CFD Online Discussion Forums

CFD Online Discussion Forums (
-   FLUENT (
-   -   hemisphere in channel flow (

kirill September 12, 2004 08:21

hemisphere in channel flow
Hi dear all,

I would like to ask for your ideas:

My problem: I need to model a hemisphere lying on the bottom of the channel with aim to obtain drag and lift coeficients for the Re-number range from 1e+4 to 2e+5. The flow is fully developed, so as inflow boundary condition i give the log velocity profile.

Here it comes: in reality the bottom of the channel is covered with big roughness elements (their size is comparable with the hemisphere size), the hemisphere is lying on the top of them. I give the modified velosity profile as inflow and consider the top of roughness elements as fictive bottom (moving wall boundary condition). Do you think I can get a reasonable estimation of forces acting on the hemisphere and is the bottom boundary condition appropriate?

Thanks a lot for any help


Titiksh Patel September 13, 2004 06:58

Re: hemisphere in channel flow

I think the B.C. at the inlet is alright. But you can change the wall roughness in the Wall B.C. panel. Why are you using moving wall. I cant understand that..?? Is your bottom wall moving?

I think u can estimate the forces even if you dont use moving wall.

Regards, Titiksh

kirill September 14, 2004 17:29

Re: hemisphere in channel flow
Hallo Titiksh,

thanks a lot for Your response.

I want to use moving wall because the roughness is to big and according to the log law (I use modified one for such big roughness) on the level of hemisphere bottom the mean velocity component has still a rather big value.

In Fluent user manual states: "Note that it is not physically meaningful to have a mesh size such that the wall-adjacent cell is smaller than the roughness height. For best results, make sure that the distance from the wall to the centroid of the wall-adjacent cell is greater than Ks".

In my case the hemisphere radius is 4 cm and roughness Ks is also of the same range, the hemisphere is lying directly on the top of roughness elements. It is mean that my mesh have to be at least 4 cm in direction normal to the wall, or am I wrong?

I tried to simulate with moving and stationary wall - I get the same drag coefficient but different lift coeffs.

What do You think?


Titiksh Patel September 16, 2004 11:51

Re: hemisphere in channel flow
Hi Kirill

Well I understand your point that yr mesh should be at minimum of 4 cm from the wall, because the scale of yr roughness is 4 cm or whatever.

But the roughness doesn't remain constant along the wall then how are you calculating effective roughness.

What is the velocity of the wall you are provinding? Acc. to me the drag coefficient may change if you change the wall velocity, then how will you make sure that which wall velocity will give accurate results and on what basis..? As in practical there is no movement of wall.

Secondaly kindly provide me how are you calculating drag coefficient, which formulae are you using.

Regards, Titiksh

All times are GMT -4. The time now is 03:30.