Static pressure of Fans - How much is X mmH20 in the real world?

May 30, 2018
6
0
10
Hello everyone,
Thanks already in advance for any help.
I have been looking into case fans for a while now, but I can not seem to get an intuitive understanding of how much static pressure is optimal for certain scenarios, I do get that 4 to 10 mmH20 is considered a lot and that anything below one is likely not good for a radiator, but this does not help in fully understanding how these mechanisms work therefore this question:
In the real world how much pressure will a fan face, in the open mesh (with a dust filter) and at a radiator or air cooler of varying thickness. What are the variables (in terms of what causes the resistance) and how can you estimate a value of mmH20.

Questions i would like to be able to answer (based on estimates) are along the lines of "when using a fan with 170m^3/h and 5mmH2O in a mesh front + dust filter scenario, would a fan with 2.5mmH20 less (halve) and 30 m^2/h more move more or less air compared to its counterpart?"

This is so far one of the most insightful articles I found:
https://www.anandtech.com/show/9737/the-corsair-sp140-led-af140-led-sp120-af120-fan-review/4
It suggests to me that assuming a nearly linear relation between resistance and airflow parameters is not too far off, therefore knowing the resistance in a certain scenario would give a quite good estimation, so can anyone provide a method for such a resistance estimation? How much is 1mmH20 in pressure resistance?

This question is only partially for practical application and partially for understanding and gaining insight into how these things work. Specific values, estimates, measurements and mathematics would be greatly appreciated if anyone is able to provide anything like that!

Thanks again for your time, effort and knowledge.
 
Solution
What you're really looking for is a P&Q chart. The "static" in static pressure means no air is moving. This makes it useless for cooling. The P&Q chart gives a curve of air flow at various resistances (pressure drop). Airflow is generally non linear. Add in a variable like fan RPM which tends to be goemetric in it's effect and things get complex in a hurry.
Here's a spec. sheet from a Delta 120x38mm 12V. fan.
http://www.delta-fan.com/Download/Spec/AUB0812VH-SP00.pdf
Here are some fans at Digi Key filtered the same. Pages of them You can see the spec. sheet for each one.
https://www.digikey.com/products/en/fans-thermal-management/dc-fans/217?k=Fan&k=PWM&k=&pkeyword=Fan&pv790=9&FV=fffc025b%2C1f140000%2Cfff40012%2Cfff80052&pageSize=
Add in...
My answer probably won't satisfy your technical curiosity ( which is admirable). I just skip the advertising BS and go for industrial cooling fans like Dell and HP use. Generally a thicker fan will support more airflow and pressure than a thinner one. I also look at the amperage rating and assume that more=more.
So I look at 120x38mm instead of 120x25mm aftermarket, and .8Amp and up instead of .3Amp "toys".
Delta, Sunon, MBT, Nidec, AVC and Foxcon.
https://www.newegg.com/Product/Product.aspx?Item=N82E16835706028&ignorebbr=1&nm_mc=KNC-GoogleAdwords-PC&cm_mmc=KNC-GoogleAdwords-PC-_-pla-_-Case+Fans-_-N82E16835706028&gclid=CjwKCAjw2_LcBRBYEiwA_XVBUzIMmWzYv3zAVjq0yiqPwEKvJ-bTJrgtlOLIzvz_gidCmtyBQxSVRhoCSdYQAvD_BwE&gclsrc=aw.ds
My favorite is the Delta AFC1512DG 150x50mm 1.8A from some Dell dual CPU workstations. Yes it's a 6"x2" fan.
https://www.ebay.com/p/Dell-NC466-Precision-490-Fan-Assembly-VZ/1141445915?iid=111530604196&chn=ps
 

dederedmi5plus

Prominent
Aug 17, 2018
257
1
560
it'll be much easier to simulate with CAD software, what you're post here need proper measurement instruments, input the data to compute, the result, well, some cooler vendor might pay you, in the end, this work will result more efficient blade design, noctua already done it and call it anti stall knob, for us who won't test will appreciate any given data.
 
May 30, 2018
6
0
10
Thanks for your answers! I have looked at a few of the industrial and very heavy duty fans - still didnt gain any insight by doing so.
To get some actual comparable test data i found this: https://us.hardware.info/category/23/case-fans/testresults
The issue is this got even more confusing as my current conclusion is that the data given by the manufacturers is completely useless and should be ignored - especially any db statements (this is blatantly obvious once taking a look at enermax' T.B.Silence PWM 140mm which is marketed as 9db fan and produces test results north of 44db - this is a systematic issue with enermax on a side note.
Similar issues occurred with the performance drops of fans when tested on a radiator in relation to their listed static pressure, there is little (i could find none) relation and fans of 10mmH20 sometimes had higher loss then 0.5mmH20 pressure fans. After more math my only conclusion about this as well is that measuring conditions are arbitrary and only comparable within one manufacturer (at best). The link above is now kinda my semi trust go to as i have verified a few of their data-points / relations between tests to be similar to other findings (within variance between fans given the sample sizes of 1)

That does not mean i have lost interest in the mathematical background of this.
I found this: https://www.captiveaire.com/manuals/airsystemdesign/designairsystems.htm#TotalPress_VelocityPress_StaticPress
I am not quite there yet to apply this or do meaningful conclusions but maybe it helps some of you (out of curiosity only as established above)
about increasing fan design - i do not think mathematics is the way to go - maybe i will sit down and create a model eventually (to run search algorithms etc) but doing so would require far more knowledge in physics then what i currently have so this will have to wait.

for the time being I'm just sad the gentle typhoon (high rpm variants) is no longer around (and will never be again) because the only thing any testing and my math did not change is that this fan was outstanding in all regards.
 

dederedmi5plus

Prominent
Aug 17, 2018
257
1
560
can never avoid sound pressure level issues, that'd be PC enclosure ability to minimize internal noise.

Some reviewers such as xtremerigs.net had tested numerous rads, readers like me only interest on the given data prior to opt best cheap rad for my build, I believe this forum's readers are also fond of certain rad is gonna good enough to cool down their build.

Hardware Labs introduced "optiFlow" push-pull rad, basically it increase the chances of radiator's core surface being pressurized by air and resolve the static pressure drop, the push fan will produce higher pressure on the surface near the pull fan, and the pull fan will produce higher pressure on the surface near the push fan,
 
What you're really looking for is a P&Q chart. The "static" in static pressure means no air is moving. This makes it useless for cooling. The P&Q chart gives a curve of air flow at various resistances (pressure drop). Airflow is generally non linear. Add in a variable like fan RPM which tends to be goemetric in it's effect and things get complex in a hurry.
Here's a spec. sheet from a Delta 120x38mm 12V. fan.
http://www.delta-fan.com/Download/Spec/AUB0812VH-SP00.pdf
Here are some fans at Digi Key filtered the same. Pages of them You can see the spec. sheet for each one.
https://www.digikey.com/products/en/fans-thermal-management/dc-fans/217?k=Fan&k=PWM&k=&pkeyword=Fan&pv790=9&FV=fffc025b%2C1f140000%2Cfff40012%2Cfff80052&pageSize=
Add in other manufacturers and other sizes and you have thousands of options.
 
Solution

dederedmi5plus

Prominent
Aug 17, 2018
257
1
560
The P&Q chart such as tested by https://www.anandtech.com/show/9737/the-corsair-sp140-led-af140-led-sp120-af120-fan-review/2 was one of piece of bigger "PC cooling" picture, you're presented with "rad optimize fan and case optimize fan", for those who seek best balance between entire loop cooling capacity and lesser noise, a 1 to 5 deg C delta over ambient would be a big deal.

Back then I even applied rad > fan shroud > fan setup just to get couple deg cooler, since I turned to tempered glass case, ditched that, now, my setup is push static pressure fan > rad > pull RGB fan
 


after reading on noctua's web, under their buyers guide that there really isn't much difference, performance wise, between their "static pressure" and "airflow" fans, i think their word was they perform equally well, i kind of lean toward your approach. And thanks for pointing out that dell 120x38 fan. But that 150x150x50.8 fan, hell, any bigger and it'd qualify as a whole house fan

 
The big fan was used in workstations with 2x 130W CPUs. But because they're workstations and not servers noise was a consideration. It's actually quieter under load than the smaller one. I use them in BTX systems where 1 fan is the design. It's actually a useful mod for people in 3rd world countries with hot climates and no A/C.
 
"It's actually a useful mod for people in 3rd world countries with hot climates and no A/C."

I have no doubt - i just wish i could have used some of those 38mm x 120mm fans but in my new rig, i'm trying to keep the "rooftop" radiator from crowding the motherboard - i wish they had some 38mm fans in 140mm size for the front radiator

i am glad you alerted me to those though
 
May 30, 2018
6
0
10
Talking about simply using larger fans, I am currently running 2 200mm Coolermaster fans, they perform incredibly well for being close to inaudible, bigger fans still beat anything else if you got the space in the case.
Im currently looking a lot into centrifugal fans, if any of you have expertise in that field as well (its technical questions again) please pay a visit to this question: http://www.tomshardware.co.uk/answers/id-3876225/centrifugal-fans-forward-backward-curve-free-area-hub-boss.html