Airflows over rotorcraft are far more complex and
more difficult to predict than those that occur on
fixed-wing aircraft. The inability to predict many of
these flows is a direct or indirect contributor to the
cost, duration, and risk of the rotorcraft development
process. For this reason, the application of computational
fluid dynamics (CFD) is an important and
growing part of the rotorcraft research process.
However, rotorcraft aerodynamics is difficult for CFD
because the limiting physical problems of rotorcraft
differ from those of fixed-wing aircraft for which CFD
was originally developed. An example of such a flow
is the wake that is shed by the rotor. Such wakes are
also generated by conventional aircraft and are made
visible in the contrails of high-altitude airliners. The
primary difference with rotorcraft is that because of
its rotary motion, a rotor never gets far from its shed
wake. This adjacent wake strongly determines the
performance and acoustics of a rotor. The ability to
predict this wake with the required accuracy has
been shown to require far greater computational
resources (with conventional methods) than is
practical for engineering analysis. The objective of
this work is to develop new CFD wake-computation
methods that are accurate and practical for everyday
use.
A new class of CFD methods, specifically
designed for engineering analysis, has been developed.
The method is a modification of the potential
solvers that were an early focus of CFD development.
The approach, called "vortex embedding," consists of
a means of inserting a freely convecting vortical wake
- a capability that such solvers originally lacked.
This approach permits the prediction of this wake
with a grid system that is orders-of-magnitude smaller
(and hence more efficient) than present methods.
Until this year, the approach lacked the ability to
predict the viscous flow on the rotor surface. This
shortcoming has been removed with the development
of an overset/hybrid solver that combines the
embedding method in the wake region and a Navier-Stokes
viscous-flow solver adjacent to the blade
surface.
The prediction of tilt-rotor flows is ideal for this
method because the high lift of these rotors results in
blade-root stall. Recent computations are demonstrating
the ability to predict the performance of the
XV-15 tilt-rotor performance. The accompanying
figure shows experimental and computed figures of
merit (a measure of rotor power) for this rotor. It is
shown that the figures of merit at higher thrust levels
(characteristic of actual operating conditions) is well
predicted. The inset in this figure is a visualization of
the computed wake that is shed by this rotor.
Point of Contact: F. Caradonna
(650) 604-5902
fcardonna@mail.arc.nasa.gov
Back To Top
Previous Paper
Return to Global Civil Aviation
Next Paper 
