Since taking delivery of my Studio XPS 8100 in mid-January (2011), I have had regular shutdowns as described by other owners here. After trying a few of the suggested (less costly) fixes here and in other posts, I think I have found the problem plaguing my unit. I had tried updating as many drivers as I could through the Dell Support website, re-installing the processor heat sink with Arctic Silver 5 paste, disabling the setting for USB Root Hub 'Allow the computer to turn off this device to save power', and a few other fixes that I won't get into now. I even ordered a PCI slot case cooler fan (VIZO PCI-101-BK 65 mm), which I have not yet received from NEWEGG, to lower temperatures. Having followed the advice in this forum to Disable Automatic Restart, I finally got a BSOD error report that led me to this article http://bit.ly/BugCheckFor8100 This led me to check my driver for my on-board NIC which is a Broadcom NetLink Gigabit Ethernet unit. After downloading their current driver from this site http://bit.ly/i9KJ06 , I found the Windows 7 (64-bit) driver to be dated August 2010 while the driver on my 8100 was dated July 2009. So I downloaded and installed, through Device Manager - Update Driver, and so far have had no more sudden shutdowns. I'll try to update this post later to let anyone interested know if this has 'totally' fixed my problem.
Sorry for the premature 'solution' here, as sadly I report another shutdown occurred today. I'm not sure why I do not get a BSOD when this restart occurs, since I have set the Disable Automatic Restart a few days ago. I'm now back to square one, any ideas would be appreciated.
Update # 2
I put a call in to Dell tech support a week ago and had a tech from Worldwide Tech Services come to my home to replace my motherboard, memory, hard drive, and video card on Tuesday, March 1. As we booted up the machine after the parts were replaced, Windows loaded about 50% of the device drivers and then suddenly shut down. It was great to have the failure occur with the tech watching. She immediately called this result into Dell tech support with her description of the latest failure. She told me that the power switch on the top of the case sounded loud to her when it suddenly shut down. So a new power switch, i5-650 processor and hard drive (another hard drive, because they had loaded Win 7 Home Premium instead of Win 7 Professional that came with my computer) were ordered and the same tech (at my request) came back on Thursday, March 3 to install these new parts. No more sudden shutdowns! The machine has run properly for 3 days now. The tech is quite sure that the power switch was the only thing wrong, though just about everything else was replaced in the process of diagnosing the problem. The processor was also changed when the switch was, but was not likely the problem, especially considering the louder noise the tech heard when she observed the failure with the old power switch in place. After 6 weeks of frustration with the Studio XPS 8100 desktop, I finally have the machine I was hoping for when I purchased it. Hope this helps anyone who may be dealing with this same problem.
I have about ten of these machines on our floor and one of them started with these symptoms suddenly. It worked fine until I started installing a Centos6 upgrade from cd and then suddenly the reboots started. I've swapped the RAM, the hard drive, the video card, the power supply, and everything except for the motherboard and CPU. It's more like a complete poweroff, I just hear <click> then it's dead.
Have you run the Dell diagnostics? That's where I would start. If you run through all the diagnostics without a restart, then it's an OS problem, not a hardware issue.
Sure I'll try it again, but it will need at least 30 min of up time without shutting down for that. I've replaced everything but the power supply and motherboard so far.
I just ran the diagnostics and let it sit for 30min, when I came back it was frozen at the screen that says "No problems have been found with this system so far, do you want to run the remaining memory tests.." but I can't proceed because it's stuck.
I checked the temperature monitoring with lv-sensors in centos and all the cores run around 40C degrees. If I touch the inside heatsink it seems fairly cold actually, but the northbridge chipset heatsink is scalding hot. I read online that this is normal.
it's shut down at random with any hardware I put in it. Right now I have swapped everything including the power supply and it still shuts down. And thats with no hard drives, just running off a Centos Live USB. Sometimes it will run for a few hours, some only a few seconds but if I leave it on overnight it is always off in the morning.
Additionally I have disconnected all the USB & front panel devices on the case from the mobo. I'm just going to order a new LGA 1156 micro ATX motherboard online and hope that solves the issue, because I've narrowed it down to the mobo / cpu and I seriously doubt it's the CPU.