The CPU Has Shut Down Due to a Thermal Event (Overheating)

[Digobick]

A while ago I upgraded my computer with a new motherboard (Intel D865GBF), processer (3.0Ghz P4, all thanks to the Intel Retail discount), and Kingston HyperX RAM.

Ever since then, my computer will shutdown if I attempt to do anything that stresses the video card (it's a Radeon 9700 Pro that's always worked beautifully before on an Asus board). After I boot the computer back up, I am greeted with a message that "the CPU has shut down Due to a thermal event (overheating)."

The thing that gets me though is that it's not really overheating. I was originally running the stock HSF from Intel, but I just replaced that with a Zalman 7000A Al-Cu HSF. Doing so lowered my CPU temps from 45°C idle, 65°C max load to a mere 35°C idle, and 47°C max load. I assumed this would solve my problem, but it didn't.

My four 120mm case fans are now running at full speed constantly, as is my Zalman fan; I've also gone ahead and left the case off. Still the same problem. There's no place in my BIOS that I can change the CPU alarm sensor, so that's out of the question.

Anybody have any ideas? I'm absolutely stumped, and I'm beginning to think I should've stuck with my P4 2.4 Ghz.


<Edit> Thought I'd reiterate that my computer ONLY does this when running a program that stresses the video card (most all 3D games, and even Windows Media Player visualizations!). The 3D demo on ATI's Catalyst Control Center crashes my computer within seconds everytime.


Cliff Notes
1. 3.0 Ghz P4 crashes when playing a game or anything that stresses video card
2. BIOS says the CPU overheated, but it never went above 47°C
3. 4 case fans, 1 Zalman HSF don't help
4. Any solutions? The rest of my system specs can be found in "My Rig" link

 

[MichaelD]

Not a temperature problem. It sounds like a weak powersupply problem. Are the connectors for mobo and videocard pushed on tightly?

ps
Your "My Rig" link is busted. Everyone had to redo them after the last FuseTalk downgrade...er upgrade.

 

[Digobick]

Fixed the "My Rig" link, thanks!

I will double check my connections and see if that helps. In the mean-time, here's what Intel Active Monitor shows for the power supply voltages:

+12V = 12.000
+5V = 5.052
+3.3V = 3.300
CPU Core: +1.52V = 1.494
CPU I/O: +1.5V = 1.484

I'm not really sure how close these should be, but hopefully someone here does.

 

[MichaelD]

Your voltages look fine, but possibly you have a weak PS that can't push the amperage a hungry rig like yours wants.

What brand/wattage PS?

 

[Double Trouble]

It could just be the video card that's causing problems -- put in another card to see if the problem persists. Also, you can try to see if there's an issue with the power supply, but that's more work as you'd have to put in a different PS to test it out.....

 

[mechBgon]

Have you checked for a BIOS update? Not sure what else to think of, the PSU is decent and then some (TruePower550). Is this a Prescott-core or Northwood-core CPU, by the way?

 

[cy7878]

did you have a heat problem BEFORE changing the heat sink?

 

[Digobick]

Thanks everyone for the suggestions. Sorry about the slow updates...my computer was shutdown while I tested the power connections and different configurations of the cables (nothing worked).

Here's some more information about my computer:
CPU: 3.0Ghz Northwood (130nm Process)
BIOS: Updated to the latest version (P19 I believe).

And yes, I was having the exact same problem before I replaced the Heatsink. In fact, my computer crashes just as often now as it always has.

I'm going to try a different video card next (ATI Radeon 64MB VIVO).

If that doesn't work, I suspect the problem lies with the power. I noticed in my Intel Active Monitor Alert Logs show that my +3.3V power feed dipped to 2.25 a couple of times in the past month, but my computer didn't crash when that happened.

I'm also wondering if a UPS back-up would help.

 

[cy7878]

you may just have a bad CPU. It's rare, but it happens. RMA the CPU by calling Intel Support (I hope it is a boxed version, otherwise Intel will make you go thru store of purchase). Tell them your CPU has problem with random shut-off's. Send the CPU back and they will check it.

 

[mindblogg]

I too think its to do with a low power source. Your config definetly requires atleast 350W smp.

 

[Digobick]

Well I did some more testing, and it appears to be tied to my video card. When I downgrade to an older video card I have no problems whatsoever.

My Radeon 9700 Pro does require an extra power connection, so I'm wondering if it's somehow drawing too much power. You would think a 550W Antec True power supply would be able to handle that, however.

I really don't want to go back to my old Radeon 64MB card, so I need to figure out where I go from here. Get a UPS to provide clean, uninterrupted power? New power supply? New video card (though I doubt the wife would approve of that)?