[Quentin_C]

Hi everyone!

I’m a bit of a newbie in Fluent, but I decided for a project to simulate the performances of a propeller blade using moving reference frames and k-eps model. I’m trying to get a similar result from my experimental data but I have some errors messages showing up during the process. Namely ‘turbulent viscosity limited to viscosity ratio for x cells’ (x going up to 3e+9!) and / or ‘reverse flow on y faces on outflow’… I don’t know what could have gone wrong and would gladly share with you what you think

I’ve divided my problem as the geometry file show: 1 wall for my propeller, a cylindrical enclosure for my rotating part and a brick enclosure for my stationary domain. Tried to make the rotating part as near as possible of the propeller and keep sufficient distant between inlet and outflow. I have then 2 interfaces (see mesh part) and put the same size for the mesh elements. I then start the resolution of my problem. I put the rotating velocity I want for my rotating domain (6000rpm). Consider at the inlet a 0.1% turbulence intensity (I’ve read that this value is set based on the wind tunnel intensity experiment measures) and a turbulence Length Scale at 0.3m (not shown in photo) (considering this link https://www.sharcnet.ca/Software/Flu...ug/node217.htm). Actually for my calculations I define and use k and epsilon using the same link (for my inlet) so k=0.015 et epsilon=7.547e-5.

For my reference values, I chose the data specific of my propeller (total area, length = diameter, enthalpy of air, basic pressure and velocity from inlet, affected to my rotating zone). Then I calculate with the value of time steps as shown as in the pictures… I have a converging solution (residuals continuity at e-4 and fluent says it’s converged) but I’ve got an enormous amount of cells with the turbulent viscosity error. My results seems good for the lift of the propeller if I take the values as soon as the solution converged and the amount of cells drops slowly as the calculation continues.

Do you think the mesh could be too much coarse? Should I keep the default reference values? How does this work for the solution? How should I define my turbulence parameters? Ever encountered this error? Can I consider my results good even with this message or should I let continue calculate until there is no cells left concerned? Thank you kindly if you share some of your ideas for I will gladly use some help ^^