1. Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

Guide to proper case airflow design

Discussion in 'Overclocking & Cooling' started by keakar, Oct 24, 2007.

  1. keakar

    keakar

    Joined:
    Mar 27, 2007
    Messages:
    2,381 (0.84/day)
    Thanks Received:
    292
    Location:
    just outside of new orleans
    Guide to Proper Case Airflow Design

    Airflow and case circulation:

    Airflow:

    Rule #1 - Airflow is best if it can flow in one direction only and it is always better off (stays stronger) flowing in a straight line of sight direction if possible. Also always cut out a hole where you mount a fan, never mount it behind those predrilled "cheese grater" holes for fans or you won’t get any kind of good airflow. cutting out the opening for the fan can increase airflow by as much as 70% more then the fan trying to get air to pass thru the holes. the newer and larger openings and fan grill designed cutouts are ok but if you aren't sure then test them by feeling the airflow with the fan outside the case then in the case and if its unrestricted you shouldn't be able to tell the difference. Redrill all those small holes in the vents at the back of the case with a 3/16" drill bit to increase air from that spot and if you have the larger hexagon shaped holes that are already larger then it should be ok the way it is. another thing to note is noise, a fan is always louder when mounted onto some type of a stamped metal grill then with it just mounted over an open hole, sometimes its way louder.

    Heat:

    Rule #2 - Heat naturally rises so to remove heat the easiest and most efficient way it is best to draw air in from the cases lowest points and exhaust it from the cases highest points.

    The Direction of Airflow:

    The airflow can be directed into the case from a few different ways as long as you remember rules 1 & 2 above. The most common setup for cases is to intake air from the bottom front and exhausted out the top and back. The reason they do it this way is it takes best advantage of rules 1 & 2 above. The air is drawn in from the coolest point (bottom front) and moves in a lazy "S" upwards to flow across the motherboard and CPU and then carry the heat from the case out the rear exhaust fans (top rear and psu fans).

    Fans:

    Side fans - side fans can and do interfere with this "S" flow of air so keep this in mind if you’re thinking of a high cfm side fan. It may improve cooling in one spot and allow a "dead spot" for airflow somewhere else. Side fans are great cooling aids for direct cooling but reek havoc on proper airflow thru the case because they just cause the air to flow in a circular pattern and the heat has trouble leaving the case so it makes one spot cooler while making everything else in the case constantly stay warmer so while side fans can be helpful, don’t go overboard with the cfms. To reduce the interference from side fans most cases use air ducts to direct the cool air right into the CPU fan so it will not interfere with airflow through the case so side fans should never be used without air ducts to prevent the whirlpooling of the case airflow.

    Video Cooling:

    VGA - the video card area tends to be a dead spot for airflow. This is why they have fans now that exhaust hot air directly out the back of the case. Adding a side fan blowing cool air into this bottom corner helps "stir" the air so heat doesn’t get trapped there and it can rise up and be exhausted by the rear fan. This works well as long as you don’t over do it to the point that it will interfere with the vga cooling fan and the airflow going through the heat sink. Just remember the fan is there to stir and mix the air in that bottom corner dead spot and that’s all. It should not blow high cfms directly across the heat sink or its fan or you could actually reduce its cooling ability by canceling out the airflow moving across and though the heat sink. you also don't want to mess with the airflow coming from the front of the case so its important to use lower cfm fans if you want to add more cooling to that area.




    Balanced Airflow:

    Push / pull or Fan In / Fan Out System - It's not as easy as just having your fan cfms and sizes matching in the front and rear because they don’t always give you evenly balanced airflow through your case due to obstructions and things like side fans or large video card fans that exhaust inside the case. This is because it is a lot harder for a fan to pull air into a case then to blow it out of the case so you need a little more cfms from your exhaust fans to equal out the airflow and help the intake fans to draw in air easier so don’t just assume two 80mm 45cfm fans in the front and two 80mm 45cfm fans in the rear will give you a balanced airflow, they wont but it should be close enough that you don't really notice it.

    Positive Pressure Airflow System:

    More Air in Than Air Out - the benefit to this type of system is supposed to be that it pulls in less dust than a negative airflow system but this is a false idea. Weather the fan draws the air in or directly blows it in it is still the same air from your room with the same amount of dust in it. If your room is dusty your case will be dusty.

    A simple air supply test is to just check the amount of air your system is exhausting through the small holes in the back of your case, if you can feel a gentle draft coming out then you are restricted and should adjust the airflow by reducing the cfms of your intake fans or making bigger exhaust openings to balance things out.

    Negative Pressure Airflow System:

    More Air Out Than Air In - the benefit to this type of system is they are quieter because they have slower fans that move less cfms so they run quieter. one main disadvantage is because of the slower airflow it can build up heat quickly if the airflow is restricted in any way so you need to be sure you don't block the case air intakes and the first time you test your system with the new airflow design its best to check your temperatures are in the safe ranges during heavy gaming activity when heat can build up quickly.

    Often people will use fan controllers to speed up fans for better but noisier cooling while gaming and yet turn them down to still have a nice quiet computer when they are just doing normal computing activities.

    Why does a positive airflow system get less dusty than a negative airflow system?

    Both systems have the exact same amount of dust moving through them but it tends to settle on things in a negative airflow system more, only because the air is moving slower. Think of a fan blowing feathers around in a small room, as long as the fan keeps blowing fast they all stay airborne but put the fan on slow speed and everything will be covered in feathers. The same applies to dust in your computer, it is the airflow "inside" the computer that determines how fast your computer gets dusty. You can reduce dust building up by putting a fan inside the case just to keep the air stirred up so dust won’t settle and it can then be exhausted out the back of the case by the fans but then everything in the case would have poorer cooling because of this air whirlpool going on.

    Air Supply Test:

    A simple air supply test is to just check the amount of air your system is exhausting (rear exhaust fan) and also checking the air your psu is exhausting is to just remove the side cover on your computer. It will always move more air with the door open so if you notice only a little difference then you are fine but if you notice a big difference then you may need to investigate where your restriction is and provide a larger or less restrictive air intake opening.


    A person with high cfm exhaust fans who doesn't have proper sized intake openings for his case can actually cause a slight vacuum inside the case and prevent the psu from getting enough air because it is being held back by that vacuum causing the psu to overheat. Some PSU that have high cfm fans can have the reverse effect and cancel out the rear exhaust fans ability to remove air from the case. Most problems people have with negative airflow systems are with the proper sizing of cfms and the size of the fans and not the negative airflow design itself.

    If you aren’t over clocking, a properly designed negative pressure airflow system will always work great with just a large 120mm rear exhaust fans only and no intake fans at all unless spot cooling is needed for a hot spot. It just needs to have an unobstructed opening (no cheese grater holes blocking the front fan spot or air filter) that is equal to or bigger than the total size and shape of the exhaust fan openings. If for example, if you have two 80mm exhaust fans (plus the 80mm psu fan) then you need an unobstructed opening that is equal to four 80mm holes with open grills. This won’t work with filters because with the extra restriction of the filters you don't get proper airflow. it also works fine with just one 120mm rear fan and the openings for the 80mm side fan and 80mm front fan will allow more then enough air in to have good balanced airflow. in this situation you would only cut out the front fan cheese grater grill not the side fan because it isn't needed and it looks better.

    Another overlooked obstruction is the front panel cover; it can restrict airflow no matter how big the opening is behind it so test with and without the front panel in place to see if you need to cut back the plastic in an out of sight spot such as the bottom edge or if need be then air must be brought in from a different place (maybe the bottom case panel).
     
    Last edited: Apr 30, 2013
    a_ump, [Ion], DrPepper and 14 others say thanks.
  2. von kain New Member

    Joined:
    Sep 13, 2006
    Messages:
    782 (0.26/day)
    Thanks Received:
    49
    Location:
    athens/greece
    very good work

    even though i din't read it all
     
  3. a111087

    a111087

    Joined:
    Apr 2, 2007
    Messages:
    2,778 (0.98/day)
    Thanks Received:
    203
    Location:
    US
    some pics for lazy people would be nice :D
     
  4. Namslas90 New Member

    Joined:
    Aug 27, 2006
    Messages:
    4,851 (1.59/day)
    Thanks Received:
    555
    Location:
    Earth
    Something that may also be a factor is;

    Avoid A fan on one component blowing hot air into the intake of the fan on another component. It's not that common but I have allready encountered this a few times.

    :toast:
     
  5. keakar

    keakar

    Joined:
    Mar 27, 2007
    Messages:
    2,381 (0.84/day)
    Thanks Received:
    292
    Location:
    just outside of new orleans
    this was mentioned in #4 and $5 but maybe i should have addressed it in its own paragraph

    this is a common mistake because people think putting a fan in the side panel duct blowing into the cpu will always help but if they are different cfms they can actually interfere with each other and cool a lot worse.

    for my case i put just one 120mm exhaust fan up on the top centered between the psu and dvd drive blowing up and i opend all the pci slots and cut out the front fan area so it draws from the front up over my ram and from the rear accross my video card then up over the board. the psu removes any heat leftover in that corner. the cpu cooler stirs everything up real good so i barely get over room temps with a prescott!! :eek: quiet as a whisper too.
     
    Last edited: Oct 27, 2007
  6. AsRock

    AsRock TPU addict

    Joined:
    Jun 23, 2007
    Messages:
    11,255 (4.10/day)
    Thanks Received:
    1,804
    Location:
    US
    Worth a thanks for effert + those who it will help

    :toast::toast:
     
  7. Namslas90 New Member

    Joined:
    Aug 27, 2006
    Messages:
    4,851 (1.59/day)
    Thanks Received:
    555
    Location:
    Earth
    One case in particular was a CPU HS feeding it's hot air directly into the air intake of vga card due to the improper instalation of a chipset fan.(hot air from cpu was deflected by chipset fan into vga card, turning chipset fan around to blow the other way solved a 7° rise in vga card temp).

    :toast:
     
    keakar says thanks.
  8. WBaS

    WBaS New Member

    Joined:
    Oct 17, 2007
    Messages:
    15 (0.01/day)
    Thanks Received:
    2
    Location:
    New York, USA
    heh, glad to see you took my advice! :) nice post!
     
  9. d44ve New Member

    Joined:
    Apr 16, 2007
    Messages:
    2,521 (0.90/day)
    Thanks Received:
    309

    Some? Screw that.... It needs to be all pictures.... with pop ups.... and flames..... and things that jump out and say "You have just won a free iPod Nano"......
     
  10. nhlrocker New Member

    Joined:
    Oct 20, 2007
    Messages:
    65 (0.02/day)
    Thanks Received:
    1
    thank you very much for this!
     
  11. Cuzza

    Cuzza New Member

    Joined:
    Jul 3, 2007
    Messages:
    1,318 (0.48/day)
    Thanks Received:
    207
    Location:
    New Zealand
    ewwwww nanos are for girls
     
  12. theonetruewill New Member

    Joined:
    Nov 12, 2006
    Messages:
    2,996 (1.01/day)
    Thanks Received:
    240
    Location:
    London - Close your eyes and you'll see me
    I have? Where?!:toast:

    :roll:
     
  13. Tan DJ

    Tan DJ New Member

    Joined:
    Sep 10, 2006
    Messages:
    103 (0.03/day)
    Thanks Received:
    0
    Location:
    Brisbane, Australia
    I don't know about the "high airflow = less dust settling" theory! The most dust in my setup is directly in the path of the fan, both inside the case for the intake fan, and on the wall behind the rear fan.

    I think that a positive internal pressure would assist in keeping the internals of the case dust free because it would enable you to control the opening that air goes in, and provide filtering. Any other opening will have air going out because of the positive air presure, and therefore, unfiltered air cannot enter the case.

    And the less you have to open your case to clean out the dust, the less likely you are to introduce faults into your system.
     
  14. Tan DJ

    Tan DJ New Member

    Joined:
    Sep 10, 2006
    Messages:
    103 (0.03/day)
    Thanks Received:
    0
    Location:
    Brisbane, Australia
    I think this thread should be made sticky
     
  15. HookeyStreet

    HookeyStreet Eat, sleep, game!

    Joined:
    Aug 29, 2004
    Messages:
    7,135 (1.89/day)
    Thanks Received:
    523
    Location:
    Great Yarmouth, England
    Nice guide, makes a lot of sense too ;) A diagram would also be pretty good ;)
     
  16. Zedicus

    Joined:
    Sep 30, 2004
    Messages:
    490 (0.13/day)
    Thanks Received:
    41
    extra info.
    positive presure inside a case will aid in cooling because theres more air in a given area for heat to dispearse into. this does not mean a positive system is the only way to go. i myself still use a negative system and there are advantages to it.

    Positive systems tend to pool up the air a bit and make the air flow not as good. especially in a case where there is a bend in the air flow. the "S" curve in cases can cause eddies and hotspots to form.

    Negative presure systems are a little easier to direct airflow, a slight vacuum will help pull air through all areas of a case.

    side panel fans can be a gracing or a curse, serious consideration needs to be taken. one good example is a side panel fan directly over the back edge of a video card. the video card must hav a rear case exhaust for this to be best. you will then pull outside air into the video card and push it out of the case. this will disrupt the air flow inside the case in the most minimal way.

    proof, my cpu temperature with my side panel off my case runs at 56 degrees celcius. my cpu temperature with my side panel on my case never goes over 53 degrees celcius. im also now running an x2 3800+ at 2.8ghz.

    i use a negative design, 2 intake 80mm fans at very bottom front of case, my first and lowest exhaust fan is on the video card, then right at procesor height i hav 2 80mm case fans exhausting. and the psu fan is exhaust. now another very important part is i hav all the little holes sealed up and around the drive bay covers and stuff all sealed up. if air comes in my case it has to come through the front fan wholes, which are covered with a filter too, and filters do restrict a bit of airflow, again im pulling negative presure on the case to help make sure air gets sucked through the filter and not through some random case opening.
     
  17. keakar

    keakar

    Joined:
    Mar 27, 2007
    Messages:
    2,381 (0.84/day)
    Thanks Received:
    292
    Location:
    just outside of new orleans
    this is not caused by positive airflow but because the rate that air absorbs heat is a constant and the difference in the case ambient air temperatures is dictated by the amount of time the air actually spends inside the case. since in a positive airflow system the air spends less time inside of the case it therefore has less time in which to absorb heat from everything in the case so simple logic follows it has to have cooler ambient air temperatures at the time it leaves the case then if it were in a negative pressure airflow case were it would be moving much slower and picking up more heat in doing so.

    this does not mean either system will cool your components better than the other, this only effects case ambient temps and will not change the efficiency of heatsinks except for a very very minor effect (we are talking in tenths of 1 degree) as far as how fast say the cpu cooler can give off heat so only serious overclockers might care about such a small difference but then they mostly go with watercooling anyway.

    this just proves that you have better airflow through your case when the side panel is on, which is proof that you have a well designed case airflow system. but i disagree there is any more or less air depending on the type of airflow you use.
     
    Last edited: Oct 27, 2007
  18. panchoman

    panchoman Sold my stars!

    Joined:
    Jul 16, 2007
    Messages:
    9,595 (3.52/day)
    Thanks Received:
    1,200
    great guide keaker. :toast::toast::toast::respect:
     
  19. Wile E

    Wile E Power User

    Joined:
    Oct 1, 2006
    Messages:
    24,324 (8.08/day)
    Thanks Received:
    3,778
    I agree, another good guide here on tpu. Now to just make some diagrams, and you're all set.
     
  20. Wile E

    Wile E Power User

    Joined:
    Oct 1, 2006
    Messages:
    24,324 (8.08/day)
    Thanks Received:
    3,778
  21. panchoman

    panchoman Sold my stars!

    Joined:
    Jul 16, 2007
    Messages:
    9,595 (3.52/day)
    Thanks Received:
    1,200
    since when has that been around wile e? google keeps coming out with really nice and new services and doesn't even tell us wtf.
     
  22. Wile E

    Wile E Power User

    Joined:
    Oct 1, 2006
    Messages:
    24,324 (8.08/day)
    Thanks Received:
    3,778
    I don't really know. I saw it mentioned in the forums here the other day. I didn't know about it either, prior to that.
     
  23. Zedicus

    Joined:
    Sep 30, 2004
    Messages:
    490 (0.13/day)
    Thanks Received:
    41
    high PRESSURE will cool something faster then low presure even if the same air temperature and humidity are taken into consideration. something at sea level will cool quicker then something in the rocky mountains, again given equal air. now you are correct with the odds of being able to build enough of a pressure difference in a case for this too matter, but i was just trying to explain a bit about air density and pressures.
     
  24. Zedicus

    Joined:
    Sep 30, 2004
    Messages:
    490 (0.13/day)
    Thanks Received:
    41
    heat will cause a raise in pressure. but then that was totally not was i was saying in the first place, possibly we just arent understanding each other, and probably we r both right in a given instance. also rapid changes in presure are what creates the hot or cold effect, this isnt exactly what im refering to still though.

    how about this, denser air at the same static temperature compared to less dense air at the same static temperature (including humidity etc, etc) the denser air will cool something better then less dense air, assuming all other variables are the same.

    (incredibly more technical then i was trying to be in my original explenation)

    and yur right we hav straid a bit from this topic.
     
  25. Good work! However , when I was a total beginner, I thought I would have 2 fans in the front sucking in air, and 2 at the rear blowing it all out. So I tried.. And uhm,I was not even amazed. On my new case I have 3 fans only, 1 rear,1 top, and one on the "window". All sucks the hot air out of the case. And now I can see a difference, one of them the Tt fan has adjustable speed, putting it on max really sucks the hot air out.
     

Currently Active Users Viewing This Thread: 1 (0 members and 1 guest)

Share This Page