Massive voltage spike... power surge?

So I sat down at my home computer this morning, and I checked my HWinfo numbers out of habit, and I was alarmed to see that while my computer was sitting idle overnight, my CPU core voltage had apparently hit 870v, and my GPU core voltage had apparently hit 643v. Considering that my normal maximum CPU voltage is in the low 1.3's and my GPU is usually less than 1.1, seeing those massive numbers is slightly concerning.
I'm on windows 10, and I don't use sleep mode or hibernate mode, so the computer was just sitting idle on the desktop all night with the screen turned off.

My city in California had some large power outages from wildfires in the last few days, so my first thought was that there had been some type of power surge that caused my system to get hit with a huge voltage spike. I don't know if that would make sense though, since the core voltages are controlled by the computer itself. I have no idea if a power grid surge could cause my core voltages to spike like that. I also think that my PC would have gone up in flames if the CPU had in fact been hit with 870 volts. I'm currently not using a surge protector, but my (cheap) power strip says that it has protection against 500v. And yes, I will be buying an actual surge protector this afternoon.

Is it possible that this was just a computer brain-fart? I've been using HWiNFO64 for a while now, and I've never seen it make any blatant errors like this before... so if it's saying that my CPU hit 870.249v, I'm inclined to believe it.

Thanks for any input, I really appreciate it!

Capture10.jpg
 
This is certainly an invalid readout from sensor. You have renamed sensors and didn't specify what CPU/GPU do you have so I'm unable to figure out where exactly to check.
Capturing this situation in HWiNFO Debug File would be great and allow me to see what exactly happened there.
 
Hey Martin, thank you for the reply!

This same type of sensor error just occured again actually, this time while I was sitting at the computer. I didn't notice anything weird happen... I just found that the Maximum readings were very high again.

I'm looking at the very high numbers in the maximum column of the chart right now (just like in the previous photo I had posted), but I did not have debug mode enabled when it just happened, so I don't think it would have been captured in a debug file this time. I was just reading your post about how to submit a report with the debug file, and if I'm understanding it correctly, I need to enable debug mode and then try to replicate the issue, which will then have the error info in the debug file.

As far as the renamed sensors go, my OCD got the best of me, and I did what I could to tidy up the sensor window into a nice tight box.
Would it help if I reset the sensor window to the full default to show everything with all the original labels? I'm also wondering if having certain sensors disabled from the window will affect the debug mode? I'm more than happy to reset everything to default if it would help.
I have not reset the labels yet, so please advise if this will be necessary.

Here's some specs: Windows 10, Intel 9700KF, Asus Z390-P, Asus DUAL-RTX2060S-O8G-EVO.
If any info from my BIOS would be helpful, I'm happy to share that as well. I have numerous things changed in the BIOS related to overclocking, so I suppose it's possible that those settings could be related to this somehow.

I really appreciate you offering some assistance in this matter. For now, I've enabled debug mode in the settings, but I have no idea if and/or when this strange error will happen again. It looks like HWiNFO has created a debug file, but I don't know if that file will be helpful at this point since the voltage reading error occured BEFORE I enabled debug mode. I went ahead and attached the debug file anyways, just incase it has anything good in it that may be helpful. I put the debug into a .zip file because the website said it was too big to upload.

Thanks again!
 

Attachments

  • HWiNFO64debug.zip
    426.3 KB · Views: 1
No need to reset any settings now, since I have the Debug File now I can see what your system configuration is and at least have some clue where to look for the issue.
Yet, it would be best if you could capture the error while running in Debug Mode. Yes, you need have it enabled first, so that the data is captured. But since this problem occurs so rare, the Debug File will most likely grow very large until the problem happens again.
 
I did some checks and believe this problem is somehow related to the fact that you have set HWiNFO to multiply the voltages by 1000 to show the result in millivolts.
Can you try to revert this setting and don't apply any multiplication to see if the issue will happen again ? Or keep it in mV for some values and others revert back to V - then you should be able to see if it happens again what will be the difference.
For now there should be no need to run in Debug Mode again as this issue seems to be of a different kind.
 
Last edited:
Ah yes, thats a good catch there! I'll try setting only some of the multipliers back to normal to see if I can catch it again to verify that is in fact the cause of this.
This strange issue has happened a couple more times since my last post, but I haven't managed to capture it while debug mode is enabled.
I'll let you know what happens with this. Thanks again! Much appreciated!
 
Hey Martin,
It looks like you were correct about the multiplier value causing the anomalous voltage readings. I set most of the multipliers back to normal and left just one of them modified, and today I have found a high reading on the one that has the multiplier of 1000. The other values without any multipliers are still normal.
Thanks again for the assistance!
 
Thanks for the feedback. I checked the code several times, but unable to find yet what might be causing this :(
 
Back
Top