Hi all,

In the 'Reference Values' panel, you can set a 'depth' variable which is applicable to 2D problems only. I would assume this is a kind of Z coordinate value?

Does anyone know how to correctly determine the value for depth? My 2D grid for flow past a cylinder has a height of 3 and a length of 6, so what is the correct value for depth in this case?

I've thoroughly searched the Fluent online help, but there is very little on this.

Thanks in advance

FLUENT assumes a "depth" for 2D simulations which is the unit of lenght you're using. If you use meters, depth is 1 meter.

Hi

ap

Thanks for responding, ap.

How do I define the correct depth for a domain 3m in the Y direction and 6m in the x direction?

I don't know if it's possible to change the depth in a 2d case. If do a 2D simulation, you decide not to consider it.

Hi

ap

Hi go for menu Report > Reference Values you can set something like that you want there but I would read the manual

The problem is that the manual does not give any indication of how to calculate the depth in the reference values panel.

I've been varying the value for my flow past a cylinder simulation and it directly effects the drag coefficient on the cylinder. In other words, I can get any drag coefficient value I like by varying this parameter!

As far as I can tell, fluent just multiplies things appropriately by the depth.

So if I ask for the total force on a boundary, it is multiplied by the "depth" (i.e. if I double the depth, the force is double), but if I ask for average pressure, it is unaffected.

Where exactly do you get your drag coefficient from? I would have thought depth should cancel out and it would remain constant.

Look at: "26.3.1 Computing Forces and Moments." in the user guide.

"The force coefficient is defined as force divided by [(1/2)rhoV^2A] , where rho, v, and A are the density, velocity, and area explicitly specified in the Reference Values panel."

In other words if you mess around with the "depth" value, (which will increase your force) then you should also mess with the "area" value.

I think this means you will only get the correct coefficient if you explicitly set the reference values to fit your case.

Rob

Cheers Rob,

I'll look into that!

[folan]

the question here is if a Fluent 2d computation with a depth allow a 3d flow? can we have a 3D turbulent structure with this kind of configurations? I dont think ... there is no mesh in this direction
the flow arround the cylindre is 3D with a hight Reynolds number