Learning Forum

[nicolo.cappello]

Dear All,

 

I need to get an info which is somehow crucial: I'm doing time-dependent simulation of an axisymmatric system using Fluent 2023R2, and I need to monitor the volume flow rate through the cross-section in the middle of a duct. Such cross-section is perpendicular to the axial direction, so in practice it is a radial line. I've defined a new Line-surface (by means of the TUI, that is surface/line-surface ...) and defined a report definition (Surface Report -> Volume Flow Rate) to get the volumetric flow rate through that surface.

The problem is that the signal I get is the opposite of the one I expect from physics. So I have to conclude that Fluent is considering as normal to the surface the opposite of the one pointing in the positive axial direction.  

 

My question is: how does Fluent define the unit normals to surfaces internal to the domain which have been created from the TUI?

 

Thanks in advance for the precious help.

Nicolò

[Rob]

Are you reporting on a line or an interior surface?  Please can you post a screen grab of the report definition?

[nicolo.cappello]

Hi Rob, I'm sorry for bothering. Have you tried take a look at my question?

Thanks a lot.

Nicolò

[nicolo.cappello]

Hi Rob, sure here they are: screenshot1 is about the internal surface, and screenshot2 shows the report def.

 

 

Thank you for your help!

[nicolo.cappello]

Just to be clear, when I run the simulation I get a volume flow rate which is positive in the convergent direction of the duct, which is wrong I think. What am I missing?

 

[Rob]

Sorry, we had a message issue & I was then off for a few days. 

There isn't a normal associated with those surface types so I'd tend to use mass averged velocity (vector) to get mass flow data. 

[nicolo.cappello]

Dear Rob, no problem, thank you for your support.

So if I've understood it correctly you suggest defining a custom vector corresponding to the velocity vector and then use Surface Report -> Custom Vector Flux ?

Am I misunderstanding your advice?

Thank you!

[Rob]

No, I'd use the x, y or z velocity depending on what's normal to the plane. 

[nicolo.cappello]

Alright, I've got it! 

Thanks a lot for your precious advices!