Sign in with
Sign up | Sign in
Your question
Solved

Need Push/Pull + exhaust fan advice :]

Last response: in Overclocking
Share
December 6, 2009 3:23:56 AM

Hello all,

I plan on replacing several fans on my antec 902, including the exhaust fan, with the following Xiggy (mostly for aesthetic reasons in going with an orange-themed rig)
http://www.newegg.com/Product/Product.aspx?Item=N82E168...

Also, I want to replace my hyper 212+'s current push/pull fans (stock fan and old temporary fan that was in my old rig) with these:
http://www.newegg.com/Product/Product.aspx?Item=N82E168...

I'm not too keen with airflow theory, but with a setup like that, would air coming from the push/pull get "clogged" on the way out of the exhaust xiggy, since the exhaust is only running at 61cfm/1500 rpm compared to the slipstream's 1900 rpm. Also, I have the 200mm big boy exhausting up top. Would that essentially help in "unclogging" any stuck air?

I'm a bit confused. Thanks in advance for the advice :]
(also, sorry for the double post. I realized I posted this on the wrong section this morning)
a c 163 K Overclocking
December 6, 2009 3:57:07 AM

37 dBa ... do they come with ear protection ? :) 

This is often a confusing question. Best to think of it this way, if I put 3 fans blowing in and 2 fans blowing out at 100 cfm each ....

how much air is blowing in ? it's not 300
how much air is blowing out ? It's not 200

If those answers were true the case would explode.

fans like pumps move their "liquid" on the basis of a curve...the higher the resistance, the further they drop down on the curve.

In the above example, that 3rd fan blowing in would increase case pressure allowing the two exhaust fan to move a bit more than their 100 cfm rating. That extra case pressure would drop their flow by a bit also and you'll prolly wind up with the two exhaust fans pushing out 120 each and the three intake fans pushing 80 each.

Keep in mind that at a certain point, pushing more air thru closely spaced fins doesn't do much because it creates too much turbulence. Prolimatech tech Support for example recommends no more than 800 -1200 rm fans.





m
0
l
December 6, 2009 8:17:04 AM

Ahh, I see.. thanks for the example and explanation, it made plenty of sense :D .

Hmm, I did have a follow up question if you don't mind. At what point (or at what rough estimate) does pushing air through closed spaced fins cause turbulence? I'm guessing that I may have to tone down the 2 cpu fans in hopes of not doing just that.

Thanks!! :) 
m
0
l
Related resources

Best solution

a b K Overclocking
December 6, 2009 3:11:28 PM

For starters don't concern yourself so much with the rpm of the fan. RPMs biggest effect will be on noise and current draw, not performance. Pay closer attention to the actual cfm rating of the fans. Fans with with the same cfm rating at lower rpms equal better. Normally a larger fan though, but also how the blades are designed effect how well they push or pull air. And some fans are better at pushing than pulling or vice versa. Too bad none of them I have seen say what they are made for.

Another thing to consider is those cfm ratings are not measured in a case, they are measured on a bench with unrestricted airflow. If you mount three fans for intake and 2 fans for exhaust (positive pressure, assuming equal cfm ratings) you will have back pressure in the case which will reduce the amount of air the fans move.

So unless you have really powerful fans I doubt you will get much more air flow than what you have for exhaust. Actually raising the pressure inside your case would require something more than your average computer fan. And if you did manage to raise the pressure inside your case it would be warmer since when you compress air it heats up. And we certainly don't want that.

I would say for good airflow you want as much cfm for exhaust if not a bit more, and possibly a fan with more area for the exhaust. And unrestricted air flow though the case.

Same thing with the fins on your cooler, if you try shove more air than can physically move through the fins you are just wasting fan power and possibly decreasing the cooling instead of increasing it. I simple way to demonstrate this is take one of your case fans before you put it in and put your hand over the front of it. It doesn't blow your hand away, it just slows down and quits moving air or reverts the air back through the fan. Pull your hand away 5 or 6 inches and the flow increases and you can feel it cooling your hand. Back pressure will choke a low powered fan out.

Something else to consider is the dynamics behind cooling. Cooling electronics doesn't work quite the same as cooling us. When we feel fan we feel it cooling our body. This doesn't really happen with electronics. When we (humans) feel cool under a fan its because the air blowing across our skin evaporates moisture off our body, and when moisture evaporates it cools the surface. This is sort of how an airconditioner or swamp cooler works. With electronics it more just moving the hot air away from the components so they don't over heat.

Sometimes you can double the amount of air that moves over a chip and not really see a big temperature drop. Now if you could find a way to spray water on it and have it evaporate off without ruining it you would have something lol.

Paul
Share
a b K Overclocking
December 6, 2009 3:35:40 PM

Paul - Excllent post! :) 
m
0
l
December 7, 2009 1:22:51 AM

Thanks everyone and Paul for the lengthy and nice read :D .

I have a clearer understanding now and will reconsider my options.
m
0
l
a b K Overclocking
December 7, 2009 4:12:43 AM

I am not known for my short to the point posts in the world wide web lol. Glad it helped.

Paul
m
0
l
December 7, 2009 7:58:33 PM

Here is a good article i found on air cooling, and, how positive/negative pressures affect the cooling and dust levels
m
0
l
a c 163 K Overclocking
December 11, 2009 12:27:15 AM

vznoobie12 said:
Hmm, I did have a follow up question if you don't mind. At what point (or at what rough estimate) does pushing air through closed spaced fins cause turbulence? I'm guessing that I may have to tone down the 2 cpu fans in hopes of not doing just that.


Depends on find thichness and spacing...suggest asking manufacturer....Prolimatech says 1200 rpm for the Megahalems
m
0
l
a c 163 K Overclocking
December 11, 2009 12:50:17 AM

paul mohr said:
So unless you have really powerful fans I doubt you will get much more air flow than what you have for exhaust. Actually raising the pressure inside your case would require something more than your average computer fan. And if you did manage to raise the pressure inside your case it would be warmer since when you compress air it heats up. And we certainly don't want that.


What goes in must = what goes out. Otherwise over time, case would explode. Since fans can't produce that kind of pressure, any oversizing or intake or exhaust is "shared" by the other and flow equalizes between istake and exhaust

In building HVAC design (my world) and PC case cooling we are talking about where people go.....the typical increase in room pressure is measured in the fractions of an inch of water not PSI. To get measurable (by touch) temperature changes, we need significant PSI differences....we are several orders of magnitude below that kinda thing

Look at this for example:

http://www.frozencpu.com/products/2492/fan-191/SilverSt...

Static Pressure: 0.91mmH2O @ 12VDC at zero airflow

At 25.2 mm, we have 1 " of water pressure or 0.03 psi

At 91mm, we have 0.035 inches of water
At 0.035 inches of water, we have 0.003 feet of water
At 0.003 feet of water, we have 0.00129 psi

Now you certainly can't feel the difference of 0.00129 psi in a room but it's enough to make a case fan produce 0 air flow.

Putting this case fan blowing into a closed case would mean that outside the case, atmospheric pressure would be 14.7 psi and inside the case it will be 14.7013 psi. It's certainly not going to cause any significant change in temperature.

Quote:
Something else to consider is the dynamics behind cooling. Cooling electronics doesn't work quite the same as cooling us. When we feel fan we feel it cooling our body. This doesn't really happen with electronics. When we (humans) feel cool under a fan its because the air blowing across our skin evaporates moisture off our body, and when moisture evaporates it cools the surface. This is sort of how an airconditioner or swamp cooler works. With electronics it more just moving the hot air away from the components so they don't over heat


Air conditioners cool the air in an empty room even w/o humans and moisture. When moisture evaporates off skin, there is a large change in temperature because of the phase change ... this is the exact same techniques as is used in modern "phase change" HS today. The reason you blow air thru a cooler is to take the heat away from the fins so that new cooler air can get in there. Cooling efficiency is based upon the "Delta T". If we depend upon convection, air moves slowly ....air entering one side of the fins might be 25C but by the time it reaches the other side, it might be 30C....so over that last half of its contact time, it's 28, 29, 30 degrees. Now move it thru there twice as fast .... and it's now 26, 27 going out .... move it faster and it's 26 going out. So since 26 gives us a better delta T than 29 T, the fins are much more efficient in transferring heat.

When you move too much air thru the fins it becomes turbulent...air friction causes eddy currents which decrease cooling efficiency both between the fins and in the case. Eddying currents bring hot air to the intake side of the fans which again decreases cooling efficiency.
m
0
l
a b K Overclocking
December 13, 2009 3:58:18 PM

I said if you DID manage to raise the pressure inside your case the temps would go up, never down. Not that you would, and you have proven that it would be near impossible to do that with a fan.

And I said an air conditioner or swamp cooler works in very simplistic terms like our bodies do. In no way did imply an air conditioner needed a human to work? You are right though an air conditioner really doesn't work this way, its much more complicated. I used to work for Tecumseh products cooling division making air conditioning pumps actually.

I think you over technified what I said lol. Our bodies as a function perspire and when this evaporates it cools us. Running a fan over our body speeds this process up. Which is why we feel cool when air hits us, even if its warm air. Computer chips don't work this way is all I was saying.
m
0
l
a c 163 K Overclocking
December 14, 2009 2:40:07 AM

paul mohr said:
I think you over technified what I said lol. Our bodies as a function perspire and when this evaporates it cools us. Running a fan over our body speeds this process up. Which is why we feel cool when air hits us, even if its warm air. Computer chips don't work this way is all I was saying.


Well, ya see that PE after my name it means Anal (posterior) Engineer .... my wife hates going to movies with me cause I go on about tires screeching on dirt roads, fire and sound from explosions in space w/ no atmosphere and tell her who did it 10 minutes into the movie :) 
m
0
l
a c 86 K Overclocking
December 14, 2009 12:39:49 PM

How many atmospheres would the case pressure needed to be raised for .5 C increase in temps? Fans don't have near the needed compressability to even begin to think about that.

Thats bullocks, never happen in a computer case.
m
0
l
a b K Overclocking
December 14, 2009 3:31:18 PM

Oh on the tire screeching on dirt roads, it can be done just so you know. Been to lots of dirt track racing and grew up in middle of no where midwest. If the dirt or clay under the gravel is packed hard enough you can get the sound. Rough pavement is actually harder to get that sound from because of the uneven texture.

Paul
m
0
l
a c 86 K Overclocking
December 14, 2009 10:40:10 PM

What about the classic 'Airplane' movie? DC-9 with prop sounds, on purpose, was funny though.
m
0
l
!