Overclock.net › Forums › Cooling › Air Cooling › Air Cooling in theory
New Posts  All Forums:Forum Nav:

Air Cooling in theory

post #1 of 4
Thread Starter 
Reading all these posts, and spending a considerable amount of time testing air cooling in general, I has some thoughts and wondered if people would think along the same lines. The thread about "how much static pressure is needed" got me thinking about the limit of an air cooler based on design, and how to classify top coolers in terms of performance only.

There are obviously hundreds of different designs for an air cooler, but in theory it all boils down to heat dissapation, which should be a function of 2 things:

#1 The ability of the cooler to carry heat away from heat source

#2 The ability of the heat radiating from the cooler to be removed (ie. blown away via the fan(s).

The first is pretty much in my mind the MAJOR player in if a cooler is on the high end or not. You see many coolers with very similar designs, but perform very differently. This seems to have much more to do with how well it can carry heat away from the heat source than it does #2 above (more on that below).

The ability of the cooler to remove the radiating heat is purely a function of the amount of surface area on the cooler, and the amount of air blowing across that surface area. So, logically, the larger the surface area, the easier it is to remove the heat, with the right amount of air flow. This part of the cooler can vary widely, as a cooler with a smaller surface area could remove the same amount of heat as one with a larger surface area if it had a more powerful fan.

So in testing an air cooler fo maximum performance, you should really isolate its heat transfer ability vs the ability to remove heat via the fans. This is given that all todays high end coolers have massive surface area to dissipate heat, and there will come a point where extra CFM or pressure capable of normal fans will not yield any additional cooling performance.

This is what I would call the maximum performance of a cooler, which I could call the "Heat transfer limit, meaning even if the heat was instantly dissipated at the surface area, the cooler will not perform any better given it just cannot transfer the heat any faster through the heat pipes.

Now, given the designs, I think some coolers can reach this "heat transfer limit" much easier (ie at a lower noise level), but if we are just talking max performance, I would love to see charts that take the dissipation out of th equation and just test for maximum heat transfer.

Given my computer lives in the basement, a floor away from my keyboard and monitor, noise is not a huge concern, but that being said, I have a feeling that the best coolers would reach the heat transfer limit before you had to strap deltas to them. The new ARCHON cooler coming out with 1.6x the surface area of the Ven X would be a great example of a cooler that could be awesome if it can carry the heat, because with that surface area, you would most likely reach max heat dissipation pretty easily if the heat pipe design is no better than the VenX. But if the heat pipe design improved, then the cooler should be dynamite.

Testing in this fashion would tell us if the companies are lazy with a new product by just strapping on more fins, or actually improving the underlying design.

Those are my thoughts anyway,

Chris
Maximus V Extreme
(14 items)
 
  
CPUMotherboardGraphicsGraphics
I-4970K @ 4.6ghz AsRock Formula OC Powercolor 7870 LE  Powercolor 7870 LE 
RAMHard DriveOSPower
32GB Crucial Ballistix 1866 500GB Samsung 840 Win 8 Pro Corsair AX 860 
Case
Corsair 800D 
  hide details  
Reply
Maximus V Extreme
(14 items)
 
  
CPUMotherboardGraphicsGraphics
I-4970K @ 4.6ghz AsRock Formula OC Powercolor 7870 LE  Powercolor 7870 LE 
RAMHard DriveOSPower
32GB Crucial Ballistix 1866 500GB Samsung 840 Win 8 Pro Corsair AX 860 
Case
Corsair 800D 
  hide details  
Reply
post #2 of 4
Perhaps put in some things about certain materials' specific heat capacities and why they're more suited for being used in heatsinks than other materials?

But from a quick glance-over, it looks good.
    
CPUMotherboardGraphicsRAM
Core i7 930 at 4.5GHz, HT on EVGA X58 SLI LE ASUS 7970 TOP Crucial Ballistix Tracers, 6GB, 1720MHz, 8-8-8-24 
Hard DriveOptical DriveOSMonitor
120GB Force 3, 2 Samsung F3 500GB in RAID 0, WD... Sony  Windows 7 64 bit LG W2452TX - 1920 x 1200 
KeyboardPowerCaseMouse
Razer Black Widow Antec TruePower New 650W (SeaSonic) Clemens 004 Logitech G500 
Mouse PadAudioAudioAudio
XTrac Ripper XXL Musliand Monitor 02 Alessandro MS1/JVC RX700 Tannoy Bookshelf Speakers 
Audio
Samson Servo 120A Amplifier 
  hide details  
Reply
    
CPUMotherboardGraphicsRAM
Core i7 930 at 4.5GHz, HT on EVGA X58 SLI LE ASUS 7970 TOP Crucial Ballistix Tracers, 6GB, 1720MHz, 8-8-8-24 
Hard DriveOptical DriveOSMonitor
120GB Force 3, 2 Samsung F3 500GB in RAID 0, WD... Sony  Windows 7 64 bit LG W2452TX - 1920 x 1200 
KeyboardPowerCaseMouse
Razer Black Widow Antec TruePower New 650W (SeaSonic) Clemens 004 Logitech G500 
Mouse PadAudioAudioAudio
XTrac Ripper XXL Musliand Monitor 02 Alessandro MS1/JVC RX700 Tannoy Bookshelf Speakers 
Audio
Samson Servo 120A Amplifier 
  hide details  
Reply
post #3 of 4
Maximum heat transfer by itself is not a good measure of a heatsink performance. The thermal resistence is also very important.


A heatsink that can disperse 1000w can be outperformed by a heatsink that can only disperse 200w if the 200w heatsink has a lower thermal resistence.

It is like fan airflow or pump performance curves.... it is a multi-dimensional graph or multi-variable equaiton. At certain loads, one heatsink may do better than another.
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
post #4 of 4
Thread Starter 
How do you not consider maximum heat transfer directly related to thermal resistence? Maybe the term "maximum heat transfer" is somewhat generic, but I define it as:

Ability of the heat sink base that contacts the heat source to carry away the heat generated by that heat source.

Thermal resistence is a direct property of that, and in the context of CPU coolers, they are all using one of maybe 2 materials, copper or aluminum. so it is somewhat out of the equation as a constant.

Also, in theory, there would never be a case where a cooler gets outperformed by another at a different level of heat, given all other constants being the same, at any load.

**EDIT** thinking about it some more, my statement above could be conceived as untrue. If you had a more efficient cooler, with lower maximum dissipation, it could outperform another cooler until it reached "dissipation saturation", ie the point at which you can not longer dissipate the heat faster than the heat pipes are carrying it to the dissipation area. Remember though, as stated above, dissipation should be taken out of the equation, as all of the high end coolers are going to have huge surface areas and I would be hard pressed to think you would not be able to out dissipate the heat transfer with airflow.

Chris

Quote:
Originally Posted by DuckieHo View Post
Maximum heat transfer by itself is not a good measure of a heatsink performance. The thermal resistence is also very important.


A heatsink that can disperse 1000w can be outperformed by a heatsink that can only disperse 200w if the 200w heatsink has a lower thermal resistence.

It is like fan airflow or pump performance curves.... it is a multi-dimensional graph or multi-variable equaiton. At certain loads, one heatsink may do better than another.

Edited by crayneogeo - 10/22/10 at 9:58am
Maximus V Extreme
(14 items)
 
  
CPUMotherboardGraphicsGraphics
I-4970K @ 4.6ghz AsRock Formula OC Powercolor 7870 LE  Powercolor 7870 LE 
RAMHard DriveOSPower
32GB Crucial Ballistix 1866 500GB Samsung 840 Win 8 Pro Corsair AX 860 
Case
Corsair 800D 
  hide details  
Reply
Maximus V Extreme
(14 items)
 
  
CPUMotherboardGraphicsGraphics
I-4970K @ 4.6ghz AsRock Formula OC Powercolor 7870 LE  Powercolor 7870 LE 
RAMHard DriveOSPower
32GB Crucial Ballistix 1866 500GB Samsung 840 Win 8 Pro Corsair AX 860 
Case
Corsair 800D 
  hide details  
Reply
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Air Cooling
Overclock.net › Forums › Cooling › Air Cooling › Air Cooling in theory