Question Multiple (or not) problem/s

Hello,

I'm having some problems with my PC, and thought i would have fixed it with a new PSU but that seems not to have been the case.

The initial issues had noticed were that the PC hangs on restarting, didn't have power to the USBs whilst shutdown and sleeping, and occasionally crashes at seemingly random times. Since the restarting issue occurred not only from Windows, but also from places like the BIOS, these together seemed like a PSU issue. So I upgraded from the original 2014-ish FSP 300W that it came with to a MWE White 450w V2. Now I do have power to the USBs during shutdown, but the other problems persist.

Crashes occur in a somewhat odd manner.
One type which I've had three times in the space of about 6 months - video output freezes, with coloured static over it (the static is stationary, like a still shot of someone wiggling a video connector). Holding the power button to turn it off, then turning it on again, I get to the manufacturer logo, the Windows logo spins and then I see the static again. After this a boot loop occurs and/or get POST beeps which are apparently relating to RAM. To remedy, I must clear the CMOS and then boot with only 1 RAM module, then after that I can put the 2nd module back in and it runs again.
Second one seems to be random BSODs which after occurring the PC is fine. I have recently had Critical Process Died and System Service Exception. Of course I do still have to hold the power button down since it doesnt automatically restart, but other than this it then boots back up normally.

I have tried different hardware combinations, including only CPU and single RAM, no SSD/HDD, yet it still doesnt restart successfully from the BIOS - only getting stuck at with a blank screen after Discarding Changes and Exiting. This led me to believe BIOS issue updating to the newest available changed nothing. I have tried Memtest86 but there was no issues detected after performing long runs. My next plan is to turn CSM off and see if that helps, but I'd have to reinstall Windows so before I do that I suppose I should check crash logs - something I am not sure how to do. In the event viewer there are critical Kernal Power events, which based on the times they occurred do line up with my crashes. Perhaps the logs would indicate a particular faulty device. But as I said earlier, this is something I have never looked at so help regarding this or any other ideas you have would be very useful.

Specs:
Acer Veriton M2631G
Windows 10 20H2 64-Bit
i5 4690
16GB (2x8GB) Crucial DDR3-1600 CT102464BA160B.M16
Samsung 850 EVO 120GB (SATA)
Seagate 1TB HDD
Intel HD 4600
Cooler Master MWE White V2 450W
 
Last edited:

Ralston18

Titan
Moderator
No GPU - correct?

How many USB devices? What are those devices?

Event Viewer is one way to discover problems.

Another way is to use Reliability History which is much more user friendly and provides a useful timeline format. E.g. mapping captured errors, etc. to crashes.

Overall, I noted two immediate concerns:

1) Is the Samsung 120 GB SSD the boot drive? If so, how full is the drive? In any case that 120 GB is too small. Minimum 240 GB.

2) PSU - also likely that 450 watts (although better than 300 watts) is still not enough to keep up with peak system power demands.

First take a closer look at the errors and crashes: varying types of errors are, in my mind, one sign of a faulty or undersized PSU.

Second - power down, unplug, open the case.

Clean out dust and debris.

Double check by sight and feel that all connectors, cards, RAM, and jumpers are fully and firmly in place.
 
Thanks for the reply.

I will have a look at Reliability History tomorrow as its getting late tonight, I will update with what i find tomorrow.

The SSD is the boot drive - while it is very small, I have really only got the Windows install and a few programs on it, still got over 30% free space (40gb) and only has a 3TB total write history on it should be healthy.
I regularly clean the PC so its clean and the thermal paste is regularly replaced too. All connections are double checked and i have tried different RAM/hardware combinations including breadboarding, but it still will not restart from the BIOS.
I'm hoping that the 450w isnt an issue - since yes, I am on the integrated graphics. I went for the MWE White V2 because it seems to be mid end tier on most lists and since this is a low end PC I believe it should be adequate. I only got it a few days ago so (as a last resort) could send it back for a replacement), but I'm not certain that's the issue.
 

Ralston18

Titan
Moderator
Having only 30% free space may be getting a bit close to the line.

Personally I prefer to keep any drive at least 30-20% free and your PC is within that margin.

However, every PC is different.

Another thing to try is using Task Manager and Resource Monitor to observe system performance.

Use both tools but only one at a time.

General idea being to discover what PC resources are being used, to what extent (%), and what is using any given resource.

I am still leaning towards a power supply problem but you are correct: check Reliability History and otherwise try to learn more about what is (or is not) happening.

Also - late thought.

Some initial issue (power or otherwise) could have caused some file corruption.

Use "sfc /scannow" and "dism" to clean things up.

https://www.lifewire.com/how-to-use-sfc-scannow-to-repair-windows-system-files-2626161

https://www.howtogeek.com/222532/ho...-system-files-with-the-sfc-and-dism-commands/

And be sure that all important data is backed up, proven recoverable, and readable. At least 2 x copies and preferable on other drives and hosts.
 
System resources are all OK.
i5 4690 currently at about 5%.
RAM at 5GB out of 16GB, that's with Edge and background programs like Steam.
SSD and HDD are both at 0% with occassional peaks (obviously get higher if I open another program)
Intel graphics are at <5%.
Everything there seems normal for practically an idling system.

I did some Windows and driver updates yesterday and sfc/scannow today, which seems to have fixed the random BSODs I was getting. They had only started a few days ago so in all honesty I should've realised that beforehand. Reliability History did catch all of those BSODs and more crashes than I had realised were happening, but today there has been none. So I'm hope the Windows side of things has been fixed.

The initial issues had noticed were that the PC hangs on restarting, didn't have power to the USBs whilst shutdown and sleeping, and occasionally crashes at seemingly random times. Since the restarting issue occurred not only from Windows, but also from places like the BIOS, these together seemed like a PSU issue. So I upgraded from the original 2014-ish FSP 300W that it came with to a MWE White 450w V2. Now I do have power to the USBs during shutdown, but the other problems persist.
However, this issue restarting is still present, and it occurs not exclusively within windows. For example, if I enter the BIOS, and then discard changes and exit, it hangs with a blank screen, eventually my monitor goes to standby. Fans are still running and the power button light is still on. Occasionally the HDD has a 'blip' of activity. This also happens if I take my drives out and hit CTRL+ALT+DEL to restart at the no boot media detect screen. In other words, the computer itself is unable to restart.

Crashes occur in a somewhat odd manner.
One type which I've had three times in the space of about 6 months - video output freezes, with coloured static over it (the static is stationary, like a still shot of someone wiggling a video connector). Holding the power button to turn it off, then turning it on again, I get to the manufacturer logo, the Windows logo spins and then I see the static again. After this a boot loop occurs and/or get POST beeps which are apparently relating to RAM. To remedy, I must clear the CMOS and then boot with only 1 RAM module, then after that I can put the 2nd module back in and it runs again.
This crash also may still occur, since it is infrequent I only assume it has been fixed by a new PSU. Such a randomly timed freeze to me seemed like a PSU issue, and since it resulted in boot loops and POST failures, it was clearly outside of Windows. I'm hoping that the new PSU has fixed it.