I have a very similar issue. The fan lock into its high RPM state, the PC is impossible to shut down and a restart from that state behaves like the BIOS flash restart ie it powers off totally for a few seconds and then turns on again - not good for the hard disk drives!
I dont have any eSATA device
Windows 7 X86
I confirm this problem.
My actual configuration is:
DG45ID + Q9400S + 4x2G Kingston HyperX DDR2 800MHz CL4 + 1 eSATA HD + LG DVD+&-R/RW
No Internal HD, 2 Fan: CPU+Front Fan.
With Old E5300 CPU And 2x512MB+2x2GB PC2-5300 667MHz and previous bios version I've no problem.
After Bios update to 117 and new CPU and ram installation, the cpu and front fan lock at high rpm after 2/3 days and after this HWMonitor can't detect motherboard data but only the cores and HD temperatures and if I shutdown the os, the system does not power off. If I force power off and restart all is normal for 2/3 day when the problem reappear.
I'm also getting this problem with my DG45ID and E8500 and the latest shiny new 127 BIOS. I've tracked it down to changes made after the 111 BIOS as 113 and newer exhibit the problem but 111 does not. I do not want to have to revert to 111 as the latest BIOS has a newer Graphics BIOS module so presumably there are improvements to be had.
What does QST mean - I suspect it's SATA related as in my case it seems to occur when the HDDs (using onboard RAID 0+1 with 4 x Samsung 1 TB F1 HDDs) are being written to (although it's hard to be sure. It'll usually happen within a day in my case and all the same behaviors occur when its in the high fan speed state such as Intel Desktop Utilities not being able to read the variables of the board, restarts which momentarily power off the board and shutdowns which wont power it off. Given this problem has been going on for months now, I'm wondering whats so hard about looking at the changes between BIOS 111 and 113 to narrow down the relevant piece of code that's throwing the spanner in the works and correcting it?
Maybe one of the Intel people can comment about it?
Removing the noisy capacitor might have affected part of the PWM control. They probably fixed that issue by changing the power scheme... I highly doubt they can fix the PWM issue. I bought another Mobo and to be honest will never buy another intel product except for CPU... This DG45ID is collecting dust in my basement!
I don't know which part was making the high pitched squeal (only narrowed it down to within a few components) but this fan high speed locking problem occurred on my old AA rev 307 board as well as the new 310 rev I have (which doesn't make the component noise) so it seems to be a clear indication of a problem with the newer than ver 111 BIOS files to me. If the older version does not exhibit the problem and newer versions do (on identical hardware), what other assumption can be drawn?
I agree that the problem appears to be with BIOS versions newer than 111. I also noticed when the problem manifests itself, CPU-Z cannot read information (temp, fan speed, voltage, etc) from the "Intel HECI" sensors. Somehow the Intel Management Engine (ME) appears to stop functioning. Since the QST is controlled by the ME, it stops functioning and causes the CPU fan locks at a high RPM rate thus creating excess noise.
Manually disabling QST in the ME setup (ctrl-p during POST) causes the CPU fan to be locked into a high RPM state. This is why I believe there is a bug in any bios version >0111 that causes the ME to become non-functional.
On a side note: With BIOS 0127 I noticed that the above issue has manifested itself (twice) within hours of my turning off my monitor (usually I let the monitor go into standby mode). My monitor is connected via DVI and has a USB hub in which a bluetooth USB adapter is connected. When I turn the monitor off, the hub and bluetooth also get turned off. I do not know if the this is coincidental as I have not tested it any further.
My MB revision is currently 310, but I had the problem on an older 307.
In the same screen that you can press ctrl-P during start up, my system with BIOS 111 (I've now reverted for greater stability - a sad indictment on those responsible for these newer Intel BIOS files) reads:
Intel Management Engine BIOS Extension 5.0.5.0004
Intel ME Firmware version 126.96.36.1999
What are the differences with BIOS 127 (or even 113) in this regard... I think that's where the problem is. The restart under the high fan RPM scenario even behaves the same as when configuration changes are made in here.