New PC randomly locking up

[turbzy]

So I built a PC a few months ago:

Asus P5N32-E SLI motherboard
Core 2 Duo E6600 Conroe @ 2.4ghz
2GB (2x1GB) G.Skill Dual Channel DDR2-800
EVGA GeForce 8800GTX 768mb
Raidmax Aurora 580w power supply

The CPU is cooled with a Zalman 9700 air cooler, idles around 20c

Anyway, it worked beautifully for the first month or so, ran every game I had at full settings with no problems. However lately, it's been completely locking up at random times, with an occasionally BSOD. At first it was only in games/stressful programs, but recently it does it even if idle. To rule out a problem with XP/drivers, I went to reinstall XP, but was unable to, because it would lock up before completing the installation.

I looked around on various forums, and people with similar situations had insufficient power. So I upgraded to a 850 watt with tons of power on the 12v rails (http://www.tigerdirect.com/applications/SearchTools/item-details.asp?EdpNo=2668642&CatId=106), but this didn't help.

Upon installing the new power supply, I noticed there was an 8 pin CPU power, and my old one had a 4 pin, which seemed fine to me because Asus had covered half the pins on the motherboard with plastic. I'm hoping using only a 4 pin connector did not cause any permanent damage.

While I was in there, I noticed that my 8800gtx sits right on top of the southbridge, and thought this might be overheating and causing the problem. I moved the card to a lower slot, which exposed more of the southbridge to the air. After that I was able to get XP installed, and had it running for a few minutes, when it locked up again. Sigh.

I rebooted and got nTune running, to check on my temps, etc., and when I was in GPU settings, I noticed that the fan on the 8800gtx runs at 59% by default. I turned it up to 100%, which caused my GPU temp to drop 10c, as well as my case temp. After this, my PC ran fine all through the night, doing some par2 verifying and rar extracting, and downloading World of Warcraft. Everything seemed peachy.

This morning I loaded up WoW, and was greeted with a BSOD. I rebooted, and was going to check if there were any bugs with the latest nVidia driver, and it locked up again. Since then I haven't been able to keep it running for more than 10 minutes without locking up again. So I'm back to where I started.

I have run out of things to try. I've run memtest86, no errors with my RAM. Knoppix locks up just as much as windows, so it's not the hard drive. Replaced the power supply, so I highly doubt that is the problem.

Any help would be greatly appreciated.

 

[tdank]

Once you reboot after a BSOD are there any post errors? I had similar issues about a month back. Turned out the northbridge was not seated right. Manufacturer had applied too much thermal ceramic on the NB, once it baked in the heatsink no longer sat tightly on the NB and no amount of tweaking could get it down tight. Like the holding pins had been stretched slightly or something. I installed an aftermarket NB cooler which held the NB in its correct seating. Problem solved.

You may also want to read through you hardware manuals looking for voltage settings. Check your bios to make sure the CPU RAM Northbridge and Southbridge are all set at the optimal voltages.

Good luck friend.




PS. Just curious here. Is your GPU water cooled? How are you getting it at 10c?

 

[turbzy]

I haven't had any POST errors period through this ordeal, which seems odd.

I'll go check the manual now. Also, how would I check if the heatsink was seated correctly?

My GPU temp dropped BY 10c, from idle 60 to 50. Sorry if my wording was fuzzy. I'm just using the stock cooler.

Update: I couldn't find anything in the manual about voltages, but I did try googling and many people with my motherboard/ram combination needed to change the ram voltage to 1.9v. I tried this, but it didn't help any.