I recently upgraded from an E6700 to a Q6700 in my system, built with a DG35EC board. Since the upgrade, I'd noticed temperatures getting a wee bit toasty, the cores of the CPU reporting upwards of 100°C (in CPUID's HWMonitor) when doing anything CPU-intensive.
Cooking the processor is highly undesirable, and normally I would have chalked it up to insufficient cooling, or a poor connection with the heatsink/fan, or even dirty stinking liars of temperature sensors, but when I went to check the heatsink temperature with my forefinger (this particular finger test caused a torrent of creative epithets describing just how blasted hot it was) I found the fan to be spinning very slowly. A few seconds of headscratching and an install of speedfan later I found the CPU fan to be reporting a speed of just ~850RPM. Attempts to increase this speed using speedfan's controls were unsuccessful.
Some googling later, and I came across this thread: describing a similar problem with the same board. Following some of what was listed there, I fired up Prime95 to heat the CPU up to around 95°C briefly, and immediately rebooted and entered the hardware monitoring section of the BIOS. There I found that the fan speed had increased to 2500RPM. I booted into Windows, and fired up speedfan again, and found I was able to control the speed between ~850 and ~2500RPM, where ~850RPM was the "0%" value, and 2500RPM was the "100%" value.
Through this testing ,I have established that the fan speed will not increase beyond the maximum speed from the last time the system booted.
I seem to recall as well, that previously, this system on boot would run its fans up to maximum for a brief moment after power on, and then settle back to appropriate values for the temperature, but I could be wrong on that one. I never really paid too much attention to it prior to noticing this problem.
Any ideas on why I'd be experiencing these issues? I'm not sure if they started as a result of installing the Q6700, or if they'd been going on with my E6700 as well. The E6700's lesser heat output may have led it to be just peachy even though the fans didn't crank up.
My BIOS is the latest revision for this board offered by Intel, and when I attempted to reinstall chipset drivers to see if this was a peculiar driver issue, I was informed that a newer version was installed than the latest version available (as of yesterday) on Intel's download page. I told it to go ahead and install, and it went ahead, but completed almost instantaneously leading me to believe that it didn't do a whole lot, and didn't prompt for a reboot. I rebooted anyway to be on the safe side, but same issue persists.
Anyway, Thanks in advance!
I recently upgraded from an E6700 to a Q6700 in my system,
When you upgraded the processor did you re-use the stock HSF or use a HSF designed for the newer processor? There are differences in the stock HSFs. If you are using something aftermarket then it's less of an issue provided the HSF is designed to handle the quad's power use.
There's also something funky with the series 35 motherboards. Try flashing the BIOS down a version and then back up to the latest, completing the process by loading default values. Others have written here that seems to get the board working correctly when a processor upgrade seems to cause problems with fan regulation.
Finally, I'm suspicious of the processor getting as hot as you describe without some kind of CPU utilization. The Windows task manager might shed some clues also as to why the chip is getting so hot. Perhaps a fresh install of the OS is in order or maybe try loading a clean install of the OS on another disk and see how the board behaves. You may have some kind of malware thing going on.
I used the HSF that I'd been previously using -- it's an Intel stock HSF, and I think it did originally come from a Core 2 Quad (though I'm not 100% sure on that -- it was secondhand, and came from a bin of unused stock Intel coolers. It was larger than the rest, so I picked it out instead.
I realized that I might have to upgrade my cooling, but I figured I'd see what the one I had would do at first. The problem is though, that the fan sticks at a low speed. Assuming the heatsink I have is too small and can't dissipate the heat generated by the Q6700, I'd expect the fan to be going at full song, whirring like a leafblower. I'd like to figure out why it's just sticking at it's slow speed before I go swapping it out.
I'll try flashing the BIOS back down, and then up again this evening, and hope that it solves my problem
Also, to clarify (and I'm sorry I wasn't as clear as I could've been in my initial post) -- the CPU idles at 30-50°C, and only climbs up north of 90°C when I put a load on it (I've been using Prime95 during my troubleshooting)
Thanks for the quick, and helpful reply!
Righto, so last night I gave flashing back down to a previous BIOS version a try, but it hung on "Flashing Processor Updates..." -- I left it sit there for a good half hour on the off chance that for some reason it might just have been taking it's sweet time to finish off. When I rebooted, the recovery image re-asserted itself (or so it would seem) and the BIOS version was the same as when I started.
I then tried flashing the latest version on, in the off chance that the previous version had some compatibility issues with my CPU, and it hung in the same place.
Tonight, unless there are other suggestions, I'm going to try replacing my E6700 to see if I can replicate the fan behavior, and see if I can flash the BIOS properly without it hanging up.
Update the second.
Last night I swapped the Q6700 for the E6700 again, and ran my prime95 test. Same results with fan speeds staying down at low levels, and not spinning up past where they were at startup.
I tried to flash the BIOS down a version, and was successful this time. Ran tests at this version of the BIOS, and encountered same results.
Flashed the BIOS back up to the most current revision, same results, no changes.
Any other thoughts?