Please Help!!

[taurus1520]

Here's my problem. Back in August I put together a system that ran flawlessly until the beginning of December. Then one day, without warning, it shut down. I rebooted, no errors popped up, and a short while later it did the same thing. I let it cool down, thinking it overheated, and waited a few hours. I turned it on again, and it did the same thing. It has shut down while in Windows, and even booting up. It did it once while I was in the BIOS trying to see if the problem was there. Since December I have returned and had replaced my MoBo and my PSU (both with identical models.) I just got my PSU back today, and after installing it, I left my computer on, not doing anything except running a screensaver, and after about four hours, it shut down again, no warning, no error messages upon rebooting. My best guess is that something may be overheating. I thought it might be a virus, but shutting down before Windows even loads threw that idea out. (And in case my BIOS was somehow affected, I have had a new MoBo installed.) My biggest problem though, is that I'm stumped, and I'm the one everyone I know comes to for their computer problems. So I am here to ask for your help. If any more details about the system or anything else are needed, just ask.

Here is my system:
CASE - Lian Li PC65B (standard case fans which came with the tower)
PSU - Antec True Blue 480 (replaced by mfg.)
CPU - Intel P4 3.0c 800mhz FSB (heatsink/fan that came with the CPU)
MoBo - ABIT IC7-G (replaced by mfg.)
RAM - Corsair TwinX 1GB 3200LL
Video - BFG Technologies GeForce FX 5900 Ultra 256MB
Audio - Creative Sound Blaster Audigy 2 Platinum
Hard Drive - Seagate 160GB SATA
CDROM - Plextor Plexwriter Premium 52/32/52
DVD-ROM - AOpen 16x
OS - Windows XP SP1

I'm about ready to go insane and take a bat to my computer. If only I wasn't still paying it off.

I have also checked the temps, and the CPU, Northbridge, and case, and they all seem to be okay.

[trev]

Hi,

I would try and unplug/remove as much as possible from the system. CD-Writer, DVD, Sound card and anything else that isn't needed to run the machine. You could try running a memory test (www.memtest86.com) to see if that chucked up anything although I don?t think that that would cause the problem.

I've had similar problems that ended up being caused by power in the person?s house. You may want to try plugging it in to another area of the house if possible or taking it to a friend?s house and trying it there.

Good luck

[silencio]

Good advice above. Unplug as much as you can and then let it sit for a long long time or until it reboots again. Since you said that it rebooted while you were in the bios one time you can probably rule out a windows issue.

[gazzak]

Power or heat, I'd be surprised if it was anything else!

[confus-ed]

Are their any logs to look at ? we just have the clue 'OS - Windows XP SP1' so which one ? ... home or pro ?

If you think its overheating, get the motherboard utility to monitor its temperature (they nearly all have one) & see ... telling us also what bios reports as 'starting' temp

Rebooting in 'bios' though sounds like motherboard or chip & since we have replaced a 'faulty' psu - I'd be looking very hard at those ... & we replaced an m/b too (was that replaced before/after 'reboot'?) - find someone who has all the necessary bits for you to test each part seperately or you'll go nutz trying to figure which part is 'to blame'

[taurus1520]

I have XP Pro. And it's not rebooting, just shutting off. I'll try the solution of removing as many devices as possible. And when I get the chance I'll post the startup temps, and the temps after it has shut down on me.

However, I don't think it's a power problem at my house. I have my monitor and speakers plugged into the same surge protector, and when the computer shuts down the monitor ans speakers are still on.

I did call BFG Tech, the mfg. of my video card. The tech support said that if the GPU was overheating a warning would pop up first. Regardless, I think I'll try that device first.

Like I said, what gets me is that it ran flawlessly for four months. I'll post the results from the suggestions and we'll see what's up.

[Tekboy]

Also, a "Line Conditioning" UPS is a very good idea in this case.
Surge protector is not enough, and the ONLY things plugged into the UPS should be the tower and the monitor.

[NooNoo]

So nothing in the event viewer?
Bios temps can be seen from windows, on the mobo cd will be a windows app to do this.

Is the sytstem overclocked anywhere?
How about the ram settings?

[dddwarp]

Are you sure there is nothing in the event viewer? If it was an "unexpected power loss" than there would be something in the event view about that. You need to isolate the "bad" hardware. Remove everything and I mean everything. That may mean taking out the mobo from the case and just have the mobo, vc, floppy and pw (I have had shorts with a case before). See if you have the same problem running hardware tester. If you do, then it is one of those pieces. If not, then start adding one piece of equipment at a time until you get the same issue. The toughest thing to troubleshoot is that on-off hardware problem.

[taurus1520]

Nothing is overclocked, and the RAM settings are the default in my BIOS. I'll grab that info when I get the initial temps.

I am trying to hunt down a video card so that when I remove my current one, I can check out the system. Something tells me that is what is overheating. Though I don't have it overclocked, there's no obstructions to impede airflow, and I have been running it with the cover off. I will try to get on and check the event viewer first, I haven't had the chance yet. But that won't be until I get home from work. I'll keep everyone posted. Thanks for the help so far.

[confus-ed]

..I am trying to hunt down a video card so that when I remove my current one, I can check out the system. Something tells me that is what is overheating ...

Well is it too hot to touch 'comfortably' ? ... If it is, then there, very probably, is your issue

Are we absolutely sure there's no log entries ? (Or do we not know where to look ?)

& of course this is all before we get around to 'services' ... which is yet another possibility ... but lets get all these 'temp' issues 'put to bed' first ...

[taurus1520]

Here's my update...

Finally got the initial temps. I started it up for the first time in two weeks (I was waiting to get that temporary video card.) The starting temps, which I got from the BIOS immediately after startup, were as follows (in Celsius):

System 18 degrees
CPU 34 degrees
PWM 20 degrees

I entered windows and opened the event viewer. I found absolutely nothing regarding power loss. In fact most data in the log was from the startup of each of my windows session, and nothing regarding sudden power loss. Any errors in the log were regarding applications. Then I opened my WinBond Hardware Doctor v2.71. All temps after ten minutes of idling, in celsius, were as follows:

Sys 23 degrees
CPU 34 degrees
PWM 26 degrees

I also opened my NVIDIA advanced settings and got the current GPU core temp in celsius, which was 41 degrees.

I surfed the internet for about an hour and things were still running smooth. I was also listening to some music on winamp for a while and then checked my temps at about 1 hour from initial startup.

Sys 28 degrees
CPU 52 degrees
PWM 41.5 degrees
GPU 43 degrees

Then I started up Knights of the Old Republic to test the GPU some. I ran it for about ten minutes and saw little change in the temps, except the GPU which went to about 60 degrees Celsius (I tasked switch back and forth to keep an eye on the temp.) I felt it was to tedious to do that so I began looking for something I could run in a window and push my GPU at the same time. I was looking around for a few minutes and thought I might try running the Tron 2.0 Demo. I was in the startup window, and tweaking a few game settings...then, shutdown. I did manage to see that the GPU temp was still around 41-43 degrees Celsius before the shutdown, and I hadn't even started the game yet.

My next goal was to quickly get into the BIOS to verify the temps at shutoff. I rebooted, got into the BIOS and checked the temps, I had no sooner verified that the temps were the same as my 1 hour temps when it shutoff again. Now that I knew overheating was no longer the issue (every heat producing hardware was only warm to the touch) I decided to go to the next step.

I quickly, yet carefully of course, removed my graphics card and put in an old NVIDIA RIVA TNT2 Model 64. I booted up, the system immediately recognized the new card and now I am here typing this post. No shutoff has occurred. Temps are all about the same as before.

My suspicion is that my 5900 Ultra must have some kind of flaw or defect. Something's happening to it that is causing the shutdown. Any ideas? I will be calling BFG Tech and try to get any info from them. Thanks for the help so far, but I'm not out of the woods yet. Not until I can specifically pinpoint the exact nature of the problem. But at least I seem to have narrowed it down to the problem component.

[arch0nmyc0n]

I had a DLink 538TX network card that caused the same thing to happen in a computer I was working on a while back. Well I dunno if it was the card specifically, but I took it out and it stopped shutting down. I hope it helps... but I doubt it

But I have seen video cards cause the same thing, actually just the other day I had one in here that was sorta the same problem. It ended up being the video card.

[ilovetheusers]

This has got to be the most annoying thing. Since you're having a hard time pegging the issue maybe install something like sisoft sandra that you can test items individually and see when the shutdown happens when you stress individual components and see if you can repeat the issue a few times. Since that's probably not feasable due to the shut downs...


If it's any help I had 2 maxtor drives that caused the system to shut down, reboot and hang. Might want to DL whatever seagate uses to test their drives for errors. This is one item you have not swapped.

Also, did you run memory test (www.memtest86.com) like was said before? It runs at boot up so windows is factored out.

Finally, I know it's probably checked but under power or health in your BIOS - the feature to shut down if the cpu or mobo overheats is turned off or is cranked up to the max temp. right?

[DocPC]

Sounds like one of the RAM modules on your video card is going bad.