higher CFM (cubic feet per minute or air moved). Look for the highest CFM but still with a reasonably low Db levels.
If there is a way to make the front fans blow inward and still function properly for the radiator, then they may also bring in cooler air.

You have to be a little bit crazy to keep you from going insane.


Reply ↓  Report •

#20
September 15, 2017 at 18:15:43
The front fans are blowing inward, sorry if that wasn't clear in the picture I put up. Imgur seemed to have warped it a bit.

I will try to swap out the fan. In the past I have set the GPU to ~70% and played a game and it still shut during that so I'm doubtful that it is really GPU heat. Could be the chipset. I'll update once i swap out fans.

Thanks again.

Edit: Getting the SilverStone FM121

message edited by _aloy


Reply ↓  Report •

#21
September 15, 2017 at 19:59:58
"Checked the PSU" how?

How To Ask Questions The Smart Way


Reply ↓  Report •

Ask Question