Dear Users,

considering incoherences between the results of the flowrates calculated and the plots given by Flow3d 9.3, I would like to know how the flux surfaces are interpolated in the mesh, and then how the flowrate is calculated.

More precicely, is the flux surface sensitive with the proximity of a limit of the mesh?

Thanks

Julien

Are you using flux planes to calculate the volumetric flowrates? What variables are you looking at? Are you looking at the mesh volumetric flowrates? What version are you using?

I am using Flow3D v9.3, trying to measure a volumic flowrate through a baffle define as a flux surface. I don't use particles, I suppose the software integrates the velocities on the plane...but how? in which cells? What about cells that are not entirely part of the flux surface?

My aim is to qualify the distribution of the flow over a crest, so that I have placed a lot of juxtaposed baffles...but the distribution seems to be strange...so comes my question.

Thx

Julien

A baffle flux surface should be placed on a cell surface, if it isn't then it will take the nearest face velocity. Also, if you have a circular baffle and if in one cell, it only occupies 1/16 or less than 1/2 then it will not use that area in the calculation.

Hi Julien,

Flux surfaces are defined with 100% open baffles and, therefore, behave like baffles in the way they interact with mesh. A baffle is located at cells faces; if it is defined not at a cell face, it will be moved to the nearest face by the pre-processor. Also, a cell face is either fully occupied by the baffle or not at all, depending on whether its original definition blocks more or less than 50% of the face.

Baffles and flux surfaces can be visualized in 3D using the Favorize tool in Meshing & Geometry.

The flow rate at a flux surface is integrated over all cell faces occupied by it.

One additional limitation in the flux surface tool is that any two flux surfaces must be at least two cells apart for the flow rates at both of them to be accurate.

Michael

Hi Michael,

I understand that, in order to make my flux surface measures accurate, i should set a fixed point at its limiters coordinates. But in that case, if I juxtapose my baffles, it should work, isn't it?

Do you mean that if my flux surface only takes place on less than 50 % of the cell face, this one is eliminated from the integration area? And that on the other hand, if more than 50 %%, the velocity is integrated on all the cell's face?

This two cells which should separate two flux surfaces only avoid integrating the velocities on the same cells for both of the flux surfaces, is that right?

Thanks a lot

Julien

Hi Julien,

I am not sure how you juxtapose baffles. Could you give more details?

You are right about the 50% logic.

I forgot what the two-cell separation is for :~) I will find out this week, let you know.

Michael

In fact, I just set the baffle like in the following example:

baf 1 : definition : x = 2 limiters : yl = 1; yh = 2

baf 2 : definition : x = 2 limiters : yl = 2; yh = 3

Setting a fixed point between the baffle (i.e. y = 2 here), I have already got better results, but should I believe in the accuracy of these ones?

Julien

[sunxiao]

Have you got the flow rate by the baffle?
if that is the case, could you tell me how to do it?
I am trying get the volume flow rate by the baffle (flux surface), but it didn't work.
Thank u!

Originally Posted by Julien Pralong
;93242

In fact, I just set the baffle like in the following example:

baf 1 : definition : x = 2 limiters : yl = 1; yh = 2

baf 2 : definition : x = 2 limiters : yl = 2; yh = 3

Setting a fixed point between the baffle (i.e. y = 2 here), I have already got better results, but should I believe in the accuracy of these ones?

Julien