Roman Li asked 4 months ago
Comparison of the small airplane (500mm wing span) T2 simulations between XFLR and FLOW 5 gives different stall speed results (final CL differs a lot and for some other simulations I did, becomes far more than theoretical airfoil CLmax).
It looks like the solver does not precisely follow the constant lift requirement. In the 1.25 kg model, the lift difference could be 0.2 N, which produces the difference in stall speed 17.3 vs. 20.8 m/s. The lift plot is not flat in flow5 (see the screenshots attached). Are there some tolerance settings which I don’t know for a T2 solver?
Furthermore, the T6 simulation gives the same wrong results, but having an additional CD spike on the curve which I cannot explain.
Can you help with this, please?
Hello,
Thanks for the feedback. I’ll look into the issue of the lift in T2 polars and will post back.
The bump in the drag curve in T6 polars is likely due to an intersection of wake panels with plane panels at this specific angle of attack. You can check this visually by activating “wake panels” in the 3d view. If this is not the explanation, please share the project file and I’ll investigate.
André
The reason why the lift is not constant in the T2 polar is that the velocity is adjusted only approximately to speed up the analysis.
In most cases the error is negligible, but not apparently in the present case.
For the next version, I’ll see if this can be improved without increasing too much the computation times.
André
Thanks for the comment. Carefully reevaluated the T6 plots, I found they follow the constant lift requirements quite precisely, so the results are comparable with XFLR T2 solver. Also, I enabled VPW and viscous loop for the costs of simulation time, and now I’m satisfied with the results, even though there is no real baseline to compare. Thank you!
Roman