A small correction regarding the laws of thermodynamics and convection. - Overclock.net - An Overclocking Community

Forum Jump: 

A small correction regarding the laws of thermodynamics and convection.

Reply
 
Thread Tools
post #1 of 108 (permalink) Old 02-14-2012, 06:33 PM - Thread Starter
New to Overclock.net
 
Snowmen's Avatar
 
Join Date: Apr 2011
Location: Montreal, Canada
Posts: 2,267
Rep: 179 (Unique: 162)
Recently, I've been seeing many people suggesting the following thing to people looking for help regarding the position of their fan in their case: "No matter what, put your top fans as exhaust because heat rises!"... Wrong! Something you must absolutely have for convection to be applicable is a totally passive environment. As soon as a fan is present, the environment is not passive and the effects of convection are pretty much irrelevant. Here are a few examples of where convection or top exhausts are relevant.

1) Passive PSUs

Some new PSUs are more and more present on the market, these are passive PSUs. These have no fans and therefore rely on passive air movement for cooling (also know as convection). Therefore, you should have the most vented part of it pointing UP (even if you have a bottom vent). This way, the heat that is coming from the PSU components will heat up the air and it will rise out of the PSU. Also, this will help cooling it because your GPU's fan should suck air from the PSU, helping in the cooling. It must noted that this will have almost no impact on the temperature of your GPU since passive PSUs are always very efficient and therefore produce a lot less heat than GPUs.

2) Strict air cooling

By strict air cooling, I mean no radiators. If all your components are cooled by air, you should normally have your top fan(s) as exhaust(s). But, this is not because of convection! This is simply because the component that is the nearest from the top fan(s) is the CPU, a major heat source. Therefore, you wan't the heat to be taken away from the heatsink as fast as possible. Also, this is a good way to keep directional airflow. The top fan will create a low pressure area near the CPU heatsink area which will cause the low pressure area to be filled by the closest source of air. That source should also normally be an intake fan providing cool air for the heatsink's fan(s). If you have a radiator on the back fan, they should always be set up as an intake for the best amount of air. Usually, they will be single radiators hooked up to a CPU cooler like the H Series from Corsair. Therefore, they are cooling the major heat producing component in that area. Because of that, the slightly heated air from the radiator will not cause the temperatures of any components to rise because there's nothing to cool. Of course, there's the VRM area that produces heat but it's a negligable source of heat and could probably be cooled passively in most cases (ie: anything but extreme overclocks on high TDP CPUs).

Now we have seen where your top fan(s) should be set up as exhausts. Sometimes though, you wan't them to be intakes and this is when most misleading advices come.

1) Watercooling

If you have a radiator (Like the H100 or most larger radiators that can only be installed on the tops of a chassis) on the top of your chassis or anywhere else for that matter, it should always be mounted as an intake. This is for various reasons. First, the radiator will only slightly increase the temperature of the air coming in which will not affect the temperature of other components by a significant margin. Also, the radiator is what's cooling the major components in terms of heat production (like a CPU or GPUs) and therefore, that slightly heated air intake will not affect important components except the VRM area which doesn't need much cooling in most cases anyway (ie: anything but extreme overclocks on high TDP CPUs).

Thanks for reading. If you have found any mistakes or believe I have forgotten something, I'd be glad to hear it from you. smile.gif


Quote:
Originally Posted by ____ go_quote.gif
Toothpaste, it will also make your computer minty fresh.
Quote:
Originally Posted by Warhaven go_quote.gif
Hire a big day-laborer for $40, rent an Italian suite for him, and then have him stand next to your PC all night frowning at passersby and shaking his head, "No."
Quote:
Originally Posted by Paradox me go_quote.gif
I think it's obvious humans weren't the lead platform. We got saddled with cat ports.


Snowmen is offline  
Sponsored Links
Advertisement
 
post #2 of 108 (permalink) Old 02-14-2012, 06:50 PM
New to Overclock.net
 
triarii3's Avatar
 
Join Date: Apr 2011
Location: MIAMI, FL. USA
Posts: 304
Rep: 12 (Unique: 12)
Good read. Thanks!

What we do in life, echoes in eternity.
triarii3 is offline  
post #3 of 108 (permalink) Old 02-14-2012, 09:23 PM
New to Overclock.net
 
Join Date: Jan 2010
Location: Florida
Posts: 8,266
Rep: 238 (Unique: 214)
chinesethunda is offline  
Sponsored Links
Advertisement
 
post #4 of 108 (permalink) Old 02-15-2012, 07:58 AM
New to Overclock.net
 
mhocker10's Avatar
 
Join Date: Sep 2011
Location: Elizabethtown , PA
Posts: 12
Rep: 1 (Unique: 1)
great post, good information.. thanks for this thread

Matt.
(my first PC was an Apple IIC in 1987)

Member of EdgeGamers Org. =(eGO)=™Silent_Predator

validation Q6600 CPU-Z
http://valid.canardpc.com/show_oc.php?id=2008757

mhocker10 is offline  
post #5 of 108 (permalink) Old 04-08-2012, 09:56 PM
New to Overclock.net
 
Join Date: Feb 2012
Posts: 1,178
Rep: 43 (Unique: 39)
+Rep. I like top intake and side exhaust, myself.
samwiches is offline  
post #6 of 108 (permalink) Old 04-08-2012, 11:31 PM
New to Overclock.net
 
Molybdenum's Avatar
 
Join Date: Jan 2012
Location: Columbus, OH
Posts: 225
Rep: 48 (Unique: 45)
Quote:
Originally Posted by Snowmen View Post

2) Strict air cooling

By strict air cooling, I mean no radiators. If all your components are cooled by air, you should normally have your top fan(s) as exhaust(s). But, this is not because of convection! This is simply because the component that is the nearest from the top fan(s) is the CPU, a major heat source. Therefore, you wan't the heat to be taken away from the heatsink as fast as possible. Also, this is a good way to keep directional airflow. The top fan will create a low pressure area near the CPU heatsink area which will cause the low pressure area to be filled by the closest source of air. That source should also normally be an intake fan providing cool air for the heatsink's fan(s). If you have a radiator on the back fan, they should always be set up as an intake for the best amount of air. Usually, they will be single radiators hooked up to a CPU cooler like the H Series from Corsair. Therefore, they are cooling the major heat producing component in that area. Because of that, the slightly heated air from the radiator will not cause the temperatures of any components to rise because there's nothing to cool. Of course, there's the VRM area that produces heat but it's a negligable source of heat and could probably be cooled passively in most cases (ie: anything but extreme overclocks on high TDP CPUs).

Now we have seen where your top fan(s) should be set up as exhausts. Sometimes though, you wan't them to be intakes and this is when most misleading advices come.

Rather than words, the best way to test this is with numbers. It's simple and rather concrete. I don't mean to sound bitter, but quite honestly, this test can be done qualitatively, and recommendations should be based on numbers.

I tested my case just now, once with a top intake fan (140 mm) mounted in the forward position (toward the front) then once with the fan in a top exhaust position, mounted in the rear. In both tests, the other side was blocked with a sheet of polycarbonate.
I let my computer idle (well, I'm watching a SC2 stream) for a few minutes, then ran Prime95 for 6 mins (1344 FFT, 5000 MB RAM, 1 min interval) while recording temperature data with realtemp 3.69. I averaged the temperature across all 4 cores, then took the average of these 'average core' temps, starting 10 seconds after beginning Prime95 each time. The results are as follows:
Top intake 54.9 ± 1.1 ºC
Top Exhaust 56.2 ± 1.1 ºC
Ambient temperature was the same for both (I have a thermometer right here), 20.7 ºC
As can be seen, it's rather minimal. But, the effect is there, at least for me in my case with my setup. An intake fan is more effective for cooling. And now that we know the answer, we can begin trying to explain it. I think that the intake fan brings fresh air directly into the CPU cooler. With an exhaust fan, I think that there's slightly less air available for the cooler, if that makes sense. Either way, these are my numbers, if people want to do their own tests, even better, because more data is ideal.

In the spoiler below is a picture of my case, for completeness. Picture of Case (Click to show)
450


Molybdenum is offline  
post #7 of 108 (permalink) Old 04-09-2012, 01:18 AM
Aussie Overclocker
 
tw33k's Avatar
 
Join Date: Aug 2010
Location: Canberra
Posts: 2,767
Rep: 178 (Unique: 144)
I've been saying for years that hot air won't rise if you have a fan(s) pushing it. I've had my case fans in every combination and top intake is the best option for my system.

x99 Test Bench
(13 items)
CPU
9900K
Motherboard
ASUS ROG Maximus XI Hero (wifi)
GPU
ASUS GeForce GTX 1070
RAM
16GB (2x8GB) G-Skill Trident Z F4-3600C16D-16GTZ
Hard Drive
Samsung 950 Pro 256GB| Crucial M4 128GB | 1TB F3 | 2TB WD Black
Optical Drive
Samsung BD-ROM
Power Supply
Seasonic 1050W Platinum
Cooling
Black Ice GTX 480
Cooling
FrozenQ Liquid Fusion X4 Quad Spiral Reservoir
Cooling
EK-XTOP Revo D5 PWM Pump
Cooling
EK Velocity RGB
Cooling
Alphacool NexXxoS GPX - Nvidia Geforce GTX 1070 M02
Case
Corsair Obsidian 900D
Operating System
Windows 10 Pro x64
Monitor
27" Acer CB271HU (2056 x 1440)
Keyboard
Microsoft Wireless 5500
Mouse
Microsoft Wireless 5000
CPU
Intel 5960X
Motherboard
ASUS X99 Deluxe
GPU
Sapphire R9 270X Toxic
RAM
G.Skill 32GB (4 x 8GB) PC4-25600 3200MHz DDR4 RAM - 16-18-18-38 - Trident Z Series
Hard Drive
Samsung 950 Pro 256GB
Power Supply
Corsair AX760i Platinum
Cooling
Swiftech H240-X
Case
Lian Li T60
Operating System
Windows 10 Pro 64bit
Monitor
ACER B273HU
Keyboard
Microsoft Wireless 5050
Mouse
Microsoft Wireless 5000
Mouse
Razer Destructor 2
▲ hide details ▲
tw33k is offline  
post #8 of 108 (permalink) Old 04-09-2012, 03:07 AM
New to Overclock.net
 
lagittaja's Avatar
 
Join Date: Oct 2010
Location: Finland
Posts: 1,955
Rep: 68 (Unique: 60)
Quote:
Originally Posted by tw33k View Post

I've been saying for years that hot air won't rise if you have a fan(s) pushing it. I've had my case fans in every combination and top intake is the best option for my system.

Same here!
I hate how ignorant some people are.

From Finland with Desire
lagittaja is offline  
post #9 of 108 (permalink) Old 04-09-2012, 03:21 AM
New to Overclock.net
 
Join Date: Oct 2011
Location: Houston
Posts: 624
Rep: 57 (Unique: 44)
Ever since I had my phantom, I've always ran both top 200mm fan as intake and it does wonders for the vrm and ram temps along with every other air cooled component on the motherboard. I can't comment on anything else since the rest is watercooled with an external radiator. A fan pointed at naked ram dimms is vastly better than any fancy heatspreader can accomplish if the top fans were exhaust.

I have Mcdonald's fries, but at home, I only have packets of Burger King ketchup.
Will the food still work even though they're not compatible models?

Cha0s_Cha0 is offline  
post #10 of 108 (permalink) Old 04-09-2012, 03:35 AM
New to Overclock.net
 
psyclum's Avatar
 
Join Date: Sep 2011
Posts: 6,034
Rep: 424 (Unique: 300)
the importance and relevance of convection cooling is often OUTSIDE the box.... to discount convection entirely is uneducated at best... yes inside a box, any kind of fan would overpower the action of convection, however, this does not negate for physics behind heat rising to the top. with a top side intake, you are often recirculating hot exhaust from PSU/GPU/CPU right back into the case and causing further inefficiency in your overall cooling solution. this problem is especially acute if the computer is in a semi enclosed area such as under a desk where hot exhaust is trapped by the desktop and then sucked back into the case from the top side intake.

recirculation of hot exhaust (thermal short circuit) creates inefficiency and as a result higher decibel levels then necessary to achieve the same goal. overall cooling performance may not be very noticeable however the real discussion in PC cooling really lies in long term decibel level. anyone can cool anything if they used enough 5000rpm delta fans, however, to keep it under acceptable decibel levels, one must search for efficiency and this is where consideration in convection comes into play

a bottom intake and top exhaust ensures maximum separation between cool intake air and hot exhaust air because OUTSIDE the box, hot air WILL rise and convection DOES work...
psyclum is offline  
Reply

Quick Reply
Message:
Options

Register Now

In order to be able to post messages on the Overclock.net - An Overclocking Community forums, you must first register.
Please enter your desired user name, your email address and other required details in the form below.
User Name:
If you do not want to register, fill this field only and the name will be used as user name for your post.
Password
Please enter a password for your user account. Note that passwords are case-sensitive.
Password:
Confirm Password:
Email Address
Please enter a valid email address for yourself.
Email Address:

Log-in



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may post new threads
You may post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off