Hi,

I had run into some problems to analyse the flow filed of an axial fan flow. The following is my questions:

1. Is this case running in the steady-state problem or transient problem in Star-CD? 2. If we use multiple rotating reference frames to simulate the rotating flow, which method to choose as analysis model ¡V explicit multiple reference frame or implicit multiple reference frame. Is it possible to use cyclic + implicit rotation to treat the flow in an axial fan (only deal with one blade row)? 3. What is turbulent model to use in axial fan case? We will need to investigate the flow in the tip clearance region. 4. As for the boundary conditions specification, what kind of boundary conditions are placed on the inlet and outlet ¡V ¡¥Inlet Boundary Condition and Outlet Boundary Condition¡¦ or ¡¥Pressure Boundary Condition¡¦ in the boundary condition terminology of Star-CD? 5. What to set up the grids (one channel between blades or grid with one blade inside.), boundary locations, boundary location, analyses methods¡K..

Thank your for your help.

chienchi

Chienchi,

Part of the response is it really depends on what you are looking for and how much detail/cpu time you want to spend in looking at the problem. I have done plenty of axial fan studies using implicit multiple reference frames. I have run them inlet/outlet and have had very good convergence. Now if you are looking for more exact solution with distribution, or, noise, a transient solution may be better. If you are looking at thermal heat transferring through the blades, explicit may actually due better then the implicit, however, it might take longer to converge.

Typical fan models have used k-e. For the gap clearance, we have used a two layer model. I have not looked to much detail on the models where we are running different gap clearance test to know all the settings.

Hope this helps.