Computer randomly shut off, now wont power on

[bignateyk]

I have an xp 2400+ tbred B, epox 8rda+, 2x512mb corsair value ram, alpha pal 8045 w/ panaflo fan, and a 380 watt antec true power

I have had it running stable at 11.5x180 for the past 3 weeks.

Today, my computer just started randomly shutting off, so I thought it was somehow an overclocking issue. Originally, overclocking it had taken alot of voltage to reach even mediocre overclocks. I had thought it was a bad stepping I got.

Now today it shut off and I got an error on boot, and it was really unstable, so I decreased the overclock from 11.5x180 @ 1.725 volts to 12x166 @ 1.675 volts, and thought it was fixed. Then it shut off again an hour or so later. This time it did not turn back on at all.

I cleared the CMOS, and am running it at 100mhz FSB right now on default voltage, just to be safe.

Could my power supply be bad? I got it refurbed from newegg if that helps... Wouldnt it at least turn the power LED on if the power supply was OK and i powered on?? Would a bad PS explain why it took such high voltages to get even moderate overclocks?

Thanks,

Nate

 

[bignateyk]

hmm... some other stuff i noticed...

The first time it randomly shut off and I restarted I checked the temps in sisoft sandra and it read like 72 degrees C, and i was like HOLY SH!T and i turned it off right away cuz i thought the cpu fan was out or something, but when i checked it was working Then I restarted again, and it read like 20 degrees C, which obviusly wasnt right, so I restarted once again and it was like 50degrees C, which is what it has been for the last 3 weeks.

I restarted once more and it read like 70 degrees again. This is the time it shut off and I couldnt get it to turn back on... What could be going on?? I dont see how my temps could be reaching that high, and it never did that over the last few weeks...

could a faulty temperature sensor cause something like this? I dont see why the computer wouldnt turn on at all after that if that was the problem tho...

 

[MercenaryForHire]

Check your HSF contact area ... your main CPU or northbridge HSF combo may have started to detach. Or you damaged a component with your OCing.

- M4H

 

[bignateyk]

hmm well i ran memtest for a while and everything seemed ok in terms of memory... could it really be a misseated heatsink? How could that have happened with the alpha pal which screws on?

 

[DAPUNISHER]

could a faulty temperature sensor cause something like this? I dont see why the computer wouldnt turn on at all after that if that was the problem tho...

Do you have it set to shutdown in PC health? If so it may be working and since you are using a bolt down HS unless you pooched the install somehow (can't see how you could and it take 3 weeks to turn up as an issue ) then it probably is a faulty sensor reading high temps and PC health shutting it down. If you don't have it set to shutdown definitely look to the PSU as it could just be flat out flakey. Perhaps if none of that is a problem the board could be having some power reg. problems so watch those voltages for spikes and dips that are drastic.

 

[bignateyk]

is there anything that could have happened that would actually cause the cpu temps to go up that high if they were reading 50 idle for the last 3 weeks, now like 60-70 idle? That doesnt make any sense... not to mention the one time i turned it on it said like 20 degrees...

edit: also, wouldnt I hear it beeping or something if it got near the shutdown temperature?

 

[DAPUNISHER]

it should warn you if you have that enabled in PC health so that's something to check out. It's also possible the heat sensor is damaged or the CPU is damaged due to overclocking among other possibilities but you have places to start anyways so let the troubleshooting begin!

 

[orion7144]

I had the same thing happen to me when I was tunning AMD XP cpu's. Ended up having to RMA two of my chips. Mine however ran fine when I bought them (winter months) with temps around 50c. As soon as summer hit the idle temps were up above 50c and started climbing utill I started getting random lockups and then they would only work stable at 100mhz fsb. They were tried on other MB's as well. AMD did not have a problem sending me new one's and even said they were oos of 1700+ and sent me 1800+.

 

[bignateyk]

ok well I tried turning off the shutdown temperatures and warnings, but it still crashed, so it must be either the processor or the power supply...

I'm going to get my old 350watt antec from my dad when he comes up on thursday so ill test that i guess, then ill have to replace processor

I went ahead and ordered a 2100+ tbred B from newegg... hopefully that will solve the problem... i guess ill just keep this 2400+ for spare parts or something

 

[alphonsus]

how did you fix this problem?