Air Cooling at its FINEST!

R

r4d

Weaksauce
Joined
Feb 24, 2004
Messages
70
Hey all, haven't posted on the forums in a long time, i haven't dont much with my computer in a while.

Anyways, I just overclocked my video card a bit so i could get a little more out of a few games and needed something to cool it.
I had no casefans lying around for my window, but i did find this, lol.

Full Load = 15ºC
fan.jpg


fan2.jpg


it sits at a chilly 10ºC idle, and its not very loud, probably just as noisy as a case with 3 fans.

PS: I'll be getting my new lian-li and psu thursday June 24, time to completely redo my computer, stay tuned ladies!
 
Good job! LOL :p

BTW what the hell temperature is your room at?
 
umm.. it must be like 50* F in your room o_0

theres no way in hell you are at normal room temp.
 
although i'm in a basement, its not that cold down here, maybe ~60/70ºF
 
So what you are saying is that under full load your temps stay at the exact temperature of your room if it's 60

and somehow if it's 70 you are able to be 6 degreese celsius below room temp under full load.
 
matoch said:
So what you are saying is that under full load your temps stay at the exact temperature of your room if it's 60

and somehow if it's 70 you are able to be 6 degreese celsius below room temp under full load.
Click to expand...

my thoughts exactly
 
Obviously the sensor cannot accurately guage temperature if air is being blown by it at 100 CFM...
 
Matrices said:
Obviously the sensor cannot accurately guage temperature if air is being blown by it at 100 CFM...
Click to expand...

the sensor is on the chip, it doesnt care what the air is


also i call bullshit, somethings wrong. theres no way that your CPU is that cold. NO WAY.
 
Maybe he puts ice cubes in front of the fan :eek:

And yea, maybe you should reconsider the fact that your cpu is running BELOW room temp, a physical impossibility with air cooling.
 
the sensor is on the chip, it doesnt care what the air is
Click to expand...
Not necessarily true. Only the intels have on-die sensors, the temp readings for AMDs come from a thermal probe located in the socket. They can be highly inaccurate, as this measures the air temps inside the socket, and then uses a formula to give the estimated processor temps. This is why in some cases people claim to see higher temps after switching to watercooling, because there is no air circulation inside the socket. Now if he has an Intel or a thermistor properly installed in there, it may be different....but until that is proven, I join the call of bullshit.
 
server_error said:
Not necessarily true. Only the intels have on-die sensors, the temp readings for AMDs come from a thermal probe located in the socket. They can be highly inaccurate, as this measures the air temps inside the socket, and then uses a formula to give the estimated processor temps. This is why in some cases people claim to see higher temps after switching to watercooling, because there is no air circulation inside the socket. Now if he has an Intel or a thermistor properly installed in there, it may be different....but until that is proven, I join the call of bullshit.
Click to expand...

um excuse me? AMD's have on-chip sensors also, and even if they didnt, the socket diode is covered by a processor and surrounded by a processor. that would have no affect whatsoever, the space from the board to the socket is insignificant
 
reconsider the fact that your cpu is running BELOW room temp, a physical impossibility with air cooling
Click to expand...
..And, I recall reading that the on-chip sensors on AMD's (XP's @ least - dunno about A64s) only have todo with the critical temperature shutdown mode thingy?
 
kush said:
..And, I recall reading that the on-chip sensors on AMD's (XP's @ least - dunno about A64s) only have todo with the critical temperature shutdown mode thingy?
Click to expand...

AMD's have no critical shutdown temp onchip, its all handled by the mobo. the sensor on AMD chips is as accurate as they can be...it can be off as many as 5C but its not unrealistic.
 
If I do that to my case my temps go up. But yeah it's makes a lot more sense to spend $10 on a box fan rather than $10 for a 120MM fan.
 
Major_A said:
...it's makes a lot more sense to spend $10 on a box fan rather than $10 for a 120MM fan.
Click to expand...

Unless you care about noise, because at full speed box fans are very loud (compared to case fans, anyway). To me it makes more sense to buy 10 80mm fans from the Egg and not worry about knocking my cooling over. Plus you can 7-volt them and it's nice and quiet.

-SEAL
 
kronchev said:
the sensor is on the chip, it doesnt care what the air is
Click to expand...

But if it's an AMD XP, the temperature readings aren't taken from the chip. They're taken from an external thermistor. The internal diode is only used for the overtemp shutdown.

At least that was what I had gathered from a post by an EPoX representative on www.amdmb.com some time ago. From what I remember, it was possible to read the on-die temperature, but it would cause the overtemp shutdown to no longer work.
 
its all to do with the pressure of the air flowing into the case.

That fan, its what, something like a 1800mm fan? and it spins at say about 5000 RPM?

Lol. and I've now discovered my maths has failed me otherwise I could have explained this.

bet it pushes something like a good 1000 or so cubic METRES a minute, so it beats all your crap HSF fans neday. and that 1000 CMM is at high pressure, leading to high turbulance within the machine, which cools things faster.

And anyway, ever stuck you head out a car window while its driving along fast? Felt cold didnt it? definately did not feel like the ambient temperature of the country around you.

Finally, I dont this myself when my HSF failed, and got a nice chilly 5 degrees centrigrade or so. Loud but it does the job till a new hsf arrives.

There. Thats that settled then. :p
 
server_error said:
Not necessarily true. Only the intels have on-die sensors, the temp readings for AMDs come from a thermal probe located in the socket. They can be highly inaccurate, as this measures the air temps inside the socket, and then uses a formula to give the estimated processor temps. This is why in some cases people claim to see higher temps after switching to watercooling, because there is no air circulation inside the socket. Now if he has an Intel or a thermistor properly installed in there, it may be different....but until that is proven, I join the call of bullshit.
Click to expand...

*sigh* idiot. People like you go around attempting to talk intelligently about comptuers to your friends. They all think you are the "guru" don't they?

My xp2500 has an "on-die" temp sensor, and my motherboard also has one in the socket. I can read both, however the diode is much more accurate.

Do us a favor, stop propogating BS information.
 
Le_Petit_Lapin said:
its all to do with the pressure of the air flowing into the case.

That fan, its what, something like a 1800mm fan? and it spins at say about 5000 RPM?

Lol. and I've now discovered my maths has failed me otherwise I could have explained this.

bet it pushes something like a good 1000 or so cubic METRES a minute, so it beats all your crap HSF fans neday. and that 1000 CMM is at high pressure, leading to high turbulance within the machine, which cools things faster.
Click to expand...

But not below ambient...unless he's getting supersonic flow through the fan...which he isn't.

And anyway, ever stuck you head out a car window while its driving along fast? Felt cold didnt it? definately did not feel like the ambient temperature of the country around you.
Click to expand...

Yup, that's due to the effect commonly known as "wind chill". Along with the convective cooling of the air passing by your head, moisture on the surface of your skin is also evaporating, so your skin drops below ambient due to latent heat of vaporization.
 
orkan said:
*sigh* idiot. People like you go around attempting to talk intelligently about comptuers to your friends. They all think you are the "guru" don't they?

My xp2500 has an "on-die" temp sensor, and my motherboard also has one in the socket. I can read both, however the diode is much more accurate.

Do us a favor, stop propogating BS information.
Click to expand...

*sigh* flamer

no need to be a fucktard.

at any rate, I , too, join the call of Bullshit...unless, he has moisture in his case, and that's being caused to evaporate?

but that's such an unlikely explanation...
 
lol bs. 12C, 15C load? it would be impossible to cool down the aluminium or cooper that low with air even if the cpu wasn't operating.. unless ambient temp is <12C.
 
Parja said:
But if it's an AMD XP, the temperature readings aren't taken from the chip. They're taken from an external thermistor. The internal diode is only used for the overtemp shutdown.

At least that was what I had gathered from a post by an EPoX representative on www.amdmb.com some time ago. From what I remember, it was possible to read the on-die temperature, but it would cause the overtemp shutdown to no longer work.
Click to expand...

NO NO NO. AMD's have an ondie sensor, that means its INSIDE THE DIE OF THE CPU, now its inaccurate because its not calibrated but its accurate enough for our purposes. The boards also have in socket sensors and ONLY use them when a chip without a sensor is put into the socket. When a chip WITH a sensor is put it, it knows to switch the sensing to the on chip sensor.
 
Le_Petit_Lapin said:
its all to do with the pressure of the air flowing into the case.

That fan, its what, something like a 1800mm fan? and it spins at say about 5000 RPM?

Lol. and I've now discovered my maths has failed me otherwise I could have explained this.

bet it pushes something like a good 1000 or so cubic METRES a minute, so it beats all your crap HSF fans neday. and that 1000 CMM is at high pressure, leading to high turbulance within the machine, which cools things faster.

And anyway, ever stuck you head out a car window while its driving along fast? Felt cold didnt it? definately did not feel like the ambient temperature of the country around you.

Finally, I dont this myself when my HSF failed, and got a nice chilly 5 degrees centrigrade or so. Loud but it does the job till a new hsf arrives.

There. Thats that settled then. :p
Click to expand...

"Finally, I dont this myself when my HSF failed, and got a nice chilly 5 degrees centrigrade or so. Loud but it does the job till a new hsf arrives. "

No you didnt. I bet you all the money I will ever make, you did not get 5C.

And I also bet you dont understand biology. As someone said its evaporation, but also your skin has a natural climate around it thats about a mm or so thick, basically it has its own little cloud of heat floating on your skin all the time. wind chill rips that off.

and a fan does NOT push 1000 CMM a minute, and I dont believe its considered high pressure at all.
 
Damn it i'm going to have to try it for myself ...lol

I'm thinking BS though.
 
stumpy said:
Damn it i'm going to have to try it for myself ...lol

I'm thinking BS though.
Click to expand...

I will slaughter you and your children.

There's no way a box fan could really be that quiet. And I won't stand for a rumbling box fan directly over my room. Damn frogs are enough.
 
air cooling at its finest? i was expecting something original :rolleyes:

i think everyone has thought of this before lol
 
Le_Petit_Lapin said:
And anyway, ever stuck you head out a car window while its driving along fast? Felt cold didnt it? definately did not feel like the ambient temperature of the country around you.
Click to expand...

Thats called the wind chill factor, it's not that you are actually getting any colder than the ambient tempature, it is that the air is cooling your head to ambient much more efficiently, a thermometer left outside would read the same as a thermometer going 50 mph outside(discounting air friction which would make it warmer!) It's the same reason people have high cfm fans on their heatsinks.
 
kronchev said:
NO NO NO. AMD's have an ondie sensor, that means its INSIDE THE DIE OF THE CPU, now its inaccurate because its not calibrated but its accurate enough for our purposes. The boards also have in socket sensors and ONLY use them when a chip without a sensor is put into the socket. When a chip WITH a sensor is put it, it knows to switch the sensing to the on chip sensor.
Click to expand...

I'm sure you probably just worded that wrong, but if you say AMD's have ondie sensors, yet your board has an "in socket sensor" to use ONLY when a chip without a sensor is in use. What other chips does your board accept? I got curious after reading your post. Thanks.
 
lol, I tried this before, my comp usually sits at 28C idle, when i put the box fan on it it ran 20C idle, but i didnt really do anymore testing than that, its loud and stupid imo.

Its 70F in my house almost all the time due to nice air conditioning :p (btw 60F is really cold from air conditioners, u must be freezing ur balls off).

Thats what the sensors showed me in mbm5, and thats what I'm going off of.
 
orkan said:
*sigh* idiot. People like you go around attempting to talk intelligently about comptuers to your friends. They all think you are the "guru" don't they?.
Click to expand...

Carnival Forces said:
*sigh* flamer
no need to be a fucktard.
Click to expand...


1st off I know Carnival Forces knows better,
and orkan if you dont you do now

there will be no more flaming

I am sadly disappointed at your lack of creativity :rolleyes:
whatever happen to the fine art of sarcasm?
the withering deluge of witty commentary? the rapier wit?
the rhetorical question? :p
are the finer points of expressing your distain unknown to you?
or are you just too lazy to compose them?
Resorting to base insults, is the refuge of the incompetent
back to school for both of you :p

r4d just venturing a guess here, but I doubt that motor has much in the way of sheilding, not to mention if its actually in contact with the case vibration is going to be a big issue, when I first joined here I ducted a 3"x3" squirrel cage blower to a square duct directly to the heatsink and then fed it AC at the intake, that was in a sheilded metal case and with a much smaller AC motor and Im still pretty sure it did in my mobo
(got fantastic temps but that didnt do me much good with a dead mobo)

that thing is washing the interior with some pretty serious EM fields that will eventually at the very least bork your data transfers through any unshielded cables given enough time

and since several members (but not all) seem to be thermodynamically challenged
General Heat Transfer @ amdmb
advanced studies A Heat Transfer Textbook Lienhard & Lienhard (MIT & U of Houston) PDF
 
kronchev said:
NO NO NO. AMD's have an ondie sensor, that means its INSIDE THE DIE OF THE CPU, now its inaccurate because its not calibrated but its accurate enough for our purposes. The boards also have in socket sensors and ONLY use them when a chip without a sensor is put into the socket. When a chip WITH a sensor is put it, it knows to switch the sensing to the on chip sensor.
Click to expand...

Hmm, interesting. Are there any documents available that explain this? I'm a bit skeptical.
 
kronchev said:
And I also bet you dont understand biology. As someone said its evaporation, but also your skin has a natural climate around it thats about a mm or so thick, basically it has its own little cloud of heat floating on your skin all the time. wind chill rips that off.
Click to expand...

Now that's some funny shit. Where do people come up with stuff like this?
 
You must log in or register to reply here.
Back
Top