[SOLVED] Pc shutting down under heavy cpu usage

[Whery]

Hello everyone, first of all I want to thanks everyone helping in this forum as I have learnt a lot and I am having lot of fun tinkering with my computer and you all have been of great help.

And here comes my problem... all started when I decided to give it a try to borderlands 3 , my pc randomly started to shut down but only when the game had finished loading and I was about to start the "in game main menu" but sometimes it didn't and I could play the game perfectly (not many fps and some suthering , but playable) without any major issues or shut downs. I read in the forums that borderlands 3 was very poorly optimized so I assumed it was this game problem and not my pc.

It still annoyed me that my pc would randomly shut down, so I cheeked for solutions and the main problems that could cause this "shut downs". After 3 days of trying different things , which I will list, I came to a dead end my pc still shut downs when I try to play heavy cpu games ( I also tried to play cs-go, which I could already play with the same pc set up at about 100+ fps without any problem and after all the borderlands 3 shutdowns the game also shut down in the main menu wich really concerned me) also some benchmarks such as heaven would shut down my pc during test.

Here is the list of the things that I tried:

1º Removing old PS for a new one ( Upgrade Thermaltake litepower 700 to Corsair RM750I ) With this change I could run Heaven benchmark and pc would not shut down, but games would still shut down when main menu started. Do I need more than 750W ? It is really weird to me as i could run it with 700 without any problem for a long time .

2º Antivirus, I ran antiviruses ( Malwarebytes, and windows full scan and nothing was found ) I still formatted win 10 removed all data and started brand new, installing all drivers and updating bios to 3.0 ( Z77 extreme4) and latest AMD drivers. Game would shut down again in main menu.

3º Checked for overheating problem with core temp to see if there was a problem of overheating. When running (playing Borderlands 3) and loading the game (100% core usage) , the core temps never went up 70º and the pc would still shut down. So I assumed that core temp could not be giving the right data and I proceed to removed cpu fan from the CPU to see if the temp increased and oh baby it did! . Temperatures reached +80º when loading the game and at that point I cut the PS as I didn't want to dmg the core and I assumed core overheating may not be the problem as it didn't shut down when reaching +80º and it did when it was at 60º.
( I want to add that my pc is pretty clean , and I also swapped the thermal paste from the core yesterday and temps didn't change)

4º I ran memtest86 to see if it could be any problems with the ram and there was 0 errors found.

So anyone have any suggestions of what could be the problem here? or what should I try next? Feel free to ask me any questions I will be glad to answer them!

Pc specs:
OS: Windows 10 pro
CPU: Intel core i7-2600 3.40 Mhz
Motherboard: Asrock z77 Extreme4
Ram: 16Gb (x2 8Gb) ram 2400Mhz ( running at DRAM 1066.5 MHz with dual channel, this main board max is 2133 Mhz)
Graphic card: Radeon Rx 480
I run OS in a 250 GB SSD (samsung 840 Evo) , and games in another 500 GB SSD (Crucial mx500), Also got 2tb solid state disk that I disconnected to save some W to see if that would help.
CPU cooler katana 4
Chasis fans x3 Noctua NF-P14s redux-900

 

[Piske]

First of all, your PSU is more than enough, unless it's faulty, it should be fine. I have a CPU with 130W TDP and an RX 580 oced to the max (1510mhz core/2250mem) pulling around 200W and my PSU is a Corsair TX550M. your GPU does not use that much, nor your CPU, soo you should be fine.

Temps should stay below 80ºC on that CPU. Which seems , according to your words that it's that case.

Normally memtest would see if the ram is faulty, or if there is a problem with voltage, which appears to not be the issue since it passed.

I think the issue is the voltage. Make sure that you leave CPU VCORE (or something like that on AUTO) in your bios. It should show around 1,25v depending on the Load line calibration on your motherboard.

If you want to deal with it yourself do this.
use HWMONITOR. See on Voltages where it says CPU VCORE. and watch minimum, current value and max. Write them down. i assume that max value on HWMONITOR on load will be less then 1,27v because if it's more then the problem maybe elsewhere.

watch the program while at max CPU usage. if the computer shuts down go to bios and put 0,05v on the CPU Vcore on the Bios. But first try leaving AUTO there. It's not the first time a motherboard applies less voltage then normal to a CPU. is your motherboard with latest bios?

If that does not solve you the problem, discarded RAM , PSU and CPU voltage....well...it seems there is overheating somewhere (VRM could be an issue). Hardware problems or incompatibilities always show up with blue screen.

 

[Whery]

First of all, your PSU is more than enough, unless it's faulty, it should be fine. I have a CPU with 130W TDP and an RX 580 oced to the max (1510mhz core/2250mem) pulling around 200W and my PSU is a Corsair TX550M. your GPU does not use that much, nor your CPU, soo you should be fine.

Temps should stay below 80ºC on that CPU. Which seems , according to your words that it's that case.

Normally memtest would see if the ram is faulty, or if there is a problem with voltage, which appears to not be the issue since it passed.

I think the issue is the voltage. Make sure that you leave CPU VCORE (or something like that on AUTO) in your bios. It should show around 1,25v depending on the Load line calibration on your motherboard.

If you want to deal with it yourself do this.
use HWMONITOR. See on Voltages where it says CPU VCORE. and watch minimum, current value and max. Write them down. i assume that max value on HWMONITOR on load will be less then 1,27v because if it's more then the problem maybe elsewhere.

watch the program while at max CPU usage. if the computer shuts down go to bios and put 0,05v on the CPU Vcore on the Bios. But first try leaving AUTO there. It's not the first time a motherboard applies less voltage then normal to a CPU. is your motherboard with latest bios?

If that does not solve you the problem, discarded RAM , PSU and CPU voltage....well...it seems there is overheating somewhere (VRM could be an issue). Hardware problems or incompatibilities always show up with blue screen.

Thx for your fast reply!

I did as you said before launching borderlands 3 the cPUVcores were: min 0,984 & max 1,216 , They stayed the same while loading game, I noticed the Voltages from the core were all using 1.271 V constantly when they were fully working ( it is the max V) , temperatures never reached above 70º , I printed screen with all the data but as soon as the game finished loading the pc shut down and I lost the image sorry for that..., Also this time it did something that it has never done before, the cpu fan started to spin at max speed as soon as the pc shut down, even when OS was down the fan kept spinning at max speed which never did when loading or playing the game, I had to cut power to make it stop.

And all the Voltage are on auto, and with the latest BIO update.

Edit: I also watched this vid

View: https://www.youtube.com/watch?v=UHXbQ_kyebw
, as u mention in the last part it could be an overheating problem in some other component or part which could be really hard to spot.

Edit2: Tried again but this time i took photos with my phone, and this time cpu fan didn't go crazy after shut down
https://drive.google.com/drive/folders/1BC0eahWSc6MaYG9mXZdunaoe5TQEcqRI?usp=sharing

 

[Piske]

Thx for your fast reply!

I did as you said before launching borderlands 3 the cPUVcores were: min 0,984 & max 1,216 , They stayed the same while loading game, I noticed the Voltages from the core were all using 1.271 V constantly when they were fully working ( it is the max V) , temperatures never reached above 70º , I printed screen with all the data but as soon as the game finished loading the pc shut down and I lost the image sorry for that..., Also this time it did something that it has never done before, the cpu fan started to spin at max speed as soon as the pc shut down, even when OS was down the fan kept spinning at max speed which never did when loading or playing the game, I had to cut power to make it stop.

And all the Voltage are on auto, and with the latest BIO update.

Edit: I also watched this vid

View: https://www.youtube.com/watch?v=UHXbQ_kyebw
, as u mention in the last part it could be an overheating problem in some other component or part which could be really hard to spot.

Edit2: Tried again but this time i took photos with my phone, and this time cpu fan didn't go crazy after shut down
https://drive.google.com/drive/folders/1BC0eahWSc6MaYG9mXZdunaoe5TQEcqRI?usp=sharing

Are you sure it was the CPU cooler? Maybe it was the GPU cooler. It has happened to me sometime.

if Vcore was auto, then i think i have to discard a cpu voltage problem. Althought 1,224v is low for that CPU. but as you said on max load there is 1,271v that should be enough, although i would monitor that just in case.

What i've seen is that your GPU goes to 76ºC but the cooler does not start (it stays below 1000rpm) there could be the issue with overheating. If the gpu cooler does not work properly, it can go to 90ºC easy and shut down the computer.

Install MSI afterburner, and tell it to run the gpu cooler at 100%. it should show in HWMONITOR around 3000rpm. i think it can be the issue

 

[Whery]

Are you sure it was the CPU cooler? Maybe it was the GPU cooler. It has happened to me sometime.

if Vcore was auto, then i think i have to discard a cpu voltage problem. Althought 1,224v is low for that CPU. but as you said on max load there is 1,271v that should be enough, although i would monitor that just in case.

What i've seen is that your GPU goes to 76ºC but the cooler does not start (it stays below 1000rpm) there could be the issue with overheating. If the gpu cooler does not work properly, it can go to 90ºC easy and shut down the computer.

Install MSI afterburner, and tell it to run the gpu cooler at 100%. it should show in HWMONITOR around 3000rpm. i think it can be the issue

Woah! I think you are completely right about this one I was monitoring the temperature of the GPU with the fans at 100% , and the moment the game usually shuts down my pc the GPU temperature spiked to 90º even with the fans at 100% , no wonder it shut down when the fans were just existing ...

here are some images with MWmonitor and afterburner while running borderlands 3 at 100% fan speed
https://drive.google.com/drive/folders/1aXGzQ0mxFuOFlUKbxr5wcocZxBTGPjIj?usp=sharing

So... it is time to upgrade GPU I guess? Because borderlands settings are at minimum and running the game at 100% fan speed seems like I will put GPU into too much stress .

 

[Piske]

Woah! I think you are completely right about this one I was monitoring the temperature of the GPU with the fans at 100% , and the moment the game usually shuts down my pc the GPU temperature spiked to 90º even with the fans at 100% , no wonder it shut down when the fans were just existing ...

here are some images with MWmonitor and afterburner while running borderlands 3 at 100% fan speed
https://drive.google.com/drive/folders/1aXGzQ0mxFuOFlUKbxr5wcocZxBTGPjIj?usp=sharing

So... it is time to upgrade GPU I guess? Because borderlands settings are at minimum and running the game at 100% fan speed seems like I will put GPU into too much stress .

Does HWMONITOR read 5000rpm on the GPU cooler? That is not possible. That is due to a GPU cooler malfunction.

So first of all i would try to see if the problem is fixable. Try researching for those GPU fans. For example, if you have a Sapphire like mine, that problem will not be an issue.
That problem should be an "easy" fix if you can get the pieces for replacement.

If not, then you´ll have to update, obviously depending on you budget.

 

[Whery]

Does HWMONITOR read 5000rpm on the GPU cooler? That is not possible. That is due to a GPU cooler malfunction.

So first of all i would try to see if the problem is fixable. Try researching for those GPU fans. For example, if you have a Sapphire like mine, that problem will not be an issue.
That problem should be an "easy" fix if you can get the pieces for replacement.

If not, then you´ll have to update, obviously depending on you budget.

Sorry I didn't answer earlier as i had my pc torn off again xD but seems like I finally solved it thanks to you!! It was a problem with the GPU when I saw the spike going to 90º I thought it has to be either the graphic card is not good enough or the GPU cooler was not working properly so I disassemble the GPU and if yet there was some dust inside the main problem I think it was the thermal paste ( I didn't even know it has thermal paste in it) So I replaced for new one and Boilá... 60º GPU while playing borderlands 3 at 100+ fps and fans not even sweating !! I took some pictures with my phone inside of the GPU as there was some kind of "stikers" that were broken not sure if they are going to be a future problem

https://drive.google.com/drive/folders/1EKAuyYQSKW8qF2rtRoZTIH_oa_wxSy5S?usp=sharing

Again thank you very much for your help Piske!