CPU overheating and shutting down within 5 mins (Please help)

Backtrap7 · Jun 24, 2016

Hello,

My home built gaming PC has been working fine for about 6 months now. Recently (After returning from a trip [son was using PC]) when I turn on my PC, within 1min, the fan is running really fast and by 5 min the PC shuts down. Below are my computer specs and also the troubleshooting I have done so far. Your help would be greatly appreciated.

Mobo - ASRock Z77 ext
CPU- i3770k
ram - 16gb
Win 8
PSU - Seasonic 650w
GPU - GTX970
Antec liquid cooling for CPU.

I went to bios and check the temp of CPU. At startup it was 42 c, within 5 min it was 79c.

that is with no other programs running.

I have done the following troubleshooting:
- Cleaned out dust (was not much at all) PC is well ventilated.
- Cleaned out old thermal paste and added new
- Removed both Ram and GPU to ensure clean and no issues.
- Checked Mobo for burnt connections or blown capacitors.
- Removed Mobo battery to reset Bios

I can not think of anything else to do. I have heard some folks say it could be a bad PSU, but also read that it would not cause the CPU to overheat.

I use the PC for gaming, and it has performed great so far. My son likes to play RUST and I know it is a resource hog... can that overwork the CPU and kill it?

Any advice you can give would be greatly appreciated.

Thank you,
B

Karsten75 · Jun 24, 2016

How old is your son?

scout_03 · Jun 24, 2016

did you check liquid cooler for leak and that the fan work also use this on sensor page to read temp voltages and fan speed http://www.hwinfo.com/download.php

nukemaster · Jun 24, 2016

At this point, but first guess would be a pump failure on the liquid cooler.

If liquid does not flow, things get pretty toasty.

Backtrap7 · Jun 24, 2016

Thanks for the responses...

The liquid cooling system is a sealed system. No leaks detected and the fan runs very well. It even changes color based on temp. Starts off as blue, then by 5mins its red.

I just dont understand why it would overheat from a cold start within 5min with NO real load on the system.

madmatt30 · Jun 24, 2016

He's not on about the fan - the block that fixes to the CPU is actually a pump , if it's not running properly the coolant will just heat up.
Which sounds like exactly what's happening.

Removed solution as I assume you clicked it by mistake.

Backtrap7 · Jun 24, 2016

madmatt30 :

Yes sorry, I click solution as a mistake. (Sorry, the "Add solution" is selected automatically and I do not see it until its to late and i hit post).....

Is there a way to tell if the Sealed coolant system is flowing? Should I try removing the coolant system from the CPU and see if it stays cooler for a longer period?

Again, thanks for the advice.

madmatt30 · Jun 24, 2016

2 fingers on the CPU block & you should feel a fair bit of pump vibration - it still doesn't mean it's working properly though.

If there's no vibration then it's fairly certain it's failed.

Backtrap7 · Jun 24, 2016

madmatt30 :

Thanks Mad, I will do as you suggested.

Appreciate the help!

Update: I am really confused now. I checked the cooling block and also the radiator. The cooling block had a slight vibration and the radiator remained cool the entire time.... the CPU continued to rise in temp, but the air and radiator remained cool.

I also took the Cooler totally off so just the CPU was showing and started from scratch. Turned on the PC and it still continued to rapidly get hotter.

I have heard that maybe bad RAM an cause this.. anyone hear that? Just not convinced its the pump.

thanks

nukemaster · Jun 24, 2016

Sounds like a failed pump even more if the radiator is not getting hot.

The heat transfers in the liquid from the cooling pump/coldplate(thing on the cpu) to the radiator to be cooled off with the fan.

You said the pump led is on, so it should be getting power. Can you tell me if it is connected to the board or the power supply?

EDIT
Did you just say you ran with no cooler? That is NOT good for the system, cpus generate LOTS of heat that needs to be removed.

You can always test by installing the stock cooler if you still have it to see how the temps are.

I forgot to connect a pump on an all in one in the past, can you guess what happened? It was just like what you are having(but mine was on a video card).

Backtrap7 · Jun 24, 2016

nukemaster :

The cooling unit is plugged into the Mobo. I tried 2 different CPU fan connections.

I guess it is the Cooling unit... I just did not think it would overheat that fast... literally from a cold start to 80c in less then 5 min (with zero programs running)

Also I only ran it without any cooling for less than 2-3 min. The temp climbed way fast so I shut it down.

(Is it possible to have some kind of malware or virus that can trick the cpu into thinking it is overheating and shut it down? reason I ask is the PC was fine while I was away. Son was on it and maybe he downloaded malware? I did notice my Bitdefender was acting weird... it was not updated and failed to connect the first time)

nukemaster · Jun 24, 2016

You may be able to check the temperatures in the bios(Bios temperatures read lower most times and depending on if the power management works, your cpu may take longer or shorter to heat up). You should also be able to monitor the pump speed(Please connect it to a fan header with NO control).

This is all that is in those coolers most of the time. This is not much surface area and without constant liquid flow things get hot pretty quick.

Image from Asetek

You can generally use the task manager to see if anything is running the cpu hard. Mining malware commonly causes high temperatures, but should not overheat the system. I have never seen any software fake an overheating cpu. Too many system variables to make it worth doing(and nothing to gain for a hacker/malware maker).

Please never run a cpu without a cooler in the future even for testing. If you did not have a modern cpu(they clock down very fast to prevent damage), it would have been toast.

Backtrap7 · Jun 24, 2016

nukemaster :

Nuke,

Appreciate the help. I was checking the temps on the bios screen. I also booted in safe mode and checked task manager...nothing was running the CPU more than 1-2%.

I will order a new cooler unit... hope it works

thanks again,

nukemaster · Jun 24, 2016

Does the board show a fan speed on the pump header by any chance?

Backtrap7 · Jun 24, 2016

nukemaster :

Sorry, I am not sure what you mean. The Antec cooler wire is hooked to the CPU Fan 1 spot (that wire goes into the Antec head attached to CPU and also to the fan connected to the radiator) There is a CPU fan 2 and CPU Pwr Fan 1 spot next to it which are empty.

nukemaster · Jun 24, 2016

When you enter the bios hardware monitor, does it show a speed for the CPU FAN 1 location.

Backtrap7 · Jun 24, 2016

nukemaster :

Yes, the RPM's were:
start- 820
1min- 1450
2min - 1912
3 min - 2170
at 4min cpu temp was 77c. I shut it down.
(aren't those RPMs just for the fan attached to the radiator and not the pump?)

I did notice that the cooler tubes were both cool to touch.. it was slightly warm near where the tube enters the CPU part. I am thinking that maybe the pump is not circulating the coolant.

nukemaster · Jun 24, 2016

In your case it is the fan(on the radiator) for sure.

Shame since some units show the pump instead. This would flag a failure right away in the bios.

Do you have the stock cooler by any chance? It would be better and work while waiting on a new cooler.

Backtrap7 · Jun 24, 2016

nukemaster :

I do not have the stock fan anymore. I will pick up a new cooler tomorrow

I thank you for your help.

nukemaster · Jun 24, 2016

Please keep us posted. I want to know how this turns out.

This is also a reason why all in one coolers are not always recommended. Pump failure is sometimes worse than fan failure since the cold plate has very little heat dissipation ability.

For this reason I keep an eye on my temps and pump/fans speeds with HWinfo64. With Afterburners included Rivatuner statistics server, I can see temperatures/fan speeds in games as well.

Make sure you get all the old thermal paste off when swapping coolers because that can cause some issues as well(poor performance). 70+(90 preferably) isopropyl alcohol works great to remove old thermal paste(q-tips, coffee filters or a lint free cloth work well for cleaning).

madmatt30 · Jun 25, 2016

im with nukemaster in all honesty -there is far far less to go wrong with air coolers - just a simple fan that is easy to see if its not working.
I understand aio's from a cosmetic point of view & for custom loop systems in high end rigs but the 120mm aio's are just not worth the money or hassle really.

you could spend $35 on a decent 120mm tower cooler & it will likely outperform that antec cooler easily.

Backtrap7 · Jun 25, 2016

nukemaster :

Nuke,

It was the cooling unit (Bad). I replaced it with a Corsair H100i V2 and the temps are holding steady at 28c.

I will also check out the HWinfo64 program you recommended.

Thanks again for all the help!

nukemaster · Jun 25, 2016

Glad you got it working.

If the H100i V2 is anything like the H80i it will show pump speeds to the bios(and get power from the power supply to ensure the pump always runs full speed.).

CPU overheating and shutting down within 5 mins (Please help)

Reputable

Judicious

Titan

Titan

Reputable

Titan

Reputable

Titan

Reputable

Titan

Reputable

Titan

Reputable

Titan

Reputable

Titan

Reputable

Titan

Reputable

Titan

Titan

Reputable

Titan

Share this page