[yansheng]

Dear experts,

I was doing a validation test of a fully turbulent plane channel flow. From the literature, it is obvious that velocity, turbulence intensity and Reynolds shear stress are the most important statistics that anyone who is doing turbulent flow analyzing will calculate. However, I was wondering how to calculate the Reynolds shear stress in Star-ccm+. Some example graphs are attached below from John Kim (1986), Turbulence statistics in fully developed channel flow at low Reynolds number.


Best regards,

Yansheng

[LuckyTran]

Reynolds stresses are absolutely paramount, complete knowledge of the Reynolds stresses at every x,y,z will close the RANS equations. The only difference between one turbulence model and another is the Reynolds stresses (and experiment). If you want to unambiguously compare results, then the Reynolds stresses are the only way to do so.

It varies widely depending on the models how to compute them:

If you are doing LES then you can brute force calculate the Reynolds stresses.
If you are using a Reynolds stress model then the Reynolds stresses are available, but they will be modeled.

If you are using an eddy viscosity model, then you will have to calculate the Reynolds stresses yourself by using the Boussinesq eddy viscosity hypothesis (+ the strain rate & the turbulent viscosity). You will need to setup new field function definitions.