Take the 2-minute tour ×
Super User is a question and answer site for computer enthusiasts and power users. It's 100% free, no registration required.

I have Windows 8 and I've been suffering from "kernel-power 41"
I tried changing the power supply but the problem persisted.
I noticed that the heat sink was unbearably hot so i removed it and removed the dust from it and installed core temp to monitor temperatures.
then the stopped shutting down suddenly but the monitor would turn off and the computer would become unresponsive.

I tested the memory and after waiting an hour and one pass occurred, there were no errors. I also did a gpu stress test as i suspected the gpu was overheating but after an hour and a half of testing the computer didn't turn off and there was no fan on the vga card to change. i only noticed that the temp of the vga reached 83c.

I was concerned that the cpu at idle was running at 63c so went and bought a new heat sink and after properly installing it i noticed that the temp fell to the 40's c.
I did a stress test and temps reached 90's but the cpu didn't shutdown. after 10 passes the cpu became idle and the temp fell to the 50's and i was just browsing a site when the monitor turned off and the comp became unresponsive. after rebooting there was it again a kernel-power 41 error code.

so as last resort, I changed the battery in the computer and now I'm running a hard stress test temp is currently at 112c but the cpu is still running. are there any ideas why this is occurring?

I'm going to go with Ubuntu's suggestion and rule out that it is a Windows 8 issue. Next I could turn it to the repair men to see if it's a motherboard issue. With all 3 sticks of memory being new you would still assume it's RAM? Maybe after ruling out Windows 8, I would try running testmem86+ a bit longer.

share|improve this question
    
1 pass on a RAM test isn't nearly enough. Run Memtest86+ for 48+ hours. –  Ƭᴇcʜιᴇ007 Oct 1 '13 at 15:31
    
i doubt it is the ram. i recently upgraded the ram with a 2 gb plank and changed the other 2 to the memory that was at work. is it normal for the cpu to reach 115c? it didn't pass the test by the way. the system shut down while the temp was at 115 in one core. –  Majed Oct 1 '13 at 16:18
    
Could it be a software issue? Try running a live cd (you boot using it and do not have to install anything) and leaving it on for several hours. I recommend trying a linux live cd, like Ubuntu. –  medigeek Oct 1 '13 at 20:55
    
"I noticed that the heat sink was unbearably hot so i removed it" First, the heatsink is -supposed- to be hot; it is taking the heat away from the CPU. Second, I'm concerned that you may have damaged something if you ran your computer at all without the heatsink installed. Third, you need to use good thermal paste between the CPU and the heatsink, so that it will transfer the heat from CPU to sink & then it dissipates. Fourth, the kernel-power error could be many things, including the power supply, mobo issue, bad RAM etc. Since you recently replaced RAM, that's #1 to suspect. –  Debra Oct 2 '13 at 3:59
    
Also, MS's article on this suggests just about everything that's been suggested here: support.microsoft.com/kb/2028504 –  Debra Oct 2 '13 at 4:01

2 Answers 2

You say you doubt it was the RAM, but unless you totally replaced all (all!) of the old RAM in your machine with new RAM after this started happening and it didn't change, this is very consistent with bad RAM. Especially if it's only happening every once in awhile, it's extremely unlikely to show up on the first pass. Let MemTest86+ go for awhile, or alternately, if you've got extra memory available to swap in, try running it with a different stick for a few days. When this happened to me, I basically pulled one of the two sticks out of my laptop, it kept crashing, I swapped them, it stopped. One new stick later, it's run fine since.

share|improve this answer

You did not say which CPU you have, but 112C is too hot. Expected average temperature is more like 70C to 90C. You can look up temperature specs through Intel ARK if you have an Intel CPU.

Make sure to clean the dust off the heat sink, and make sure it is clamped down tight. Use a thin layer of thermal paste where the CPU touches the heat sink.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

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