What is forced air pc cooling?

[sticknstone]

revision 1

 

-removed passive

-removed intake temp formula because it has a different application

 

To my fellow occ members and welcome guests because of my fondness for pc's and diy and occ.

 

Forced air cooling of the pc case can use a positive pressure system, negative pressure system or balanced pressure system. The positive pressure system will only use intake fans and have grilles which allow the air blown in to exhaust . The negative pressure system will only use exhaust fans and have grilles which allow the air blown out to intake. The balanced pressure system will use intake and exhaust fans in a manner that the total cfm brought into the case is equal to the total cfm blown out of the case.

 

The following examples are created with the idea that warm air will rise and cool air will sink.

 

The positive pressure system will contain intake fans in the lower front and side panel of the case. The grilles on the rear and top of the case will not have any fans installed. If the 250 mm side panel fan blows 110 cfm and the 200

mm lower front fan blows 100 cfm then the total air flow of this system will be 210 cfm. The grille velocity for the grille should be 500-700 fpm (feet per minute) as used in other industries. The formula for grille sizing is fpm = cfm

(air flow in cubic feet per minute) x 144 / free area in inches squared. The 'free area' does not include a grille mesh or filter both of which will reduce the amount of air blown through. The larger the mesh spacing the less the air flow reduction

with 1/2" mesh being 90% efficient, 1/4" mesh being 80% efficient, and insect screen being 50% efficient. Therefore, for the desired air to flow through the grille an increased grille size would be required based on the mesh spacing. With a

rear 120 mm grille and two top 120 mm grilles the fpm would be 210 x 144 / (pi x 2.36 squared x 3) = 576 fpm.

Let us say that more air flow is desired in the case and the rear 120 mm grille has an additional intake fan of 60 cfm the two top 120 mm grilles would be exhausting at a fpm of 270 x 144 / (pi x 2.36 squared x 2) = 1111 fpm.

The grille should be sized to be quiet and have a comfortable air flow. The size of the grille area will affect the velocity of the air moving through it. With a small grille area the air will have a high velocity, and a large grille

area will have low velocity. When air velocities begin to get higher the sound of the air moving will be audible and the air felt will annoy the computer user. The recommended range represents these two discomforts in the high end.

The other negative effect is that the fan does not have enough static pressure to exhaust the air. If this occurs then no air will be felt or no air will be seen moving a strip of paper placed about an inch from the grille. Someone may

question Is the low velocity from a large grille or the fan not having enough static pressure? can be tested by covering the existing grille area to reduce it and cause the air velocity to increase, and therefore, proving the fan has

enough static pressure. Adding additional intake fans will increase the fan sp (static pressure).

 

The negative pressure system will contain exhaust fans in the rear and top of the case. The grilles on the lower front and side panel of the case will not have any fans installed. With a 120 mm rear fan blowing 60 cfm and two 120 mm

top fans blowing 60 cfm each the total air flow of the system will be 180 cfm. Grille sizing and fan considerations can be found in positive pressure system paragraph. The grille velocity for the grille should be 400-600 fpm (feet per

minute) as used in other industries. With a lower front grille size of 200 mm the area is pi x 4 squared = 50.3 and the side panel grille size of 250 mm the area is pi x 5 squared = 78.5. The intake air velocity at the grille would be

180 x 144 / (50.3 + 78.5) = 201 fpm. While the fpm is below the established range it is not a negative in terms of hearing the air being brought in through the grille.

Let us examine a system which will use the lower front exhaust fan of 100 cfm in the system as well. The intake air at the grille velocity would be 280 x 144 / 78.5 = 514 fpm.

 

The balanced pressure system will contain intake fans in the lower front and side panel of the case and exhaust fans in the rear and top of the case. The 200 mm lower front fan blows 100 cfm and the 250 mm side panel fan blows 110 cfm

for a total of 210 intake cfm. The 120 mm rear fan blowing 60 cfm and two 120 mm top fans blowing 60 cfm each giving a total 180 exhaust cfm. To achieve a balanced pressure in the case the intake cfm would need to be reduced with a fan

controller or an exhaust grille of a suitable size would be needed to allow the excess intake air to exhaust.

With the fan specification information for the side panel fan of 110 cfm at 2500 rpm (revolutions per minute) and the use of one of the fan laws will give us the new fan controlled rpm of the side panel fan. The fan law formula states

desired rpm = (desired cfm / manufacturer rated cfm) x manufacturer rated rpm. The side panel fan would need to operate at (80 / 110) x 2500 = 1818 rpm.

The exhaust grille of 30 cfm would be 30 cfm x 144 / 500 fpm (low end of recommended range) = 8.6 inches squared. If a grille was placed in the 5.25" slot the area would be 5.875" x 1.625" = 9.6 inches squared. The resulting fpm would be 30 x 144 / 9.6 = 450.

 

The fpm velocity sound of the moving air at the grille is not the sound of the fan motor which is what is listed in dBA on the fan specification information.

Edited June 12, 2011 by sticknstone

[Black64]

Dude, all you need to know is that the more fans you have the cooler your PC is.

[Locutus]

Dude, all you need to know is that the more fans you have the cooler your PC is.

 

While some people argue that positive pressure is better, and others argue that negative is better, it doesn't really matter which one as long as there's a lot of air moving it's all good.

 

I personally find that a negative pressure setup is better. I have both my huge fans that came with my 932, and a 120MM exhaust.

[VaporX]

Dude, all you need to know is that the more fans you have the cooler your PC is.

 

That is NOT accurate. I have seen a number of cases that cool better than cases with more fans due to better placement.

 

The best case design for cooling I think can be found in the concepts used in the Raven series. By allowing the natural tendency of heat to rise and couple this with air pressure pushing the air in the upward direction I think you would get your best cooling effects.

 

The advanatge of positive air pressure is that you can have a bit better dust control since you are directly controlling the air intake flow. Negative pressure works but it can full air from ares not filters and increase dust potential. Another big plus for postiive pressure is the removal of eddys in the aior. Pockets of heated air that are held in place by the air flow patterns of the case. Positive pressure essentially is forcing new air into all areas and thus forcing out the heat.

 

At the end of the day however with the excpetion of extreme overclocks or a poor case design any of the modern cases offers more than sufficent cooling.

[sticknstone]

Well I have seen other members have questions on the content in this topic, and it is difficult to answer the question, at times, right on the spot. It is better to make an informed choice.

[Munkypoo7]

Dude, all you need to know is that the more fans you have the cooler your PC is.

Nope...you can get a HAFX or a TT Speedo with as many fans as you want... temps still will be higher than other cases if you don't have a proper designated airflow.. as in too much turbulence. Quality over Quantity works best

 

Well I have seen other members have questions on the content in this topic, and it is difficult to answer the question, at times, right on the spot. It is better to make an informed choice.

 

Even then... no real amount of information can really substitute a test run between positive and negative in your own situation. Literally the placement of your case [as in, is it near a wall on one side or does a draft hit it near the rear etc] changes which method works best. Generally neutral is almost impossible to be perfectly set up, it's always at least a bit positive or negative. As for personal opinion, I'm more of a fan of positive pressure just to avoid extra dust from being sucked in through any pores, while with positive I just need to worry about my fan filters.

 

In the end... it requires a test run... there's no real substitute for that imo.

 

Also, while your original post is great, it kinda generalizes a bit much. Each case varies, for example my FT01 [positive pressure] uses a front and top mounted intake fan, no side fan, yet it keeps my temps lower that way then the top being exhaust....case design / construction also has a huge factor.

 

Unfortunately, there will never be a perfect pressure system, it's all individualized

[sticknstone]

Neutral systems that contain purely fans are almost impossible because of all the grille space used on the cases. It is meant to be general information to be applied by each user.

[Shalaan]

Hello,

 

Thanks for this post, it might be a bit of help for me, though I think I need some more guidance.

 

I have a relative new computer, it's soon turning 2 years old, but it's still very fast and plays games very smoothly. The chassis have 3 fans installed, all 120mm, while the fan on my processor is 70mm. My major problem with my computer is the intense noise coming from the fan on the processor after a couple of hours. The display on my chassis says that its around 50 celcius, which isnt that bad, but my noob sensors tell me that my processor fan isnt supposed to be so noisy. I tried to take off the plate on my chassis to put my hand in to feel the air inside my computer, and it was rather warm, and after a couple of minutes the fans begin to "slow down" and the temperature fall a bit. I've been trying to fiddle around with the BIOS settings for the fans, but it doesnt seem to change much. I noticed that the processor fan was running at a whopping 4700rpm while the other fans ran at around 1300rpm.

 

So I would like to know what you guys might suggest I should do. I've been looking at some fans with the same size, but with a faster rpm. (My currents fans are at around 1100-1300rpm, these I found had 2000rpm.) As far as I am concerned, my front fan is sucking in, the fan at the loose plate is sucking in as well, while the one at the back is blowing it out, but it is like it doesnt really blow it properly out.

 

For orders sake, here is my specs:

 

Processor:

AMD Phenom II X4 955 Processor (4 CPUs), ~3.2GHz AM3+

 

Memory:

8190MB RAM

 

Video Card:

Sapphire TOXIC HD 4870 1GB

 

Motherboard:

Asus M4A79T Deluxe

 

Computer Case:

NZXT Guardian 921

 

 

I hope there is a good soul out there who can help me, I am a rather big noob when it comes to the cooling part, so it would be nice if someone could guide me in the right direction .

[Black64]

Hello,

 

Thanks for this post, it might be a bit of help for me, though I think I need some more guidance.

 

I have a relative new computer, it's soon turning 2 years old, but it's still very fast and plays games very smoothly. The chassis have 3 fans installed, all 120mm, while the fan on my processor is 70mm. My major problem with my computer is the intense noise coming from the fan on the processor after a couple of hours. The display on my chassis says that its around 50 celcius, which isnt that bad, but my noob sensors tell me that my processor fan isnt supposed to be so noisy. I tried to take off the plate on my chassis to put my hand in to feel the air inside my computer, and it was rather warm, and after a couple of minutes the fans begin to "slow down" and the temperature fall a bit. I've been trying to fiddle around with the BIOS settings for the fans, but it doesnt seem to change much. I noticed that the processor fan was running at a whopping 4700rpm while the other fans ran at around 1300rpm.

 

So I would like to know what you guys might suggest I should do. I've been looking at some fans with the same size, but with a faster rpm. (My currents fans are at around 1100-1300rpm, these I found had 2000rpm.) As far as I am concerned, my front fan is sucking in, the fan at the loose plate is sucking in as well, while the one at the back is blowing it out, but it is like it doesnt really blow it properly out.

 

For orders sake, here is my specs:

 

Processor:

AMD Phenom II X4 955 Processor (4 CPUs), ~3.2GHz AM3+

 

Memory:

8190MB RAM

 

Video Card:

Sapphire TOXIC HD 4870 1GB

 

Motherboard:

Asus M4A79T Deluxe

 

Computer Case:

NZXT Guardian 921

 

 

I hope there is a good soul out there who can help me, I am a rather big noob when it comes to the cooling part, so it would be nice if someone could guide me in the right direction .

 

I can help. I am guessing that you are running the stock heatsink? If you are, get an aftermarket one. It will be cooler and quieter as well.

Edited May 16, 2011 by Black6464

[VaporX]

A CPU fan that ramps up and is noisey can be caused by a heatsink with dust blocking fins. I would look there first.

 

Now what do you mean the rear fan is not "blowing out" properly?

[krazypoloc]

I used to think along the lines of "More air flowing through the case must be better". This is not the case at all. In my experience I have found that positive air pressure is best. Think of it as saturating the inside of the case with fresh air for the components to use, if the air is being flushed out too quickly only the components that are sucking in air with a fan will get cooling.

[sticknstone]

The cpu fan operation is normal for a case temp of 50 c. My phenom x4 9550 has a STOCK max case temp of 61 c. While you have a phenom II I am reasonably sure the tech hasen't changed much it should be at the same max case temp or maybe 62, 63 c.

 

If you are looking to quiet the stock heatsink reach a case temp below 40 c.

 

That case has those three fans included. The case temp is 50 c. It is high for the amd cpu, but still allowable.

 

My analysis of the case cooling would be a poor grade. The front fan has a drive bay directly in front of it impeding air flow. The grille area and mesh spacing for the 120mm front fan is unknown to me but it appears more asthetic than functional. Low speed fans for a high end computer is pretty silly, but I have no idea what the fan cfm is.

 

If that were my case and in consideration of the 50 c case temp I would be measuring the front grille area and the mesh spacing. Taking off the fans and reading the fan specs on the fan sticker. Putting the fan model number into an internet search engine to determine what cfm and rpm the fans are rated at. I would remove the drive bay in front of the front fan; however, it may not be getting enough air through the front bezel.

 

To determine what fan performs better than a current fan installed. The manufacturer fan rpm/cfm data need to be compared. For example the installed 120 mm fan is rated at 60 cfm and the fan intended for upgrade is rated at 70 cfm then because the greater cfm indicates more air flow it is better. I would use the fan law cfm2 = (rpm2/rpm1) x cfm1 when the fan will be operating at different rpms, for example when the fan will be speed controlled for quiet operation or light pc use. The fan law numbers the fan operation 1 and 2 which represent 1 for the manufacturer tested state and 2 for the speed controlled state. Now the installed 120 mm fan is rated at 60 cfm and 1500 rpm, but is speed controlled to 1100 rpm then by the fan law (1100/1500) x 60 is 44 cfm air flow at the speed controlled state. The fan intended for upgrade is rated at 70 cfm and 2000 rpm and will be speed controlled to 1100 rpm then by the fan law (1100/2000) x 70 is 38.5 cfm air flow at the speed controlled state. Therefore, the installed fan will provide greater air flow than the upgrade option.

 

I would consider a fan grille mod of the round wire type. This type of grille allows the most air flow through it.

 

While the display on the chassis monitors temps with the included sensors it is important to place them in the optimal position. Double check the temps in bios, or software.

After the case displays 50 c restart the os and enter bios and go to the pc health screen to read the component temps.

My gigabyte mobo came with an oc utility which has a health monitor screen to display temps perhaps asus has a simular utility which will display temps, or find a different utility which will display the temps.