Positive Air Pressure?

[Mower]

Adjusting my case cooling and wondering if this a valid issue, positive versus negative air pressure in the case...positive would force dust out while negative could pull dust in.

However, being that my new case is a Corsair 500R and is pretty leak proof, do I even need to worry about air pressure? Maybe I only need to focus on good airflow through?

 

[SMOKEU]

Positive pressure is better than negative pressure, as you should be able to put dust filters in front of your intake fans. With negative air pressure, you don't have as much control for the incoming air, and you'll end up with more dust for the same amount of air flow.

 

[doyll]

/\ /\ /\ /\ What he said.

 

[magellan]

Considering how leaky a modern case is, the whole idea of any significant positive pressure at STP conditions with muffin fans sounds ludicrous. Has anyone actually used a pressure gauge to measure case pressure? I'll bet it would be hard to measure any increase whatsoever.

 

[doyll]

Positive pressure = air leaks out of case.
Like positive pressure in room or building to keep dust contaminates out.. open door and air flows out.

Negative pressure = air leak into case.
Like negative pressure in containment lab to keep contaminates in.. open door and air flows in.

 

[d1nky]

ive had both set ups, had around 6 intake to 3 exhaust and then switched it to 3 intake and 6 exhaust.

I now have it in negative pressure 5 high rpm exhausts 3 intake.

I didn't see one bit of difference in the amount of dust!

theres a good way to keep dust out! have the surrounding area clean, rig off the floor and away from carpet etc

 

[magellan]

On the two rack mount servers I'm familiar with, the IBM x3550 and IBM blades, there are no chassis intake fans. All the installed fans are purposed to cooling either the CPU or the DIMM's and have baffles to insure that's exactly where the airflow goes. The only exhaust fans would be those found in the PSU's. Of course, these servers normally live in nearly dust free, continuously air-conditioned environments.

On my old case, I had a filtered, powerful 120mm x 38mm fan as an intake in the front bottom of the chassis. The thing is it never seemed to make any difference in my CPU
or GPU temps whether it was on or off.

 

[EarthDog]

There is a HUUUUUUUUUUUUUUUUUGE thread on this subject (will dig it up...LINK!) already. In that thread it has gone baaaaaaaaaaack and forth for YEARS of people claiming to know the right answer but in reality, we do not have one. As far as dust goes, positive pressure (and I use the term pressure VERY loose as a PC case is not remotely sealed so the pressure changes are really not much it is airFLOW that is key) does prevent air from coming in the cracks versus negative pressure that will allow that to happen.

The generally accepted rule of thumb is this:

Front/Sides = intake
Top/Rear = exhaust

AirFLOW is the KEY to good temps. Now if you ask me what MY belief is for more intake than exhaust... I say a bit more exhaust than intake. I could care less about dust (I have filtered intakes) in the cracks. What I *think* this does is prevent any 'hotspots' from building in the case since you are always taking out a bit more air than coming in. Again this is negligible as well. The real answer to me is to try that rule of thumb and play with the airflow and see what works best for your case and setup.

As far as servers go, magellan, I *think* you and I talked about that in the recent past as well. A rack mount server and especially a blade chassis is soooooooooooooo much different than a computer case that, IMHO, it isn't even worth discussing how they work since their internal properties are vastly different. They also, generally, do not intentionally/directly cool DIMMs either. Ram doesn't need it. The fans in those chassis are lined up across the width of the server IIRC and will put some air there, but, even that much ram will not get warm even when they are being beat on.

 

[magellan]

IBM System x3550 Type 7978User’s Guide:

"The server supports up to six hot-swap cooling fans. The server comes with five fans that service the primary microprocessor and the DIMMs."

You claim positive pressure, all I want to see is proof that anyone is actually
achieving this. Let's see proof that the pressure in a computer case is actually
above 14.696 psi. That's what positive pressure means right? You seem to be
trying to re-define it in terms of airflow.

 

[EarthDog]

I said this specifically... we are on the same page.

(and I use the term pressure VERY loose as a PC case is not remotely sealed so the pressure changes are really not much it is airFLOW that is key)

Airflow is really what we are talking here (as mentioned above). To be clear, I mean more intake than exhaust airflow. People use, as the thread is titled, Positive and negative pressure terms so I stuck with that while (trying) to clarify that the pressure differences are negligible at best.

As far as the server.. great. Remove the fans in front of the DIMMs and watch that thing run until its lifecycle is complete. DIMMs do not need actively cooled (outside of perhaps a blade environment). But it seems you are missing the point or at least going back to something that has little/nothing to do with PC cases. Let's just drop the semantics on the off topic server stuff. But, if you would like to talk about it, my PM box is always open.

 

[magellan]

I see your point about air flow, but if this is the deciding factor, then why mention positive pressure at all? If I sell a new car, I can't advertise it as such w/the caveat of using the term "loosely". Either the case is positively pressurized or its not.

 

[EarthDog]

...then why mention positive pressure at all?

Again, I said...

People use, as the thread is titled, Positive and negative pressure terms so I stuck with that while (trying) to clarify that the pressure differences are negligible/non existent at best.

 

[Soulcatcher668]

Look at it like this.

Take a single fan. They create airflow by making a negative pressure area behind them, more air rushes in to fill that gap.... presto... airflow.

We are not talking about 1000psi oxygen tanks here.

By the nature of how fans work, if you have more fans blowing into a case than out, you have created a positive pressure environment where the positively pressurized air exits through all the gaps in the case.

The same thing works in the reverse.

Pressure is pressure regardless of how much or little of it there is.

 

[doyll]

The pressure differential at play here is the same as is in play in our weather. Look at a barometric pressure weather map. High pressure areas push air into low pressure areas. The resulting effect is wind.

Case pressure or lack of same has the same effect.
Low pressure in case and higher pressure outside pushes / leaks air into case.
Higher pressure in case and air pushes / leaks out of case.

 

[RGE]

You can indirectly see the effects of positive or negative case pressure in terms of air resistance.

This is a thread I started back in 2009 on Xtreme.

Bottom line, I have similar temp testing equipment to vapor and martin, but always measured 40% less rad cooling, compared to both of them, even using same testing equipment, same rads, etc. Then after vapor confirmed they do all testing with external rads and no filters and that air resistance was really that high....(I had 6 fans in push/pull all pushing air inside the case via 6x120 rad, and only 1 exhaust fan)....I then removed my air filters and gained 10% back (already knew that). Then I removed both side panels on case so equivalent of external rads and then was within 5% of vapor/martins testing with similar rads/fans/specs. Ie, I was losing 30% cooling from current configuration.

Adding a second rear exhaust improved cooling to only losing 25%. Adding some vents in back did nothing, nor adding a fan behind them, sort of hit diminishing returns.

I lose 10-11% cooling from air filters, and another 15% from pushing air into a case, so I always know I need 25% more cooling than vapor/martins with current filters/configuration.

 

[magellan]

Look at it like this.

Take a single fan. They create airflow by making a negative pressure area behind them, more air rushes in to fill that gap.... presto... airflow.

We are not talking about 1000psi oxygen tanks here.

By the nature of how fans work, if you have more fans blowing into a case than out, you have created a positive pressure environment where the positively pressurized air exits through all the gaps in the case.

The same thing works in the reverse.

Pressure is pressure regardless of how much or little of it there is.

If you have more fans blowing into a case than out is absolutely no guarantee you have positive pressure in a case. It all depends on how leaky the case is. If I have a bucket with a large hole in it that leaks at a rate a of 5 gallons per minute, it doesn't make a difference if I fill it at 2 gallons per minute or 4 gallons per minute, it still won't fill.
Similarly, if a case has a capacity of 9 cubic feet, but leaks at a maximum rate
of 300 CFM, you're going to be needing some powerful fans to overcome that
leakage and introduce any pressure at all.

 

[doyll]

If you have more fans blowing into a case than out is absolutely no guarantee you have positive pressure in a case. It all depends on how leaky the case is. If I have a bucket with a large hole in it that leaks at a rate a of 5 gallons per minute, it doesn't make a difference if I fill it at 2 gallons per minute or 4 gallons per minute, it still won't fill.
Similarly, if a case has a capacity of 9 cubic feet, but leaks at a maximum rate
of 300 CFM, you're going to be needing some powerful fans to overcome that
leakage and introduce any pressure at all.

Sorry but incorrect.
* if case leaks air out it has more pressure inside than outside.. Positive pressure case
* if case leaks air in it has less pressure inside than outside.. Negative pressure case

Water comparison is not applicable.. unless you submerge the case in water, it is full of water and you add / remove water from case.

 

[EarthDog]

Sorry but incorrect.
* if case leaks air out it has more pressure inside than outside.. Positive pressure case
* if case leaks air in it has less pressure inside than outside.. Negative pressure case

Water comparison is not applicable.. unless you submerge the case in water... it is full of water and you add / remove water from case.

+1. To add to that, air never 'goes away' like the water would in your (magellan) example, it is replaced by intake air or air from cracks, but the amount of air does not change for all intents and purposes.

 

[marjamar]

OK, what's this thread about? Other then alot of rambling, it seems sorta like people are debating which is better, pushing air in, or drawing air out of a case. From any practical aspects I can imagine, where any heat is being generated, better to exhaust that heat out as close to the source as possible. Any place else, it would be better to inlet air. I would also surmise the cfm going into the case, would help augment the discharge of component heated air leaving the case. How could any other way then this be advantageous?

As a contractor, I can tell you added air volume in one enclosed space, will move that excess volume into adjoining space, or any open space it can find. I can also tell you that cold air is more dense and heavier then hot, which would cause a potential for more air to come into the case at a lower level as the warmer air within the case would tend to leave the case at a higher level. For this reason, moving heat away and out of the case, would by necessity, bring cooler air into the case, given there is a colder potential difference between the heat source and ambient air outside the case. It would further mean that fans would work better drawing or pushing air into the case at a lower point, and fans drawing or pushing air out of the case should be at a higher point. This would have a higher potential for better cooling the case interior then if reversed.

Therefore, it is always wise to have more fan power pushing/pulling air in the case and at a lower point, then pushing/pulling air out of the case and at a higher point. And regardless of how leaky any case is, when more volume of air is induced into that case, there will absolutely be a pressure differential with a higher pressure in the case the outside the case. Like the poster earlier with the water idea, you could also say water always runs downhill, unless and added volume of water is induced and forces it up and over obstacles it finds in its path. In physics, all matter seeks to stabilize itself with all surrounding matter. Working with it, is easier then working against it.

 

[EarthDog]

LOL mar... thanks for your two cents added to the pile . Like I mentioned from the get go, there was already a 487 post thread about this and we are about to get into the same. damn. bull****. again. Nobody here has proved anything frankly. Like in that other thread, we are just spinning our wheels. You essentially agree with what I feel though and the general consensus...

Front/sides = intake
top/rear = exhaust

As far as if you should have more intake than exhaust that depends.

This physical setup helps with the natural properties of convection/heat rises. AirFLOW is always the key regardless of this never ending discussion about 'pressure'.