Positive pressure airflow working!

[fuzzymath10]

This may not be relevant to all users here, but after having cooling issues with my i5 system due to a custom mod (passively cooling a 5750 in a half-height case), I found a solution involving switching to positive pressure.

In a vertical orientation, my NSK-1480 has two 80mm fans at the base (one for the 350w PSU), and two 80mm fans at the top. Conventional wisdom suggests intake on the bottom, exhaust at the top.

For awhile, this appeared to work, but I began developing BSODs very frequently a few months ago. I made sure nothing had changed, but something was still wrong.

By changing all four 80mm openings to intake, the majority of the air gets pushed along the video card (whose modded Accelero S1 occupies three slots having been bent into a C shape) and out the expansion slot openings.

My BSODs have disappeared, and I've returned to overclocking the 5750. I can't do much more than 800/1200 with the fans all at the lowest voltage, but at mid-voltage, 860/1260 works fine. At peak (I've only tried with SC2 at 1080p high), the CPU hits around 50C and the GPU hits around 75C, with very little noise produced.

 

[Idontcare]

In terms of heat-transfer, positive versus negative airflow is expected to favor positive airflow for a good reason - turbulence.

The turbulence created by the airflow on the output side of the fan is much much higher than the turbulence created in the negative pressure region on the input side of the fan.

Convective heat transfer relies critically on the thickness of the diffusion zone near the surface of the hot component you are attempting to cool. Turbulence reduces the thickness of this diffusion zone.

This is why it feels cooler, and in fact the evaporation rate is higher, when you stand in front of a fan versus standing behind it. The CFM flowing past you is the same whether you are in front or behind the fan, but the air dynamics itself is very different. The airflow in front of the fan is extremely turbulent (swirling, eddies, etc) whereas the airflow behind the fan feeding into the fan will be much more laminar and less turbulent.

http://encyclopedia2.thefreedictionary.com/Convective+boundary+layer

In fluid mechanics, a thin layer of flowing gas or liquid in contact with a surface (e.g., of an airplane wing or the inside of a pipe). The fluid in the boundary layer is subjected to shear forces. A range of velocities is established across the boundary layer, from zero (provided the fluid is in contact with the surface) to maximum. Flow in boundary layers is more easily described mathematically than is flow in the free stream. Boundary layers are thinner at the leading edge of an aircraft wing and thicker toward the trailing edge; such boundary layers generally have laminar flow in the leading (upstream) portion and turbulent flow in the trailing (downstream) portion. See also drag.

http://www.enigmatic-consulting.com/semiconductor_processing/CVD_Fundamentals/xprt/Boundary_lyr.html

http://en.wikipedia.org/wiki/Turbulence#Heat_and_momentum_transfer

When flow is turbulent, particles exhibit additional transverse motion which enhances the rate of energy and momentum exchange between them thus increasing the heat transfer and the friction coefficient.

 

[EarthwormJim]

This is why it feels cooler, and in fact the evaporation rate is higher, when you stand in front of a fan versus standing behind it. The CFM flowing past you is the same whether you are in front or behind the fan, but the air dynamics itself is very different. The airflow in front of the fan is extremely turbulent (swirling, eddies, etc) whereas the airflow behind the fan feeding into the fan will be much more laminar and less turbulent.

Isn't it mainly due to how much more directional the output air is vs. the input? If you're standing in front of the output of a fan, almost 100% of the airflow will go past you. The air is being accelerated by the fan blades, normal to their rotation plane.

The input air could be sourced from a very wide area around the input side, so a person standing behind a fan will not necessarily feel 100% of the air flow since some of it could be source around them. Nothing ensures that the air travels a straight path into the input side.

 

[Idontcare]

Isn't it mainly due to how much more directional the output air is vs. the input? If you're standing in front of the output of a fan, almost 100% of the airflow will go past you. The air is being accelerated by the fan blades, normal to their rotation plane.

The input air could be sourced from a very wide area around the input side, so a person standing behind a fan will not necessarily feel 100% of the air flow since some of it could be source around them. Nothing ensures that the air travels a straight path into the input side.

That's the laminar part of the airflow that is going into the fan.

The point is that the laminar flow is converted into turbulent flow on the output side, 50 CFM of laminar flow will not cool as well as 50 CFM of turbulent flow.

Put yourself into a wind-tunnel, stand on the input side versus the output side. There's a difference and the difference is because of the Reynolds number, turbulence, and the diffusion barrier that impedes heat transfer in convection mechanics.

 

[aigomorla]

say it with me...

unless u spend a lot of time sealing every nook and cranny of your PC, or you got a monster blower pushing air into your case.... you will never get positive or negative pressure to work in a standard computer case.

 

[Idontcare]

Aigo, maybe you are misunderstanding the terms?

A fan, all on its own sitting out in the open, is generating positive/negative pressure. Boxed up or not.

If a fan is blowing air into a computer case then there is positive air pressure inside the case, if there wasn't then the air would not be exiting the case, at any point from the case, at all, ever.

That's why/how the air moves in the first place, pressure gradients.

 

[T_Yamamoto]

theres a company that makes them though

they say its proven to have positive airflow.

 

[Brakner]

I always thought my Silverstone FT02 was positive pressure. Wouldn't 3 x 18cm intake and only single 12cm exhaust be positive pressure? I know it keeps pretty much every component 5c to 10c cooler that when they were in my P180 which had more exhaust than intake.

 

[aigomorla]

Aigo, maybe you are misunderstanding the terms?

A fan, all on its own sitting out in the open, is generating positive/negative pressure. Boxed up or not.

If a fan is blowing air into a computer case then there is positive air pressure inside the case, if there wasn't then the air would not be exiting the case, at any point from the case, at all, ever.

That's why/how the air moves in the first place, pressure gradients.

Positive pressure and Negative pressure used in the actual pressure states of the case.

In order to get a working positive or negative pressure, the air vol. inside the case has to be different to the outside environment.

The pressure values your talking about in a standard case with just 1 fan is negligible, as i said, you need a pressure difference between outside and inside which is just not possible in a standard case.

This is only possible with a high powered blower pushing air into the case, OR, having your case completely sealed with an filtered air duct from an outside source and having the case pull air out.

Once again... that pressure difference is close to impossible to yield in a standard case unless you use a high powered blower into the case.


Positive pressures works well in high dust enviorments if your blower is connected to a filter.

As for temp difference from the two, its almost negligble as the ambient temp inside the case determines the final value, and ambients inside a case is not difficult to control unless ur trying to hold pressure.


EDIT ive only seen one positive pressure unit which was great.... and it was tied to an Air Conditioner.. lol..
Other then that... real possitive pressure systems are very difficult to make.... trust me.. ive made probably only 2 in all my time at building PC's.

By changing all four 80mm openings to intake, the majority of the air gets pushed along the video card(whose modded Accelero S1 occupies three slots having been bent into a C shape) and out the expansion slot openings..

That being said... Fuzzy's improvement for swaping the direction of fan is mostly directed at a cold air pull source, in short he was able to duct the air more efficiently, and not use pressure ur are talking about... because i can bet you, the pressure difference between his inside and outside case hasnt changed much unless those are 4x80's are vantec tornado's which then he is now effectively using positive pressure and should even feel air bleed out the sides of his case.

hence why they work in great high dust enviorments if u have them on a filter... cuz clean air even gets blown out though all the holes, so your PC stays effectively dust free.

 

[fuzzymath10]

Interesting conversation happening here!

At low fan settings, I do feel heat coming from the cracks in my case. I also feel the hot air coming out in the expansion slots near my video card (as desired).

When I turn the fans up to max (3xNoctua NF-R8 + 1 Yate Loon, all around 1800rpm), the hot air coming out of the expansion slots is quite noticeable

 

[lehtv]

Hmm.. I've thought that negative pressure gets you better airflow because the air is sucked in and out better, instead of flowing all over the place. But I guess such streamlining only serves to make the air exit the case cold.

 

[joetekubi]

Thanks IdontCare for the great technical explanation of air flow.
As far as postive pressure cases go, I modded an old Antec tower into a positive pressure case, using several Scythe Sflex input fans sucking air in through a micron furnace filter, and no output fans. Left the case closed for a year, finally opened it up, and there was NO dust inside. That case was some real ugly mod, with lots of duct tape and such. Just rebuilt the rig into a new Rosewill mini-tower, sorry no pics yet.

But I am absolutely sure that positive pressure cases are not that hard to make and can work great.

 

[Michael Meio]

Positive airflow or not, you'll know if it's working for you by using this method:
(You need a lighter or something light enough to be visibly disturbed by small air current).
1. Locate a fanless perforation on your computer case.
2. Turn everything on.
3. (please be careful! do not burn your case) Move the lighter close to the perforation located on point 1. Please remember the lighter must be lit!

Simple put: if the air is flowing towards the perforation, your fans are set incorrectly which usually means you have more air output than input. If the air is blowing away from the perforation, then there is more pressure inside, which is the desirable way to have it.

When it comes to specialty fan settings and speed control via software, it is often impossible to maintain more pressure inside the case under full speed operation. This is applicable to water cooling systems like the ANTEC KÜHLER 920 (and pretty much all exhaust fan setup which happen to be controlled via software) because suddenly you will have a greater exhaust which will revert the pressure inside the case.

This is why it's recommended that under normal operation, there should be positive (more inside the case) pressure WITH an extra fanless opening that will supply air once the water cooling system's fan kick in. Otherwise, you will encounter lack of cooling efficiency on the water cooling system. Think of it as an overload vent... without it, the cfm fans are designed for will be quite restricted.

Good luck!