[SOLVED] PC freezing with no indication of cause

[AJStellios]

For the past half a year I have been experiencing random whole computer freezing, which can happening when playing an intensive game or simply being idle on desktop. The freezing will start with being unable to click on buttons, my keyboard will then stop responding, mouse will stop moving, and eventually audio will cut out. There is no BSOD or error message. The only fix is to hard reboot the pc. Unplugging the hdmi cable and replugging it will just show a black screen. The freezing also doesn't occur after a set length of time. I've had my PC run fine one week, then the next crash 3 times in a day.

Having this issue for the past 6 months I've tried countless ideas, none which have worked:

• replaced m.2 SSD and reinstalled Windows 10 multiple times.
• replaced PSU.
• got my hands on a 3070ti, my PC didn't crash for a week after so I thought it may have been a GPU issue. Since then my PC has crashed everyday for the past week.
• ran windows memory diagnostic and memtest86 and had no errors.
• all temps are good.
• there are no errors in event viewer, apart from the forced shutdown I have to do.

My PC specs are:

• Ryzen 5 5600x
• RTX 3070ti
• MSI B500i Gaming Edge WiFi
• 16GB DDR4 Memory
• 1TB Sabrent Rocket m.2
• Corsair RM750X PSU

I'm completely out of ideas and money for replacing parts, so any suggestions would be greatly appreciated. Is there any chance this could be a CPU or motherboard issue and how would I know?

Thank you

 

[Ralston18]

Also look in Reliability History for error codes, warnings, etc. that may precede or correspond with the computer freezes.

Power down, unplug, open the case.

Clean out dust and debris.

Ensure by sight and feel that all connectors, cards, RAM, and jumpers are fully and firmly in place.

Check connections to case switches as well.

Look for signs of damage: bare conductor showing, melted insulation, kinked or pinched wires, blackened, browned, or swollen components. Any signs of damage.
Also check around the I/O panel: is it secure? Will gently wiggling any of the connected cables/plugs cause a freeze?

Check power connection path to the computer: wall outlet, surge protectors, power cords, etc.. Could be a problem where there is a very brief loss of power.

What was the original PSU that you replaced?

Is the Corsair RM750X PSU new, used, refurbished?

 

[AJStellios]

Thanks for the reply.

Having checked reliability history, the last few freezes have came up with just a "Windows was not shut down properly" error. However, a week ago a had a hardware error. I'm unsure if this is related but here is the error:

Source
Windows

Summary
Hardware error

Date
‎28/‎08/‎2021 18:32

Status
Report sent

Description
A problem with your hardware caused Windows to stop working correctly.

Problem signature
Problem Event Name: LiveKernelEvent
Code: ffab
Parameter 1: 9
Parameter 2: ffffe387333562f0
Parameter 3: d20
Parameter 4: 0
OS version: 10_0_19043
Service Pack: 0_0
Product: 768_1
OS Version: 10.0.19043.2.0.0.768.101
Locale ID: 2057

Extra information about the problem
Bucket ID: LKD_0xFFAB_ibtusb!BtFdCreateLiveKernelMinidump
Server information: 7e29bc22-3ddb-40c6-bfa5-3c17cc8a59c1

I've also checked inside the case. Everything is clean and by sight and feel all the connectors are secure. There is no visual damage to the motherboard. The IO is very secure, with the only thing being a bit loose is the 2 WiFi antennas, but wiggling these around like you suggested doesn't cause a crash.

I have my PC plugged into an extension bar, I have a couple of other devices (monitor, clock, phone charger) also plugged in that aren't affected when the PC crashes so I don't believe a small power outage would be causing this.

My previous PSU was a Corsair CX550M (before getting the 3070ti so wattage was fine). The RM750X is new.

Thank you

 

[pwrreset]

Do you have any overclocking settings enabled in the bios? have you tried turning them all off? even the ones that are automated?

 

[AJStellios]

Do you have any overclocking settings enabled in the bios? have you tried turning them all off? even the ones that are automated?

I haven't overclocked anything except for enabling XMP. I've double checked that the XMP Profile matches what the manufacture says. RAM is running at 1.2V as it should. Speed is 2133MHz. Only thing I've noticed different is that the manufacturers timings say 14-14-14. The XMP profile says 14-14-14-35. I'm not very knowledgeable about RAM speeds and timings so don't know what that 35 means and if it is normal. This is my RAM (I have 2 8GB modules) if it helps.

Thank you

 

[Ralston18]

And:

"I have a couple of other devices (monitor, clock, phone charger) also plugged in that aren't affected when the PC crashes ".

Those other devices are far more tolerant of power variations. Monitor might blink, the clock may miss a tick or a tock, the phone charger LED (if there is one) might flicker.

The PSU, on the other hand, without power, will simply stop powering the PC. Which is seen as an improper shutdown....

Remember that a PSU provides different voltages (3, 5, 12) to varying PC components. A failure to provide any of those voltages, however briefly, is all that is necessary for a problem to occur.

Failing/faltering PSU's tend to present a variety of increasing and varying problems.

Do you have all important data backed up at least 2 x to other locations? And those backups proven to be recoverable and readable?

 

[pwrreset]

try disabling xmp.

 

[AJStellios]

try disabling xmp.

I've disabled XMP. Hopefully this will fix the issue but I probably won't know for a few weeks with the irregularity of the freezes.

And:

"I have a couple of other devices (monitor, clock, phone charger) also plugged in that aren't affected when the PC crashes ".

Those other devices are far more tolerant of power variations. Monitor might blink, the clock may miss a tick or a tock, the phone charger LED (if there is one) might flicker.

The PSU, on the other hand, without power, will simply stop powering the PC. Which is seen as an improper shutdown....

Remember that a PSU provides different voltages (3, 5, 12) to varying PC components. A failure to provide any of those voltages, however briefly, is all that is necessary for a problem to occur.

Failing/faltering PSU's tend to present a variety of increasing and varying problems.

Do you have all important data backed up at least 2 x to other locations? And those backups proven to be recoverable and readable?

I haven't noticed any of the other devices acting strangely when my PC has froze. Would a PSU failure not cause the PC to immediately shutdown instead of freeze with audio still running? And I have all my important work backed up on onedrive so everything is safe.

Thanks

 

[Ralston18]

The PSU may not be failing as a whole.

All that is necessary is glitch in any given voltage.

= = = =

As for One Drive - having backups there is good.

However, you should have other backups as well where you have control of the media and direct access.

And do remember to test recovering that important work.....

 

[AJStellios]

Having already replaced the PSU with a good quality new one I'm just doubtful the issue could be caused by 2 PSUs in a row.

What I have just noticed is that my RAM has a slightly different model for each stick. One is Kingston KHX2133C14D4. The other is KHX2133C14. I can't find any difference in these searching online as they both point back to the same Kingston webpage, plus I've had them both for 5+ years with no issues.

 

[AJStellios]

Just had another freeze, so XMP wasn't the problem.