Solved My computer randomly crashes and shuts down without warning.

April 28, 2017 at 14:26:52
Specs: Windows 10, Radeon/ 8GB of RAM
Randomly, particularly in intense situations (gaming), but after it crashes I try to reboot but it shuts down almost immediately even after trying multiple times. I just installed a new motherboard and it was still crashing, I tried putting thermal paste but it did not help. I'm not quite sure what to try next.

See More: My computer randomly crashes and shuts down without warning.

Report •

✔ Best Answer
May 5, 2017 at 11:24:22
There is another possibility that I have seen here a time or two. Under load, some systems increase the voltage (VCore) to the CPU to increase stability. This can be good BUT it also increases the CPU temperature during this time. You can try monitoring the VCore in software and see how much it rises under gaming loads in say 10 minutes. If it is not a factory made machine then you may be able to go into BIOS set up and try manually locking down the CVore to the same as idle. If it becomes unstable but cool, you can raise the VCore by .01 or .02 and try again. If you post the make and models of your hardware we probably can look up a normal VCore and what overclockers have found works stable but cool enough.

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



#1
April 28, 2017 at 16:35:42
The thermal paste must be applied in the exact manner that is specified by the CPU manufacturer. Any other way might still result in over heating.

What temperatures are you getting?

Always pop back and let us know the outcome - thanks


Report •

#2
April 28, 2017 at 17:26:03
Select your CPU for the proper thermal paste application: http://www.arcticsilver.com/methods...

Also, post the make/model & wattage of your power supply.


Report •

#3
April 28, 2017 at 21:02:01
If the reason you replaced the motherboard is due to the crashing then you jumped to conclusions and probably did this unnecessarily.
Post all of your hardware old and new.
List in order the problems you were having initially and the things you first tried as well as the results.
Install HWMonitor and post the temperature readings you are showing, immediately on boot and after a few minutes.
Remove the heat sink, clean off all thermal compound with a high quality rubbing alcohol (95%+ Isopropyl,) and follow instructions for your CPU (link above) to apply the thermal compound properly. Report all temps again, right after boot, after 15minutes idle, and after 10 minutes gaming.
Also while you are at it, please post the primary voltages (12V, 5V, 3.3V) at the same times as temps.

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


Report •

Related Solutions

#4
April 29, 2017 at 05:47:22
Ok thanks for the speedy reply, I'll be sure to go through all the parts and get the spends, technically it's a friends computer but I'll be sure to make a full list. The temperature was displayed in the boot menu of the motherboards system and was displaying 85-90 after a few minutes of being on.

Report •

#5
April 29, 2017 at 06:07:05
85-90 what? Fahrenheit? Celsius? . . . Kelvin?

How To Ask Questions The Smart Way


Report •

#6
April 29, 2017 at 10:14:30
Celsius it says on the boot menu

Report •

#7
April 29, 2017 at 12:39:05
Way too hot - almost certainly the cause of your shutdowns.

Always pop back and let us know the outcome - thanks


Report •

#8
April 29, 2017 at 14:27:36
Okay that was what I figured, hence the thermal paste, I'll try to clean it correctly and apply it correctly based on the CPU instructions and get back with the results.

Report •

#9
April 30, 2017 at 13:25:59
Ok correctly cleaned and applied the paste, about to test.

Report •

#10
April 30, 2017 at 13:33:16
Still crashing, I did use 3% hydrogen peroxide rubbing alcohol instead of hypropotyl alcohol to clean it. I'm not sure what difference that would make however.

Report •

#11
April 30, 2017 at 21:36:51
Hydrogen Peroxide is H2O2 (mixed with H2O) and not alcohol at all. Alcohol will dissolve the compound and degrease so the surface is clean and oil free. Peroxide is just oxygenated water and will not do the job properly.

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


Report •

#12
May 1, 2017 at 04:24:21
Ok got it, I'll try it with the correct stuff next time I have a chance to get some.

Report •

#13
May 2, 2017 at 17:52:32
Thoroughly and correctly cleaned with isopropyl alcohol (91%) but it's still running super hot, like still 90 Celsius.

Report •

#14
May 3, 2017 at 03:22:24
Have you got the right CPU fan on there and is it running properly? In some instances running with the case off can cause overheating if that's what you happen to be doing. Otherwise there must be something else wrong with the fixing of the heat sink to the CPU.

Always pop back and let us know the outcome - thanks


Report •

#15
May 3, 2017 at 04:14:53
Alright, the heading is my next object of observation, also, something of note: the pc seems to crash when a loading screen appears or finishes for a game, I don't think it crashes just in the middle of regular gameplay. I'm not sure what connotations this has but I thought it was worth mentioning.

Report •

#16
May 3, 2017 at 05:13:00
The heatsink I meant to say is my next object of observation*

Report •

#17
May 3, 2017 at 07:25:38
You might want to check the temperature inside the case while it's running. Electronic probe thermometers work well for such measurements.

How To Ask Questions The Smart Way


Report •

#18
May 3, 2017 at 09:04:00
It runs 47 just on regular running, at least that's what it says in the boot menu.

Report •

#19
May 3, 2017 at 13:35:11
Yeah, that's still the temperature of the CPU, not the temperature of the air inside the case.

Look, one of two things is happening here. Either the CPU heat sink isn't working, or it's working perfectly but the heat from the CPU has no where to go. You need to figure out which it is before you can move forward.

How To Ask Questions The Smart Way


Report •

#20
May 3, 2017 at 14:39:47
Okay got it, I'll try to determine that soon.

Report •

#21
May 3, 2017 at 14:40:31
I think it's the heatsink that isn't working, because the thermal paste has been properly applied, of that I'm quite sure.

Report •

#22
May 3, 2017 at 14:51:33
There's nothing much about the heatsink itself that can fail. Could be the surrounding heat is too high.

Always pop back and let us know the outcome - thanks


Report •

#23
May 3, 2017 at 16:48:00
If you're sure the heatsink is bad, buy a new one. Make sure it supports your CPU chipset.

Otherwise, as Derek said, there's not much to the CPU cooling system. It's a chunk of heat conductive metal with as much surface area as reasonable, usually a fan of some sort, and thermally conductive soft material to help fill the gaps between the CPU and the chunk of metal. Assuming the heatsink and thermal compound has been applied correctly, there's not much left. If the fan's working and the CPU contact surface isn't damaged, you're better off making sure the inside of the PC is clean-ish and the air temperatures are reasonable.

How To Ask Questions The Smart Way


Report •

#24
May 4, 2017 at 04:24:34
Alright thank for the advise, I'll order a new heatsink at some point soon, by CPU chipset, do you mean the socket the the motherboard has or is that different?

Report •

#25
May 4, 2017 at 04:56:57
CPUs can have different thermal requirements but still fit in to the same socket. You need to make sure you can both physically mount the thing, and that it provides the required cooling.

How To Ask Questions The Smart Way


Report •

#26
May 4, 2017 at 07:06:47
Got it, I'll make sure to figure it out.

Report •

#27
May 5, 2017 at 11:24:22
✔ Best Answer
There is another possibility that I have seen here a time or two. Under load, some systems increase the voltage (VCore) to the CPU to increase stability. This can be good BUT it also increases the CPU temperature during this time. You can try monitoring the VCore in software and see how much it rises under gaming loads in say 10 minutes. If it is not a factory made machine then you may be able to go into BIOS set up and try manually locking down the CVore to the same as idle. If it becomes unstable but cool, you can raise the VCore by .01 or .02 and try again. If you post the make and models of your hardware we probably can look up a normal VCore and what overclockers have found works stable but cool enough.

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


Report •

#28
May 5, 2017 at 17:24:49
Alright I'll make sure I take a look at that too, thanks.

Report •

#29
May 6, 2017 at 10:39:45
I think the problem was solved after going through some BIOS settings and increasing and decreasing some stuff from a guide, thanks for all the help.

Report •

#30
May 6, 2017 at 22:01:05
Thanks for dropping back to let us know, it may help others at some point.

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


Report •

Ask Question