• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

CPU temp sensor reporting at 107°C

joshyz

n00b
Joined
Feb 6, 2006
Messages
9
My brother just turned on my family's PC and got a message before the post screen about a CPU over temperature error (the PC had been powered off all night). I checked in the BIOS, and it was reporting the CPU temperature to be 107°C/224°F. I immediately shut the PC down, and disconnected it from the wall.
I then opened up the case, and plugged it back in to the wall. The CPU cooling fan came on, and the heatsink was just slightly warm to the touch. Then I shut it back down, and connected the keyboard, mouse and monitor to it and powered it back up. I got the warning again, and checked the BIOS and received the same temperature reading (and it does fluctuate about 2 or 3 degrees up or down as I watch it).

Does it look like something is wrong with the temperature sensor?
I have not removed the heatsink since the CPU was installed (and it is the original HSF), and the PC had been off for over 12 hours when it was first turned on, it never got a chance to overheat. Nothing changed on the PC at all. It was fine when it was shut down, and then came back up with the error. Any ideas?
 
whats is the voltage?

i think its an error by the temp sensor...try getting a few fans and blowing them into the computer (take the side panels off get desktop fans to blow inside)...if the temperature doesnt drop you know its a faulty temp sensor
 
Try replacing the thermal compound and ensuring the HSF has proper contact with the CPU...

I had an old socket A system that (amazingly) ran at over 100 degrees celcius for about 2.5 years. Unfortunately, the heatsink was warm, but not hot. I figured the temps were fine and did not bother to check them. One day, I decided to try to overclock the proc, so I installed Speedfan to see a temp of about 90 degrees idle. I reapplied thermal paste and the temps dropped to around 70, but the heatsink was HOT. I mean, the damn thing burned me... Anyways, I stuck a massive socket 478 cooler on instead of the tiny stock fan. Temps are now 50 idle 60 load with no overclock.

Long story short, old AMD socket A's run really hot. I had about the same temps in my P4 system too, which was only a few months older than the AMD, so old P4s run hot too. I really doubt it is the sensor... Just get an aftermarket hsf and some as5 to get the temps back down...
 
The processor is a Celeron D 310 Socket 478 processor (sorry, forgot to mention that). I don't think that they normally run hot, and it hasn't had any issues before since it was installed.
 
If the heatsink is making poor contact with the CPU then you wont get a lot of heat transfer. This causes the CPU to stay hot and the heatsink to remain relatively cool.

So you have one of two situations, a faulty temp sensor as suggested, or bad contact between the CPU and the heatsink. The simple way to find which is which is to simply remove the heatsink, clean the processor/heatsink with isopropyl alcohol. Then reapply the heatsink with new thermal paste. If you still have problems then you have a faulty temp sensor (assuming you didnt screw up the install).
 
Back
Top