[pisykl]

Hello,

I want to simulate the flow around a NACA profile for different stagger and incidence angles. Especially the loss coefficients and velocity ratios behind the profile are of interest and will be matched with experimental results I'll obtain in the next weeks.

For very small incidence angles I'm getting bad results: With the previous settings of my case I obtain large losses as a result of a flow separation on the full length of my blade. The results of the experimental investigation of my colleagues suggest that these enormous losses won't appear for my configuration too.

Cause the residuals and some monitors oscillated I switched to a transient calculation. The residuals haven't got better (maybe a problem of timestep?), but the flow separation is much smaller and fits better in the loss bucket created by the values of higher incidence angles.

Cause I'm not sure if the outflow boundary condition is appropriate I repeated the steady simulation with an pressure outlet condition. This led not to better residuals, but the losses are smaller than in my first simulation.

Now I'm confused – what would be the right approach?

I've attached some velocity plots and the residuals for the first steady simulation with the outflow bc (blending first-to-second-order).

Thanks for your help!

[pisykl]

Residuals for outflow bc (Blending first-to-second-order 0.5 -> 1)