Moved My Tower While On Causing Crash, then Infinite Boot

jra716

Distinguished
Dec 20, 2007
31
0
18,530
So. My computer is pretty new, just built it in December.
Windows Vista Ultimate 64-Bit
Q6600
GIGABYTE GA-P35-DS3R
CORSAIR XMS2 4GB (2 x 2GB) 240-Pin DDR2 SDRAM DDR2 800
SILVERSTONE ST85ZF 850W
Western Digital Caviar SE16 WD3200AAKS 320GB 7200 RPM SATA 3.0Gb/s Hard Drive
EVGA 512-P3-N841-A3 GeForce 8800GTS (G92) 512MB

It's been working completely fine. I had to re-plug in the audio-in in the back. I turned it on and it was working fine in windows, then I decided to move the computer back in place. While pushing it in place, I think I pushed the tower over the PSU cord, it fell off the cord a little (causing it to shake slightly) and the computer crashed. When I tried to turn it back on, it just keeps going into infinite boot loop and doesn't show any sign on the monitor. Fans and everything turns on for a few seconds, then it shuts off, then repeats.

My guess is that the power may have been pulled out slightly or the abrupt slight impact may have damaged something. The thing is, I have no idea what could be damaged, or even how to test what's wrong since I don't have any extra hardware to swap anything out. Could have I burnt out the CPU? Maybe damaged the hard drive? Or maybe I messed up the PSU?

I was thinking of buying a new hard drive in the future but don't want to purchase any new hardware at the moment. Any advice would be greatly appreciated.

Thanks!
 

Zorg

Splendid
May 31, 2004
6,732
0
25,790
There is probably no damaged hardware. Unplug the PSU and check to be sure eveything is seated properly on the mobo, then plug it in and boot.

If it fails, unplug the PSU and move the CMOS reset jumper to the reset position for 5 seconds and then back to the original position, plug the PSU back in and boot.
 

mman74

Distinguished
Mar 22, 2006
403
0
18,790
Can you get into the BIOS? Most likely it would have been the HDD you messed up, I can't see the motherboard shorting out because of what you did. Does the BIOS detect the HDD? If you can't even get into the BIOS then the culprit is most likely the motherboard.
What exactly is it doing when you turn it on?
 

blashyrkh

Distinguished
Jul 4, 2007
350
0
18,780
Why would a HDD cause constant reboot?? That's silly. A messed up PSU can cause constant reboots or maybe some shortcircuit which occured after you pushed your tower back to its position. Check everything out inside your box and do what Zorg said.
 

jra716

Distinguished
Dec 20, 2007
31
0
18,530
I tried resetting the CMOS and nothing happened. Can't get into the BIOS at all. I checked and everything seems to be seated in place. Checking online about my PSU, I'm pretty sure its the root of the problem now based on the LED light color.

"A small LED located next to the large power On-Off switch indicates the operational status of the power supply depending on its color.
Orange = Standby Mode
Green = Normal Operation
Red = Fault Condition"

I'm pretty sure its orange when I boot up from what I recall, but I don't understand why it stays in standby mode, unless it was actually red. My computer is sitting at home unplugged so maybe it just needs to sit for awhile. Well see. I just hope its not my hard drive! I don't know what I'd do if I lost everything. I guess it really pays to backup important files.
 

jra716

Distinguished
Dec 20, 2007
31
0
18,530
When I got home from work yesterday, I plugged everything in, it went into BIOS, and I reset my previous settings. Everything was working fine, except my idle temps were 10 degrees higher and my load temps were about 5 degrees higher. After 5 minutes, it crashed and went back into infinite boot loop. Turned it off, then it was able to turn on fine again.

What would cause it to run hotter? I need to check CPU-z again and see what the voltage is like compared to what it was before. Is it possible its sending more voltage than before or maybe less? Should I try reseating the HSF? I hope I don't have to replace anything.
 

Zorg

Splendid
May 31, 2004
6,732
0
25,790
No, don't remove the HS, it is not the cause of the Temp change. Measure your idle temps with either Core Temp or Real Temp and post them. Real Temp has a default TjMax of 95C instead of 100C in Core Temp, so your temps will be 5C lower.

Also post the HS information and the VID. I assume 2.4G/1066.


 

jra716

Distinguished
Dec 20, 2007
31
0
18,530
I'm using a Tuniq Tower 120. VID is 1.225. Q6600 is overclocked to 3.6GHz (400x9). DDR2 800 is running 1:1.

Before, my temps (measured by CoreTemp) were about 33º idle and 61º load. After the crash, it went up to about 43º idle and 65º load.

BIOS cpu voltage was set to 1.41875, CPU-z showed 1.360 on idle, then I can't recall what it was under load. This is from memory of the notes I wrote down. I'll have to check at home to verify.
 

jra716

Distinguished
Dec 20, 2007
31
0
18,530
Each core is usually pretty much at the same temp. I use CoreTemp. Can't post any screenshots because i'm at work. Load temps measured after using two Orthos instances with the affinities manually set in the task manager.
 

jra716

Distinguished
Dec 20, 2007
31
0
18,530
No particular reason. I just use the computer mainly for media and gaming (UT3, Crysis, Bioshock, etc). I want to run the RAM at 1:1 so I guess my only other option would be to drop the multiplier to 8 for 3.2GHz. Or I guess I could just change the memory multiplier to something else.
 

jra716

Distinguished
Dec 20, 2007
31
0
18,530
The last chip I upgraded from was an E2200 and it was struggling to get overclocked to 3.0GHz with even higher temps than my quadcore. The VID on that one was 1.325.
 

Zorg

Splendid
May 31, 2004
6,732
0
25,790
Try setting 3G/1333 with a 2.4 multiplier. It will be slower i.e., not as snappy, but you might find that it is easily livable. Your CPU will last a lot longer. If you don't care about longevity, or the reduction in speed bothers you, then go back to 3.6. There is no real loss in not having 1:1. The RAM speed/multiplier has much less of an effect than you might think on the core2. The difference from 3.6 to 3.0 may aggravate you though. There is a lot of room between the two, maybe 3.2 or 3.4.

If you do a lot of encoding then it will take more time, obviously.

If you decide that 3G is livable you can certainly reduce the Vcore and heat significantly.
 

jra716

Distinguished
Dec 20, 2007
31
0
18,530
So after the computer was sitting for a few days. Turned it on. Temps were down to what they were before, but got blue screen on stress testing. Upped the voltage one notch and everything seems fine now.