How accurate is XFoil/XFLR5 at Reynolds numbers between 2 million and 10 million?

From my understanding it's only accurate at lower Re values.

  • 3
    $\begingroup$ I believe it's quite the opposite: XFOIL would be more accurate at higher Reynolds number. The higher the Reynolds number, the smaller the displacement thickness and the closer the displacement body is to the original airfoil shape. $\endgroup$ – JZYL Mar 29 '20 at 16:08
  • $\begingroup$ I edited a little and from the context I assumed that "2" meant "2 million"; if I got something wrong then you can just roll back my changes or edit again. $\endgroup$ – Pondlife Mar 29 '20 at 20:17

Before version 6.9 (30 Nov 2001), XFOIL had some problems at Re > 3 million, but that's now been fixed. Its documentation says:

The BL equations are normally discretized with two-point central differencing (i.e. the Trapezoidal Scheme), which is second-order accurate, but only marginally stable. In particular, it has problems with the relatively stiff shape parameter and lag equations at transition, where at high Reynolds number the shape parameter must change very rapidly. Oscillations and overshoots in the shape parameter will occur with the Trapezoidal Scheme if the grid cannot resolve this rapid change. To avoid this nasty behavior, upwinding must be introduced, resulting in the Backward Euler Scheme, which is very stable, but has only first-order accuracy. Previous versions of XFOIL allowed a specific constant amount of upwinding to be user-specified. Currently, XFOIL automatically introduces upwinding into the equations only in regions of rapid change (typically transition). This ensures that the overall scheme is stable and as accurate as possible.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.