Tech support needed ;(

[c0nnery]

Before I describe my problem, here are the stats for the box in question:
1.2 AMD Athlon Thunderbird OEM
MSI K7T Pro 2-A
256megs Crucial PC100
Western Digital ATA100 5400rpm 20.0gig
Linksys LNE100TX (or something similar) NIC
Voodoo3 2000 PCI
Generic CDROM/floppy

Ok, my computer crashes within two minutes of powering on. The mouse and keyboard lock up, leaving me with a still image of whatever was on the screen at the time. I have absolutely no idea what would cause this. I haven't installed any software that could cause something like this to happen. The problem isn't OS specific; it affects Windows 2000, my Windows 98 rescue disk, multiple Linux/unix installation CD's, even the BIOS config screen. I have an old Compaq available to me, and I've tried swapping the RAM and hard drive with it, but the problem remains. I've taken out the NIC but that doesn't solve the problem. The only pieces of hardware that I haven't swapped are the video card and the motherboard itself. The video card, although old, has never caused me any trouble. The onboard sound card has been a problem, however, but it usually only caused me to reboot and never caused the system to totally lock up. I tried removing the battery that backs up the CMOS ram in hopes of clearing some obscure setting that might be causing the problem, but that didn't help. I am totally out of leads, and I'd really rather not have to buy a new mainboard (although snapping the MSI in half would bring me great pleasure). Do you guys have any ideas as to what would cause this type of problem? Many thanks,

Connery

 

[Geekbabe]

What are you using for a power supply ?

 

[h0vic]

If your computer crashes consistently like this no matter what you're doing, then it could be that your CPU is overheating. Check to make sure your HSF is seated on the CPU properly. Is your case getting enough cooling? It's summer, so the ambient (room) temperature can contribute to this.

 

[Butler]

My first guess would be your processor. If it's locking up within a given time in all those diferent OS's, and even the BIOS, it sounds like the processor's getting too hot. I would check the heat-sink to make sure it's snug with the processor and that the fan's running. It could also be the video card, maybe it's overheating.

If it hangs in BIOS like you say, you can rule out the hard drive, and probably even the RAM, the processor is about the only thing that's being used when you're in the BIOS that could be causing such a problem.

Try timing how long it takes to lock...if you've had the machine on it may lock a bit faster than if it's had a chance to cool for a bit, this would heighten my suspicion of the processor.


----------------------------------------------------------------
Great minds think alike eh' hOv?
B

 

[c0nnery]

I have a Cooler Master heatsink with the Arctic Silver thermal goop. I got into the BIOS and looked at the temps and everything seemed normal. When it crashed, the CPU was at 36 degrees celsius (about 97F). I haven't touched the heatsink setup since I put the computer together in January and it's worked fine all this time. I leave the case open, but it doesn't have any case fans. The power supply is an awfully cheap 300w one that came with my cheap case. I think the power supply theory is interesting; I once had to take the case I bought back because the power supply failed on me (the motherboard was getting no power). Unfortunately the only option I had was to exchange for the exact same case, so I was stuck with another cheap power supply. I'll have a look at the heatsink/cpu setup and see if I notice anything out of the ordinary, though. If it's not that, do you think the power supply is causing this problem?

 

[Butler]

I'd never rule out the PS, but I'd check the CPU first...

B

 

[Saltin]

Hope its the power supply. You can fix that for peanuts

 

[c0nnery]

I took the heatsink off the cpu and everything looked fine, but I reapplied the arctic silver just in case. I was intrigued by the power supply theory so I checked the voltage area of the bios and noticed a trend. When the computer powers on, the vcore is at 1.62 (or thereabouts). After a minute or so, the vcore starts to drop steadily. When it falls past 1.45, the computer stops responding. What is the "normal" value of the vcore? And would the falling vcore be indicative of a power supply failure or just a side effect of a bad CPU?

 

[Shack70]

2 questions...
1.2 AMD Athlon Thunderbird OEM...it that a 266FSB or 200FSB??
If it's 266, your PC100 ram may be acting up.
Voodoo3 2000 PCI....why are you using a PCI video card??? Did you set the BIOS for a PCI card?

 

[Darkluck]

your vcore should be somewhere in the neighborhood of 1.78 to 1.82 V (this depends on how accurate your system readings are and stuff), I believe the AMD specs say 1.80V...you can find out on their website for sure. I'd buy a new powersupply for sure, an enermax if you have the choice. It's been a long time since I had an electronics class but I think your PS is dying.