By how many CFM is it necessary to increase intake over exhaust to have an effective positive airflow?
quick question about postive airflow
- Thread starter revanrules
- Start date
The advantage of a positive pressure system is that air flows out of every hole, including the optical drive and the floppy drive. The result is that the only air that enters the case does so through the intake fans, which you could filter.
If that is what you are after, the smallest delta (intake > exhaust) will do.
The delta will have to come out of every other hole (other than the intake and exhaust fans).
Edit:
You have to take into account any impedance in the way of the fans to come up with the effective CFM. For example, if you have a filter in front of your intake fan, its effective CFM will be less than that of an identical fan with no filter in front of it.
Even one CFM differential would be adequate, but you can't just calculate the intake and exhaust fans. You also have to consider leakage. Using the raw CFM values of the fans, I'd like to have at least twice the CFM capacity of the exhaust(s) on the intake fan(s). How much you'll actually need to use will be determined by how well you've plugged the leaks and how much total airflow you need to cool your system the way you want it, so I put my fans on a fan controller so I can tweak to my heart's content. I use a fluffy feather to detect air coming out from around the burner bezels, etc. to see if I've got air coming out everywhere there's still a leak.
.bh.
.bh.
BonzaiDuck
Lifer
- Jun 30, 2004
- 16,341
- 1,891
- 126
Zepper's got this one nailed down.
There have been many prevalent myths surrounding this issue. I'm the "Bonzai-Duck-t-ster," and ducting works better with any positive buildup of pressure inside the case. I pretty much do what Zepper does. So the matter of deploying fans is a slightly complicated tradeoff between N of intake fans, total CFM of intake, N of exhaust fans and total CFM of exhaust, and CFM reductions due to filtering. Any positive difference between intake and exhaust (greater intake CFM) will create some pressure, and assure an airflow that, at least, is not "rarified."
So, like Dennis Hopper in "Apocalypse Now," I'll say "With that, Mannnnnnn, I'm gonna split," and finish plugging a few unneeded vents and holes in my own system.
There have been many prevalent myths surrounding this issue. I'm the "Bonzai-Duck-t-ster," and ducting works better with any positive buildup of pressure inside the case. I pretty much do what Zepper does. So the matter of deploying fans is a slightly complicated tradeoff between N of intake fans, total CFM of intake, N of exhaust fans and total CFM of exhaust, and CFM reductions due to filtering. Any positive difference between intake and exhaust (greater intake CFM) will create some pressure, and assure an airflow that, at least, is not "rarified."
So, like Dennis Hopper in "Apocalypse Now," I'll say "With that, Mannnnnnn, I'm gonna split," and finish plugging a few unneeded vents and holes in my own system.
There is one certain way to assure you will be completely positive pressure......don't use exhaust fans. Replace the PSU with a fanless model like a Fortron. With all your fans blowing filtered air into the computer, it will stay clean. The exhaust will vent passively.....some through the PSU, and through any other vents you modify or install. The heated air will tend to rise, so venting at the top of the case will work best.
There are few cases made to work this way.......so you'll have to make some modifications to an existing case. Constructed properly, this setup works well. If you're a fan of hot graphic cards......you're on your own.
There are few cases made to work this way.......so you'll have to make some modifications to an existing case. Constructed properly, this setup works well. If you're a fan of hot graphic cards......you're on your own.
BonzaiDuck
Lifer
- Jun 30, 2004
- 16,341
- 1,891
- 126
We've been down this route before. I can almost imagine a pressurized system with no exhaust fans, but it would really need to be sealed off thoroughly barring the exhaust ports available for exhaust fans and the PSU. In my situation, I want pressure to build up inside the case and force air through low-volume passages covering the heat-generating components at a high airspeed. The exhaust fans help promote the high airflow.
This is the fundamental difference with Bluefront's scenario, and why he suggests one should be satisfied with higher VGA temperatures. My VGA card is a BFG 8800 GTS, and with the stock heatpipe cooler and vented PCI-plate design, it would rise to 70C-plus under heavy load. Replacing it with a ThermalRight HR-03-Plus reduced peak temperature to between 57C and 60C (depending on room-ambient). Adding a fan to it brought the temperature down to 53C at load.
Ducting the VGA and HR-03 kept the temperatures in a range between 43C and 54C -- even without the fan. Putting a fan in the ducting equation and in one orientation made idle and load peg at 42C -- no matter what. But that choice interfered with other priorities, and I replaced a "puller" fan design with a pusher -- and a smaller, 80mm fan. So now, the temperature bounces between 42 and 47C -- it might rise to 51 if room temperature is approaching 80F.
Keeping the intake CFMs higher and out of balance with the exhaust helps with all of this, and I can tell when my filters are getting clogged, because the temps will have risen maybe 4C degrees "across the board" (ha!)
This is the fundamental difference with Bluefront's scenario, and why he suggests one should be satisfied with higher VGA temperatures. My VGA card is a BFG 8800 GTS, and with the stock heatpipe cooler and vented PCI-plate design, it would rise to 70C-plus under heavy load. Replacing it with a ThermalRight HR-03-Plus reduced peak temperature to between 57C and 60C (depending on room-ambient). Adding a fan to it brought the temperature down to 53C at load.
Ducting the VGA and HR-03 kept the temperatures in a range between 43C and 54C -- even without the fan. Putting a fan in the ducting equation and in one orientation made idle and load peg at 42C -- no matter what. But that choice interfered with other priorities, and I replaced a "puller" fan design with a pusher -- and a smaller, 80mm fan. So now, the temperature bounces between 42 and 47C -- it might rise to 51 if room temperature is approaching 80F.
Keeping the intake CFMs higher and out of balance with the exhaust helps with all of this, and I can tell when my filters are getting clogged, because the temps will have risen maybe 4C degrees "across the board" (ha!)
- Sep 28, 2005
- 21,044
- 3,524
- 126
okey unless you completely sealed that case, air tight, postive vs negitive airflow is absolute BS.
Its the same crap because the inside of the case isnt a perfect vacume.
I dont know how or where all these possitive vs negitive airflow topics came. But just think about it. How is it possible for the inside of your case to even experience high degree's of negitive pressure? And same can be said for possitive pressure.
The only true question for these topics should be: Push or PULL fan configs. That is what your basically asking. Not Pressure.
Its the same crap because the inside of the case isnt a perfect vacume.
I dont know how or where all these possitive vs negitive airflow topics came. But just think about it. How is it possible for the inside of your case to even experience high degree's of negitive pressure? And same can be said for possitive pressure.
The only true question for these topics should be: Push or PULL fan configs. That is what your basically asking. Not Pressure.
Positive and negative are only signs. They have no magnitude.
1000 is a positive figure. So is 0.00001.
If you only have exhaust fans, why do you think air moves into the case? Because the outside air has a higher pressure than the inside air. That does not mean that the difference in pressure is 1PSI though!
No matter how small, there is a difference in pressure.
1000 is a positive figure. So is 0.00001.
If you only have exhaust fans, why do you think air moves into the case? Because the outside air has a higher pressure than the inside air. That does not mean that the difference in pressure is 1PSI though!
No matter how small, there is a difference in pressure.
That's true.....there's no massive pressure either way. The case does not have to be totally sealed. If you blow all the fans inward, that's a completely positive pressure case.......meaning that the air pressure inside your case is slightly above the ambient air pressure. It doesn't need to be much higher than ambient.....just enough to keep any airflow other than by the fans, from entering the case.
By carefully arranging the exhaust vents, you can have airflow over any/all of the hot spots. When you drop a super-hot video card into the mix, you have a new set of problems, and you may need extra internal fans to cool the thing. There's no reason to think a positive pressure case cannot handle the heat......it's a matter of having enough air being blown inward, and being directed properly. The benefit.....a cleaner, quieter computer.
Take a look at this setup. That's a P4-3.4 in there.....plenty of heat. The case has one Scythe intake fan, and no exhaust fans. Cool, quiet, clean.
By carefully arranging the exhaust vents, you can have airflow over any/all of the hot spots. When you drop a super-hot video card into the mix, you have a new set of problems, and you may need extra internal fans to cool the thing. There's no reason to think a positive pressure case cannot handle the heat......it's a matter of having enough air being blown inward, and being directed properly. The benefit.....a cleaner, quieter computer.
Take a look at this setup. That's a P4-3.4 in there.....plenty of heat. The case has one Scythe intake fan, and no exhaust fans. Cool, quiet, clean.
- Sep 28, 2005
- 21,044
- 3,524
- 126
and how would this aid or hinder convection? The same amount of air is being passed though the case and then exhausted.
If you want to take it into a physic standpoint, then negitive pressure would be the worst because we all know, in space/vacume there can be no thermal transfer.
- Sep 28, 2005
- 21,044
- 3,524
- 126
nonono... im not getting mad or anything.
im just asking how it would help anything?
The only time i read where you need to take into aspect of possitive vs negitives are dust conditions. But thats about it. EVeryone knows if you use negitive pressure with filters as intakes then your case will be cleaner.
But i dont see how any of this would matter on a typical setting. So unless someone wants to show me proof or links to where it does help, i dont see why this thread keeps poping in and out every year.
All I said was that positive pressure helped keep the case cleaner as long as all the intake fans are properly filtered.
I add to that now that to have positive pressure in a conventional case, where the main intake fan is in front (closer to the user) and the main exhaust fans are at the back (farther from the user), a positive pressure case ends up being louder where the user sits.
And I am not touching the subject of which one is going to be cooler with a 10-foot pole!
That is one of those subjects that brings out all the baseless "negative is cooler" & "positive is cooler" claims.
Sometimes these forums are so funny.
I remember when every day we had a fight over "ATI is better" vs. "NVIDIA is better".
I add to that now that to have positive pressure in a conventional case, where the main intake fan is in front (closer to the user) and the main exhaust fans are at the back (farther from the user), a positive pressure case ends up being louder where the user sits.
And I am not touching the subject of which one is going to be cooler with a 10-foot pole!
That is one of those subjects that brings out all the baseless "negative is cooler" & "positive is cooler" claims.
Sometimes these forums are so funny.
I remember when every day we had a fight over "ATI is better" vs. "NVIDIA is better".
BonzaiDuck
Lifer
- Jun 30, 2004
- 16,341
- 1,891
- 126
Well, this is all about "air-molecule-and-density-accounting," isn't it? If there are small leaks -- like the little vents in my graphics card's PCI-plate, and the little 3/8" hole I need to patch in the front chassis (nevertheless covered by the front bezel), then it is still a question as to the amount of air that will leak from those holes per unit of time versus the difference between the intake CFMs and exhaust CFMs.
On my own graphics card question, I've got an 80x15mm fan built into the duct cover for the card, only spinning now at about 1,400 rpm, because So-Cal just got a reprieve to 63F evening temperatures, and the mobo option-fan thermal control works perfectly. But that VGA card is never going to go over 50C unless we kick back to 85F room-ambients tomorrow, and then -- not by much (we're talking about load temperatures there.)
This seems like all straightforward stuff, even if you include the factor for kruft building up on fan filters as you watch your temperatures edge upward.
My living room looks like an electronics-store Hiroshima -- a computer-parts Nagasaki. I'm overwhelmed with the prospective task of cleaning up. . .
On my own graphics card question, I've got an 80x15mm fan built into the duct cover for the card, only spinning now at about 1,400 rpm, because So-Cal just got a reprieve to 63F evening temperatures, and the mobo option-fan thermal control works perfectly. But that VGA card is never going to go over 50C unless we kick back to 85F room-ambients tomorrow, and then -- not by much (we're talking about load temperatures there.)
This seems like all straightforward stuff, even if you include the factor for kruft building up on fan filters as you watch your temperatures edge upward.
My living room looks like an electronics-store Hiroshima -- a computer-parts Nagasaki. I'm overwhelmed with the prospective task of cleaning up. . .
Sorry I didn't respond, I just like reading everyones thoughts on the subject. Thanks for all the replies. You guys gave me some good ideas and I think I'm going to end up adding some ducting and then have positive airflow in the ducting while not worrying too much about over all positive/negativeness of the case.
VinDSL
Diamond Member
Yes, we have.
It always reminds me of the ol' Monty Python skit "...if you want to know the air speed velocity of an unladen swallow..." yada, yada...
To me, it's as simply as this - Ventilation, i.e. the number of times the whole interior volume of air is replaced, e.g. good air in, bad air out!
This can be accomplished in a zillion different ways - temperature readings being the determining factor as to which way is best...
Modelworks
Lifer
- Feb 22, 2007
- 16,240
- 7
- 76
Not the home pc world but in the mainframe world there are several systems that are designed for positive pressure cooling.
Its done more as a convenience than a dust catcher, but it achieves both fairly well.
In the floor of the room there are ducts, similar to what you have in a home, except they are designed so that when you place a rack over the opening it connects.
In the rack at the back there is a channel that extends all the way to the top and at key points there are opening in the channel to blow that air out onto the motherboards and out the front or sometimes into another channel where it is ducted away through a return channel.
On site you have a system that similar to a home AC system that produces air at 55F and removes humidity before being blown into the racks.
I suppose if someone really wanted to cool a pc and have it absolutely quiet they could put a similar duct in the floor of the home and run it off to a basement, where all the blower, cooling unit would exist. Cut a hole in the bottom of a tower case and place it over the top of the duct
Ultra quiet and you could send very cold air if you wanted.
Its done more as a convenience than a dust catcher, but it achieves both fairly well.
In the floor of the room there are ducts, similar to what you have in a home, except they are designed so that when you place a rack over the opening it connects.
In the rack at the back there is a channel that extends all the way to the top and at key points there are opening in the channel to blow that air out onto the motherboards and out the front or sometimes into another channel where it is ducted away through a return channel.
On site you have a system that similar to a home AC system that produces air at 55F and removes humidity before being blown into the racks.
I suppose if someone really wanted to cool a pc and have it absolutely quiet they could put a similar duct in the floor of the home and run it off to a basement, where all the blower, cooling unit would exist. Cut a hole in the bottom of a tower case and place it over the top of the duct
Ultra quiet and you could send very cold air if you wanted.
TRENDING THREADS
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 25K
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 21K
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
-
Facebook
Twitter