[Paradai]

Evening all,

I'm currently running a CFD case similar to a channel flow but with blocky obstacles on the ground, a moving wall (V=80 m/s) at the top and the inlet and outlet are set at periodic (translational) conditions with no Mass Flow Rate or Pressure Gradient determined.

The strategy was to determine the Mass Flow Rate by running this case using a RANS model until convergence and note the MFR using the Report Flux Tool and then update the periodic condition to Specify the MFR at that value when using the DES model.

However the outcome doesn't seem to be turbulent as the pressure gradient set as zero originally. Is there a way to determine the MFR correctly or to determine the Pressure Gradient for the RANS case to get the correct parameters and produce realistic turbulent flow?

Thank you in advance,
Paradai

[LuckyTran]

You mean the RANS result is not turbulent or the DES result is not turbulent? For the DES you should provide some perturbations in order to help the flow transition to a turbulent state.

[Paradai]

Referring to the DES result, I loaded the RANS result and switched the model to DES both which uses Realisable k-e model.

Is the lack of turbulence caused by loading the RANS results rather than re-initialise?

[LuckyTran]

If you simply switch from RANS to DES/LES, you have only the mean flow without velocity fluctuations in it and this tends to stay laminar. You have to give it a good whack.

While still in RANS, you should run the cmd:

/solve/initialize/init-instantaneous-vel

This will take give some perturbations based on k. It needs the k field so you have to do this while that field is available.