Low reynolds number airfoils simulation - COMSOL Multiphysics
 

Hello!

I'm working on a simulation of low reynolds number airfoils in software COMSOL Multiphysics 4.3, using the k-e turbulence model and extra-fine triangular mesh.

I have three basic questions that I could not solve yet. I know that in other simulation software these values is "ready", like in CFX. The questions are:

- How can I get the value of "y+" in COMSOL?
- How can I get the characteristic graphs cl and cd in this software?
- There is some method in COMSOL to analyze the drag and lift through the pressure and velocity fields? Since this is the only result that I arrived.

Thank you in advance.

P.S.: Is the first time that I work with numerical simulation in this software, and it's really important because my "term paper" is based on this simulation.

I don't know much about COMSOL, but: What does low mean? What's the Reynolds number? Is a turbulence model even necessary? In either case, k-e is the wrong model for "low Reynolds number" turbulent flow

Adrin

Hello,

Thanks for answer adrin...

The reynolds number is Re = 4,73x10^5 in this case, and the turbulence model is necessary cause we working with "critical airfoils" (S1223 and E423), the laminar flow don't run correctly, show errors about non convergence points, etc.

As said, I don't have much experience with numerical simulation, and applied the k-e model taking as a basis other papers of undergraduate and master's. The COMSOL have two other options: K-omega and Spallart Almaras models.

Have you any suggestion about it?

And about the obtaining of cl and cd, an idea?

Thanks and regards.

You will not get accurate results with ordinary k-omega or sa since in these models flow will be turbulent everywhere. They cannot feel laminar part of the flow and transitional region.

Re=470000 doesn't seem "low" to me…

I am not working with COMSOL myself but I have people in my group doing so from time to time. I have seen them perform all of the above questions, so either it exists in the code or they use their own functions. You can easily define your own functions/variables in COMSOL. For instance, drag can be found by integrating the appropriate function of drag across the boundary of interest.

This might help: www.comsol.com/showroom/downlo...e/models.mph.cylinder_flow.pdf

Re=1e9 might be low and Re=1e3 might be high. Without knowing the system it is impossible to tell. Comparing to pipe flow or flow over a cylinder I would agree it doesn't seem low. I have seen many airfoil calculations/experiments with substantially higher Re though.