Sign in with
Sign up | Sign in
Your question

CPU Overheating/overvoltage !?

Last response: in Systems
Share
December 3, 2007 6:58:40 PM

I just built a new desktop system with the following specs:

AMD X2 6000+
GeForce 7900 GT OC 256MB (BFG Tech)
Foxconn n570sm2aa-8ekrs2h (nForce570i SLI)
2GB DDR2 800 Patriot Extreme Performance 4-4-4-12
320 GB SATA Seagate
SoundBlaster Audigy2 ZS Gamer
PSU: Coolermaster eXtreme Power Duo 600W
Running XP Pro SP2

This is not my first system and I haven't had any overheating problems with anything in the past, but I am getting some unfortunate results with this build. The computer will just shut off under heavy loads to the CPU (CPU tests in 3DMark06, after ~30min of Prime95, etc). Using CoreTemp 0.95/Everest I read:
Core1: 55C
Core2: 51C all while cpu is idle.
The program bundled with my motherboard (Tiger One) is giving me much hotter, 60-62C and voltage of 1.5V.

Everything else in the system seems to be working okay.

I used the heatsink/fan that came in the retail box with the CPU assuming that would be sufficient. Is there some issue that I was unaware about with this cooler or could it be another problem, such as the motherboard supplying too much voltage to the CPU?

Any help would be greatly appreciated.
December 4, 2007 4:16:29 AM

Yes, that is my CPU and I agree that 60C is unacceptable. I have a full tower Antec case with 3 case fans, 120mm in the PSU and the retail cooler on the CPU and GPU. I am not overclocking and the high today was 34 degrees F (stays around 72-74 in my room).

I found in the BIOS settings this morning that the clock stepping was set on performance, I changed this to auto and it lowered the Core voltage to 1.38V and the temps dropped about to about 55C, still too hot for idling I would say.

Also, I don't know which program I can trust as an accurate reading of the core temperatures because they differ by so much. Core Temp is giving me 28C and 25C right now, which would be great. But the Tiger One program is giving me 43C, which is still high while idling.

Could the motherboard possibly have defective sensors that are tripping the auto-shutdowns because they are reading the temps as 10C+ greater than what it actually is?

I only have about 10 days left to troubleshoot this and have any hope of RMA'ing the motherboard and/or CPU to Newegg unfortunately.

Thanks for the reply.
Related resources
December 4, 2007 5:06:39 AM

Send the MB back and if you can't get a DFI then at least get an Asus...but if it was me I would have gotten a DFI MB to start with.

Befor doing any of that you can try reseating the cooler first and/or check temp of HS with a infared temp probe.
You can also try setting the voltage even lower (try 1.2) and if it boots retest it. The MB may not be providing the voltage your programs/bios say it is.
December 4, 2007 11:29:15 AM

I agree. Get something else. Foxconn isn't the best brand. Biostar at the minimum.

Yes, if your cpu cooler isn't installed properly, the temps will be high. I know AM2/939 coolers lock down real hard to the point it almost breaks. I have gotten rid of the AM2 socket retention & bolt a large cooler down.

The bios temps should be the one to trust. If it's above room temp, it should be right.
December 4, 2007 11:57:03 AM

Can you decrease the Voltage? 1.5v seems awful high for a non OC'd chip. See if you can set it to something like 1.4v.

The 6000+ Chips, however, do run quite hot and take a bit of voltage.
You may just happen to have on of the hotter 6000+ chips which is already among one of the hotter types of chips.
December 4, 2007 12:06:44 PM

sehret2010 said:
Yes, that is my CPU and I agree that 60C is unacceptable. I have a full tower Antec case with 3 case fans, 120mm in the PSU and the retail cooler on the CPU and GPU. I am not overclocking and the high today was 34 degrees F (stays around 72-74 in my room).

I found in the BIOS settings this morning that the clock stepping was set on performance, I changed this to auto and it lowered the Core voltage to 1.38V and the temps dropped about to about 55C, still too hot for idling I would say.

Also, I don't know which program I can trust as an accurate reading of the core temperatures because they differ by so much. Core Temp is giving me 28C and 25C right now, which would be great. But the Tiger One program is giving me 43C, which is still high while idling.

Could the motherboard possibly have defective sensors that are tripping the auto-shutdowns because they are reading the temps as 10C+ greater than what it actually is?

I only have about 10 days left to troubleshoot this and have any hope of RMA'ing the motherboard and/or CPU to Newegg unfortunately.

Thanks for the reply.


You left out the main thing. Is it still crashing? 43c doesn't sound too high for idling, off the top of my head, though I am not familiar with this chip.
December 4, 2007 4:47:43 PM

Wow, thanks for all of the advice.
Notherdude: Yes the system is still crashing.
Zenmaster: I cannot underclock the system or lower the voltage, only raise it unfortunately.
akhilles: I have a Foxconn motherboard now... Bios temps seem high even at startup (40C+)

As soon as I get some alcohol to clean the heatsink with I will reseat the heatsink and use some Artic Silver and see if that works. If it doesn't, sounds like I'll have to RMA the motherboard from all of your advice.

Thanks again and I'll post back if reseating fixes the problem.
!