Hello everyone,
I am trying to simulate cavitating flow over the NACA66(mod) hydrofoil by interPhaseChangeFoam solver. SST k-w turbulence model and Kunz model is applied. The initial setting of k and w is small when I use the calculator by setting the turbulence length scale 7% of the hydraulic diameter.
Boundary conditions in nut file are:
nut —> nutUSpaldingWallFunction
inlet —> calculated
outlet —> calculated
The simulating goes well but the unexpected cavity shedding appears since attached cavitation is expected.Bounded variables omega_max has great value during the simulation.
However when I change the k and w to a great number,there is no cavity shedding and the bounded value seems good to me.
Does anyone know why k and w have such influence on the simulation?
Attached files are the residuals plotted by pyFoam and fvSchemes, fvSolutions files.
https://www.cfd-online.com/Tools/turbulence.php
Hi,
I think, changes in k-w inlet conditions can lead to next flow behaviour in your computational domain:
1) Change in pressure distribution along airfoil which can lead in different vapor production/destruction rates around airfoil.
2) Increase in nut can lead to more dissipative momentum exchange, so maybe you should try keeping nut=k/w constant.
3) Rapid increase of velocity at zero time (U=0 at t0=0, U 0 at t0+dt). Try to increase velocity at inlet for some time
Hello Matvey,
Thank you for your impressive answer! Since I am not familiar with the turbulence theory, I still have some doubts.
1)The small value for k=0.000171,w=0.02918 is calculated from turbulence intensity=0.5% and length-sclae=0.07*(inlet hydraulic diameter), and the large value for k=0.0172 ,w=1852.6 is calculated from turbulence intensity=5% and eddy viscosity ratio=10.I haven't got the exact parameters in the experiment, and I don't know the nut=k/w in the experiment.Then which approach is suitable for the simulation?
3)The inlet velocity and internalField velocity at zero time is 2.1 m/s.If I want to increase velocity velocity at inlet for some time, should I have time-varying speed boundary condition at inlet?
Hello, Ziming Zhu
I'm not an expert in turbulence also, but i guessed several reasons which can lead to behaviour, described by you. So i can't say which turbulent viscosity ratio more realistic, but i can ask you to change intensity at inlet whithout changing ratio. And if results are same for both cases then the reason in eddy viscosity ratio.
Regarding Q.2 - yes, you need time varying B.C. Also, if you have initial velocity internal field with value, equal to inlet, then flow "hits" the hydrofoil with dynamic pressure rho*w*w/2 and this can create large pressure drop and vapor generation term, Try to increase velocity field from 0 to 2.1m/s.
- Badges
- Science topic
- Similar topics
- Engineering
- Chemical Engineering