203F -- Overheating CPU =:0

DeusExMachina · 03-07-2006, 09:58 PM

Been using my sister's ancient computer, and now on her laptop, until I could get some parts to get my own computer up.

Been recognizing the CPU temps rising slowly. This morning while doing routine maintenance, the CPU temp shot upto 203F!! Yeah, I immediately turned the computer off!!.

This is an slot-1 P3 1Ghz processor on a P3F-B (Version 1.4/5 PCI model) with a typical slot cooler. Asus Probe shows the voltage as being fine (it's a Enermax 350 PSU). Fan speed is at 5000+ RPM (normal).

What can be causing the overheating when the processor fan is working, plenty of case cooling (6 fans - X-Alien case), and zero overclocking? Never had a processor seriously overheat before -- and I have computers from all the way back to 1991.

Help!

*Kage* · 03-07-2006, 10:05 PM

What is that in C?
Have you inspected the HSF to see if there is a large quantity of dust and gunk in the Heatsinks fins?
Are you sure the reading wasn't an error?

DeusExMachina · 03-07-2006, 11:25 PM

Metric conversion: 203F/95C. 212F is the temp of boiling water. =:0

Haven't pulled the slot card to eyeball the whole processor unit yet (largely because I never worked with a Slot-1 before, and if I break one of those retention arms, Sis is going to kill me -- had to promise to stay off her laptop after I get my parts in as it is <-- it's coming Thursday). :/ But I recently blew all the dust off.

Part of the problem in dxing this computer is the P3F-B is intolerant of low RPMs on PSUs on boot (Enermax can spin real slow on a cold boot). So on almost every boot (especially in winter) there's going to be a hardware problem alert at boot. Nothing is wrong at boot though -- voltages were fine (and only recently has the processor temps rise). Temps normally were alround 66F to 78F idle to 130F (sorry I use F instead of C) on load. Now it's normally 78F to 104F idle, and 130-160F on load.

The temps are true (shows in the BIOS and on Asus Probe), unless the on die temp sensor is bad. The high temps only show on load though. If I don't do anything it'll stay anywhere from 66 to 78F. Open a MP3 player and it'll go upto 120F. Do any CPU intensive activity, then the insane temps show up.

Can dust alone be the problem? The HSF fins are very open, and I could see through the window if there's a "gray mask" developing.

Can it be something like some short (weary of them due to what happened to my P4T533's voltage regulator)? Reason I ask is the funkiness of the memory slot placement -- a stick will fail in one slot, work fine in another. :/

In short, can it just be a m/b failure all together? Never had one before the P4T533 went bad, and then I didn't notice the problems until the reboots occured.

DXM

*Kage* · 03-07-2006, 11:54 PM

The reason I asked about dust is that here it is so dusty that in a well ventilated case, it would build up so thick under the CPU fan that no air reached the fins...
Do you have any sort of Thermal Probe to verify that the reported temps are accurate? A case temp sensor would be enough to see if the reported temps are way off.

DeusExMachina · 03-08-2006, 12:12 AM

Have an Enermax 6 fan controller with IIRC 2 temp probes. Weary of using it, as I don't know if it's that accurate or not (and if I fry Sis's board I'm in BIG trouble), and frankly don't know how to use one of those flat leads on a CPU die itself (won't it melt)? Only thing I've ever done with a P3 Slot-1 was remove one once, years ago.

H-e-l-p!

DXM

3 of 7 · 03-08-2006, 12:45 AM

So clamp an aligator clip on the cpu fan blades so it doesn't overspin and blow that sucker out...Canned air is the best, but I just use a vaccuum cleaner on blow (hint, don't get it anywhere near the computer, they make a whack of static)

WarpSpeed · 03-08-2006, 06:36 AM

Make sure all the fans are actually spinning when you turn it on. Instead of simply blowing off the heatsink, I would be inclined to disassemble it to clean it out completely and to reseat the heatsink with fresh thermal compound.

3 of 7 · 03-08-2006, 07:58 AM

This is a slot 1 computer....IIRC the CPU and HS are a pita to get off.

3 of 7 · 03-08-2006, 11:46 AM

Another thought here, open your task manager and see if something has grabbed your CPU, with a bunch of dust on the HS and a bad driver running the CPU 100% it'll spike the temps..

DeusExMachina · 03-10-2006, 04:39 AM

Back on this Slot-1....Yeah!

Took out the processor. Dissassembled it (wasn't very dirty, amazingly). Cleaned the old thermal compound. Put some Arctic Silver on it instead. Added an extra powerful fan (50mm instead of 40mm <-- additional 9cfu). Put it back in and temps are a respectable 85F with Winamp playing, 68F/20c while idle.

Maybe it was the thermal compound that needed switching, since it hasn't been changed for 7 years. :/

BUT.... Rechecking the voltage, and the volts are alittle on the high side on the +12v (12.418) and +3.3 (3.488) rails. The +5v rail is fine, just a tad above +5v.

Is it just the PSU model causing the higher voltages, or just a too good PSU? It's just a 350w and it's operating 2x256mb sticks of memory; old GeForce 2 MX v/c; Creative 7.1 s/c; LAN card; and 8 fans (adding the CPU and door fans), two HDDs, DVD/RW and CD-RW. Would've thought the voltages would dip under than load. :/

DXM

03-07-2006, 09:58 PM	#1
DeusExMachina ABX Folder Join Date: Mar 2006 Location: The Empire State Posts: 477	203F -- Overheating CPU =:0 Been using my sister's ancient computer, and now on her laptop, until I could get some parts to get my own computer up. Been recognizing the CPU temps rising slowly. This morning while doing routine maintenance, the CPU temp shot upto 203F!! Yeah, I immediately turned the computer off!!. This is an slot-1 P3 1Ghz processor on a P3F-B (Version 1.4/5 PCI model) with a typical slot cooler. Asus Probe shows the voltage as being fine (it's a Enermax 350 PSU). Fan speed is at 5000+ RPM (normal). What can be causing the overheating when the processor fan is working, plenty of case cooling (6 fans - X-Alien case), and zero overclocking? Never had a processor seriously overheat before -- and I have computers from all the way back to 1991. Help!
(Offline)

03-07-2006, 10:05 PM	#2
Kage Man-Foe! Join Date: Nov 2004 Location: Fiery depths of Hell Posts: 3,607	What is that in C? Have you inspected the HSF to see if there is a large quantity of dust and gunk in the Heatsinks fins? Are you sure the reading wasn't an error? __________________
(Online)

03-07-2006, 11:25 PM	#3
DeusExMachina ABX Folder Join Date: Mar 2006 Location: The Empire State Posts: 477	Metric conversion: 203F/95C. 212F is the temp of boiling water. =:0 Haven't pulled the slot card to eyeball the whole processor unit yet (largely because I never worked with a Slot-1 before, and if I break one of those retention arms, Sis is going to kill me -- had to promise to stay off her laptop after I get my parts in as it is <-- it's coming Thursday). :/ But I recently blew all the dust off. Part of the problem in dxing this computer is the P3F-B is intolerant of low RPMs on PSUs on boot (Enermax can spin real slow on a cold boot). So on almost every boot (especially in winter) there's going to be a hardware problem alert at boot. Nothing is wrong at boot though -- voltages were fine (and only recently has the processor temps rise). Temps normally were alround 66F to 78F idle to 130F (sorry I use F instead of C) on load. Now it's normally 78F to 104F idle, and 130-160F on load. The temps are true (shows in the BIOS and on Asus Probe), unless the on die temp sensor is bad. The high temps only show on load though. If I don't do anything it'll stay anywhere from 66 to 78F. Open a MP3 player and it'll go upto 120F. Do any CPU intensive activity, then the insane temps show up. Can dust alone be the problem? The HSF fins are very open, and I could see through the window if there's a "gray mask" developing. Can it be something like some short (weary of them due to what happened to my P4T533's voltage regulator)? Reason I ask is the funkiness of the memory slot placement -- a stick will fail in one slot, work fine in another. :/ In short, can it just be a m/b failure all together? Never had one before the P4T533 went bad, and then I didn't notice the problems until the reboots occured. DXM
(Offline)

03-07-2006, 11:54 PM	#4
Kage Man-Foe! Join Date: Nov 2004 Location: Fiery depths of Hell Posts: 3,607	The reason I asked about dust is that here it is so dusty that in a well ventilated case, it would build up so thick under the CPU fan that no air reached the fins... Do you have any sort of Thermal Probe to verify that the reported temps are accurate? A case temp sensor would be enough to see if the reported temps are way off. __________________
(Online)

03-08-2006, 12:12 AM	#5
DeusExMachina ABX Folder Join Date: Mar 2006 Location: The Empire State Posts: 477	Have an Enermax 6 fan controller with IIRC 2 temp probes. Weary of using it, as I don't know if it's that accurate or not (and if I fry Sis's board I'm in BIG trouble), and frankly don't know how to use one of those flat leads on a CPU die itself (won't it melt)? Only thing I've ever done with a P3 Slot-1 was remove one once, years ago. H-e-l-p! DXM
(Offline)

03-08-2006, 12:45 AM	#6
3 of 7 You can run..... Join Date: Feb 2004 Posts: 4,660	So clamp an aligator clip on the cpu fan blades so it doesn't overspin and blow that sucker out...Canned air is the best, but I just use a vaccuum cleaner on blow (hint, don't get it anywhere near the computer, they make a whack of static) __________________
(Offline)

03-08-2006, 06:36 AM	#7
WarpSpeed Quod erat demonstratum Join Date: Apr 2002 Posts: 194	Make sure all the fans are actually spinning when you turn it on. Instead of simply blowing off the heatsink, I would be inclined to disassemble it to clean it out completely and to reseat the heatsink with fresh thermal compound. __________________ IS-7, 2.6C @ 3.45GHz, 2GB OCZ PC3200, Antec 500, W2KSP4
(Offline)

03-08-2006, 07:58 AM	#8
3 of 7 You can run..... Join Date: Feb 2004 Posts: 4,660	This is a slot 1 computer....IIRC the CPU and HS are a pita to get off. __________________
(Offline)

03-08-2006, 11:46 AM	#9
3 of 7 You can run..... Join Date: Feb 2004 Posts: 4,660	Another thought here, open your task manager and see if something has grabbed your CPU, with a bunch of dust on the HS and a bad driver running the CPU 100% it'll spike the temps.. __________________
(Offline)

03-10-2006, 04:39 AM	#10
DeusExMachina ABX Folder Join Date: Mar 2006 Location: The Empire State Posts: 477	Back on this Slot-1....Yeah! Took out the processor. Dissassembled it (wasn't very dirty, amazingly). Cleaned the old thermal compound. Put some Arctic Silver on it instead. Added an extra powerful fan (50mm instead of 40mm <-- additional 9cfu). Put it back in and temps are a respectable 85F with Winamp playing, 68F/20c while idle. Maybe it was the thermal compound that needed switching, since it hasn't been changed for 7 years. :/ BUT.... Rechecking the voltage, and the volts are alittle on the high side on the +12v (12.418) and +3.3 (3.488) rails. The +5v rail is fine, just a tad above +5v. Is it just the PSU model causing the higher voltages, or just a too good PSU? It's just a 350w and it's operating 2x256mb sticks of memory; old GeForce 2 MX v/c; Creative 7.1 s/c; LAN card; and 8 fans (adding the CPU and door fans), two HDDs, DVD/RW and CD-RW. Would've thought the voltages would dip under than load. :/ DXM
(Offline)

Thread Tools
Show Printable Version Email this Page
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode