Fans and case pressure

[biggoofy]

Here and there I hear about people putting a bunch of fans in there case to increase cooling, but with that I also hear alot of talk about case pressure and terms like that. I was under the assumption that if you have, for example, 1x 120mm fan blowing in then you should have one blowing out, so on and so forth. Is this what they mean when talking about air cooling and case pressure?
Thanks in advance.

[Trace]

Yes, but you can also do Positive pressure or negative pressure.

[SXRguyinMA]

^^ exactly. I've seen a lot of talk and it doesn't really seem that one works better than the other (if someone has definitive proof show us!) Basically Positive airflow would be 2 120mm fans blowing in and 1 120mm fan blowing out. Negative would be 1 120mm blowing in and 2 120mm fans blowing out. Basically as long as you have air moving things will cool properly

[mDust]

I think the further you get from neutral in either direction, the less efficient the overall airflow would be and thus less heat moving from the components to outside the case.

e.g. 4 intake, 0 exhaust = bad; 0 intake, 4 exhaust = bad; 2 intake, 2 exhaust = good.

I've heard though, that slight positive pressure keeps a cleaner case as air is forced out the nooks and crannies instead of drawing dust into the case.

e.g. 3 intake, 1 well-placed exhaust.

However, if you have pets, you smoke near the computer, or have an otherwise unclean environment you can forget about a clean case and should aim for neutral airflow.

[Couch3ater]

Bear in mind, most case manufacturers ship their cases with an 80mm intake and a 120mm exhaust. Lower pressure means lower temperature.

[x88x]

However, if you have pets, you smoke near the computer, or have an otherwise unclean environment you can forget about a clean case and should aim for neutral airflow.
One thing I always recommend for smokers is to reverse the normal airflow (make it rear-intake/front-exhaust instead of front-intake/rear-exhaust). This will likely get more dust in your PC, but better dust than smoke...ever tried to clean cigarette smoke grime off a heatsink? ...don't... :dead:


Bear in mind, most case manufacturers ship their cases with an 80mm intake and a 120mm exhaust. Lower pressure means lower temperature.
Not really, no. Lower pressure means less air moving through the case, which means less heat can be dissipated, which means higher temps. In that arrangement you're probably got a fair bit of negative pressure.

Another thing to keep in mind in addition to the number of fans is the airflow each of those fans creates. For example, I have 6 140mm 60cfm fans on the top of my case pulling air out, so I should have both the front (6x 120mm 40cfm) and rear (2x 120mm 40cfm) fans pulling air in (currently the rear is pushing out...I need to fix that).

[mDust]

Another thing to keep in mind in addition to the number of fans is the airflow each of those fans creates. For example, I have 6 140mm 60cfm fans on the top of my case pulling air out, so I should have both the front (6x 120mm 40cfm) and rear (2x 120mm 40cfm) fans pulling air in (currently the rear is pushing out...I need to fix that).
Good point. I was assuming identical fans throughout because that's what I generally do, but that's not always the case. So total CFM (or CMS, for you metric weirdos in the rest of the world :D) in vs. out would be a more accurate measurement.

[slaveofconvention]

Big thing to bear in mind regarding dust.... Positive pressure will mean that almost all of the air entering the case will do so via the fans. If you then put filters on them, it will help keep the case cleaner inside. Negative pressure will result in air being drawn in thru vents, gaps, anywhere it can so filters will be much less effective. Oh and don't forget your PSU most likely has a decent size extractor fan on it when trying to figure out your balance.

Cheap and easy way to check if you have positive, negative or balanced air pressure. Get a sheet of bathroom tissue, tear it down to about an inch square, single ply (if you're posh and use multi-ply tissue) so it's small and weighs as little as possible. Turn on the system, and place the tissue on a vent on the side (if you have one - any gap will work as long as it's horizontal and away from a fan) - if it blows off, there's positive airflow. If it stays put, you have negative. If it just lazily falls down, you're close to neutral....

[x88x]

Oh and don't forget your PSU most likely has a decent size extractor fan on it when trying to figure out your balance.

That's one of the things I love about my case. PSU fan pulls air in the bottom, through the PSU, and straight out the back, so it doesn't affect my airflow at all. :D

[Couch3ater]

LOL, I was thinking air pressure. >_<; Fail. Hahaha, I was thinking Charles Law. --facepalm--

[SXRguyinMA]

don't forget most higher-end video cards vent directly out the back as well. my current setup has 3 120mm intake fans, the 140mm fan on the PSU working as an exhaust, and the fans on both of my 3870x2's acting as exhaust as well as they blow right out the back. One of my 120mm intake fans is on the side panel blowing directly onto the CPU cooler, so the front 2 120mm fans pretty much supply cooler air to the memory cooler, hard drive and video cards.

The Tempest EVO case I'm working on will have 1 120mm side intake, 2 120mm front intake (or maybe just 1, we'll see what happens), the PSU is bottom mounted so it won't count, and it will have the 2 video cards exhausting as well as 2 140mm fans on the top exhausting.

The other thing to think about besides CFM is fan speed. All of my fans (including the CPU cooler, but not the video card fans, are hooked to my NZXT Sentry 2 fan controller, which automatically changes fan speed based on temp. The video card fans do this by themselves, as well as the PSU fan. So with all the fans in the case constantly changing RPM based on load/temp, that will change your airflow significantly. For instance, at idle, most of the fans in my case are running about 1/2 speed, save the VGA and PSU fans, which are maybe 20-30% speed. Under gaming conditions, the 2 fronts get to 90-100% speed, the VGA fans are constantly changing, and well I'm not really sure how much the PSU fan changes because I never pay attention to it lol (Ultra X3 800w PSU)

[biggoofy]

Thanks for all the feedback. I just needed to know because I'm going to have a 120mm intake on the front of the case, an 80mm intake on the side and an 80mm as an exhaust on the back. I want to watercool just because I have an 8800 GT that gets super hot. Anybody know how to keep that thing cool?

[mDust]

Thanks for all the feedback. I just needed to know because I'm going to have a 120mm intake on the front of the case, an 80mm intake on the side and an 80mm as an exhaust on the back. I want to watercool just because I have an 8800 GT that gets super hot. Anybody know how to keep that thing cool?

A full coverage water-block and H20. Water will be able to pipe the heat away more efficiently than air so it will likely idle a few degrees cooler and won't spike as high.

[biggoofy]

Yeah I've been looking into it trying to get some used watercooling parts off the forums here.

[Konrad]

I know this post is a bit of a necro, sorry.

Question: in a positive-pressure system, does the heat generated by the fans themselves contribute to overall temp? It's not much, but still a 12V/0.15A fan is 1.8W of heat, so say 3 intake fans would continuously add another 4-6W of undissipated heat into the system. I would guess that this heat gets spread all around the interior of the case, but these fans tend to be placed exactly in the spots you don't want heat to accumulate. Does this matter in practical terms?

[Trace]

Not all of that input energy is converted into heat. A majority of it is transferred into mechanical energy. I.e. the spinning fan and moving air.

[mDust]

I know this post is a bit of a necro, sorry.

Question: in a positive-pressure system, does the heat generated by the fans themselves contribute to overall temp? It's not much, but still a 12V/0.15A fan is 1.8W of heat, so say 3 intake fans would continuously add another 4-6W of undissipated heat into the system. I would guess that this heat gets spread all around the interior of the case, but these fans tend to be placed exactly in the spots you don't want heat to accumulate. Does this matter in practical terms?

Waste heat is generated due to inefficiency. It's not possible to have 100% efficiency so there will always be wasted energy in any 'system'. However, the entire amount of power to a device isn't wasted as heat...usually it's only a small portion of that energy. If a fan worked at 0% efficiency then it would generate that 1.8W of heat...and the manufacturer seriously screwed up at some point. I don't know the efficiency of a generic case fan, but it will never produce enough heat to affect the ambient temp of your case, so no worries.

[knowledgegranted]

You guys are missing something, another example. Quantify the problem a little bit larger, and make it into something we can measure easily. My solution is a hot summer day and a small 10' x 10' room with three windows.

1. No airflow = sweating, to a point where it's dripping off of my nose

2. one fan intake, two outake = Great airflow, little breeze temperature decreased dramatically, and no perspiration

3. Two intake, one outake = Excellent airflow, increased breeze. However the temperature and humidity is still taking its toll. I can feel the perspiration film on my skin

4. All outake = Awesome airflow, can almost feel the air movement throughout the room, humidity feels as if it's gone. A comfortable temperature; the air is basically "pulled" through the room

5. All intake = Superb airflow, where the air stream is felt, the air is cool. However stepping outside of the airflow brings a humidity sweep

[SXRguyinMA]

well now you're getting into wind chill. wind chill does not effect inanimate objects (i.e. not alive). The reason for this is sweat. when the wind blows across your skin it makes your sweat evaporate (or any moisture on the surface if you're not sweating) and gives the sensation of cold (evaporative cooling). Now a computer doesn't sweat, so that doesn't apply in quite the same manner.

When I was testing my arduino setup for my louvers for project tempest, I worked the temp sensor by placing it in my mouth (hot) then putting it in front of a fan. The fan created wind chill (and thus evaporative cooling) and the louvers closed quite a bit, then as any saliva evaporated, the temp sensor warmed to room temp (well air temp coming through the fan from outside) and the louvers opened back up slightly.

[knowledgegranted]

well now you're getting into wind chill. wind chill does not effect inanimate objects (i.e. not alive). The reason for this is sweat. when the wind blows across your skin it makes your sweat evaporate (or any moisture on the surface if you're not sweating) and gives the sensation of cold (evaporative cooling). Now a computer doesn't sweat, so that doesn't apply in quite the same manner.

When I was testing my arduino setup for my louvers for project tempest, I worked the temp sensor by placing it in my mouth (hot) then putting it in front of a fan. The fan created wind chill (and thus evaporative cooling) and the louvers closed quite a bit, then as any saliva evaporated, the temp sensor warmed to room temp (well air temp coming through the fan from outside) and the louvers opened back up slightly.

Thank you for the clarification I probably should have stated that. It was just an example for people to test their theories.

[Konrad]

I didn't consider fan efficiency, yeah it's obviously true that they don't convert all the electrical power into radiated heat (the way a resistor would), plus of course their purpose is to introduce airflow. My dumb.

What effect is gained by placing two (presumably identical) fans back-to-front, so airflow is forced through both as a single unit?

Does this (nearly) double the CFM airflow or static pressure? Terribly increase noise? (As compared to using both fans in a more conventional side-by-side arrangement.) Would a single double-thickness fan be a better choice?

[x88x]

What effect is gained by placing two (presumably identical) fans back-to-front, so airflow is forced through both as a single unit?

Does this (nearly) double the CFM airflow or static pressure? Terribly increase noise? (As compared to using both fans in a more conventional side-by-side arrangement.) Would a single double-thickness fan be a better choice?

IIRC, there are special fans designed specifically for this. It'll work with any fans, but sometimes you'll get audible resonance, occasionally you'll get physical resonance.. IDK, my personal stance has always just been that if you need to do that, you need a better fan. :P

[Konrad]

Actually, the reason I asked was that I'm considering the option of putting another 120mm fan overtop the 120mm built into my PSU; it happens to be the only fan in this system and the only channel for exhaust. I'm willing to swap the PSU fan so both are perfectly matched for rpm/etc (don't want to decrease efficiency or increase noise with mismatched audio resonance) but I'd prefer to keep the PSU form factor as-is (and removable) instead of extending it with an integrated double-thick fan.

I've thought about using a ducted method, basically a tube with fans at the ends, but thought this probably won't be as airflow-efficient as mating the fans directly.

So do you still advise a single better fan?

[x88x]

I would have to see the setup, and I am by no means an expert on the topic, but yeah, I would think that you would get more airflow with less noise with a single beefier fan (eg, Scythe Ultra Kaze).

[Luke122]

^ +1. Upgrade the fan in the PSU to a higher CFM rated unit, and you'll see a benefit with little change in noise.

[Konrad]

Two votes yes is good enough for me. Thanx ;)