Homebuilt PC Randomly Freezes and Black Screen after POST

Goober Trooper

Reputable
Jun 9, 2014
47
0
4,540
I'll just cut straight to the chase. At seemingly random times my computer will freeze. Just a hard freeze, no sounds playing in the background or anything. It's just a still frame, forcing me to restart. Further, when the computer is booting up, it'll get past the enter BIOS prompt, then promptly blackscreen. It won't load Windows, or even do what I assume is search for a secondary boot, like a flash drive or disc. That's what I assume the flashing white underscore usually means, but my memory might not be the best.

However, if I restart my system at the blackscreen, it'll load Windows fine, and I won't have a problem until the computer freezes again, or a display driver crashes and recovers. (Which I'll save for another post if I have to.)

I've heard that the freezing issue could be caused by the power supply, the CPU overheating, and RAM. I've managed to rule out the PSU, considering it's happened with two different PSU's in place, and it's certainly not the CPU overheating, or just overheating in general. Considering that it's frozen up while the computer is in sleep overnight in a cool room. Even then, system load seems to be irrelevant. It's just as likely to freeze while watching a Youtube video as it is during serious gaming.

RAM is the only component that I'm not certain about. I read that reseating the RAM solved it for someone who had a similar problem. I reseated the RAM while switching out PSU's and when I booted back up, I was met with no blackscreen after the BIOS prompt. However, it wasn't long before it froze up again, however this time when I restarted it seemed several USB ports on the Motherboard temporarily stopped functioning, as I had lost use of my mouse and wireless adapter. I was able to restart it again which seemed to restore the USB ports.

In any case, what I have done so far is updating the BIOS in hopes of fixing... something, and today I've updated the chipset as well. I suppose I could run a memtest, but I'm not entirely confident on how to do that or if it'd even be worth doing.

And of course, here are my PC Specs. Hopefully you guys can come up with some reasons as to why this is happening!


Operating System: Windows 8.1 Pro 64-Bit
Processor: Intel Pentium CPU G2030 @ 3.0Ghz (2 CPUs), ~3Ghz
RAM: CORSAIR 8GB (2 x 4GB) 240-Pin DDR3 SDRAM DDR3 1333 Desktop Memory Model CMV8GX3M2A1333C9
GPU: NVIDIA GeForce GTX 660 EVGA Superclocked
Motherboard: ASUS H61M-A/USB3 LGA 1155 Intel H61 HDMI USB 3.0 Micro ATX Intel Motherboard with UEFI
Hard Drive: WD BLACK SERIES WD1003FZEX 1TB 7200 RPM 64MB Cache SATA 6.0Gb/s 3.5" Internal Hard Drive

I know I have a 500W power supply, but beyond that I don't know the brand or anything. If there's a way to find out without taking it out of the PC I'd be happy to know.
 
Memtest86+ is a great way to check for RAM errors. I would run an overnight test. If it comes back with any errors then I would get your RAM replaced under warranty. Properly functioning RAM will not have any errors during an overnight test. There are a number of ways to run a Memtest86+ test. IMO the easiest way is to download the "Download - Pre-Compiled Bootable ISO (.zip)" file and burn it to a CD. You can then boot the system from the CD and the test will start automatically. You can also use a USB thumb drive if you don't have a CD drive.

I also don't think you can fully eliminate overheating or PSU issues without some basic troubleshooting. A program like HWMonitor can monitor temps. You will want to post your idle/load CPU temps. A program like Prime95 can be used to put the CPU under load.

Trying two low-quality PSUs doesn't rule out the PSU as the possible cause of the problem. A quality PSU is the most important component to a stable system. The fact that you don't know what model/brand the PSU is likely means it's a cheap one that should be replaced immediately. There should be a sticker on the PSU with manufacturer/model/voltage output info.
 

Goober Trooper

Reputable
Jun 9, 2014
47
0
4,540


I'll give this all a shot. Thanks man!
 

werberman

Honorable
Jun 17, 2014
136
2
10,715
First and foremost, if possible, boot the system with only one stick of RAM installed (trying each individually). The likelihood of both sticks having problems is slim to none. I've been fixing computers for over 14 years and this case is... strange. PSU is unlikely the cause, cheep PSUs do just fine and seldom fail (though, when they do, the results can be spectacular XD). Honestly, my gut says motherboard if for no other reason than the uniformity of the problem (particularly the symptoms immediately after POST), but the graphics card could also be the source of your symptoms; if the connection is faulted in any way, even the vibrations caused by the HDD spinning up could cause failure.
 

Goober Trooper

Reputable
Jun 9, 2014
47
0
4,540
It's been a while, but I've managed to do some work on this issue.

Been keeping an eye on the temperatures just to make sure it's not overheating. Highest temperature I've ever gotten was 80 Celsius on the GPU from gaming, but the fan kicked up to a higher speed and kept it cool (Around 65C). When PC is idling it's between 28C and 40C across all components, which seems to be within acceptable parameters.

I've also managed to fix the black screen after POST. While I was doing a monthly cleaning of the PC, I noticed a front port that wasnt plugged in. Unsure if it was an audio jack or the HDD light I hooked it up to see, but to do that I had to take the GPU out and put it back in. Ever since then the black screen was fixed. I'm doubting that plugging in the front audio jack (since the HDD light still doesnt seem to be on) fixed the problem, and I'm leaning more towards the fact that I had to take out and reinsert the GPU. Not sure how that's connected but I'll take any fixed problem.

However, the random hard freezes are still no longer fixed, in fact I'd say they might have gotten worse. Recently the freezes have only been occurring about once or twice within a period of two days. (Which is better than what I was dealing with initially.) After aforementioned cleaning session and hooking up the GPU to the monitor via HDMI, I've experienced 4 hard freezes within 24 hours, forcing a hard reboot.

Memory test came back with no errors either. I still want to blame the Motherboard, but if I wanted to label a secondary suspect I would point at the GPU. I plan on taking it out for a day or two and running on integrated graphics to see if the issue still occurs. Do I need to uninstall current GPU drivers before doing this?
 

werberman

Honorable
Jun 17, 2014
136
2
10,715


Well, I'm sure you've done this already (considering how long it's been), but for sake of completion-ism, no, you do not need to uninstall the current GPU drivers before removing the card - OS will pick up that the device is no longer plugged and disable them automatically.

In case this still hasn't resolved, I should point out that there is a chance (albeit a smaller one) that there is an issue with the PCI bus - if you have a spare video card kicking around I would test that in the system as well before writing off the graphics card (assuming the issues were fixed by its removal).
 

Goober Trooper

Reputable
Jun 9, 2014
47
0
4,540


That's some good advice for the future! But this issue has actually been resolved. Was a faulty motherboard causing the issue. Upgrading to an ASRock Mobo fixed it for me :p
 

werberman

Honorable
Jun 17, 2014
136
2
10,715


Glad to hear the issue was resolved :)