Solved

Optimal case fan configuration for CPU water cooler?

I know conventional wisdom says to have the intake fans on the front and side of the case, and the exhaust fans on the top and rear of your case. But when using a water cooler that has the radiator secured to the rear of the case, might it be better to switch the top case fan to intake, so cool air is sucked into the case at the top and directly sucked into the radiator and out the back?

For reference, check out the picture on this link about half-way down the page:
http://hardforum.com/showthread.php?t=1559139
6 answers Last reply Best Answer
More about optimal case fan configuration cpu water cooler
  1. Best answer



    get this and side fans as intake
  2. Postive / Negative pressure has become a bogus concept as many with little to no understanding of fluid dynamics oft parrot statements and take them out of context. The air in ALWAYS equals the air out .... if it didn't your case would either implode or explode. Dust collection is simply a matter of whether or not you have filters and how fine they are. If you have more static pressure and air flow supplied by intake fans, then the excess air escapes out the little vent openings, nooks and crannies. If you have more going out via fans, then in, then air gets in through those nooks and crannies.

    Whether its positive or negative in the context used is meaningless. It's simply whether the intakes are filtered or not. Air coming in brings in dust with it. Positive pressure (in the context used by most) without filters on intake fans will fill your rig with dust. Negative pressure (again in context) without filters on intake fans AND vents will bring in dust. Negative pressure with filters on fans and not on vents and other openings will let in a small amount of dust.

    I always aim for as neutral an air balance as possible for efficiencies and always use filters. When you have more air being forced out smaller openings, you increase static pressure, the result of which is reduced air flow.....cooling is a simply a matter is how many cfm ya bring in and out (assuming its properly circulated). All things being equal, trying to push 4 intake fans out thru 2 exhaust fans requires the air through those two exhaust ports to go out at twice the velocity. Hold ya hand out a car window at 20 mph and then again at 40 mph....what causes more hand resistance ? Keep in mind that one large fan can equal the flow and static pressure of two smaller fans..... One 200mm ~ two 140mm

    I would recommend against forcing all air out thru the radiator as the static pressure resulting from such practice would greatly reduce air flow and therefore make you system hotter.
  3. JackNaylorPE said:
    Postive / Negative pressure has become a bogus concept as many with little to no understanding of fluid dynamics oft parrot statements and take them out of context. The air in ALWAYS equals the air out .... if it didn't your case would either implode or explode. Dust collection is simply a matter of whether or not you have filters and how fine they are. If you have more static pressure and air flow supplied by intake fans, then the excess air escapes out the little vent openings, nooks and crannies. If you have more going out via fans, then in, then air gets in through those nooks and crannies.

    Whether its positive or negative in the context used is meaningless. It's simply whether the intakes are filtered or not. Air coming in brings in dust with it. Positive pressure (in the context used by most) without filters on intake fans will fill your rig with dust. Negative pressure (again in context) without filters on intake fans AND vents will bring in dust. Negative pressure with filters on fans and not on vents and other openings will let in a small amount of dust.

    I always aim for as neutral an air balance as possible for efficiencies and always use filters. When you have more air being forced out smaller openings, you increase static pressure, the result of which is reduced air flow.....cooling is a simply a matter is how many cfm ya bring in and out (assuming its properly circulated). All things being equal, trying to push 4 intake fans out thru 2 exhaust fans requires the air through those two exhaust ports to go out at twice the velocity. Hold ya hand out a car window at 20 mph and then again at 40 mph....what causes more hand resistance ? Keep in mind that one large fan can equal the flow and static pressure of two smaller fans..... One 200mm ~ two 140mm

    I would recommend against forcing all air out thru the radiator as the static pressure resulting from such practice would greatly reduce air flow and therefore make you system hotter.



    HI JackNaylorPE,

    I have got a Zebronics Defender cabinet with front, side & rear fans points..
    Can I go with the below setup:
    Front Fans = Intake, 120 mm fan
    Side Fan = Intake, 120 mm fan
    Rear Fan = Exhaust, "please suggest a fan size" as I am not sure about that if it can fit a 80mm or 120 mm or higher...
    ********************************
    MY SYSTEM configuration:
    i3-4130 cpu
    4gb Ram
    Gigabyte- GA-H87M-D3H mobo
    450 W psu with rear exhaust
    --NO GRAPHICS CARD TO BE USED IN FUTURE BUT MIGHT USE PCI SLOTS IN FUTURE.--
    ********************************

    Thanks & regards
    Paresh
  4. Use the fans on Radiator to Exhaust. All air coming into the case should be filtered and your Rad isn't a filter. I came to the conclusion that " Hot Air in the case will heat up your CPU" is a load of crap. Air coolers operate inside the case and the 30-35 degrees inside the case is a lot cooler than the 70-80 degrees the CPU is at so a Rad is just fine exhausting. Besides, I wouldn't want 60 degree air blowing into the case.
  5. Quote:
    The air in ALWAYS equals the air out


    Sure, but air is highly compressible which means that air inside can be at a higher pressure than the air outside the case. When you have a difference in air pressure, you have air flow. WIth the case at higher internal pressure, all of the hundreds of air leaks will become exhaust leaks. Since the air source is the intake fans, filters on these fans will block all dust. If you were to have all fans in the case be exhaust fans, they'd still exhaust air, but they'd create a pressure drop lower than outside air pressure. The source of incoming air would be the hundreds of cracks. If there's no filters over those cracks (there aren't) you get a ton of dust and debris incoming. So, it's not bogus. Source: I'm an aerospace engineer.
  6. I am a licensed professional engineer, taught at college level and, among other things, I design building ventilation systems and compressed air systems for wastewater and power plants.... air in always equals air out. Yes air is compressible but you need significant pressure to accomplish this. You are not going to pressurize air to any significant extent with fans that have a Maximum Static Pressure of 1" of water. And what happens when you pressurize a closed system (tire) and keep putting air in faster than it can get out ....kaboom !

    Ventilation systems whether in buildings or in equipment provide no significant change in air density. A change in air pressure of 1 psi, increases air density by 1.066 %. The inch of water your case fan is capable of providing is 0.036 psi ....so your fan is capable of increasing air density by 0.038% or about 1/2600 th. And once it reaches the point where it can't compress it any more (maybe 2 seconds) ....air in must air out.

    Look at it this way ..... you have a pressure vessel .... it's rated for 100 psi ..... it has a pressure relief valve set at 90 psi. You keep pumping air in and it will compress .... 90 psi being 2500 times more pressure than the fan on your case can provide. Once it reaches 90 psi, the relief valve opens and it this point we have something analogous to a PC case. The fans pressurize the case to whopping 1" of water (0.036 psi) at which point the "head" provided by the fans, exactly equals the head loss thru the filters and grilles. This is known as a "steady state" condition.

    Now back to our pressure vessel..... if what you saw was true and somehow we were able to break the laws of physics, if more air was coming into that pressure vessel than was going out thru the relief valve, the air would have to be compressed further and the internal pressure would continue to rise above 90 psi..... and them it would pass 100 and create a hazard as it would be above its rating and eventually it would explode. Same with your house, same with your office, same with your PC case.

    Ever open a door and find it hard to open ? That's because air being blown into a building raises the internal pressure..... after ventilation is turned on it equalizes and reaches steady state condition....what would you say is the pressure you had to apply to those doors to open them ? 20 pounds ? 30 pounds ? let's go crazy and say you had to apply 50 pounds of force to leave the building .... that amounts to a difference in air pressure of 0.016 psi or 0.45 inches of water SP.


    Memhorder said:
    Use the fans on Radiator to Exhaust. All air coming into the case should be filtered and your Rad isn't a filter. I came to the conclusion that " Hot Air in the case will heat up your CPU" is a load of crap. Air coolers operate inside the case and the 30-35 degrees inside the case is a lot cooler than the 70-80 degrees the CPU is at so a Rad is just fine exhausting. Besides, I wouldn't want 60 degree air blowing into the case.


    No the rad isn't filtered but the grille in front of it usually is. See page 19.... Top Bottom, front and side rad mounts all have removable filters.
    http://phanteks.com/assets/manuals/Enthoo-Primo.pdf


    The rest is just simple thermodynamics:

    Outside Air Temp = 23C
    Inside Air temp = 28C
    Water Temp = 33C

    The GFX cards and COU and whatever else is being cooled transfer their heat to the water. The rate of transfer is directly proportional to the Delta T

    Fourier's Law express conductive heat transfer as q = k A dT / s, where

    q = heat transfer
    A = heat transfer area
    k = thermal conductivity of the material
    DT = temperature delta across the material
    s = material thickness

    OK so the area of the rad (A) remains constant, the material remains constant, the thickness remains constant .... so we could combine all those into a single constant (C) for the purposes of our analysis an rewrite that to:

    Heat Transferred = Constant x DT

    So now lets look at our 2 conditions

    1. Using 28C inside case air to cool the liquid coolant .....

    Heat Transferred = Constant x DT
    Heat Transferred = Constant x (33C coolant - 28C interior case air)
    Heat Transferred = Constant x 5

    2. Using 23C ambient air to cool the liquid coolant .....

    Heat Transferred = Constant x DT
    Heat Transferred = Constant x (33C coolant - 23C ambient air)
    Heat Transferred = Constant x 10

    Clearly, the amount of heat transfer using the outside ambient air will remove twice the amount of heat from the coolant.
Ask a new question

Read More

Cases Configuration Fan Cooling CPUs