Sudden system freezes. No BSOD

Within a couple of hours after replacing my CPU cooler, my system froze on three separate occasions. No sound. No response from mouse or keyboard. Monitors kept showing a frozen image. On one of these occasions I waited a couple of minutes before resetting, but it stayed frozen.

The freezes were sudden. No graphical glitches. No BSOD. Nothing in the event viewer, as far as I can tell. On two of these occasions I had HWMonitor visible, and no temperature was above 60°C.

This happened two days ago and I had no freezes yesterday, nor so far today. Outside of those three freezes, the system is working just fine. I wasn't doing anything in particular when the freezes occurred that I haven't done since without issue.

While replacing the cooler, I unseated the CPU to more easily clean it around the sides. I was careful not to touch or damage the pins, but could this potentially have something to do with the problem? I was surprised the BIOS detected a hardware change on startup.

I'm at a loss to know how to diagnose this problem and the worry of it happening again and not knowing how to prevent it is eating me up. I could go into more detail about several of the circumstances, but since I have no idea what's causing this problem I don't know what would be relevant. If anyone could point me in the right direction or has any idea of what might be going on, I'd be grateful.

Hello Lenticular,

Welcome to TenForums!

Did you clean and replace the thermal paste when you unseated the CPU? If not you may have just temporarily disturbed the thermal contact and it has taken a little while to reform.

Hi Philc, and thanks!

I removed the old thermal paste using isopropyl alcohol, q-tips and paper towels. The replacement cooler came with a pre-applied layer of paste that I chose to leave in place.

Can I infer from your reply that these freezes could have been a result of overheating? Like I wrote in the original post, HWMonitor did not show any excessive temperatures.


In case it's relevant, a little more details about the cooler replacement: I built the system just about a month ago. The included wraith spire was noisier than I expected so I had a replacement shipped. I swapped a noisy wraith spire with a less noisy wraith spire.

Yes, I was thinking that localised overheating might have occurred on the CPU if the paste hadn't completely been removed or the contact with the cooler was not perfect.

Not sure if this was the reason now since you have explained that you did clean off all the old paste and used the new patch provided. If it does not happen again and temps are keeping below 60°C then you should be OK.

Thanks again for taking time out of your day to help me with this, Philc

I understand we can't be completely certain of anything, given the lack of information and amount of variables, but how sure that overheating was the cause of these freezes would you say that you are and are there any other likely explanations? I have never had a computer issue like this before so I have nothing to go on, and it's difficult to google without any sort of error codes.

Is there anything you recommend I do to limit the chances of this being a problem in the future?
Should I replace the thermal paste, for example?
Did I make a mistake by removing the CPU to clean it?
How come HWMonitor showed no problems?
If it happens again, what would you recommend I do?

When there is no error message or blue screen of death (BSOD) event it is very difficult to pinpoint causes. It could have been a thermal problem but might also have been a software glitch. The fact that it has not happened again is suggestive that it was due to the CPU bedding in again after being disturbed. If there were traces of solvent contaminant on any of the pins the contact might not have been perfect during those first 2 hours but then improved with time and heat.

If you go to the BSOD section of this forum you will see some posting instructions - towards the end there is a beta log collector that can find and collect logfiles which I can look through to see if there were any possible events or logs created during the time when the freezes occurred. If you want you can attach them here and I will take a look. Let me know the approximate time and date when the freezes occurred.

Thanks again, Philc

Regarding software glitches. Right after the last freeze, one of the things I did as a hail mary was uninstall MSI's Command Center application which allows you to control things like fan speeds and overclocking. I had had this application running for all freezes, and I thought maybe it didn't like my reseating the CPU. Most likely a coincidence, of course

You're suggesting that while cleaning the unseated CPU, some isopropyl alcohol snuck around the edge to the underside, contaminating the pins? Sounds entirely plausible and I now thoroughly regret unseating the CPU. If this is indeed what happened, what are the dangers? Does the fact that I haven't had freezes for two days mean that the danger is over, or is this something that could return at some point?

I'm looking for assurances here, but I understand that you can't give any. I hope you can forgive me, I've been stressed out by this whole thing.

Attaching logs from the beta collector. Event Viewer lists four Kernel-Power Critical events on Sunday October 28th. I thought I had three freezes but perhaps I misremember. Times are in CET (UTC+1), if that matters:
16:00:23
16:07:46
16:20:09
19:04:14

I could not find anything in the logfiles that would expain the freezes at the times they occurred.

I would be inclined to continue using your system as normal and treat the events as one-off and likely to have something to do with the CPU pins getting a bit of contamination. Solvents evaporate but it can take a little while for all traces to go.

Hi AndreTen. I struggled a bit with attaching the cooler as the backplate had come loose and the screws couldn't reach the holes. Eventually found a way to keep the plate in place, which made the process simple. I have no idea whether I applied too much force during this process. How much is too much? Once I managed to get the screws to reach, I screwed them in a little bit at a time, jumping diagonally between them, until they bottomed out and immediately stopped. What is the danger here? What might I have done wrong? Am I not supposed to bottom out the screws, and if not, how do I get the heatsink level with the CPU?

Thanks for taking the time to look through the logs, philc. If it happpens to be contamination, can you say anything about the chances of this problem reappearing? I hate having the feeling that my computer will lock up at any moment. Also, should it happen again, what would you recommend I do?