Pressure developed by 80 mm axial fans

SpyderCat · Post by **SpyderCat** » Wed Apr 09, 2003 9:29 am

I was interested in what pressure undervolted fans would develop.
The graph shows some results from different manufacturers, and the purple points represent the 80 mm Panaflo fans.
I think the plotted points are sufficiently close that we can accept this as a reliable graph.
I'm especially interested in fans that plot above this line, and to the left side. When you know of any, please let me know.

dukla2000 · Post by **dukla2000** » Fri Apr 11, 2003 8:22 am

SpyderCat - I am not sure you can assume anything about pressure when you undervolt a fan. Panaflo are the only folk I have seen who show different curves at different voltages. Except those graphs are really small (imprecise) but it does seem to me that the shape of the curve is different at different voltages.

For a range that is well above your line look at Papst 8312: OK, so I am cheating because they are 32mm

But L is 2000rpm +-2.2mm (max), M is 3000rpm +-5mm etc.

Also, I am uncomfortable looking at 'max pressure' specifications: this is the point at which the fan stalls (moves no air) which is a useless result in cooling. Hence all our fetish on rounded cables, neat layouts, good airflow paths with open entry & exit points. I know it is equally naive to assume zero static pressure and the facilities required to measure any of our actual boxes is non-trivial.

For convenience (or out of ignorance), I figure most of us assume our setups are close to zero static pressure which is a lot more optimistic than assuming our fans are stalling? Like most of the addicts on this forum I have spent a fair amount of time trying to analyse fan specs to find my perfect fan: in my records I have tried to note the cfm at 10Pa - again an arbitrary number but trying to be a more educated guess than either extreme. The scary thing about this is that at 10Pa static pressure, my Papst 8412NGLs are blowing 0cfm

Which also made me realise there comes a time that your super slow, super quiet fan is actually useless!! Oh well, back to more specs and experiments!

SpyderCat · Post by **SpyderCat** » Fri Apr 11, 2003 10:41 am

dukla2000 wrote:SpyderCat - I am not sure you can assume anything about pressure when you undervolt a fan. Panaflo are the only folk I have seen who show different curves at different voltages. Except those graphs are really small (imprecise) but it does seem to me that the shape of the curve is different at different voltages.

For a range that is well above your line look at Papst 8312: OK, so I am cheating because they are 32mm But L is 2000rpm +-2.2mm (max), M is 3000rpm +-5mm etc.

Hi Dukla,

Okay, I included the Papst fans in the graph.

dukla2000 wrote:For a range that is well above your line look at Papst 8312

well above? Lets say "above", and in the rpm range we are interested in, the difference is almost non exsistent.
Still, you are right. Thicker fans develop more pressure

dukla2000 wrote:Also, I am uncomfortable looking at 'max pressure' specifications: this is the point at which the fan stalls (moves no air) which is a useless result in cooling. Hence all our fetish on rounded cables, neat layouts, good airflow paths with open entry & exit points. I know it is equally naive to assume zero static pressure and the facilities required to measure any of our actual boxes is non-trivial.

For convenience (or out of ignorance), I figure most of us assume our setups are close to zero static pressure which is a lot more optimistic than assuming our fans are stalling? Like most of the addicts on this forum I have spent a fair amount of time trying to analyse fan specs to find my perfect fan: in my records I have tried to note the cfm at 10Pa - again an arbitrary number but trying to be a more educated guess than either extreme. The scary thing about this is that at 10Pa static pressure, my Papst 8412NGLs are blowing 0cfm Which also made me realise there comes a time that your super slow, super quiet fan is actually useless!! Oh well, back to more specs and experiments!

I noticed that almost every fan produces (discounting scale) the same graph. An almost straight line with a hump at 70% of it's rated CFM.
I think the fans are stalling at the hump.
Also the pressure at zero CFM seems a good indicator for the pressure developed at the hump.
There is a big difference in blade area as a percentage of the swept area among the fans, but I haven't found any characteristics to belong to "wide bladed fans" not found with "narrow bladed fans".
The volume we displace with low-rpm fans is sufficient to cool the rigs, provided we can get enough pressure build up from those fans.
I want to do some tests with fans "in tandem" (behind each other), to see if my feeling is right.
Another advantage of "tandem fans" is that almost by definition one of the fans is located inside the case, so it's noise is somewhat muffled.
Another "gut-feeling" of mine: Fans with large hubs are better for developing a high pressure.

Thanks for contributing.

Regards, Han.

dukla2000 · Post by **dukla2000** » Fri Apr 11, 2003 11:44 am

I want to do some tests with fans "in tandem" (behind each other), to see if my feeling is right.

You mean "The static pressure capacity of two fans in series can be doubled at zero air flow condition, but do not increase the airflow in the free-air situation. An additional fan in series increases the volume flow in a higher static pressure enclosure.

Thus, in series operation, the best results are achieved in systems with high resistance. " ?

OK - so I was cheating again. The curve you want is at www.sunon.com and go to the "Engineering Info" page, about 2/3 of the way down. Maybe this link gets you straight there?
http://www.sunon.com.tw/device.htm#The% ... 0Operation

SpyderCat · Post by **SpyderCat** » Fri Apr 11, 2003 12:44 pm

dukla2000 wrote:
I want to do some tests with fans "in tandem" (behind each other), to see if my feeling is right.
You mean "The static pressure capacity of two fans in series can be doubled at zero air flow condition, but do not increase the airflow in the free-air situation. An additional fan in series increases the volume flow in a higher static pressure enclosure.

Thus, in series operation, the best results are achieved in systems with high resistance. " ?

OK - so I was cheating again. The curve you want is at www.sunon.com and go to the "Engineering Info" page, about 2/3 of the way down. Maybe this link gets you straight there?
http://www.sunon.com.tw/device.htm#The% ... 0Operation

Yep, perhaps not litterally, but that's more or less what I mean.
I want the ability to overcome a high resistance, with enough volume, without the noise that normally comes with high pressure buildup.
Why I want this? I want to suck enough air through the heatsink (high resistance) without using a high revving fan.

The picture at the Sunon-site is a little deceiving. When the rotation of both fans is the same, the pressure isn't doubled as the air comes swirling/rotating to the second fan. You need some distance (or a fence) to straighten the flow.

I haven't checked, but I suppose all 80 fans rotate in the same direction.
We would have been very lucky when we could just buy just 2 counter-rotating silent fans. No, I don't think we are THAT lucky.

Thanks for the link to the Sunon-site. Hadn't been there

Regards, Han

SpyderCat · Post by **SpyderCat** » Fri Apr 11, 2003 1:02 pm

Here, from the Cooling-forum (posted just a few minutes ago)

jinu117 wrote:Oh... for radiator, I found putting fan in front of and behind radi instead of on either and running it a lower speed decreases noise level and improved performance.

He speaks of 2 fans "in tandem" with a fence between them