Case fans and air pressure

Control: management of fans, temp/rpm monitoring via soft/hardware

Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee

Post Reply
Hatch
Posts: 4
Joined: Mon May 07, 2007 7:27 pm

Case fans and air pressure

Post by Hatch » Tue May 08, 2007 8:08 pm

I'm curious about the negative case perssure that seems to be the norm. Typically you find more CFM out of the case than coming in. This results in a negative gradient of pressure from the outside of the case to the inside.

Why is that? I know in industrial "clean room" environments the opposite is true. Positive pressure is usually maintained so that no contaminants are drawn in. Is there a thermodynamic reason for the choice to maintain a lower pressure inside the case than out?

derekchinese
Posts: 255
Joined: Fri Mar 02, 2007 6:57 pm
Location: Nowhere.

Post by derekchinese » Tue May 08, 2007 9:34 pm

I am not sure why. I would guess that it is because silent pc'ers want to use as few fans as they can and see exhaust fans as more important than intake fans resulting in more exhaust then intakes and therefore negative pressure.

What I have heard confirms what you have said, negative pressure leads to a really dusty case. Intuitevly it does not make sense to me, but everyone tells me that this is how it is.

Derek

jaganath
Posts: 5085
Joined: Tue Sep 20, 2005 6:55 am
Location: UK

Post by jaganath » Tue May 08, 2007 11:21 pm

Typically you find more CFM out of the case than coming in.
how can this be? surely 1 CFM in = 1 CFM out?
Is there a thermodynamic reason for the choice to maintain a lower pressure inside the case than out?
no, that was just the way the ATX standard was laid out. Many SPCR members (for ex. Bluefront, Felger Carbon) have had superb results with positive pressure.

disphenoidal
Posts: 333
Joined: Wed Apr 05, 2006 9:54 pm
Location: USA

Post by disphenoidal » Tue May 08, 2007 11:33 pm

I think the reason negative pressure cases accumulate more dust is a negative pressure case will pull air in from any openings. A positive pressure case only pulls air in through the fans, where dust filters can easily be installed. Filtering negative pressure cases is harder, because every opening is an intake.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Wed May 09, 2007 1:41 am

Yeah it is much harder to keep negative pressure cases clean. When I ran negative pressure I had to seal every little crack around the I/O ports, every opening on every card I used, the back of every optical drive, the cracks between the external drives, etc. Practically impossible..... But when you go positive pressure, keeping the computer clean is easy. Unless you use PSUs with exhaust fans, you do not need any exhaust fans on your case. It's simply a matter of having, or installing vent holes in the right places. Unfortunately the ATX standard of most cases, is not setup for positive pressure......modding is the solution, although some people seem to get by without any cutting.

NeilBlanchard
Moderator
Posts: 7681
Joined: Mon Dec 09, 2002 7:11 pm
Location: Maynard, MA, Eaarth
Contact:

Post by NeilBlanchard » Wed May 09, 2007 1:53 am

Greetings,

Do you think the air that a fan blows into a case is cleaner than the air that gets sucked into a case?

It takes filters to clean the dust out of the air. All else being equal, the same amount of dust & air will move through a case, no matter how you have your case set up.

hubs99
Posts: 4
Joined: Thu Apr 26, 2007 5:18 am

Positive pressuve vs Negative Pressure

Post by hubs99 » Wed May 09, 2007 5:38 am

Yes, unless you filter the in-blowing air there will be the same amount of dust brought in by the fans. However, you also have to consider electro-magnetic attraction of dust into the case. Depending on your placement of fans the Negative pressure allows this dust to be attracted to and then into the case through all your crevices.

Ideally to make the case dust free you should have a filtered-input positive pressure case. This way all the air getting into the case will be filtered and hopefully dust free.

jhhoffma
Posts: 2131
Joined: Mon Apr 25, 2005 10:00 am
Location: Grand Rapids, MI

Re: Positive pressuve vs Negative Pressure

Post by jhhoffma » Wed May 09, 2007 6:40 am

hubs99 wrote:electro-magnetic attraction of dust into the case.
The word you're looking for is "static". :P
hubs99 wrote: Ideally to make the case dust free you should have a filtered-input positive pressure case. This way all the air getting into the case will be filtered and hopefully dust free.
Or have a completely sealed case with no airflow, like a Zalman TNN case, with no fans and a completely passive or external PSU. But this is not practical, so I digress...

qviri
Posts: 2465
Joined: Tue May 24, 2005 8:22 pm
Location: Berlin
Contact:

Post by qviri » Wed May 09, 2007 6:44 am

NeilBlanchard wrote:Do you think the air that a fan blows into a case is cleaner than the air that gets sucked into a case?

It takes filters to clean the dust out of the air. All else being equal, the same amount of dust & air will move through a case, no matter how you have your case set up.
This is obviously true, however, all else is very much not equal, as it is significantly easier to filter all of the incoming air with a positive pressure setup.

Hatch
Posts: 4
Joined: Mon May 07, 2007 7:27 pm

Post by Hatch » Wed May 09, 2007 10:29 am

jagnath, yes, ultimately CFM out = CFM in. Often you see multiple fans moving air out of a case and fewer (sometimes none) fans moving air into a case. And I'm simplifying greatly, I realize, since different sizes and speeds all have a bearing on the actual numbers. In actuality, a steady state is reached where outflow = inflow no matter how many are installed or which way they're pointing. WIthout equilibrium being reached you'd end up with a total vacuum inside!

Bluefront, your description is what I have always seen. Every little nook and cranny eventually gets packed with dust, right down to the space around the tray button and LEDs on my CD/DVD drives.

Felger Carbon
Posts: 2049
Joined: Thu Dec 15, 2005 11:06 am
Location: Klamath Falls, OR

Post by Felger Carbon » Wed May 09, 2007 11:11 am

NeilBlanchard wrote:Do you think the air that a fan blows into a case is cleaner than the air that gets sucked into a case?
Neil, I believe Bluefront's point was, you can put one filter on a fan that blows into a case, and dust = gone. With air being sucked into a case, you need to put a filter on every hole - not practical. :)

Rusty075
SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ
Contact:

Post by Rusty075 » Wed May 09, 2007 12:29 pm

Disregarding dust for a moment, the differences between negative and positive pressure case airflow schemes are really pretty small, assuming that each is implemented with similar care and forethought. A quiet negative pressure system is often easier to built however. "Easier" being defined as something that you can build with off-the-shelf parts. If you're sticking to off-the-shelf cases negative pressure has another advantage over positive: it typically puts the noise-makers at the back of the case, furthest away from the user.


Either way, positive or negative, you want unbalanced airflow. The only way to get the same CFM "in" as "out" would be to have both intake and exhaust fans. Despite what Tt and some "gamerz" seem to think, having 200CFM blowing into a case, and 200CFM blowing out of it does not get you 400CFM flowing through it...it gets you 200, only with twice the noise. :wink:

IsaacKuo
Posts: 1705
Joined: Fri Jan 23, 2004 7:50 am
Location: Baton Rouge, Louisiana

Post by IsaacKuo » Wed May 09, 2007 12:44 pm

Rusty075 wrote:"Easier" being defined as something that you can build with off-the-shelf parts.
In particular, a fanned power supply in its off-the-shelf form will have its fan(s) oriented to pull air out of the case. It's easy enough to modify a PSU to either flip the fan or simply remove it, but that's not how it was meant to be used.

Theoretically, you can just remove the PSU fan and rely upon positive pressure within the case to push air through it. Unfortunately, most stock computer cases have too many stupid openings for intake fans to work efficiently. Typically, the front wall has all sorts of openings other than just the intake fan. These openings allow air to leak right back into the front bezel, instead of getting ultimately forced out the rear exhaust openings (including the PSU). With all these stupid openings, intake fans are almost worthless for anything other than spot cooling of hard drives.

Felger Carbon
Posts: 2049
Joined: Thu Dec 15, 2005 11:06 am
Location: Klamath Falls, OR

Post by Felger Carbon » Wed May 09, 2007 6:12 pm

IsaacKuo wrote:In particular, a fanned power supply in its off-the-shelf form will have its fan(s) oriented to pull air out of the case. It's easy enough to modify a PSU to either flip the fan or simply remove it, but that's not how it was meant to be used.

Theoretically, you can just remove the PSU fan and rely upon positive pressure within the case to push air through it. Unfortunately, most stock computer cases have too many stupid openings for intake fans to work efficiently. Typically, the front wall has all sorts of openings other than just the intake fan. These openings allow air to leak right back into the front bezel, instead of getting ultimately forced out the rear exhaust openings (including the PSU). With all these stupid openings, intake fans are almost worthless for anything other than spot cooling of hard drives.
Isaac, I endorse everything you say above. However, I'd like to put a different spin on it (practicing my political skillz).

With a 220mm intake fan on the side, no other intake or exhaust fans are needed. That means no exhaust fans are needed. The PSU fan is an exhaust fan. There's an exception to this rule: The ovclk/gamer guys pulling hundreds of watts still need the PSU fan. Us lower-power guys don't.

I take it back, I'll disagree with one thing: front intake fans aren't almost useless, they are plain totally useless. Leave them off. Anybody who really wants a quiet computer, really quiet, will seal all the myriad holes in the front panel (in the sheet metal behind the bezel). Use Dynamat Extreme or equivalent. No airflow is needed.

So, besides the 220mm side fan, the only fans needed are a spot fan for the HDDs, a spot fan for the HSF, and for those who have to have a hot video board, a spot fan for it. [Not them. If you have 2 video cards, please depart for your ultimate destination forthwith! :twisted: ]

And, since the big fan is on the left side panel, that means the computer case goes to the left of the keyboard and monitor.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Wed May 09, 2007 10:35 pm

In case anyone hasn't seen this picture, here's how I keep my totally positive pressure 220mm fan case.......clean as a pin. Since the photo was taken, I changed the filter material to a more restrictive type. There is no exhaust fan in this setup at all. Even the psu is fanless. All the openings into the front bezel are sealed with tape and acoustic foam. Every bit of air entering the computer goes through that filter, which is easily removed for cleaning. I've considered re-doing the setup so the filter pulls out the top or front.....making it even easier to clean. Just when will we see an OEM case with a real filtering setup? Thus far, the filters that come with some computer cases are pathetic...

Image

The photo shows a fanned psu, now replaced with a fanless 400w Zen. The rear case fan blows out, but only turns on under extreme benchmarking. Normally it doesn't spin, with the positive pressure in the case using the fan opening as a passive vent. This setup is working perfectly from a filtering stand-point. I haven't found a spec of dust inside the case after several months of constant use. :D

jaganath
Posts: 5085
Joined: Tue Sep 20, 2005 6:55 am
Location: UK

Post by jaganath » Thu May 10, 2007 2:14 am

I haven't found a spec of dust inside the case after several months of constant use.
another upside of filtering is you keep the heatsinks working at maximum efficiency; no thick layer of dust to impede heat transfer. Filtering maybe deserves more attention on SPCR than it is currently getting; if we ever need someone to teach Filtering 101, we know who to call. :wink:

Rusty075
SPCR Reviewer
Posts: 4000
Joined: Sun Aug 11, 2002 3:26 pm
Location: Phoenix, AZ
Contact:

Post by Rusty075 » Thu May 10, 2007 1:52 pm

Felger Carbon wrote:So, besides the 220mm side fan, the only fans needed are a spot fan for the HDDs, a spot fan for the HSF, and for those who have to have a hot video board, a spot fan for it.
Let me get this straight, in your positive pressure setup you need:
One 220mm fan on the side.
One fan blowing on the HDD's.
One fan on the CPU HSF.
And maybe one fan on the VGA.

How is that quieter than a conventional negative pressure system where you have:
One 120mm fan in the PSU.
One 120mm exhaust fan that draws intake air in over the HDD's.
One fan on the CPU HSF
And maybe one fan on the VGA.

I fail to see the point in cutting a big giant noise-escaping hole in the side of the case when there's a perfectly good hole already in the back that could be used instead. Yes, in theory 220mm fans can provide more CFM/dB than 120mm, but when you can already get all the CFM you need from a 120mm fan at a noise level that is below ambient, what are you really gaining?

My quietest system is completely cooled by a single 7volted Nexus 120mm fan: PSU, CPU, VGA and a pair of 3.5" HDD's. This entire system was built with off-the-shelf bits (in fact it all came from a single trip to Fry's) and with no ducting, and no case mods beyond slapping some packing tape over the exhaust grill on the back.
jaganath wrote:another upside of filtering is you keep the heatsinks working at maximum efficiency; no thick layer of dust to impede heat transfer. Filtering maybe deserves more attention on SPCR than it is currently getting

I think it would be interesting to actually study/test what the effect of dust on HS efficiency really is. I've had systems that I've built run for literally years without being opened, and the temps haven't gone nuclear yet. At worst they're a degree or two higher than the they were the day they were screwed shut. Considering that most of us run systems that have 20-30°'s of cooling system safety, having the temps raise a degree or two due to dust isn't a dangerous thing.

I think that for most people, dust control schemes are a tiny niche. Its important for people that for whatever reason have a lot of dust in their environments and can't reduce it, or that are totally OCD about having the insides of their computers squeaky clean. :wink: (smile Carl, I'm just poking fun)

For most of us, I think, dust inside a PC is just like the dust on top of your kitchen cabinets: I can't see it, and it doesn't really affect me, so there's no need to worry about it. That is, until I have to be up there to change a light-bulb, or be in there to do an upgrade. Then I'll clean up the dust. Until then, eh, no big deal.

Bluefront
*Lifetime Patron*
Posts: 5316
Joined: Sat Jan 18, 2003 2:19 pm
Location: St Louis (county) Missouri USA

Post by Bluefront » Thu May 10, 2007 2:42 pm

I think all of this hinges on just how hot you are willing to run your system. As for myself.....I want the things to be as quiet as possible, but with temperatures that I feel comfortable with. With that case in the photo, the only fan that is attached to the outside of the case (other than the back-up Yate Loon) is the 220mm fan, blowing inward. This positive pressure allows a fanless PSU to run cool. And if I were running a relatively cool CPU (I'm not), and a relatively cool GPU(I'm not), I could run the whole thing off that one 220mm fan at 4-5V.......keeping everything cool, and clean. I am certain I could not do that with a std negative pressure case, with a rear fan blowing outward......anywhere nearly as quiet.

jaganath
Posts: 5085
Joined: Tue Sep 20, 2005 6:55 am
Location: UK

Post by jaganath » Thu May 10, 2007 2:52 pm

Let me get this straight, in your positive pressure setup you need:
One 220mm fan on the side.
One fan blowing on the HDD's.
One fan on the CPU HSF.
And maybe one fan on the VGA.
I don't know why FC said they are needed when I know for a fact he only uses one fan (the big one). Maybe he was referring to uber-hot setups. As you say, running multiple fans negates the CFM/dB advantage of large diameter fans.
I think it would be interesting to actually study/test what the effect of dust on HS efficiency really is.
results would probably be very predictable,heat transfer probably inversely proportional to thickness of dust layer.

flarkit
Posts: 45
Joined: Fri May 20, 2005 9:21 pm
Location: ZA

Post by flarkit » Thu May 17, 2007 10:17 pm

Very interesting topic, since I live in a dusty place and have 2 PC's in our work-room.

My Antec SLK3000B has a 120mm YL intake fan cooling 2 HDD's suspended in-front of the fan, with a filter on the fan. There's a rear 120mm Antec Tricool on minimum speed exhausting air, as well as my Antec SP2-450's 80mm fan. I hooked the PSU's temp sensor to a fan header, so it should run higher as temps increase.

Overall, the idea was that there should be either a neutral or slightly positive pressure differential. But my case clearly has more dust than I'd like, mostly on the HSF of my gfx-card and CPU.

This leads me to suspect that the pressure is closer to negative, since I doubt dust would flow from the intake (through the filter) and settle on the CPU and gfx-card. Perhaps the filter + 2 HDD's are reducing the effective air-flow from the intake, allowing the rear and PSU fans to pull air out faster.
:roll:

RBBOT
Posts: 93
Joined: Thu May 17, 2007 9:02 am

Post by RBBOT » Fri May 18, 2007 6:43 am

Without modding the case:

- With ATX negative presure design, the points of greatest airflow are right next to the exhaust fans as the intake airflow comes from a variety of unsealed holes and ducts

- These exhaust fans are strategically next to the CPU heatsink giving best cooling to the CPU

- Therefore to keep best CPU airflow with positive pressure you should turn these fans around to be the intakes and have the front bottom fans (or vents where fans could be) as exhausts

- Heat rises

Therefore if you switch to positive pressure, its best to turn your case upside down?

MikeC
Site Admin
Posts: 12283
Joined: Sun Aug 11, 2002 3:26 pm
Location: Vancouver, BC, Canada
Contact:

Post by MikeC » Fri May 18, 2007 7:53 am

IRRC, one reason Carl (bluefront) is so into filters for his PCs is that a large, feather/dust-shedding parrot (?) is a beloved pet and permanent member of the household. (or is there more than one bird?) ;)

I also have a pet -- a large, fluffy, fur-shedding Persian/Siamese cat who's mostly indoors. Her fur ends up everywhere, including filters on the forced air central heating system and any computers with fans.

I do take a vacuum to the front intake and/or filters (and very occasionally, HSF) of the PCs from time to time, but I am certainly not religious about it. However, as per the comments by Russ, I've never seen any significant change in CPU (or other) temperatures that can be attributed to dust buildup.

Whether you have a filter or not, if you're in a dusty environment, you should probably take a vacuum to your PC periodically. But unless you design your case to take the filter impedance into account (as Carl does), a filter does usually act as an added impedance, and it may mean slightly higher base levels of noise (because you have to spin the fan(s) a little faster for the same level of cooling flow).

One big downside of filters is that if they get clogged (which will happen much sooner than an unfiltered vent), the airflow to the whole system will suffer, with potential overheating problems. So having a filter usually means you have to be more diligent about regular cleaning of that filter.

I also think dust control with filters in computers is of limited interest and benefit for most folks -- especially for SPCR enthusiasts who use few fans and only at low speed.

Felger Carbon
Posts: 2049
Joined: Thu Dec 15, 2005 11:06 am
Location: Klamath Falls, OR

Post by Felger Carbon » Fri May 18, 2007 12:00 pm

jaganath wrote:
Let me get this straight, in your positive pressure setup you need:
One 220mm fan on the side.
One fan blowing on the HDD's.
One fan on the CPU HSF.
And maybe one fan on the VGA.
I don't know why FC said they are needed when I know for a fact he only uses one fan (the big one).
Jaganath, you apparently know more about my current systems than I do. :D Because, in addition to the big-fan, I have a really slow 120mm fan on my CPU heat sinks, and an ~800RPM 100mm Scythe fan inside the rubber box containing the two HDDs. Those three fans are the only ones in my systems (at the moment). Ultimately, I'd like to use passive cooling on the CPU HS, but I'm not there yet. The spot cooler in the HDD box stays.

cpemma
Posts: 351
Joined: Sun May 11, 2003 2:31 pm
Location: S Yorks, OK
Contact:

Post by cpemma » Tue May 22, 2007 3:58 pm

Rusty075 wrote:Despite what Tt and some "gamerz" seem to think, having 200CFM blowing into a case, and 200CFM blowing out of it does not get you 400CFM flowing through it...it gets you 200, only with twice the noise. :wink:
But from the fan flow/pressure curves, particularly with the slow, heavily under-volted fans in common use by our readers, any significant positive pressure in the case will have a significant hit on air flow.

I find a front intake doing a useful job like cooling the drive cage, balanced by a rear exhaust, gives better cooling than turning the exhaust fan off. Pressure inside the case will be nearer atmospheric. It's not "double the flow", just nearer to the flow advertised for each single fan in free air.

To achieve the same cooling with intake only, in the hope of creating a dust-free interior, I need to greatly increase the power to that intake fan; perhaps to more than twice the noise... ;)

Again, filters restrict air flow; the price of a dust-free interior is more noise. I'll settle for hoovering out the inside occasionally - there's not much to clean out with a short-pile carpet under the case.

Tresstump
Posts: 22
Joined: Tue Jan 16, 2007 12:50 pm
Location: Tasmania, Australia

Post by Tresstump » Tue May 22, 2007 10:27 pm

Surely if you clean your computer once in a while dust doesn't really matter, or maybe it does if you live in the desert or maybe a sawmill...

Post Reply