Intel E6600 Core 2 Duo 2.4Ghz CPU
EVGA nForce 680i SLI Motherboard
EVGA GeForce 8800GTX Superclocked Video Card
CORSAIR XMS2 2GB (2 x 1GB) DDR2 SDRAM 1066 (PC2 8500) Memory
PC Power & Cooling Silencer 750W Quad PSU
Seagate 320GB SATA Hard Drive
X-Fi Sound Card
Here is the problem. When I leave my system on for an extended period of time, like greater than 24 hours, the system will get increasingly unstable and lock up and/or blue screen (with various error codes, never the same). Eventually, the system will be unable to get past POST. The POST process will begin, but will hang about 25% of the way through. If I then turn off my system for awhile, like several hours, the problems go away for a day or so and then start back up again.
I have completely reformatted my system with a fresh install of Windows XP Pro SP2, and the problem continued. This says to me that this is absolutely not a software issue.
I have tested the memory for 24+ hours and have ZERO errors.
I have replaced the motherboard, the video card, the power supply unit and the memory. I have removed the X-Fi sound card.
I have monitored my temperatures and they are normal.
I am completely bewildered. I simply don't understand what could be causing this system instability that ONLY occurs after the system has been on for awhile.
The only thing that I have to possibly go on is my voltages. In the BIOS, they appear perfect. I'm talking within .04V of the actual rails. However in Windows, the voltages appear much worse. Here are the voltages I see when looking at them through SciSoft Sandra...
CPU DC Line - 2.45V
Aux DC Line - 3.54V
+3.3V DC Line - 3.23V
+5V DC Line - 5.43V
+12V DC Line - 9.06V
-12V DC Line - -10.74V
-5V DC Line - -5.66V
Those 12 volt rails seem very off, but like I said, in the BIOS they appear perfect. Also, this is actually the reason I replaced my PSU in the first place. I saw these very bizarre power readings, and so I thought the PSU was the issue. Replaced it. And at first, the readings were better, but then as I began to experience instability, they went back to these really strange readings again.
Anyone out there care to take a stab at this?