Strange crashing problem.

[happybelly]

So, I'm trying to figure out the reason why my computer has been randomly crashing the past 2-3 weeks. It occurs most often if I'm playing a game, but it also happens during normal web browsing and other simple tasks. In the game is would lock up, sometimes the screen would scramble, then it would go black.

My chipset fan broke 2-3 months ago, I got that replaced, but my motherboard temperature read in the high 50's to low 60's C even after fixing the fan. I originally thought this was the most likely cause of the problem. I don't know if the temperature prob is accurate since the fan is working fine now.

The last 2-3 days the problem seemed to be getting worse. I would lock up in a game or browsing the web more often and it wouldn't even post(It would turn on, but not post) several times. a couple of times the post screen looked garbled and unreadable. The next time I locked up, I reset and could hear the post beep, but I had no video. I have an SLI setup so I decided to take out the 2nd card and see what happened. Well, since then I haven't had any crashes and it seems to be running ok. The motherboard temperature is reading about 40C now as well.

Now, does anybody know if the SLI could have been the cause, or is it just a coincidence? If it is the cause I don't know why it would start doing this after more than 2 years of running ok.
System specs:
WinXP
Athlon 64 3500+
Asus A8N SLI Deluxe Motherboard
2 BFG Geforce 6800 GT video cards
2 GB Corsair RAM
550W Antec PSU

Thanks for any input.

 

[A554SS1N]

If the screen is garbled, and the fact taking out a card helped, it means the graphics card was the problem. Check if the cards are clogged up with dust? Also check the case in general, lower airflow overall in the case could be starting to tip thee cards over the edge. It's also quite possible one of the cards simply failed.

 

[goatjc]

I would swap out the power supply and see if that helps.

 

[trexpesto]

Check your PSU voltages ( maybe with MBM5 hi/lo for a while).

Don't forget to ground yourself to the chassis to eliminate static. Try your other card by itself.

>>> Before swapping out the PSU, check that the new one is running up to specs itself. Probly the BIOS will tell you.

>>> Again, before swapping out the PSU, open both cases and run the molex from the other to your cards, or just the questionable card, without unhooking either PSU, your mobo, and everything. Time-saver.

If it works you have a pretty good indication thats the problem. If not, try your PSU to the cards, borrowed PSU to your mobo and drives, etc.

THEN, if the borrowed PSU is believed good enough, I guess you go ahead and install it so you can monitor your mobo and it together, testing for a while under load.

 

[Roguestar]

http://forums.anandtech.com/categories.aspx?catid=32

 

[happybelly]

I dust the cards on a regular basis, so that's not an issue. The power supply voltage is also fine.

The odd thing is that I took out the second video card, the one not connected to the monitor. It just seems strange to me that it would be the reason I was crashing.

The motherboard temperature has gone down from high 50's to low 60's to now the low to mid 40's since taking out the second card.

I guess I'll swap the video cards to see if I start crashing again.

 

[trexpesto]

Try both cards in each slot to check the slots as well.

could triple check NB temps, that the mounting buttons didn't loosen, since that was a recent change. Temps not changing is suspicious, no? Remount with recent thermal paste, or point a temporary fan at it to see what happes

also make sure you didnt bang anything when working on that, CPU heatsink, connections tight...

sorry if this is all obvious.

 

[happybelly]

Originally posted by: trexpesto
Try both cards in each slot to check the slots as well.

could triple check NB temps, that the mounting buttons didn't loosen, since that was a recent change. Temps not changing is suspicious, no? Remount with recent thermal paste, or point a temporary fan at it to see what happes

also make sure you didnt bang anything when working on that, CPU heatsink, connections tight...

sorry if this is all obvious.

I did say in my last post that the temperature of the motherboard has gone down since taking out the 2nd card by about 15-18C. I don't know why that would be.

The CPU heatsink is fine and the temperature is as well.

 

[trexpesto]

I thought you had said replacement of the broken nb fan had no effect on nb temps? Edit: where is yer mobo sensor?

My chipset fan broke 2-3 months ago, I got that replaced, but my motherboard temperature read in the high 50's to low 60's C even after fixing the fan.

 

[happybelly]

Originally posted by: trexpesto
I thought you had said replacement of the broken nb fan had no effect on nb temps? Edit: where is yer mobo sensor?

My chipset fan broke 2-3 months ago, I got that replaced, but my motherboard temperature read in the high 50's to low 60's C even after fixing the fan.

It didn't by itself. A couple of days ago when I took the second card out, the motherboard temperature then went down.

I'm not exactly sure where the sensor is located.