• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Athlon XP users having CPU heat issues - you might want to read this...

The Sauce

The Sauce

Diamond Member
Ever since I installed my XP-1700 with the SK-6 I was having heat problems - temps upwards of 50C at idle and crashes. I did a complete breakdown and rebuild and when I took off the SK-6 i found condensation between the h/s and CPU and all over the surface of the CPU. Initially I thought it was a problem with my Arctic Silver compound. Then I realized that it was actually the little foam pads on the CPU that were the problem...they were too stiff and not allowing good contact between the h/s and CPU. This resulted in condensation between the h/s and CPU! BAD BAD BAD! The SK-6 clip is pretty damn stiff too. I took a razor and shaved off some of the pads and now temps are about 15 degrees cooler...37C after running 3dmark2001 for about a half hour. Suddenly everything is stable again 😉

Just thought some of you might find this information useful...I never would have thought of it had I not completely broken down the system.
 
The pads are there to help protect the CPU core from being crushed, sounds like the SK6 was not seated right,anyway glad you got it fixed.I did not have that problem with my Taisol CGK760092 it fits like a glove 😉.
 
Agree with ToBeMe, my 1600 running at 1900 runs at 34C at 100% (copper glad.) and 30C idle.

Cool chips! 😎
 


<< ToBeMe,

what motherboard? what ambient temp?



Mike >>


Mike, it's on a KR7-RAID and the ambient temp is at 28C - 29C all the time. I keep this sytem in my "systems room" which I've cut off the heat to (my wife loves the house to warm for my tastes) and the room stays about 21C - 22C year round!🙂 It also sits in a Tornado 3000 case so, it has no problems with ventilation!😉
 
I have a sk6 on my XP2000+ and it runs @ 49~50C under full load (8KHA+). House temp. is 23C. I never noticed any condensation on my cpu, plus I reapplied ASII many times and no change. Snatch, those foam pads cannot "lift" the heatsink up to cause the poor seating you had. I think you probably had the HSF installed wrong. In addition, 50C is not really hot enuff to cause instabilities.

My .02 cents
 


<< I have a sk6 on my XP2000+ and it runs @ 49C under full load. House temp. is 23C. I have never noticed any condensation on my cpu. I have removed and reapplied ASII many times and no change. Snatch, those foam pads cannot "lift" the heatsink up to cause the poor seating you had. I think you probably had the HSF installed wrong. >>


John............when did those become available?????? LOL! I can usually garner items early as possible, but, I've not seen those anywhere yet!:Q Apparently I need to talk to your supplier!😉 LOL!🙂
 
ToBeMe,

Your readings are clearly waay, way, way too low. 3C over ambient case temp is impossible. Is the kR7A thermistor in good contact with CPU back(not that it means anything, cause AthlonXP pcb further malfs up socket-thermistor readings)?

I would refrain from posting your temps, even for comparisons involving kr7a's until you figure out why you're under-reading temps that severely. It isn't even in the "normal" kr7a temp range.

CRV,

Are you basing that reading on socket-thermistor's too?



Mike
 
Mike, I too was wondering how those guys have such low cpu temps. I figured CRV was using a peltier or water bong, lol.
 


<< John............when did those become available?????? LOL! I can usually garner items early as possible, but, I've not seen those anywhere yet!:Q Apparently I need to talk to your supplier!😉 LOL!🙂 >>



Hey, it's a XP1800+ overclocked to 1667MHz, lol. 😉
 
SK-6 i found condensation between the h/s and CPU and all over the surface of the CPU

Umm, are you using a peltier? How are you condensing air with an air cooler?
 


<< ToBeMe,

Your readings are clearly waay, way, way too low. 3C over ambient case temp is impossible. Is the kR7A thermistor in good contact with CPU back(not that it means anything, cause AthlonXP pcb further malfs up socket-thermistor readings)?

I would refrain from posting your temps, even for comparisons involving kr7a's until you figure out why you're under-reading temps that severely. It isn't even in the "normal" kr7a temp range.

CRV,

Are you basing that reading on socket-thermistor's too?



Mike >>


Hmmm, don't know but the other two systems in that room have/are running in the same range.......a 1.2 T-Bird, and a 1.8 P4 all run within 3 - 4 degrees of case temps.. My kids' system in their room does not, nor does my wifes in the den, but, I have always attributed this to the room being in my basement and the temps being so constant as well as all three systems I keep in that room being in super ventilated full tower cases! The bios and MBM5 all read identical also......The other AMD system is built with a KG7-RAID and the P4 is on a TH7II!
 


<<

<< John............when did those become available?????? LOL! I can usually garner items early as possible, but, I've not seen those anywhere yet!:Q Apparently I need to talk to your supplier!😉 LOL!🙂 >>



Hey, it's a XP1800+ overclocked to 1667MHz, lol. 😉 >>


LOL John.......I just figured since it was you, you had found a way to get ahold of something "ahead of the pack"! LOL!🙂
 
If they are all within a few C of ambient case temp when under load, then something is wrong.

For example, a tested C/W of .32 for the 8045, and a 1900+ (which is rated at ~61 Typical thermal output):

Under load, this results in a temperature of: ~19C over Ambient Case Temp. Add onto this your ambient system temp of 28C, you've got a "die" temp that should be in the range of 48C. If you're getting readings of 32C, then something is clearly wrong. Something is interfering with the reading, be in that its just a socket-thermistor issue, or something else. Good Case ventilation can tamper with socket-thermistor readings, especially with fans directed at the socket-area.

FOr your T-bird, the same situation applies. Socket-thermistor = most inaccurate, POS form of temp measurement.

Your P4 should be on internal diode temps, so it should be running hotter as well...

Also, BIOS temps are neither full load nor idle.. if MBM readings are matching BIOS temps, then your windows isn't at full idle or load either.


Mike
 
Right now I am running ambient/24C and CPU/35.5 just finished a bunch of 3dmark2001. System overclocked to 140 FSB (AK31 mobo) and CPU to >1800.

To answer some questions:

Mem: I know the pads are there to protect the CPU...I think water is probably more dangerous than the extra pressure at this point.

neutralizer: I don't think more thermal paste is ever the right answer.

John: Dude - there's only one way to install it...trust me on this one. I even tried with the h/s and cpu uninstalled and pushing up on the back of the CPU you could see the edges bowing downward before the core contacted the h/s. That's what gave it away for me. Oh and 50C was at idle...crashes were while gaming (i never measured that temp.

merlocka: No peltier (I tried that once and had a puddle in my case). Its pretty cold out where I am now...cold air, hot CPU core = condensation.
 
I have a MSI K7T266 Pro2, an Athlon XP1600+, a Thermaltake Volcano 6cu (I only use the thermal pad that came with it) and my temps are : Processor 36C-42C, case 30C-33C. I use Pc Alert III from MSI for my readings, readings at idle from the BIOS and Pc Alert III are similar so I think it's reliable.
 


<< John: Dude - there's only one way to install it...trust me on this one. I even tried with the h/s and cpu uninstalled and pushing up on the back of the CPU you could see the edges bowing downward before the core contacted the h/s. >>



As I stated earlier, I have the SK6 as well, and yes you can install it incorrectly. Once I was in a hurry and the edge of the SK6 was resting on the raised portion of the ziff socket. It was difficult to see, but I gave it a once over before I dropped the mobo back in the case. You will never convince me that the 4 little rubber pads can keep the heatsink from making 100% contact with the core.
 


<< I have a MSI K7T266 Pro2, an Athlon XP1600+, a Thermaltake Volcano 6cu (I only use the thermal pad that came with it) and my temps are : Processor 36C-42C, case 30C-33C. I use Pc Alert III from MSI for my readings, readings at idle from the BIOS and Pc Alert III are similar so I think it's reliable. >>



How do you believe its reliable? No SOcket-thermistor is reliable enough for anything more thna a "fire alarm" and this fire alarm level is different for every single setup out there... there's not enough continuity or consistency ins ocket-thermistor readings.

for example, based on you rcase temp of 30C, your CPU "die" temp should be around 55C. In Which case, your socket-thermistor is severely under-reading temps as well.



Mike
 
cold air, hot CPU core = condensation

Actually, condensation occurs when warm (relatively) air (of a certain humidity level) comes in contact with a surface which cools it below it's dew point. The air can no longer hold the water, and condenses on the colder surface.

As air gets colder, it can hold less and less water vapor but it must still find a *colder* surface to condensate on.

So, a "properly installed" air cooled heatsink will probably never condense water out of air. I say probably because hell, I've seen stranger things...

On the other hand, when warm air (like the air from a very hot heatsink) rises and mixes with cold air (like the ambient air in your case, "Its pretty cold out where I am now") it starts to rain.

In your case (pardon the pun)... that's not condensation, it's RAINING in your case!!! I suggest Thermaltake's new CPU umbrella 🙂
 
Some of those temps seem way low... each motherboard can be different....


But here are mine:

22 room temp
25-27 case temp
42 cpu idle
47 cpu full load

SK-6 with a 80mm-60mm converter with a sunon fan


Kyle


 


<< Some of those temps seem way low... each motherboard can be different....


But here are mine:

22 room temp
25-27 case temp
42 cpu idle
47 cpu full load

SK-6 with a 80mm-60mm converter with a sunon fan


Kyle >>




Not Only is each MB different, but each heatsink has a different effect on the socket-thermistor, too....



Mike
 


<< I suggest Thermaltake's new CPU umbrella >>



Ok...that gets an LOL. Yah...right after I wrote that I realized the same thing...doesn't make sense but dag-naggit...there was water there! Look, I didn't mean to start a controversy here...just cluing people in on a potential problem they might not have thought of.

And John...well maybe you can install the sink wrong but still, as I mentioned, outside the case and uninstalled I could push up underneath the core and see the corners of the CPU bowing downward from where the pads were contacting the sink and it never made contact with the surface of the sink (i had paste on the core so i could tell). You have to admit that is a pretty impressive finding. Inside the case with the CPU in the socket it can't bow like that at the edges, making contact between the surfaces even less likely without a lot of force....and as you know the SK-6 retention mechanism is pretty damn sturdy. Maybe this is not the case for all Athlons...maybe there are some out there with really stiff pads on them...i don't know. I do know that i shaved the pads down a little, installed my h/s the exact same way, and now I don't have a problem.

Just trying to help because, for a while, i had no idea why my temps were so high.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top