Swirl-Velocity Specification of 2-D Fan Boundary Condition
 

I opened a relevant thread before at the student community forums of ANSYS but did not find a closure about the subject (i.e. https://studentcommunity.ansys.com/t...ary-condition/ ). I'll appreciate any help from the cfd-online community.

I am defined a 2-D fan model using the fan boundary condition on a square surface in a 3-D simulation of a room. I am puzzled by how Fluent interprets the tangential velocity since the tangential velocity I get as a result is always less than what I input, and typically half of the constant value I input as tangential velocity (i.e. when I use values on the order of axial velocity).

The results below correspond to the case with a constant tangential velocity of 1 m/s. According to the plot of tangential velocity as a function of y direction along the midline of the fan surface, the values plateaued at around 0.5 m/s outside hub (half of the input). How should I interpret this?

https://studentcommunity.ansys.com/c...&upscale=false

https://studentcommunity.ansys.com/c...&upscale=false

I'll greatly appreciate your help in clarifying how Fluent uses inputs for tangential velocity in the 2-D fan boundary condition? Why don't I see the tangential velocity imposed between the hub radius and outer dimension of the boundary surface as I define? Is there a physical reasoning behind this?

Which boundary condition are you using? Fluent doesn't have a "fan boundary condition" Just post a screenshot of your boundary condition input.

Here is the screenshot of what I use.

Any help from the community would be appreciated.

Can you check velocities slightly downstream/upstream of the fan surface?

1. Tangential velocity
2. X velocity
3. Z velocity

at the velocity inlet and fan bc

OK, but can you change the contour plane to another plane that is slightly downstream/upstream of the fan plane? Like, 5mm away from the fan.

Also, why do you have a velocity inlet BC?

No special reason for using velocity inlet BC. I just used a uniform velocity to be able to see the effect of swirl function in the fan BC.

Attached are contours for 5mm upstream, fan and 5mm downstream with the same scale.

Thanks

So, the max. velocity 5mm downstream is higher than max. velocity on fan plane, right?

Can you find a plane with the velocity value that you input (1m/s I guess)?

Try to turn OFF the node values.

Correct.

I have attached the contour at 25 mm downstream of the fan, where the tangential velocity becomes 1 m/s (node values off). I guess I understand where you are going. Does the fan BC impose the tangential velocity within the first cell downstream of the fan, since the cell size is 25 mm?

I don't know how the BC works, and couldn't find any explanation on the theory/user guide. But there might be an averaging problem, that's why I asked for turned off nodes.

As I understand, this is not a "problem" for you, you just wanted to make sure that the fan BC works the way it intended to.