AMD X2 Stock Cooler not doing the job

While answering another poster's question about temps, I noticed mine
have gotten too high. I'm running a socket 939 X2 3800+ (now at stock
settings) that idles near 50C. Ambient room temp right now is 81F/27C,
so I expect idle temps to be slightly elevated, but 50C seems a bit much
and full load temps are in the upper 60s.

System info:
Antec case w/ 120mm exhaust fan on low setting. The case includes an
air duct which does a fair job of ducting fresh exterior air onto the
CPU. Stock AMD cooler. Three IDE hard disks, one optical drive,
Geforce3 Ti200 AGP graphics. The system lays on its side. See the case
here:
http://www.newegg.com/Product/Produc...82E16811129152

My first thought was airflow. I've done a decent job of routing
cabling, so the interior is fairly open. I put the 120mm exhaust fan on
full speed with no real change in idle temps. The case sits in the
corner of the room, so I thought maybe the ambient airflow was the
problem. I set up a regular house fan near the PC to increase the
amount of cool, fresh air moving around and again, no real difference.
Thinking that maybe the single exhaust fan wasnt enough, I pulled the
side (top) off the case, leaving the house fan running to keep fresh air
flowing through the area. Still, idle temps are steady around 49C and
load temps are 67C. It appears that airflow is not causing the problem.

My next step is to pull the HSF and reapply the AS Ceramique between it
and the CPU. I just did this recently when I removed and cleaned the
heatsink, so I doubt that will solve the problem. In the meantime I'm
thinking a new heatsink might be in order and I'd like to know what
models are not only effective, but also reasonalby priced and quiet.

Thanks in advance for your input.

-Dylan C

Dylan C wrote:
> While answering another poster's question about temps, I noticed mine
> have gotten too high. I'm running a socket 939 X2 3800+ (now at stock
> settings) that idles near 50C. Ambient room temp right now is 81F/27C,
> so I expect idle temps to be slightly elevated, but 50C seems a bit much
> and full load temps are in the upper 60s.
>
> System info:
> Antec case w/ 120mm exhaust fan on low setting. The case includes an
> air duct which does a fair job of ducting fresh exterior air onto the
> CPU. Stock AMD cooler. Three IDE hard disks, one optical drive,
> Geforce3 Ti200 AGP graphics. The system lays on its side. See the case
> here:
> http://www.newegg.com/Product/Produc...82E16811129152
>
> My first thought was airflow. I've done a decent job of routing
> cabling, so the interior is fairly open. I put the 120mm exhaust fan on
> full speed with no real change in idle temps. The case sits in the
> corner of the room, so I thought maybe the ambient airflow was the
> problem. I set up a regular house fan near the PC to increase the
> amount of cool, fresh air moving around and again, no real difference.
> Thinking that maybe the single exhaust fan wasnt enough, I pulled the
> side (top) off the case, leaving the house fan running to keep fresh air
> flowing through the area. Still, idle temps are steady around 49C and
> load temps are 67C. It appears that airflow is not causing the problem.
>
> My next step is to pull the HSF and reapply the AS Ceramique between it
> and the CPU. I just did this recently when I removed and cleaned the
> heatsink, so I doubt that will solve the problem. In the meantime I'm
> thinking a new heatsink might be in order and I'd like to know what
> models are not only effective, but also reasonalby priced and quiet.
>
> Thanks in advance for your input.
>
> -Dylan C

Pulled the HSF and the thermal compund had dried so much it pulled the
CPU right out of the socket with it. Luckily, no pins were bent and it
was a simple matter of separating the two and cleaning them up. At that
point, I decided to test a theory of another regular in this group and
use some white lithium grease from NAPA instead of my AS Ceramique.

http://tinyurl.com/lgbql

Idle temps dropped to 38C and load temps to 61C. Thats with the side on
and the 120mm fan at its middle setting. I'll let the grease burn in
for a few days before I make a final judgement on its effectiveness, but
I think its safe to say spending $5-$6 on 3 grams of thermal compound is
basically a waste of money.

My load temps of 61C are still a bit high. I run SETI@Home which keeps
my CPU at a steady 50% load and a current temp of 55C. I'd like to get
that under 50. If a new heatsink could do that for under $30, that'd be
great.

-Dylan C

Looks like we're on the same boat, Dylan (now I don't feel so bad about
my cooler)
>From an engineering standpoint, I think first you should determine how
efficient the heat transfer to the heat sink is. Measure how quickly
the CPU core temp returns to idle levels from max load, i.e....In my
case the temperature drops 10 degrees in 2 seconds...that's pretty damn
fast and it tells me that I've optimum heat transfer between the CPU
and the heatsink. If that's the case and the delta (T) idle/max load is
more than 30 C, then it just means that the CPU heat output is
exceeding the design cooling capacity for the particular heatsink.
Personally, I'm beginning to think that high 30's to high 50s/low60s is
acceptable for AMD X2's...Your lithium grease comment is
interesting...given that lithium is added to the base lubricate oil
just as a thickening agent to make it into a grease (to increase
resistance to shear forces in bearings, etc.) and not as a heat
transfer additive.

Cheers,

JC

Dylan C wrote:
> Dylan C wrote:
> > While answering another poster's question about temps, I noticed mine
> > have gotten too high. I'm running a socket 939 X2 3800+ (now at stock
> > settings) that idles near 50C. Ambient room temp right now is 81F/27C,
> > so I expect idle temps to be slightly elevated, but 50C seems a bit much
> > and full load temps are in the upper 60s.
> >
> > System info:
> > Antec case w/ 120mm exhaust fan on low setting. The case includes an
> > air duct which does a fair job of ducting fresh exterior air onto the
> > CPU. Stock AMD cooler. Three IDE hard disks, one optical drive,
> > Geforce3 Ti200 AGP graphics. The system lays on its side. See the case
> > here:
> > http://www.newegg.com/Product/Produc...82E16811129152
> >
> > My first thought was airflow. I've done a decent job of routing
> > cabling, so the interior is fairly open. I put the 120mm exhaust fan on
> > full speed with no real change in idle temps. The case sits in the
> > corner of the room, so I thought maybe the ambient airflow was the
> > problem. I set up a regular house fan near the PC to increase the
> > amount of cool, fresh air moving around and again, no real difference.
> > Thinking that maybe the single exhaust fan wasnt enough, I pulled the
> > side (top) off the case, leaving the house fan running to keep fresh air
> > flowing through the area. Still, idle temps are steady around 49C and
> > load temps are 67C. It appears that airflow is not causing the problem.
> >
> > My next step is to pull the HSF and reapply the AS Ceramique between it
> > and the CPU. I just did this recently when I removed and cleaned the
> > heatsink, so I doubt that will solve the problem. In the meantime I'm
> > thinking a new heatsink might be in order and I'd like to know what
> > models are not only effective, but also reasonalby priced and quiet.
> >
> > Thanks in advance for your input.
> >
> > -Dylan C
>
> Pulled the HSF and the thermal compund had dried so much it pulled the
> CPU right out of the socket with it. Luckily, no pins were bent and it
> was a simple matter of separating the two and cleaning them up. At that
> point, I decided to test a theory of another regular in this group and
> use some white lithium grease from NAPA instead of my AS Ceramique.
>
> http://tinyurl.com/lgbql
>
> Idle temps dropped to 38C and load temps to 61C. Thats with the side on
> and the 120mm fan at its middle setting. I'll let the grease burn in
> for a few days before I make a final judgement on its effectiveness, but
> I think its safe to say spending $5-$6 on 3 grams of thermal compound is
> basically a waste of money.
>
> My load temps of 61C are still a bit high. I run SETI@Home which keeps
> my CPU at a steady 50% load and a current temp of 55C. I'd like to get
> that under 50. If a new heatsink could do that for under $30, that'd be
> great.
>
> -Dylan C

On Sat, 07 Oct 2006 21:11:16 -0500, Dylan C wrote:

> Pulled the HSF and the thermal compund had dried so much it pulled the
> CPU right out of the socket with it. Luckily, no pins were bent and it
> was a simple matter of separating the two and cleaning them up. At that
> point, I decided to test a theory of another regular in this group and
> use some white lithium grease from NAPA instead of my AS Ceramique.
>
> http://tinyurl.com/lgbql
>
Didn't see any temp specs for it. Their wheel bearing grease (what i used)
is rated for 500F. Careful what you bought doesn't dry out on you or run
when it gets hot.

> Idle temps dropped to 38C and load temps to 61C. Thats with the side on
> and the 120mm fan at its middle setting. I'll let the grease burn in
> for a few days before I make a final judgement on its effectiveness, but
> I think its safe to say spending $5-$6 on 3 grams of thermal compound is
> basically a waste of money.
>
That's been proven many times over.:-)

> My load temps of 61C are still a bit high. I run SETI@Home which keeps
> my CPU at a steady 50% load and a current temp of 55C. I'd like to get
> that under 50. If a new heatsink could do that for under $30, that'd be
> great.
>
I'm using the stock cooler with my AM2 X2 3800+ and it tops out about 50
something running both cores at 100%, and about 42C running 1 at 100%. But
most of time mine is using less than 5% even with the cores running at
half speed. Yeah, a waste, but I wanted dual core and at least there's
plenty of reserve power when I need it. I also disabled smartfan an the
little fan runs at 3000rpm constant. One of these days I'll put a fan
adapter on it with an 80x25mm fan. that alone should help the cooling.

--
Want the ultimate in free OTA SD/HDTV Recorder? http://mythtv.org
http://mysettopbox.tv/knoppmyth.html Usenet alt.video.ptv.mythtv
My server http://wesnewell.no-ip.com/cpu.php
HD Tivo S3 compared http://wesnewell.no-ip.com/mythtivo.htm

Quite a good cooler is the Arctic Cooling, Freezer64 pro (Ithink it is). Got
it running on my x2 3800 and idles at ~25C, loads at ~45C. I have no
problems with it at all. Easy to install and no looking after.



"Dylan C" <> wrote in message
news:...
While answering another poster's question about temps, I noticed mine
have gotten too high. I'm running a socket 939 X2 3800+ (now at stock
settings) that idles near 50C. Ambient room temp right now is 81F/27C,
so I expect idle temps to be slightly elevated, but 50C seems a bit much
and full load temps are in the upper 60s.

System info:
Antec case w/ 120mm exhaust fan on low setting. The case includes an
air duct which does a fair job of ducting fresh exterior air onto the
CPU. Stock AMD cooler. Three IDE hard disks, one optical drive,
Geforce3 Ti200 AGP graphics. The system lays on its side. See the case
here:
http://www.newegg.com/Product/Produc...82E16811129152

My first thought was airflow. I've done a decent job of routing
cabling, so the interior is fairly open. I put the 120mm exhaust fan on
full speed with no real change in idle temps. The case sits in the
corner of the room, so I thought maybe the ambient airflow was the
problem. I set up a regular house fan near the PC to increase the
amount of cool, fresh air moving around and again, no real difference.
Thinking that maybe the single exhaust fan wasnt enough, I pulled the
side (top) off the case, leaving the house fan running to keep fresh air
flowing through the area. Still, idle temps are steady around 49C and
load temps are 67C. It appears that airflow is not causing the problem.

My next step is to pull the HSF and reapply the AS Ceramique between it
and the CPU. I just did this recently when I removed and cleaned the
heatsink, so I doubt that will solve the problem. In the meantime I'm
thinking a new heatsink might be in order and I'd like to know what
models are not only effective, but also reasonalby priced and quiet.

Thanks in advance for your input.

-Dylan C

wrote:
> Looks like we're on the same boat, Dylan (now I don't feel so bad about
> my cooler)
>>From an engineering standpoint, I think first you should determine how
> efficient the heat transfer to the heat sink is. Measure how quickly
> the CPU core temp returns to idle levels from max load, i.e....In my
> case the temperature drops 10 degrees in 2 seconds...that's pretty damn
> fast and it tells me that I've optimum heat transfer between the CPU
> and the heatsink.
Even before I put the new thermal grease on my temps would drop from
~68C to ~55C instantly. That would indicate decent thermal transfer, as
you point out. It was my high idle temps (20C over ambient) that really
signaled a problem. That indicated very poor thermal transfer. Since
your idle temps are reasonable, I'd assume thermal grease isn't a
problem in your system.

> If that's the case and the delta (T) idle/max load is
> more than 30 C, then it just means that the CPU heat output is
> exceeding the design cooling capacity for the particular heatsink.
> Personally, I'm beginning to think that high 30's to high 50s/low60s is
> acceptable for AMD X2's...Your lithium grease comment is
> interesting...given that lithium is added to the base lubricate oil
> just as a thickening agent to make it into a grease (to increase
> resistance to shear forces in bearings, etc.) and not as a heat
> transfer additive.
>
> Cheers,
>
> JC
Its my understanding that lithium increases the working temperature of
the grease. I did a little more research (thanks Wes) and found out
that my grease has a max lube temp of 170C. The point of using lithium
grease wasn't to test the thermal conuctivity of the lithium, but more
to demonstrate that "heat transfer additives" like micronized silver
only add to the cost of thermal grease, and not its effectiveness.
Grease, in this case lithium, already conucts heat well enough.

-Dylan C


>
> Dylan C wrote:
>
>>Dylan C wrote:
>>
>>>While answering another poster's question about temps, I noticed mine
>>>have gotten too high. I'm running a socket 939 X2 3800+ (now at stock
>>>settings) that idles near 50C. Ambient room temp right now is 81F/27C,
>>>so I expect idle temps to be slightly elevated, but 50C seems a bit much
>>>and full load temps are in the upper 60s.
>>>
>>>System info:
>>>Antec case w/ 120mm exhaust fan on low setting. The case includes an
>>>air duct which does a fair job of ducting fresh exterior air onto the
>>>CPU. Stock AMD cooler. Three IDE hard disks, one optical drive,
>>>Geforce3 Ti200 AGP graphics. The system lays on its side. See the case
>>>here:
>>>http://www.newegg.com/Product/Produc...82E16811129152
>>>
>>>My first thought was airflow. I've done a decent job of routing
>>>cabling, so the interior is fairly open. I put the 120mm exhaust fan on
>>>full speed with no real change in idle temps. The case sits in the
>>>corner of the room, so I thought maybe the ambient airflow was the
>>>problem. I set up a regular house fan near the PC to increase the
>>>amount of cool, fresh air moving around and again, no real difference.
>>>Thinking that maybe the single exhaust fan wasnt enough, I pulled the
>>>side (top) off the case, leaving the house fan running to keep fresh air
>>>flowing through the area. Still, idle temps are steady around 49C and
>>>load temps are 67C. It appears that airflow is not causing the problem.
>>>
>>>My next step is to pull the HSF and reapply the AS Ceramique between it
>>>and the CPU. I just did this recently when I removed and cleaned the
>>>heatsink, so I doubt that will solve the problem. In the meantime I'm
>>>thinking a new heatsink might be in order and I'd like to know what
>>>models are not only effective, but also reasonalby priced and quiet.
>>>
>>>Thanks in advance for your input.
>>>
>>>-Dylan C
>>
>>Pulled the HSF and the thermal compund had dried so much it pulled the
>>CPU right out of the socket with it. Luckily, no pins were bent and it
>>was a simple matter of separating the two and cleaning them up. At that
>>point, I decided to test a theory of another regular in this group and
>>use some white lithium grease from NAPA instead of my AS Ceramique.
>>
>>http://tinyurl.com/lgbql
>>
>>Idle temps dropped to 38C and load temps to 61C. Thats with the side on
>>and the 120mm fan at its middle setting. I'll let the grease burn in
>>for a few days before I make a final judgement on its effectiveness, but
>>I think its safe to say spending $5-$6 on 3 grams of thermal compound is
>>basically a waste of money.
>>
>>My load temps of 61C are still a bit high. I run SETI@Home which keeps
>>my CPU at a steady 50% load and a current temp of 55C. I'd like to get
>>that under 50. If a new heatsink could do that for under $30, that'd be
>>great.
>>
>>-Dylan C
>
>

Wes Newell wrote:
> On Sat, 07 Oct 2006 21:11:16 -0500, Dylan C wrote:
>
>
>>Pulled the HSF and the thermal compund had dried so much it pulled the
>>CPU right out of the socket with it. Luckily, no pins were bent and it
>>was a simple matter of separating the two and cleaning them up. At that
>>point, I decided to test a theory of another regular in this group and
>>use some white lithium grease from NAPA instead of my AS Ceramique.
>>
>>http://tinyurl.com/lgbql
>>
>
> Didn't see any temp specs for it. Their wheel bearing grease (what i used)
> is rated for 500F. Careful what you bought doesn't dry out on you or run
> when it gets hot.
>
>
Did some more digging and found max lube temp for this grease is
335F/170C and drip temp is 360F/180C. Thanks for the tip.

-Dylan C

>>Idle temps dropped to 38C and load temps to 61C. Thats with the side on
>>and the 120mm fan at its middle setting. I'll let the grease burn in
>>for a few days before I make a final judgement on its effectiveness, but
>>I think its safe to say spending $5-$6 on 3 grams of thermal compound is
>>basically a waste of money.
>>
>
> That's been proven many times over.:-)
>
>
>>My load temps of 61C are still a bit high. I run SETI@Home which keeps
>>my CPU at a steady 50% load and a current temp of 55C. I'd like to get
>>that under 50. If a new heatsink could do that for under $30, that'd be
>>great.
>>
>
> I'm using the stock cooler with my AM2 X2 3800+ and it tops out about 50
> something running both cores at 100%, and about 42C running 1 at 100%. But
> most of time mine is using less than 5% even with the cores running at
> half speed. Yeah, a waste, but I wanted dual core and at least there's
> plenty of reserve power when I need it. I also disabled smartfan an the
> little fan runs at 3000rpm constant. One of these days I'll put a fan
> adapter on it with an 80x25mm fan. that alone should help the cooling.
>

"Dylan C" <> wrote in message
news:...

> Even before I put the new thermal grease on my temps would drop from ~68C
> to ~55C instantly. That would indicate decent thermal transfer, as you
> point out. It was my high idle temps (20C over ambient) that really
> signaled a problem. That indicated very poor thermal transfer. Since
> your idle temps are reasonable, I'd assume thermal grease isn't a problem
> in your system.
>
Well let's see...we can't have it both ways, can we? You have good
thermal transfer at 68 C.
On the other hand you have elevated temperatures at idle...You say that
this "indicated very poor thermal transfer"
This is how I see it: Your temperature drops from 68 to 55 fast enough
to tell me the interface between the heatsink and
the CPU is ok, what's happening is that the heat stored in the cooler
isn't being transported off the heatsink fast enough.
If you say this is a recent condition that would mean one or more of
several things:
1. The CPU is producing more heat (?)
2. The case ambient temperature is higher (explaining why at higher heat
loads -higher delta (T)- the heat transfer is more efficient)
3. The fan is no longer pulling the heat off as eficiently (bad fan/fan
controller or encrusted/dusty heatsink)
4. (related to 2.) Air circulation inside the case has changed or ambient
(room) temperature has increased.

In my case I think it means that the CPU cooler I have is barely adequate,
and certain thermal stress conditions will
outstrip its thermal capacity to cool this class of CPU. That's all.

BTW I'm looking at the Thermaltake Venus 12; looks very powerful for the
price and AMD approved.

> Its my understanding that lithium increases the working temperature of the
> grease. I did a little more research (thanks Wes) and found out that my
> grease has a max lube temp of 170C. The point of using lithium grease
> wasn't to test the thermal conuctivity of the lithium, but more to
> demonstrate that "heat transfer additives" like micronized silver only add
> to the cost of thermal grease, and not its effectiveness. Grease, in this
> case lithium, already conucts heat well enough.
>
> -Dylan C
>

I suppose before the days of Arctic Silver...thermal grease had no other
heat transfer additives?
Good question...I don't know. But I think performance-wise, micronized
silver suspended in
modern thermal grease does increse the efficiency in the grease, no doubt
about it. Whether
there's price-gouging going on is another matter.

Cheers,

JC

On Sun, 08 Oct 2006 12:53:49 -0700, Alfredo Pantani wrote:

> I suppose before the days of Arctic Silver...thermal grease had no other
> heat transfer additives?
> Good question...I don't know. But I think performance-wise, micronized
> silver suspended in
> modern thermal grease does increse the efficiency in the grease, no doubt
> about it. Whether
> there's price-gouging going on is another matter.

It may well improve the performance. But the margin of improvement is so
small it makes very little to no noticable difference. The actual main
ingredient has a much greater affect. But that has to be averaged out to
include prolonged performance. The A64 3000+ I used wheel bearing grease
on is still working fine after 13+ months. based on the properties of the
grease it should virtually last 50+ years. it was 30 years old when I used
it. It had been open to the air in the gargage for about 20 of those years
where temps varied from about -15C to over 60C each year. Wheel bearing
grease has been around for maybe 200 years now and was specically designed
for harse high temp environments. I'd use it over AS any day. And if
someone wants to send me some AS to compare, I'll do it, but I'm certainly
not going to buy any to do it.:-)

http://www.dansdata.com/goop.htm

--
Want the ultimate in free OTA SD/HDTV Recorder? http://mythtv.org
http://mysettopbox.tv/knoppmyth.html Usenet alt.video.ptv.mythtv
My server http://wesnewell.no-ip.com/cpu.php
HD Tivo S3 compared http://wesnewell.no-ip.com/mythtivo.htm

Alfredo Pantani wrote:
> "Dylan C" <> wrote in message
> news:...
>
>
>>Even before I put the new thermal grease on my temps would drop from ~68C
>>to ~55C instantly. That would indicate decent thermal transfer, as you
>>point out. It was my high idle temps (20C over ambient) that really
>>signaled a problem. That indicated very poor thermal transfer. Since
>>your idle temps are reasonable, I'd assume thermal grease isn't a problem
>>in your system.
>>
>
> Well let's see...we can't have it both ways, can we? You have good
> thermal transfer at 68 C.
> On the other hand you have elevated temperatures at idle...You say that
> this "indicated very poor thermal transfer"
> This is how I see it: Your temperature drops from 68 to 55 fast enough
> to tell me the interface between the heatsink and
> the CPU is ok, what's happening is that the heat stored in the cooler
> isn't being transported off the heatsink fast enough.
> If you say this is a recent condition that would mean one or more of
> several things:
> 1. The CPU is producing more heat (?)
> 2. The case ambient temperature is higher (explaining why at higher heat
> loads -higher delta (T)- the heat transfer is more efficient)
> 3. The fan is no longer pulling the heat off as eficiently (bad fan/fan
> controller or encrusted/dusty heatsink)
> 4. (related to 2.) Air circulation inside the case has changed or ambient
> (room) temperature has increased.
>
Dried out thermal compound was the culprit. That suprised me becuase I
had just reapplied it a few weeks ago. I guess I either did a poor job
or the AS Ceramique I bought almost three years ago has exceeded its
shelf life.

-Dylan C

> In my case I think it means that the CPU cooler I have is barely adequate,
> and certain thermal stress conditions will
> outstrip its thermal capacity to cool this class of CPU. That's all.
>
> BTW I'm looking at the Thermaltake Venus 12; looks very powerful for the
> price and AMD approved.
>
>
>>Its my understanding that lithium increases the working temperature of the
>>grease. I did a little more research (thanks Wes) and found out that my
>>grease has a max lube temp of 170C. The point of using lithium grease
>>wasn't to test the thermal conuctivity of the lithium, but more to
>>demonstrate that "heat transfer additives" like micronized silver only add
>>to the cost of thermal grease, and not its effectiveness. Grease, in this
>>case lithium, already conucts heat well enough.
>>
>>-Dylan C
>>
>
>
> I suppose before the days of Arctic Silver...thermal grease had no other
> heat transfer additives?
> Good question...I don't know. But I think performance-wise, micronized
> silver suspended in
> modern thermal grease does increse the efficiency in the grease, no doubt
> about it. Whether
> there's price-gouging going on is another matter.
>
> Cheers,
>
> JC
>
>