I have a Problem.

I modeled a square duct 25mm high and wide and 100mm long. The secondary flow in the outlet is very small in the corner. When the duct is longer, the secondary flow will be bigger. in Fluent I can see the secondary Flow full after 20-30 mm from inlet. in cfx in a longer duct after 400-500mm. I have the same boundary conditions. In Fluent I used the RSM. In cfx 5.6 I used all Reynolds stress models but every model has circa the same results. In literature the secondary flow is full after a very short distance (2*h). I think fluent calculate correct.

what is my mistake in cfx I think cfx can calculate good secondary flows too.

Can anybody help me?

Did you use the same grid, and was it structured or unstructured?

If it was unstructured you need to remember that the resolution in CFX depends on the number of nodes and not the number of elements, so the same mesh would actually be coarser.

It is the same mesh.

I made a structured Mesh in Icem CFD and exported it to fluent and cfx.

I had made a longer Duct (4*h). There ist the secondary Flow very good. It is nearly the same like in Literature.

[Mahmudrochy]

Could you please tell me how to plot the secondary flow in a plane normal to the flow direction in fluent/cfx?

Originally Posted by Marko
;68904

I have a Problem.

I modeled a square duct 25mm high and wide and 100mm long. The secondary flow in the outlet is very small in the corner. When the duct is longer, the secondary flow will be bigger. in Fluent I can see the secondary Flow full after 20-30 mm from inlet. in cfx in a longer duct after 400-500mm. I have the same boundary conditions. In Fluent I used the RSM. In cfx 5.6 I used all Reynolds stress models but every model has circa the same results. In literature the secondary flow is full after a very short distance (2*h). I think fluent calculate correct.

what is my mistake in cfx I think cfx can calculate good secondary flows too.

Can anybody help me?