I have recently built a new system and I am having a lot of problems with it crashing while under load.

 

MB: Asrock Z270Pro4
CPU: I5-7600K (Not overclocked)

Fan: Corsair H100 water cooler
GFX: Sapphire R9 Fury
Mem: 16Gb HyperX Savage 4x4Gb
PSU: Corsair 750W

OS: Windows 10

I do seem to be seeing high temperatures reported for the motherboard even though I have 5 case fans installed and the ambient room temp is <21°C. Under load I am seeing motherboard temperatures of >45°C. Is this normal/anything to worry about.

 

Any suggestions appreciated?

 

Thank you 

I had a notebook with random crashes.. in the end it was a faulty memory.

Changed from 8 to 4GB, then it was rock solid.

 

Memory - I will try memtest on my memory tonight and see if any errors are reported on the memory

 

VRM - I was hoping that with 5 separate fans I would have sufficient airflow inside the case but I will try as you have suggested.
I am at work at the moment and cannot visualise the layout inside the case but why do you think the watercooler will have obstructed the VRM?

 

Regards

 

 

 

Hi Wardog,

 

Would you recommend I sell my liquid cooling and install a conventional air cooler?

I already have a 120mm mounted near the top rear of the case which is blowing in the direction of the VRM. Do you think this will be insufficient?

 

Thanks again

 

1. Moved fans from the H100 onto the outside of the case so pulling air from the case through the radiator (Previously they were on the inside and very cramped)

2. Changed direction of fan at top rear of case to vent (Previously blowing into case)

3. Changed 2 fans on front of case to pump air into the case.

 

I don't know if it is a coincidence but since doing this I have managed to play Total War for 3 hrs without any crashes when last night I could not play for 15 minutes without crashing.

 

When I check the temperatures using the Asrock A-Tuning app it still says my motherboard temperature is 41°C under load but maybe this is normal? (Seems hot to me)

 

Thanks again

 

" rel="nofollow - Parsec - I thought my earlier post described the air flow but obviously I was not specific enough.

Case - Corsair Carbide Spec 03

We finally have a description of the air flow in this PC, although it is incomplete. We don't even know what PC case is being used. - (Corsair Carbide Spec 03)
Still not sure where the radiator is mounted, the top I assume? (Radiator is top mounted)

Radiator fans pulling hot are from the GPU onto the board and VRM heat sink? I would be shocked if my board temperature was 45C, mine is 24C currently. (Radiator has two fans mounted above the radiator pulling air from the case)

The description sounds like way to many exhaust fans, with who know what source of fresh air, hopefully the case has some vents on the top and bottom. But bottom vents will not help with the VRM temperature. The radiator and front fans as exhaust would be fighting each other for air, without an equal area of air intake somewhere. (2 fans front mounted pumping air into the case; 2 fans above the radiator pulling air out of the case; 1 fan in the rear of the case pulling air out of the case) 

Where is the PSU in this PC? (Mounted in the bottom rear of the case)

Z270 Pro4, one small heat sink on a six phase CPU VRM stage, so not optimal under high load conditions and limited ventilation.

Micro ATX board, probably small PC case, obviously the radiator and case fan placement is not optimal.

(Z270Pro4 is classed as an ATX board I believe; Carbide Spec 03 is not a small case; Where else can you fit a radiator if not at the top?) 

Video card dumping hot air into the case is one of the main problems, and not being removed from the case. (If there are 3 x 120mm fans venting from the case why would the hot air not be removed?)

What PC case are you using? Where is the radiator mounted? Radiator fans are intake or exhaust? What other case fans besides the top rear fan?

It has been on for nearly 8 hrs under load without any crashes so maybe I don't need to worry about the temperature as the most important thing is it isn't crashing anymore

 

 

 

I have tried a range of different stress test utilities on the system and they all pass apart from MEMTEST86 which reports multiple errors within minutes.

 

I have used Memtest on previous builds and never seen errors so I assume this is not normal.

Do you think this could be the cause of my random crashes and actually nothing to do with temperature?

 

I have reported the issue to Kingston technical support as the memory is only 2 weeks old so hopefully if it is the memory I can get a replacement ASAP

 

 

 

With 1 pair I was able to leave Memtest running for 2 hrs with no errors detected with either of the other 2 I had errors within seconds.

 

They were bought as a kit of 4 x 4gb but it seems 2 of them are not working correctly so I will report my findings back to Kingston and see if I can get a replacement.