Intake fans...next to useless?
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
- Posts: 87
- Joined: Sun Apr 29, 2007 1:20 am
Intake fans...next to useless?
I been pondering of recent and have come to the thinking that intake fans, especially at the kind of rpms required to be quiet are pretty useless and any case would be better served with fans extracting hot air from the case, even if it's extracting air in the opposite direction of another fan in the case.
Seems logical enough, anyone got something to back it up?.
Seems logical enough, anyone got something to back it up?.
I had another post on here about some Arctic Cooling PWM fans I was running as intake fans. When controlled by PWM, they're near-silent... but push very little air. The main temperature differences I see compared to them being off are NB and mobo temps, with a few degrees difference, so they still help.
When in non-PWM mode, running loudly at 1500RPM, they make a much larger difference to temps, knocking up to another 5 degrees off NB temps - I'm keeping them for now, but would consider replacing them with something like a Scythe 1200RPM fan to push more air quietly - eventually I'll be adding more hard disks, and the AC fans blow directly across the HDDs in their bays, so for me there's value in having them.
At the moment I don't really need them, but I reckon they will help in summer (I'm in Australia), especially when more HDDs are added.
I guess the effect is from the airflow patterns across the motherboard, rather than exhaust-only that would suck air from any old direction wherever there's an opening.
When in non-PWM mode, running loudly at 1500RPM, they make a much larger difference to temps, knocking up to another 5 degrees off NB temps - I'm keeping them for now, but would consider replacing them with something like a Scythe 1200RPM fan to push more air quietly - eventually I'll be adding more hard disks, and the AC fans blow directly across the HDDs in their bays, so for me there's value in having them.
At the moment I don't really need them, but I reckon they will help in summer (I'm in Australia), especially when more HDDs are added.
I guess the effect is from the airflow patterns across the motherboard, rather than exhaust-only that would suck air from any old direction wherever there's an opening.
You need to consider the dynamics of moving air though an enclosed space and it's relationship to air pressure. You might think that fans just sucking air out is more efficient, but it isn't. The manufacturers and enthusiasts spend huge amounts of time on research and development into this and it's merits are well documented.
Even a slight movement of air can reduce the temperature of a PC component, so the idea that a slow rotating fan is useless, is not really correct. You can try it yourself - get a hard drive and use it alone with no air moving and note it's temperature. Then try it again with a slow moving fan blowing air across it - the temperature will be significantly reduced.
Instead of pondering something, I find it's always better to try it out - you can then see the results for yourself.
Even a slight movement of air can reduce the temperature of a PC component, so the idea that a slow rotating fan is useless, is not really correct. You can try it yourself - get a hard drive and use it alone with no air moving and note it's temperature. Then try it again with a slow moving fan blowing air across it - the temperature will be significantly reduced.
Instead of pondering something, I find it's always better to try it out - you can then see the results for yourself.
-
- Posts: 87
- Joined: Sun Apr 29, 2007 1:20 am
If you could URL up some links to these documents, i'd be grateful.Great Gig wrote:You need to consider the dynamics of moving air though an enclosed space and it's relationship to air pressure. You might think that fans just sucking air out is more efficient, but it isn't. The manufacturers and enthusiasts spend huge amounts of time on research and development into this and it's merits are well documented.
Great Gig wrote:Even a slight movement of air can reduce the temperature of a PC component, so the idea that a slow rotating fan is useless, is not really correct. You can try it yourself - get a hard drive and use it alone with no air moving and note it's temperature. Then try it again with a slow moving fan blowing air across it - the temperature will be significantly reduced.
I'm not arguing that a slow moving fan is useless, but that any intake fan, especially slow moving, is probably better served pulling hot air out of the case rather than feeding cool air in.
That's true enough and I will do at some point, just wanted to examine the theory for the moment.Great Gig wrote:Instead of pondering something, I find it's always better to try it out - you can then see the results for yourself.
-
- Patron of SPCR
- Posts: 857
- Joined: Fri Dec 27, 2002 1:49 pm
- Location: Somerset, WI - USA
- Contact:
Like most things, I would say it depends on your situation. Intake fans will not generally increase the overall airflow through the case. But they can add additional airflow in places that might not normally get airflow. For instance, I've built a handful of PCs for my workplace using Antex NSK3400/3480 cases. These are generally lower-powered PCs that don't generate a lot of heat. So the 120mm tri-cool fan on low is more than enough to exhaust the warm air. The problem area though is the little bit of room between the hard drive mounted on the floor of the case and the PCIe graphics cards. These are generally HD2400/HD3400 series cards which don't generate a lot of heat, but some. The rear fan really does not create any airflow in this area. So a nice slow moving 92mm intake fan helps to add some airflow which keeps the heat from the graphics card from getting the drive too warm.
Well... It depends on how and why you use the intake fans. As Great Gig already said, one have to take note of the dynamics involved. First let me ask a question.
Which set up will move most air through the case? (in the hypothetical situation that no air enters/exits the case other than through the fans and the fans all are similar and set to the same speed)
1) 1 fan in, 1 out
2) 2 in, 1 out
3) 1 in, 2 out
The correct answer is that all three set ups move as much air through the case. Why? A fan is constructed to move X cfm and it will do this. Not less, not more. So airflow is restricted by the bottleneck, the direction with the lowest number of fans. Set up 2 and 3 would move more air if the single fan is disabled.
What intake fans CAN do, however, is to direct air. Or rather force air to move in different ways over the motherboard and thus be a part of the airflow management in the rig. And it is intelligent airflow management which allow us to reduce the need of cfm to maintain acceptable temperatures and thus run more silent computers. Wether or not intake fans should be part of your plan depends on how you have planned your airflow and the effect you get by adding one.
Which set up will move most air through the case? (in the hypothetical situation that no air enters/exits the case other than through the fans and the fans all are similar and set to the same speed)
1) 1 fan in, 1 out
2) 2 in, 1 out
3) 1 in, 2 out
The correct answer is that all three set ups move as much air through the case. Why? A fan is constructed to move X cfm and it will do this. Not less, not more. So airflow is restricted by the bottleneck, the direction with the lowest number of fans. Set up 2 and 3 would move more air if the single fan is disabled.
What intake fans CAN do, however, is to direct air. Or rather force air to move in different ways over the motherboard and thus be a part of the airflow management in the rig. And it is intelligent airflow management which allow us to reduce the need of cfm to maintain acceptable temperatures and thus run more silent computers. Wether or not intake fans should be part of your plan depends on how you have planned your airflow and the effect you get by adding one.
Ahem. Air flow is function of fan performance and pressure [difference between fan sides]. Every fan (at specified RPM) can be characterised with some pressure value, when airflow is exactly zero. Specified airflow is usually given for zero pressure. Airflow and pressure relationship is unfortunately not linear.Tobias wrote:1) 1 fan in, 1 out
2) 2 in, 1 out
3) 1 in, 2 out
The correct answer is that all three set ups move as much air through the case. Why? A fan is constructed to move X cfm and it will do this. Not less, not more. So airflow is restricted by the bottleneck, the direction with the lowest number of fans. Set up 2 and 3 would move more air if the single fan is disabled.
You can't say about setups 2 and 3, does removing single fan increase or decrease airflow - it depends on fans pressure-airflow relationship. Considering that single fan does alter in-case air pressure in favourable direction for other fans, most probably removing it will decrease airflow.
---
Back to topic.
Like many uers said already, the slow intake fan has influence to HDD temperature and does alter flow direction. When HDDs are suspended, then sometimes intake fan is the best solution to keep HDD temps low.
And here I was with a perfectly fine hypothetical reasoning and in you come and slash my entire reasoning with how things actually works in practice:P In a very low power scenario with fans set to minimal speeds and plenty of airvents, preassure differences will not be that big. On the other hand that scenario lies outside my set up premises as well:PAhem. Air flow is function of fan performance and pressure [difference between fan sides]. Every fan (at specified RPM) can be characterised with some pressure value, when airflow is exactly zero. Specified airflow is usually given for zero pressure. Airflow and pressure relationship is unfortunately not linear.
Last edited by Tobias on Sun Nov 16, 2008 10:49 am, edited 1 time in total.
Exhaust only fan(s) is the most efficient for performance/noise. First, any noise sources should be as far away as possible. Fans do not just work on pressure differential. They actually accelerate the air and the easiest path away from the fan is directly out from the center (like a blower). With intake fans you are limiting this path with the case. You are basically increasing noise and heat by forcing air to bounce against the case and slowing the air down.
I did try the one big fan positive pressure idea once. The one nice thing is that you can just cut a hole in the case by anything you want to cool.
It would at least be quieter with the intake fan located at the back or bottom of the case.
I did try the one big fan positive pressure idea once. The one nice thing is that you can just cut a hole in the case by anything you want to cool.
It would at least be quieter with the intake fan located at the back or bottom of the case.
-
- Posts: 25
- Joined: Thu Apr 05, 2007 4:23 pm
-
- *Lifetime Patron*
- Posts: 2269
- Joined: Sun May 21, 2006 9:09 am
- Location: Northern California.
If we're talking pure silence i would guess( and haven't had the time to try yet, but have always wanted too...) using only middle mounted fans, with a partition separating the front from back of the case.
With the fan(s) well inside the case it should be much quieter than having fans near the front or rear openings of the case.
With the fan(s) well inside the case it should be much quieter than having fans near the front or rear openings of the case.
This is actually a valid argument about the most efficient use of airflow. This article may be old, but the principles are sound. I like to restrict the airflow into my case, but do prefer a small low speed 80mm at the front to assist cooling the hard drives. Have a read http://icrontic.com/articles/pc_airflow ... ling_guidesantacruzbob wrote:lol...QuietOC wrote:With intake fans you are limiting this path with the case. You are basically increasing noise and heat by forcing air to bounce against the case and slowing the air down.
I also didn't mention that fans create heat themselves. So with intake fans you are adding heat into the case.
The best location I've found is to mount the CPU/exhaust fan in a duct between the CPU heatsink and the rear of the case. As far back as possible (I perfer small cases) sucking air through the heatsink.
The most troublesome component is the video card. The standard 2-slot blower arrangement isn't too bad, but I've done ducts with a 120 or 92mm fan blowing into the duct. Really it would be nice to have it done the same as the CPU with the fan sucking through the GPU heatsink using a duct.
The best location I've found is to mount the CPU/exhaust fan in a duct between the CPU heatsink and the rear of the case. As far back as possible (I perfer small cases) sucking air through the heatsink.
The most troublesome component is the video card. The standard 2-slot blower arrangement isn't too bad, but I've done ducts with a 120 or 92mm fan blowing into the duct. Really it would be nice to have it done the same as the CPU with the fan sucking through the GPU heatsink using a duct.
-
- Posts: 135
- Joined: Tue Jun 01, 2004 12:41 pm
- Location: Yonder
Yup, that's how I understand it. Positive pressure!xan_user wrote:with intake fans and a positive air pressure setup you can filter dust.
with only exhaust fans the dust comes in at every opening, like optical drives ect.
Besides, many cases have intake fans blowing directly onto hard drives, and even the slightest air flow along a hard drive helps with lowering temps.
My opinion about this:Tobias wrote:Well... It depends on how and why you use the intake fans. As Great Gig already said, one have to take note of the dynamics involved. First let me ask a question.
Which set up will move most air through the case? (in the hypothetical situation that no air enters/exits the case other than through the fans and the fans all are similar and set to the same speed)
1) 1 fan in, 1 out
2) 2 in, 1 out
3) 1 in, 2 out
The correct answer is that all three set ups move as much air through the case. Why? A fan is constructed to move X cfm and it will do this. Not less, not more. So airflow is restricted by the bottleneck, the direction with the lowest number of fans. Set up 2 and 3 would move more air if the single fan is disabled.
This would be true only if there are not additional holes in the case and it is completely hermetic (imagin the pressure in the fans if there are no additional holes or places for air to go out or in except fans themselves...). If there are holes, then air will go out or in through them. The basic principle is that the air that enters the case and the air that goes out is exactly the same (so that the pressure in the case is constant, it can not increase or decrease continuously, an equilibrium must be met). At least after a small estabilizing time.
So if you put for example 2 fans blowing in x cfm each, and 1 fan sucking out x cfm, the total air (forced) entering the case is 2x. Then the same quantity needs to go out. As there is another fan blowing out x cfm, the additional air quantity (another x cfm, because 2x-x=x) will go out through the holes.
Similar case for the situiation with 2 fans out and 1 in.
In this examples I have not considered the pressure that the fans have to support, if there is a lot of airflow impedance, then the cfm that fans move is obviously smaller than the rated one. That is why holes are important (for example the ones in the front of the case, in a negative pressure situation with fans at the back of the case blowing air out).
And as some have said in the thread, it makes sense to add a fan blowing air in, even if we already have 2 fans blowing out, because it helps to direct airflow (force the input point, so air goes through this fan instead of through other holes in the case).
PS. Sorry for my English... if there is some language mistake please correct me
When I built my new workstation in August I experimented with my P182 using the default 4 fan positions in the case.
For me I wanted a balance of quiet, cool and low dust. I ended up with 2 Scythe Slipstream 1200 rpm intakes and 2 Slipstream 800 rpm exhaust.
My final solution is probably a bit noisier than many SPCR patrons would like but it is still fairly quiet.
For me I wanted a balance of quiet, cool and low dust. I ended up with 2 Scythe Slipstream 1200 rpm intakes and 2 Slipstream 800 rpm exhaust.
My final solution is probably a bit noisier than many SPCR patrons would like but it is still fairly quiet.
The 1200's are noticeably louder than the 800's but not oppressively so (for my setup anyway).
When the house is at it's quietest there is a low hum from the case when placed under the edge of the desk. Most of the time it is at or below the ambient sound levels.
Noise generation in the system:
2 800 rpm 120mm Slipstream exhaust
2 1200 rpm 120mm Slipstream intake
1 800 rpm SLipstream 120mm on HR-01+ (only runs when overclocking)
Corsair memory cooler (3x40mm) (only runs when overclocking)
80mm in PSU (has not ramped up since I got it even when overclocking)
HD4870 stock cooler (runs at 30% except when gaming)
4xHDD's in lower compartment using grommet mounts.
When the house is at it's quietest there is a low hum from the case when placed under the edge of the desk. Most of the time it is at or below the ambient sound levels.
Noise generation in the system:
2 800 rpm 120mm Slipstream exhaust
2 1200 rpm 120mm Slipstream intake
1 800 rpm SLipstream 120mm on HR-01+ (only runs when overclocking)
Corsair memory cooler (3x40mm) (only runs when overclocking)
80mm in PSU (has not ramped up since I got it even when overclocking)
HD4870 stock cooler (runs at 30% except when gaming)
4xHDD's in lower compartment using grommet mounts.
In your typical front to back airflow configuration, yes. Intake fans are basically useless. The only time they are really needed in this situation is if you have many hard drives tightly packed together and they require active cooling to keep them from overheating.
There are airflow configurations however that make intake fans a good option. Like positive airpressure cases, mainly for air filtration purposes. Do a search for bluefront, he has discussed this option in length. Basically you take that rear 120mm fan, and instead of using it to exhaust heat, you use it to intake cool air into your case, and then put a good high quality pleated filter in front of it to keep dust out of your case.
There are airflow configurations however that make intake fans a good option. Like positive airpressure cases, mainly for air filtration purposes. Do a search for bluefront, he has discussed this option in length. Basically you take that rear 120mm fan, and instead of using it to exhaust heat, you use it to intake cool air into your case, and then put a good high quality pleated filter in front of it to keep dust out of your case.
I once think that intake fans were useless. But when I upgraded my friend computer it proved otherwise. It was normal mid tower case with one 12 cm exhaust opening, one intake opening at same size, two smaller openings at side and PSU on top. I put there PSU with 12cm fan, 12cm slow fan at back and ninja for Prescot with slow fan directed to back. What happened was that the computer wasn't stable unless we run exhaust at 12 volts which was too loud. Then we added one slow fan as intake. Then we could run all fans at 5 volt and PSU didn't ramp up and the computer was stable. Result was the quietest at we could achieve with that setup. And the fan slots were non restrictive.
You can't put a blanket statement out saying that intake fans are useless. There are far too many variables to consider.
Without getting into it, I can point out that without an intake fan, my video car runs MUCH warmer.
Intake fans can provide direct, however, turbulant flow/pressure. Using a fan on a heat sink is essentially like running an intake fan for that heat sink. This directs the flow over the heat sink but is turbulant. Less turbulant air is desired for any type of cooling, however, you need the air directed. To get the directed flow, you'd have to put in ducts and such to direct the air flow.
Without getting into it, I can point out that without an intake fan, my video car runs MUCH warmer.
Intake fans can provide direct, however, turbulant flow/pressure. Using a fan on a heat sink is essentially like running an intake fan for that heat sink. This directs the flow over the heat sink but is turbulant. Less turbulant air is desired for any type of cooling, however, you need the air directed. To get the directed flow, you'd have to put in ducts and such to direct the air flow.
Turbulence=better cooling
With forced convection turbulent flow is much better for heat transfer. It's basic thermodynamics.
-
- Posts: 1608
- Joined: Tue Jan 04, 2005 4:02 pm
- Location: United States
This is why I'm using a 92mm on my NSK3300. It doesn't move a lot of air, but it seems to help with HDD, VGA, and chipset temps. It really just depends on the system I think, with some clever ducting to an exhaust you can pull air over these components, but in my situation I just opted for an intake fan blowing air over them directly. Much easier way to get the job done, although probably not as nifty or quiet.BillyBuerger wrote:Like most things, I would say it depends on your situation. Intake fans will not generally increase the overall airflow through the case. But they can add additional airflow in places that might not normally get airflow. For instance, I've built a handful of PCs for my workplace using Antex NSK3400/3480 cases. These are generally lower-powered PCs that don't generate a lot of heat. So the 120mm tri-cool fan on low is more than enough to exhaust the warm air. The problem area though is the little bit of room between the hard drive mounted on the floor of the case and the PCIe graphics cards. These are generally HD2400/HD3400 series cards which don't generate a lot of heat, but some. The rear fan really does not create any airflow in this area. So a nice slow moving 92mm intake fan helps to add some airflow which keeps the heat from the graphics card from getting the drive too warm.
-
- Posts: 50
- Joined: Wed Oct 15, 2008 5:01 pm
- Location: Midwest
Ah, the perfect case that is quiet and cool with hard drives almost frosty!
My last case resulted in a bit of hacking to get a decent combination of quiet operation with all devices in cool air stream.
Some of my findings:
- Side-intake vents tend to short-circuit air flow - leaving HD in a hot spot.
- Convenient HD cages are a major air dam impeding front-to-back air flow.
- Supplied front-intake air filters are close to useless.
I now have quiet front-to-back air flow over all devices with dual Panaflos at 7.5V.
Side-intake air input taped-off.
All fan grills hacked open.
HD drive cage (removed 19 rivets) is gone, HD is elastic-suspended directly in air-stream.
Front intake air filter is now surface-mounted, using 3M spun-glass furnace/AC filter media cut-to-fit. This media is corrugated-surface for large surface area, low impedance, decent noise-trapping and filters particles down to 1 micron. Case interior stays hospital-clean.
Hope this helps!
My last case resulted in a bit of hacking to get a decent combination of quiet operation with all devices in cool air stream.
Some of my findings:
- Side-intake vents tend to short-circuit air flow - leaving HD in a hot spot.
- Convenient HD cages are a major air dam impeding front-to-back air flow.
- Supplied front-intake air filters are close to useless.
I now have quiet front-to-back air flow over all devices with dual Panaflos at 7.5V.
Side-intake air input taped-off.
All fan grills hacked open.
HD drive cage (removed 19 rivets) is gone, HD is elastic-suspended directly in air-stream.
Front intake air filter is now surface-mounted, using 3M spun-glass furnace/AC filter media cut-to-fit. This media is corrugated-surface for large surface area, low impedance, decent noise-trapping and filters particles down to 1 micron. Case interior stays hospital-clean.
Hope this helps!
-
- Posts: 44
- Joined: Tue Mar 24, 2009 8:05 pm
- Location: San Francisco