I know there are a lot of threads about this across the internet but none of them really point to a particular cause or a fix.

Over the last few weeks, with increasing frequency, my computer has been randomly shutting down at non-specific intervals, while performing low power tasks. As far as I can tell, it is not overheating.

It gives no warning, does not run shutdown scripts, it's as if I've had a powercut (I haven't).

What I've tried: I thought initially it was a problem with Windows 10 so I have done a clean install of Windows 7 on my 128GB SSD.

There was some mention of a faulty power strip causing power issues. I plugged the computer into a wall socket and had the same issue.

After doing this I had several problems, which may or may not be relevant. Windows Update did not run. It hangs for ages and ages. I left it for a while, and when I came back, the computer had once again shut down.

I also tried to install the drivers for my audio interface (a Native Instruments Komplete Audio 6) and they constantly fail to install, and I have no sound. The computer will still randomly restart when watching a youtube video, unzipping a .RAR, installing software, and others. I really can't pinpoint it.

Every time it happens, I get a Kernel-Power 41 error in the Windows Event Viewer. Lots of other people have reported this problem - eg. http://www.tomshardware.co.uk/forum/46250-63-random-restarts-critical-kernel-power-event-task

The way I see it, I have some hardware which is causing this error, but I don't know which part. My main guesses are, in order of likelihood:

  1. The PSU (a Corsair CX450M)

  2. The USB headers. The first time this problem happened to me, I got my leg caught in a USB cable connected to the front panel and pulled it out, cleanly but sharply. The computer instantly shut down. That was a few weeks ago. After that, sometimes when my computer went into sleep mode, it wouldn't come back on when I used the keyboard or mouse. And then about a week ago I started experiencing the shutdown problem. I have to manually restart. Did I break something here?

  3. The RAM - I've read a few places saying badly seated RAM is to blame, and to try just 1 stick (it's a pair of 4GBs)

  4. The motherboard - something on the main board (Gigabyte GA-F2898XN-WIFI) is failing. In which case I'd most likely end up building a new machine.

  5. The GPU - a (cheap) PowerColor R7 260. It's always worked fine.

Another thing about the USB headers. I mentioned I ripped out a cable by mistake. Could this also explain why my computer is failing to communicate with my Komplete Audio 6? Also, when I was looking to change the boot sequence in the BIOS to the (bootable) Windows 7 USB flash drive I had inserted, it didn't show in the list. I ended up having to run it from within Windows.

Can anyone offer me any guidance on this? Any similar issues?

The PSU (a Corsair CX450M)

Could be, but normally the PSU is just the "bad guy" if it has to less power (once happen with new graphic card). But your system, doesn't seem like a big energy waster.

The USB headers.

If it happens and no USB device is connected, you can exclude this suspect. Maybe if you have a some cheap china device connected, it can cause an short-circuit (once happend with cheap vent).

The RAM - I've read a few places saying badly seated RAM is to blame, and to try just 1 stick (it's a pair of 4GBs)

Does your OS show that it uses the 4GB (e.g. windows under "system"). If the installed memory is only 2GB maybe one is broken. Maybe asked a friend, who can lend some Ram for checking. Also you maybe change the socket.

The motherboard - something on the main board (Gigabyte GA-F2898XN-WIFI) is failing. In which case I'd most likely end up

building a new machine.

Seems very likely to me. Maybe you damaged something, while cleaning it or changing parts. Or my last Gigabyte lost more and more SATA-Ports overtime. And once had a cheap asrock, which had some randomly voltage peaks, after a year, which often ended into a shuted down pc. Here is a tool to watch your voltage level: http://openhardwaremonitor.org/

The GPU - a (cheap) PowerColor R7 260. It's always worked fine.

Disable (and maybe dismount) your graphic card and use the onboard gpu. So you can exclude your gpu as suspect.

But I think, the Mainboard is damaged (but this kind of remote maintance is not very relaible). This was often the fault, by me and friends.

Last thing check if, every cable is connected right and there is no short-circuit (some case parts fell off, screw on the mainboard or in my case once an acupuncture needle in my master slave connector strip.

good luck finding the fault!!!

  • Thank you for your response! I will try some of the above methods, and try open hardware monitor too, later when I get home from work. Many thanks! – TCassa Nov 24 '16 at 10:37
  • Another thing I thought about could be my watercooler (for the CPU) failing. How would I know if this wasn't working? – TCassa Nov 24 '16 at 14:52
  • Update - I disconnected the GPU first to see if that was the problem. Managed to run for a while without crashes, tried a few games (most didn't work without a GPU) and I thought I had solved it! But I still have a lot of problems like not being able to run Windows Update, or open boot priority menu, or install drivers for my sound card, or format a Win 7 install USB to make it bootable. So I'm going to reformat the HD this time (I did clean install but without wiping the HD last time), give it a try, then probably build a new computer! – TCassa Nov 24 '16 at 23:18
  • and what happend? – LosKartoflos Nov 30 '16 at 12:25
  • 1
    Sounds like a Mainboardproblem then. But cannot tell 100%. – LosKartoflos Dec 1 '16 at 13:16

Update: After completely rebuilding, part by part, the problem was my water CPU cooler - the part that everyone I asked thought was LEAST likely to be at fault! I put the stock cooler that came with my CPU and all is well. So I bought all those parts for nothing. Oh well, such is the nature of building from scratch!

I suppose my advice would be to test EVERY component - even the ones you don't suspect!


In my case, the reason for the "Kernel Power 41" error in the Windows Event Log was the video card. My laptop (Lenovo Flex 2-15 Pro) turned off randomly after 20-30 minutes without any notice.

I had updated all drivers before. But Windows Update seems to distribute a wrong driver for some NVIDIA video cards, as I read on https://www.computerbase.de/forum/threads/nach-neusten-treiber-installation-kernel-power-41.1703758/

After I deactived the NVIDIA video card, the problem did not appear any more:

enter image description here

I will now see how I can permanently fix the problem. I will probably re-install the original driver again and prevent driver updates for the video card.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.