VRM running too hot - system unstable

Jun 28, 2018
2
0
10
Since I got this new PC, I noticed that while playing heavier games at higher resolutions, the PC would eventually turn off. It turns completely off and then turns on again. windows event viewer registers Kernel Power error 41 which, afaik, isn't really helpful since it wouldn't register a mobo shutdown as a different event.

After doing some digging I decided to run stress tests. GPU is working mighty fine at full load, however when using prime95 I noticed that while CPU temps didn't reach above 75 celsius, the VRM sensors were registering 94 celsius (that's when I stopped the test) in HWINFO.

Not sure how to proceed from here. I'll still check the psu with a multimeter but is that 94 celsius temp for the VRM good? It seems extremely high and i decided to stop the test until I get some answers on this. Could not replicate the turning off problem during those tests.

PC Specs:

Operating System
Windows 10 Pro 64-bit
CPU
Intel Core i7 6700K @ 4.00GHz 24 °C
RAM
16,0GB Dual-Channel Corsair
Motherboard
Gigabyte Technology Co. Ltd. Z170MX-Gaming 5 (U3E1) 34 °C
Graphics
SONY TV *00 (3840x2160@60Hz)
4095MB NVIDIA GeForce GTX 1080 (Gigabyte) 32 °C
Storage
447GB SanDisk SDSSDA480G (SSD) 42 °C
931GB Seagate ST1000DM003-1SB102 (SATA) 42 °C


Case is a carbide air 240

While i'll get a multimeter and measure the psu outputs, could this be causing the shutdowns? Am I missing something else?

While we're at this, those HD temps are in idle. Should i be worried?

Many thanks!



 
Jun 28, 2018
2
0
10


Well it doesn't reach those temps unless running Prime95, otherwise the max i've logged while gaming is around 50.

Case fans are actively controlled by mobo temps, yeah. I ran tests with open lid, extra vans, and the shutting off keeps happening. After extensive research I found a ton of suggestions, from turning off turbo boost in bios (did not help) to uninstalling geforce experience (!). So far, I have no idea what to make of this.

Windows event viewer registers kernel power 41, but HWINFO does not show any voltage fluctuation. Where to go from here?
 

luckymatt42

Upstanding
May 23, 2018
446
1
360
94 C might be slightly "warm" for VRM, but most decent quality VRM components should be good to 120 C or more. Not sure exactly what might be causing your issue, but I highly, HIGHLY doubt that VRM temps are the culprit.
 

luckymatt42

Upstanding
May 23, 2018
446
1
360


You do not need to populate every fan in the case. Most decent cases come with at least a couple, you might add another front intake fan, or maybe an exhaust fan, but you do NOT need 9 fans for proper cooling...again considering that 94 degrees on the VRM is not an issue.