Describe the bug
The (freestream) turbulence intensity is defined as u'/U_mean, and it is used like that to define e.g. freestream tke at far field boundaries. However, transition models use the percentage value of turbulent intensity, so 10 x u'/U_mean. This factor of 100 is missing in SU2, e.g. in transitionLM or the SST transition model. This means that you cannot consistently use a transition model (input turbulence intensity in %) together with an inlet boundary for instance (turbulence intensity at inlet should be a fraction, so 0.01*percentage-value) Actually, in the description of the LM model, the factor of 100 is
explicitly mentioned when computing intensity from tke:
https://turbmodels.larc.nasa.gov/langtrymenter_4eqn.html
I think the best is to keep the standard definition that the input turbulence intensity is a fractional value, e.g. 0.05 or 0.10, and that we convert to the percentage value (in this example 5 or 10) in the transition models.
see also the discussion on this issue on:
https://www.cfd-online.com/Forums/su2/235365-free-stream-turbulence-intensity-option.html
Describe the bug
The (freestream) turbulence intensity is defined as u'/U_mean, and it is used like that to define e.g. freestream tke at far field boundaries. However, transition models use the percentage value of turbulent intensity, so 10 x u'/U_mean. This factor of 100 is missing in SU2, e.g. in transitionLM or the SST transition model. This means that you cannot consistently use a transition model (input turbulence intensity in %) together with an inlet boundary for instance (turbulence intensity at inlet should be a fraction, so 0.01*percentage-value) Actually, in the description of the LM model, the factor of 100 is
explicitly mentioned when computing intensity from tke:
https://turbmodels.larc.nasa.gov/langtrymenter_4eqn.html
I think the best is to keep the standard definition that the input turbulence intensity is a fractional value, e.g. 0.05 or 0.10, and that we convert to the percentage value (in this example 5 or 10) in the transition models.
see also the discussion on this issue on:
https://www.cfd-online.com/Forums/su2/235365-free-stream-turbulence-intensity-option.html