Why PC keeps freezing?

Jun 5, 2019
2
0
10
0
Hi all, I know my way around a PC but when it comes to diagnosing issues, I’m a bit helpless. The issue at hand is: I have a PC which posts and boots to windows and at any random point, the PC will completely lock up (nothing works, not even the physical reset button on the PC or Ctrl + Alt + Del) and a hard shutdown is required (by holding the power button or flipping the switch on the PSU). To put this into a bit of context, this was my partners PC which stopped being used about 6 months ago. In that time, I swapped out a few parts (fans, SSD, etc.) to use in a different build. That other build is now completely scrap. (second hand Chinese PSU shorted and burnt out everything in the process. I know, I cheeped out big time.)

Anyhow, I hooked everything up in my partner’s build (everything was working as far as I can remember) to use as my own and upon booting to windows, the above mentions issue occurs (No BSOD, no black screen, just freezes). Now, just a few things to note. Firstly, the PC will freeze at any time, whether it’s under load or not, temps looks to be normal, and no weird processes that I can see running. I haven’t even been able to install games yet. I have benched the PC and although the system does crash during normal use, it has not crashed in the 2 times I have benched it (CPU temp reaches ~49 degrees).

I did notice something very strange though:
I had hooked my AIO pump to the ‘CPU_FAN’ header and the radiator fan to the ‘SYS_FAN2’ header and left the MB fan speeds as default. The first crash happened with these default settings within 10 mins of the PC booting. I then checked the fan settings in the BIOS and noticed the pump was set to gradually increase speed with temp/workload. Although it would have made much difference anyway, I switched the AIO pump to the SYS_FAN2 header and the radiator fan to the CPU_FAN header. I then went through bios and set the pump to increase speed from 80% gradually to 100% with temp/workload and left radiator fan as CPU defaults. To my surprise, the PC didn’t freeze instantly on boot. I left the PC to download fortnite overnight and woke up to find the PC had frozen again, but with ~6 hours of up-time. This was a lot more than my initial 10 mins of up-time.

Upon some further research, I found my pump is rated at 2.4Amps while the motherboard fan headers are rated to a maximum of 2.0Amps. But even in this case, it doesn’t make much sense as the PC was freezing more often when the pump was set to 30% power with gradual increase. If the CPU pump was overloading the MB, the crashes should happen more often when the pump is set to high, not low power. I know this has something to do with the AIO cooler, I’m just not sure what it is/how to fix it. Also, would an over-current even cause the type of crash that I’m seeing? I would expect a full system shutdown if there was power issues? I’m almost certain it’s a power issue relating to the pump/fans. Let me know what you all think.
PC Specs:
CPU: i7-7700 3.6GHz (Turbo to 4.2GHz) SR3383 (Non K)
GPU: Gigabyte GTX 970 G1 (No OC)
RAM: 4 x 4GB SkHynix 2133p (Non-ECC)(No OC)
MB: Gigabyte GA-H270-HD3
PSU: Antec HCG-620 (80plus Bronze 620W (non-modular), 12V rail @ 48A & 576W)
Boot Device: Samsung 860 EVO 250GB m.2 SSD (5 Hours ‘ON’ time)
CPU Cooler: Asetek 550LC AIO (Corsair equivalent would be the Hydro H55 I guess?)
Fans: 3 x 120mm Fans (1 x Radiator Exhaust, 1 x Case Exhaust and 1 x Case Intake)
OS: Windows 10 x64 (Build 1809)
 
well with my H110 tells you to turn off all temp related stuff in the bios probably the motherboard causing the problems and you should be in cpu fan header but with such a small cooler you would be better with air
 
Jun 5, 2019
2
0
10
0
UPDATE - Till now I have tried

Remove SSD, Re-Insert, and Re-Install Windows
Reset BIOS to defaults.
Change windows power setting to high performance and never turn off anything;
Re-Seating RAM;
Memtest with 0 errors at pass 2/4;
Change fan settings in BIOS;
Connect AIO pump to PSU molex +12v instead of motherboard headers;
Checked GPU thermal paste (was unsurprisingly not enough). Cleaned and re-applied;
Removed GPU altogether and use onboard graphics;
Completely disable windows update through GPEDIT;
Checked CPU thermal paste. Was enough but re-applied anyway;
Increase CPU voltage from MB default 1.75v to 2.2v (1.75 seemed a bit too low);

WITH ABSOLUTELY NO LUCK!! Someone please help me!

All temps look normal, no weird processes that I can see. No BSOD or event logs.
Crashes are becoming more random now, could be 6, 3, 2 hours or even 10 mins before a crash happens.

I feel the crashes occur when the PC is connected on the network, but I don't have a a good way to check this theory. I have tried USB WiFi and Ethernet cable, and after a couple hours of downloading, the PC usually locks up. Thoughts?
 
Last edited:

ASK THE COMMUNITY

TRENDING THREADS