Computer Shuts Off / Locks Up With No Warning (long)

Personally I'd remove the video card and use the on board version to see if thats the problem. I would also try a new power supply and re-format the computer.

Fry.
 
Wow....

On the surface it sounds like processor overheating, I know you've said you have checked but look at the actual processor, make sure the heatsync is sitting properly. You want to make sure the plastic clips are in place on either side of the heatsync holding it on. I've seen some people have it turned wrong too where the bottom of the heatsync has a lip and that lip was sitting on the processor instead of off to the side like it was designed to be. You want the heatsync to lay flat on the processor. And for it to be CLEAN.

BTW 60c scares me, getting that close could actually be your problem, clean out your fan and heatsync with canned air (don't spin up the fan though as it is built to only spin so fast and it will do funky stuff if it gets going faster then that (In other words don't go get an air compressor and leave a steady stream of air on it, or atleast turn on the regulator down quite a bit)

I'm really doubting it's memory unless you increased your usage on sunday from saturday. I could be wrong on this, it's just unlikely.

It's not the cpu fan stopping or sizing up. I used a fan throttling utility by nVidia to test it and it was fine (4400 rpm).
The case has two 120mm fans with variable speed controls (front and back), a fan on the cpu. I cleared the fans of any dust and debris, and they were a little dirty, but spin freely.

Reading that it SOUNDS like you only cleaned the case fans and didn't open up the case, if I assumed a bit much still be wise to clean it up and check the heatsync.

It's not the hard drive. I used chkdsk, fdsk to correct 2 errors, then I reformatted the SATA drive, reinstalled windows xp. twice. I installed a HD monitoring utility, the temps are great.

Chkdsk finding errors isn't a Hard Drive problem 99% of the time, it's a "I shut down windows incorrectly" problem and this part of the pagefile didn't get removed or this file was written funny or didn't match the file allocation table. What the other gentlemen was thinking was what S.M.A.R.T. calls "Reallocated Sectors" typically when those start getting used, it's time to backup your hard drive and get a new one because you are going to see more of them, not a matter of IF, a matter of WHEN.

What you had, could have been fixed with "chkdsk C: /f" and rebooting let it run upon bootup and that would have been fine, 2 isn't all that bad either, I've seen some that literally took an hour and a half to run chkdsk because it was fixing so much crap.

Another potential problem is bad capacitors. Basically they look like little battaries with either X K or a peace sign on top of them. Typically the one's I see go bad are the one's with X's or peace signs (X's more then anything). I'd be really surprised if that was the case though, you can tell if that's the problem if the top is starting to pop out some. That would change the voltage to whatever part is down line from it and would grow worse as time went on. (www.badcaps.net)

I highly doubt it's your video card and I've never seen a power supply cause those kinds of symptoms before, typically it's either working or it is not. Even had a person bring in a computer that the capacitor had blown out on the power supply and the computer still worked (though smelled like really nasty ammonia)

Dismissing overheating because the system is set to cut off for safety purposes at 75c and you are only reaching 52c isn't wise. Open up your case and look at it.

DIRTY_computer.png


Told me their computer wasn't overheating too... (and it spun too!)

Love,
Charles~
www.CharlesTheTech.net
 
Last edited:
By the way, your processor is only rated to run at 65c

http://www.heatsink-guide.com/maxtemp.htm

Above Linked Site said:
The temperatures specified for AMD CPUs max case surface temperatures. These CPUs do not have an internal diode to measure CPU temperature. The accuracy of the CPU temperature measurement depends on the motherboard; therefore, it is possible that the CPU overheats even though the CPU temperature reported by the motherboard is below the specified maximal temperature.
 
Heh, the strangest deal I had one time was a computer rebooting itself at random intervals, that I eventually traced down to the MONITOR causing it. I guess the monitor was sending back some type of spike to the system and causing it to reboot at weird intervals. I eventually replaced all of the components, including the MB, CPU, PS, video card, etc. It was a Pentium system, so this was a while back. I would replace one component, then let it run, with a timestamp utility inserted into the autoexec.bat, to write a logfile entry every time it rebooted itself. I would get up in the morning, and look at how many times it had rebooted itself overnight, and then knew it wasn't fixed yet. That was a long troubleshooting process :)
 
I had a problem a little while ago of a computer restarting randomly.

It ended up being the Ipod that was connected to the computer. Took me hours to figure out.
 
I had an 80mm case fan do this to me in my early days of playing EQ, would only happen when I loaded a game though - my guess was that the fan was somewhat faulty and drawing excessive amounts of power and when the system needed the extra power to run a game it died and hence rebooted !

Since then I have seen even keyboards / printers cause this sort if thing though.
 
Sorry to interject and I by no means mean to hijack this thread, but I have sort of a similar problem first off I know or 1 thing I leave my comp on way too much (pretty much 24/7/365 no exaggeration), and now recently when i go to shut if off or reboot it stays off, when I hit the power button I acts like its gonna start normally but then clicks off and the front lights change from green/yellow to red (I have to hold the switch for a few seconds and then it "turns off" but then repeats: (boot up for 1/2 a second -> lights from green/yellow -> red -> hold till reset ->repeat

I think it might be my power source but unsure, I'm not quite versed in the hardware dept yet.
 
That happens because its running too hot, and when you shut it down its too warm to boot. I'd bet there's a ton of dust in your case as well as poor air circulation. Get a can of compressed air and blow all the dust out. Check your power supply if possible because it may be faulty, but more likely its your CPU that is overheating.
 
I should have taken a picture of my Nvidia video card before I used the canned air on it. Was so covered in dust that the built in fan was not working, and you could've fried an egg on it. Still ended up putting the new ATI video card in over the Nvidia one though, so much more reliable and less "quarks" with games afaik.
 
Might as well lump my tech issue on see what people think :p

Ever since I upgraded my NVidia drivers/software, their little control panel when you try to launch the display properties takes FOREVER to actually load. You can click it 40 thousand times and randomly anywhere from 2min-5min later it will pop up. I tend to leave it open out of annoyance. Dunno if anyone else has this issue.
 
I've never had a failing power supply give me the BSoD, unless it was in a system that also had another problem. The main power supply would certainly insta-off the system if bad tho, my last one like this was an out-of-box VPR Matrix (which still kicks beautifully multi-boxing at 2.67GHz).
As for the video card, I have had more than one system that gave me either lockups, or even a crash to desktop, then shutdown (w/out prompt) when the video card overheated. One time it even showed up as multiple-PnP-found video cards when I rebooted quickly after a shutdown (so a swap-to-new was the fix). That was an Nvidia 256meg (forget exact model, sorry), but wasn't an overheating problem, just bad chips/card.

The other one was an ATI 9800 video card w/ a louder-than-normal cooling fan, and that one would boot me to desktop, then shutdown the computer only after a few (like 3-4) hours of on time (or a bit longer if just sitting on desktop w/out graphic-intensive use). I had a hard time finding a replacement cooling fan w/ exact specs, so put a slot-cooler directly next to the vid.card, which didn't solve the problem but made it less occuring (sometimes days w/out a shutdown). Had to replace vid. card in the end tho of course.