Case Airflow: Positive or Negative Pressure?

[darksidedragon]

That's a bit beyond me (or perhaps I "mis-heard" it in a lecture). Ah well, I guess I better learn that module properly before IC engines next year (although engines are generally much simplier)

 

[olgraybeard]

Was not relevant to the thread, was mostly being silly. But I actually teach heat transfer and fluid dynamics, but in an industrial setting, not college. And I have forgotten a lot of it because the level I teach at is not near the level I received in college.

 

[GeorgeH]

Positive pressure will increase the air density in the case. This leads to greater heat transfer from the heat generating components.

---

When heat is a concern, I just leave the side off the case and hang a couple 120mm fans that blow into the case. (Blowing in has longer range effects than sucking out.)

 

[T8RR8R]

I know what you mean about the window fan blowing directly on a component, sure it'll cool things better. Also there aren't 2 front intake fans on the Armor. Anyway like I said way back when before I did any of this stuff, whether or not it's positive or negative pressure doesn't matter as much as placement. So if you were to suggest that my not so scientfic study is thrown off a bit my fan blowing directly on components I would have to agree, but I will also say that would probably mean negative or positive air cooling doesn't make enough difference to really matter. Depending on your case, the size and shape of components inside, cable management and thermal paste, one combination or another may throw off results of any real solution that will work for all systems. So once again I will conclude that placement is more important that PSI.

 

[blue68f100]

My vote is negative with a twist.

I like the CPU fan to have shroud so it can directly pull outside air in. That way you get max cooling to the cpu.

Fans have the tendency to pull what they need. If you using positive pressure you are forcing air out the exhaust fans at a higher rate than normal, due to High static.

You get dust no matter which way you go. Filter restrict air flow and can knock a 120 mm fan down to the airflow of a 80mm.

Most fan have a max static the can push. So if you go overboard, no gain.

 

[senor_bob]

Senior_Bob says you can not get positive pressure nor an air flow front to rear in the CM830. I for one would disagree. Also, the Stacker only comes with 2 fans, a front and a rear. You could put a total of 9 fans on this monster, just do not know why you would (have to try that and see if my case can fly, hehe).

You missed the parenthetical (without side fans). I would like to see you try to get positive pressure in the Stacker with the optional side fans removed but the side still open. Now if you install the side fans yes you can get pos pressure.

By the way, I have a Stacker 830 and the front-to-rear airflow works well if the intake and exhaust fans are balanced, but try putting three front intake fans, one rear exhaust fan, and no side fans and see how much of the air just goes straight from the intake out the side.

@T8RR8R: I'll take your word for it (since you have one...)the Armor doesn't have two intake fans. I read this on TT's website and must have misinterpreted:

High efficiency ventilation: Dual 12cm silent fans in front & rear, 9cm fans in rear and top, 25cm fan on side

http://www.thermaltakeusa.com/product/Chassis/fulltower/armor/va8003bws.asp

 

[senor_bob]

Positive pressure will increase the air density in the case. This leads to greater heat transfer from the heat generating components.

While this is technically true, 120mm fans can't really generate enough pressure to substantially change the density of air.

 

[JonnyDough]

Based on comments in another thread, I now have a question about the ideal way to set up airflow through a case. It seems that there are two camps in this debate -- those that think the air pressure in the case should be positive, and those that think it should be negative. For the sake of clarification, I'll define what I think each of those terms mean:

Positive - there is more fresh air coming into the case than there is hot air exiting the case. Basically, the CFM from intake fans exceeds the CFM from exhaust fans.

Negative - there is more air exiting the case via exhaust fans than fresh air entering the case via intake fans and vents. The CFM for exhaust fans exceeds the CFM from intake fans.

Now I know that very few of us have taken the time to run silicon caulking on all the seams of our computers to make them completely air tight, so there is more involved here than just the fan CFM -- positive pressure cases won't explode over a period of time and negative pressure cases won't implode. Also, the direction of airflow within a case has a lot to do with the effectiveness of the cooling, but let's ignore that for now.

The real question is this, I suppose. Is it better to pull air through the system or to push it? Or does it not really matter? Anyone have links to some good testing on this, or some empirical data to back up theories?

EDIT: I turned this into a poll with a slightly different flavor. Balancing a system's airflow exactly is probably impossible, so when designing your fan setup, should you make it more negative, or more positive?

In response to OP. First of all your terminology is wack. Unless the case is sealed there's no negative or positive pressure. Pressure will be whatever the barometer outside reads, unless you live in an airtight house or your case is SEALED. "Pressure" is completely the wrong way of saying it. AIRFLOW DIRECTION (inward or outward) is what you're looking for I believe.

If your case WAS sealed (for whatever reason, most likely OCing to hit record benchmarks) you'd want LOW pressure. Pressure cookers trap heat, (this is how diamonds and other things in the earth's core are made too) and condense air molecules which keeps them from moving around as much. If you've got airflow you want the air to be able to funnel in and out.

Anyway...
Regardless of how you set up your fans, you can only pull in so much air and expel so much air. There will be a balance regardless of direction. It's called displacement. Look it up. Try dictionary.com.

 

[T8RR8R]

That's not true. Pressure happens in a case whether it's sealed or not, just that the more opening you have the less the pressure will maintain itself. For example if you start to fill a bike tire that has a hole in it the tire will start to fill up but only to the point where the pump(or a fan in our situation) will no be able to keep up with the amount of air being released. Also if you have a very small hole in a car tire you might be able to fill it up and drive around for a week before you notice it's starting to lose pressure. So my point is that something doesn't have to be sealed all that well to hold pressure. In our instance though, the amount of intake needs to be higher than exhaust for us to consider it as positive pressure and the amount of exhaust would need to be higher than intake for us to call it negative pressure. I know our cases aren't sealed but they do hold a certain amount of pressure when the fans are on.

Technically speaking, if our cases were air tight except for a where a fan was placed. Then after a few seconds the case would fill up with air and then there wouldn't be any more air going in because no matter how much the fan can push the air inside will eventually push back with equal force. Just the opposite would happen if we exhausted all the air out of an air tight case.

 

[arafay1]

always negative pressure

 

[Findail]

Anyway...
Regardless of how you set up your fans, you can only pull in so much air and expel so much air. There will be a balance regardless of direction. It's called displacement. Look it up. Try dictionary.com.

Well, thank you for the enlightenment. I wasn't aware that air could move without a change in pressure. Regardless of the terms I used, it appears that the majority of the folks reading this thread at least understood my intent, even if they didn't agree with my terminology. Apparently, you did as well but felt compelled to voice your disagreement with the semantics.

It's called condescending. Look it up. Try dictionary.com.

 

[SpeedyVV]

This has been a very interesting intelectual discussion and I enjoyed it, but I am a bit new at this.

Also I rather learn by doing, so maybe you can help me here ;-)

This is my case:
http://www.tigerdirect.ca/applications/SearchTools/item-details.asp?EdpNo=2337816&CatId=32

It has the following "holes":
1 on rear top for the PSU
1 on rear middle 120mm
1 on front bottom 120mm
1 on bottom side (for GPU)
1 on top side

While the Spec sheet says it comes with a 120mm fan it does not. So I went to buy 2 120mm fans.

So my question, is (n00b alert), I assume that once installed the front bottom fan should be blowing air in, and the rear middle should be blowing air out?

What about those 2 side ones? I am going to have to remove the top side cup thing, because I installed a aftermarket HSF that blow air from the front to the back of the case on top of the CPU. I am worried that the 2 side vents are going to mess up the correct flow.

Should I buy a small fan for the top side vent? It looks like a 80mm fan would fit. If so, should it blow out or in?

And last but not least, how does this fit into this +ve -ve pressure discussion ;-)

 

[Gneisenau]

It seems to me that using most of todays cases the chances of building enough positive pressure or negative pressure to measure is slight.
If having a negative pressure will reduce hotspots by sucking air in through all the little holes, then a positive pressure will do the same thing by blowing out all the little holes.

I don't think it makes any difference whether your case is slightly negative or positive in pressure. What counts is the quality of the air flow. Air properly circulating through the case will cool better than blowing huge amounts of air in or sucking huge amounts of air out.

If I had to choose one, it would be positive. (Dust, I hate it.)

 

[Gneisenau]

Given, what the placement of fans you have, and the ones you have already bought, I would do:
front: air in,
Rear: air out
Top: air out
PSU: air out
Side: air in. Two side air in would be better, but if you have to lose one to the hsf then so be it. The hole will probably pull air in to the hsf anyway.

 

[Findail]

This is my case:
http://www.tigerdirect.ca/applications/SearchTools/item-details.asp?EdpNo=2337816&CatId=32

It has the following "holes":
1 on rear top for the PSU
1 on rear middle 120mm
1 on front bottom 120mm
1 on bottom side (for GPU)
1 on top side

While the Spec sheet says it comes with a 120mm fan it does not. So I went to buy 2 120mm fans.

That's a pretty neat looking case. I think they got you with the wording on the fans, though the spec sheet does indeed say that one is included. At least this way you get to pick your own brands, though.

I'm with Gneisenau on the orientation of the fans. Just from the highly scientific Eyeball Method, it looks like that top duct (when removed) will mount an 80mm fan. A quick measure from hole center to hole center should tell you. The lower side vent, though... that thing is not square, and I think you'd have to do some custom work to get a fan on there. Not a big issue if you don't mind doing the chop-chop on a new case. If you'd rather not cut it up, it can serve as a vent.

As to how this applies to the topic, I'd say that your case was set up to have more exhaust than intake CFM. Looks like the manufacturer wants enough exhaust to draw air IN from the two side vents and the rear VGA vent. What kind of GPU did you put in it? Is it directly under that rectangular side vent? I suppose that fresh air couldn't hurt.

I'd set it up like this if I were in your shoes:
Front cooling port: IN
Rear cooling port: OUT
Top-side panel vent: 80mm fan blowing IN

I'd leave the rest of it alone, unless you run into some kind of funky overheating issues. Also, from comments earlier in this thread, having that side 80mm blowing onto your HSF should cause a bit of turbulence, which should help the cooling -- though I don't think we've empirically proven that yet.

 

[jitpublisher]

I agree with you that something doesn't have to be air tight to build pressure. On the other hand though, when you are talking about a car tire, you are talking about 35psi of pressure to fill that tire.
No simple case fan in anyones PC is going to be able to move enough air to negatively or positively pressurize anything. Pressurizing air requires a pump, that positively displaces and moves the air, or a fan design like a jet engine spinning at 30,000 rpm.
You can say "I have 2 fans blowing in, and 1 blowing out". One might say this is positive pressure. But in reality, the 2 fans blowing in will only blow as much air into the case as the 1 fan can exhaust. Although with 2 fans blowing in, the 1 fan WILL exhaust more air than if only 1 fan was blowing in, but there will be no amount of significant pressure change what so ever.

 

[tribalartgod]

the absolute easiest way that i have found to show which is better is with a shop vac. take a digital thermometer and hold it in place. have the shop vac blow air onto it. note the temperature. now, do the same with suction. the vacuum test shows that negative pressure is far superior to that of positive pressure.

just thought i would add my 2 cents.

 

[yzzib]

Both negative and positive airflow have their pros and cons.

Negative
Pros:
■Hot air is quickly taken out of the case allowing cool air to replace it.
Cons:
■This creates a lower pressure inside the case, the lower the pressure the harder it is for heat to transfer all together.

Positive
Pros:
■Cool air is always being blown onto specific components.
■This creates a slightly higher pressure inside the case which could help with heat transfer.
Cons:
■Warm air gets recycled in the case for a bit before being exhausted.

The pressure differences inside the case might not be enough to actually notice that much of a difference however, it is completely theoretical.

 

[Paperdoc]

I'm for positive BUT with an important condition: all your intake fans MUST have dust filters on them, and they gotta be kept clean.

Start by recognizing a few facts:
What goes in must come out, and vice versa. Total air exhausted will ALWAYS exactly equal air intake from all sources. If you don't supply it with fans, it will be sucked in anywhere, like case cracks and especially the slots on the front of DVD and floppy drives. You can't put filters there, so that air will always be bringing in dust. The only way you can control that is to ensure you supply more filtered air input capacity (with intake fans that include filters) than the exhaust fans can take out. Then any imbalance will be blowing clean air out of the cracks, instead of sucking in dusty air.

Although "hot air rises", that effect is important only in places where there are almost no air currents flowing. With all of the fan-forced air flow inside a case, the air will go where the fans blow it; the air temperature will have almost no influence on the air flow patterns.

WHERE the fans blow is at least as important as how many of them you have. You really want to place them (and air ducts, if you have them) so that you concentrate air flow over the major heat sources. For efficient heat removal from a hot surface, rapid air flow over the surface is more important than the air's temperature in the general vicinity.

Balancing the air intakes and exhausts is tricky, and cannot be done, really, from fan sizes or even air flow specs. What really screws it all up is the filters you need! Fan flow specs are done for totally unimpeded fans. But we deliberately (and for good reasons I support) impede intake fans with filters to remove dust, and never do that on exhaust fans. Thus the intake fans are doing significantly less than their specs and if you calculate a match of intake and exhaust specs, the truth is a mismatch that produces negative internal pressure. Other than creating a system to measure actual internal case air pressure (possible, but a bit tricky because that pressure differential is very small), the only tool we have is to deliberately design for excess air intake, install all the filters, and HOPE that the filter-reduced intake capacity is still a little more than the exhaust.

 

[foobaz]

I'm for positive BUT with an important condition: all your intake fans MUST have dust filters on them, and they gotta be kept clean.
. . .
Total air exhausted will ALWAYS exactly equal air intake from all sources. If you don't supply it with fans, it will be sucked in anywhere, like case cracks and especially the slots on the front of DVD and floppy drives. You can't put filters there, so that air will always be bringing in dust. The only way you can control that is to ensure you supply more filtered air input capacity (with intake fans that include filters) than the exhaust fans can take out. Then any imbalance will be blowing clean air out of the cracks, instead of sucking in dusty air.

This is a true observation and I would agree that positive pressure is desirable. I worked for a company that developed software for the USPS. The software was run on expensive servers right inside the sorting room right next to huge mail sorting machines. I'll never forget the time a field tech returned from a postal site with several defunct, expensive at that time, CD-R drives. They were completely caked with what looked like flat white paint. They were completely coated inside and out with paper dust from the sorters. Clearly the server case had negative pressure that sucked in particles of paper dust right through the cracks in the removable drive. A strong argument in favor of positive pressure I'd say.

WHERE the fans blow is at least as important as how many of them you have. You really want to place them (and air ducts, if you have them) so that you concentrate air flow over the major heat sources. For efficient heat removal from a hot surface, rapid air flow over the surface is more important than the air's temperature in the general vicinity.

Also true. I remember discussing case cooling with a vendor who told me that the case company he represented had done engineering studies that showed that the internal structure and placement of fans and components were key to effective cooling. They placed thermal sensors in many locations in the case to do a complete temperature profile. They eliminated structures that impeded air flow and optimized the placement of components as a result of the tests. I think any case manufacturer worth its salt should do the same. Every case is different. I wish I could recall the name of the manufacturer that did this. Turbulence is a major factor when it comes to air flow. It can seriously impede the flow. Any components directly in the path of the air flow can cause turbulence.

the only tool we have is to deliberately design for excess air intake, install all the filters, and HOPE that the filter-reduced intake capacity is still a little more than the exhaust.

You can easily determine if you have achieved this by observing which way smoke from a burning incense stick is going -- in or out -- of various cracks and crevices of the case.

 

[delxiii]

If you ever get the opportunity, look inside a rackmount server case, esp 1 or 2U, you will notice negative pressure, and this is based purely on airflow, from front to back, as that is how all rack cabinets in data centres are designed and placed, cool air in the front, hot air out the back. That is thanks to the huge amounts invested in cooling and hot air extraction of these controlled enviroments. BUT!!! Look at most industrial chassis, those designed to work in dusty places (take cement works or railway platforms as an example, none controllable enviroments) then they are all designed as positive pressure systems. Positive pressure systems ALWAYS have filters on the intake fans, always, no exception. This creates an abundance of cool air inside the chassis which has to find a way out. One way is via the exhaust fan, another is through every other crack and cranny in the case. Through the drive openings, screw holes if not sealed, PSU, forcing cool air through it.

I used to work for an industrial PC maker, and all except our 1 & 2U offerings were positive pressure. We had systems come back for routine maintenance after being on a weighbridge of a well known "fortified" cement supplier for 12 months. Outside, they were disgusting!!!! Inside, like the day they left the workshop. Others that came in were "shoebox" type systems , used on public transport, gps, cctv recording etc. all fixed in public areas, under seats etc. and they were just the same, spotless inside.

Make from that your own conclusion. All I know is that even when I have bought a "boxed" product PC, I have made modifications to the case, with filtered intake fans, to increase the internal pressure. And to date, fingers crossed, I have never had a PC die on me due to heat or crap inside.

 

[iode]

I believe that positive airflow is the way to go (huh, that rhymes). Here's my conjecture to support my opinion:
We have pretty much all agreed that we're not talking about a completely sealed case, so therefore, any positive or negative pressure will not cause the case explode or implode, respectively. The air will likely balance itself out through other means (little nooks and crannies) either way.
Now, if you have negative airflow, there is more air being output than input. The argument in this case is that the hot air buildup in the case is more quickly removed by the high air output than by a positive airflow setup. Furthermore, the outside (colder) air will enter anyway through the seams and holes riddled throughout the case. Therefore, the negative airflow is better than positive, where a positive configuration seems to input a lot of outside air, but fails to expel it as quickly. And since the negative airflow gets its outside air anyway through the means mentioned above, the negative airflow must be better right? Wrong. Unfortunately, there is a flaw to this argument.
Again, the basis of this discussion is to ultimately find out the best configuration to DECREASE the temperature of SELECT components (cpu, gpu, etc.) that is, after all, the idea of computer cooling is it not?
Anywho, this is why I believe a positive airflow is better. With a positive airflow, you have many fans strategically placed to blow outside (and thus cooler) air at components which need it most critically (and will therefore, be the most effective). This beats a negative airflow setup which relies on the nooks and crannies of the case to provide the outside air. Even then, the nooks and crannies are NOT strategically placed to cool the SELECT components on the motherboard. Therefore, in order to match the efficiency of the positive airflow (having fans directed at critical components) the negative airflow setup will have to work much much much harder, incorporating more fans in its setup, whereas the positive airflow (with its superior efficiency for reasons described above) will only incorporate a fraction of the fans to provide the same end results. Another prevalent argument against a positive airflow is that it will fail to expel the hot air inside the computer casing quickly enough, therefore, being a detriment to the cooling configuration. This point, though insightful, is false. As with the preordained circumstances, the case can work the same way by expelling the hot air through the seams and holes in the case. If air can travel into it as in a negative airflow setup, why can't it be pushed out of it in a positive airflow setup? It can. And in a positive airflow configuration, the hot air buildup of concern, though not able to expelled as quickly by the fan blowing outwards, is easily dissipated by the nooks and crannies of the case.
With the negative airflow setup, the majority of the cold air is coming in from the nooks and crannies, but the cold air is hardly put to good use because they are not strategically aligned to critical components of the computer. With the positive airflow, however, the nooks and crannies are used to remove the hot air inside the case, and who cares if the hot air is not strategically removed. It's hardly important. For the reasons listed above, this is why I firmly believe that a positive airflow configuration in a desktop is ultimately the most effective in providing an optimal computing performance.

 

[keith93]

Surely if you get enough negative pressure you don't have to worry about dust as it would get sucked straight through...
Just a thought.

 

[Paperdoc]

Surely if you get enough negative pressure you don't have to worry about dust as it would get sucked straight through...
Just a thought.

Not quite true . Even with extreme air flow rates, a bit of dust will still collect inside the case, But it might well be so small nobody would care. To get that flow, though, would require enormous and noisy fans sucking air out. I still think it is cheaper, quieter and cleaner to use enough supply fans with intake filters to ensure the case is under tiny positive pressure, forcing air to flow OUT of the cracks and uncontrolled leak points.

 

[yoshx]

GUYS i take the copy right for this ,here is my invention, get a nylon plastic bag , put ur computer in it , then make 2 wholes on the bag ( should be sealed ) , so the 2 external holes will connect to 2 pipes, 1st pipe is where the air goes in 2nd out , now at the end of pipe u going to put 2 fans , direction should be the same so one will take in the air one will take it out, important the first fan should have the dust filter so u save ur computer , the nylon will seal the box so no air will goes in or out without the fans, sorry for my English but i believe that this improvisation can simulate good circumstances that will benefit ur computer all u need to take care about how cold is the air what will going into the sealed nylon box, Good luck and don't forget this is my invention and i copyrighted and any other modification of my box is copyrighted too ! honestly i need lots of money ciao