[MaLa]

Hello

I need some help. I wish to run a channel flow simulation. I am trying to set my flow conditions to Re_tau = 590. To achieve that, I have kept the half channel width = 1, u_tau =1 and nu =1/590.

To ensure that u_tau =1, I have used the relation that
2 * rho * u_tau ² =(dP/dx) * ( 2 * half-channel-width)
i.e.
I need to specify an (explicit) pressure gradient source to make sure that u_tau = 1.

Do you think this approach is correct so far?

Now the second question: To specify a pressure gradient explicit source, I looked up the file pressureGradientExplicitSource.C. What I could make out from it was that, the user specifies a U_bar. At each time step, the current U_bar is calculated by averaging across all cell volumes, and depending on the difference between the averaged and the specified U_bar, the pressure gradient is adjusted.

If this is true, then the pressure gradient is NOT constant but adjusts itself to meet the U_bar. This is not what I want. I want to enforce a constant pressure gradient source, so that the u_tau (as mentioned above) stays fixed.

What do you think? Comments and thoughts would be greatly appreciated.

[mturcios777]

What about setting the pressure BC's at the start and end of the channel, using fixedValue or totalPressure?

[MaLa]

Yes mturcios777, that would be an option. But I wanted to use the cyclic boundary condition on the inlet and outlet patches. If I do that, I have to change the solver explicitly I guess, in order to include the (constant) pressure gradient term. What do you think?

[turbu]

hi,

I met the same problem, do you find a solution to set the constant pressure gradient value?

thank you

[Kbshariff]

Hi, I have a similar problem, Do you have the solution to fixed the pressure gradient?


Thanks


Kabir