Simulating a fan using a cylinder

Kaptene · March 23, 2021, 13:45

Hi All
I am simulating a fan that cools an electronic equipment inside a column. To simulate the fan On FLUENT, I have used a thin cylindrical geometry with fan boundary conditions from both sides of the cylinder. In the inlet it's "Fan Intake" and the outlet is "Fan Exhaust).

I am not sure if someone could share his experience on the best way of modeling this. The purpose of my study is to see how the cooling is performed, it does not really matter to see the flow across the fan.

Thanks a lot

Kaptene · March 23, 2021, 21:12

Need help please

flotus1 · March 24, 2021, 03:10

You are currently the only person who knows which software you are using. The "right" way to model a fan inside the computational domain strongly depends on that.

Kaptene · March 24, 2021, 22:55

I am using Fluent Ansys

Kaptene · March 29, 2021, 11:47

Any help is really appreciated

Continuum · March 29, 2021, 12:38

Hi,

I have done this. It is a simple and fast way to introduce recirulation into an interior domain. I would suppose you'd want some fan swirl. Plus you'd want some throughput. The way my software works allows me to add BC to any interior control surface. So if I draw mass out one surface, I can put it back in on a another surface (other side of the fan) for a net zero mass. For the swirl, you'd need to define the fan axis and compute omega.spin ^(r-r.axis.fan) to get the spin velocity set on the surface.

You can break the fan surface into 4 quadrants for simplicity as I have done this too. You can patch 4 different swirl velocities without the above spin calculation.

If I get some time, I will add a picture.

Regards,

nima_nz · March 29, 2021, 12:42

Hi

Please take a look at this
https://youtu.be/UXaeyCGE3nk

March 24, 2021, 03:10		#3
flotus1 Super Moderator Alex Join Date: Jun 2012 Location: Germany Posts: 3,400 Rep Power: 47	You are currently the only person who knows which software you are using. The "right" way to model a fan inside the computational domain strongly depends on that. aero_head likes this.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Creating & Meshing 2D Planar Surface (for simulating 2D Centrifugal Fan)	Arrows	FLUENT	0	January 21, 2017 12:02
Simulating Fan in Ansys CFX	Cabi365	CFX	1	November 4, 2016 12:20
Trouble modelling a jet fan	josee	OpenFOAM Pre-Processing	0	September 23, 2016 13:06
Drag force coefficient too high for a flow past a cylinder using komega sst	Scabbard	OpenFOAM Running, Solving & CFD	37	March 21, 2016 16:16
benchmark: flow over a circular cylinder	goodegg	Main CFD Forum	12	January 22, 2013 11:47

March 23, 2021, 21:12		#2
Kaptene New Member Join Date: Mar 2021 Posts: 4 Rep Power: 5	Need help please

March 24, 2021, 22:55		#4
Kaptene New Member Join Date: Mar 2021 Posts: 4 Rep Power: 5	I am using Fluent Ansys

March 29, 2021, 11:47		#5
Kaptene New Member Join Date: Mar 2021 Posts: 4 Rep Power: 5	Any help is really appreciated

March 29, 2021, 12:38		#6
Continuum New Member Continuum Join Date: Aug 2009 Posts: 19 Rep Power: 16	Hi, I have done this. It is a simple and fast way to introduce recirulation into an interior domain. I would suppose you'd want some fan swirl. Plus you'd want some throughput. The way my software works allows me to add BC to any interior control surface. So if I draw mass out one surface, I can put it back in on a another surface (other side of the fan) for a net zero mass. For the swirl, you'd need to define the fan axis and compute omega.spin ^(r-r.axis.fan) to get the spin velocity set on the surface. You can break the fan surface into 4 quadrants for simplicity as I have done this too. You can patch 4 different swirl velocities without the above spin calculation. If I get some time, I will add a picture. Regards,

March 29, 2021, 12:42		#7
nima_nz Member Nima Join Date: May 2014 Location: Ireland Posts: 51 Rep Power: 12	Hi Please take a look at this https://youtu.be/UXaeyCGE3nk