Aeroelasticity

Lecture 5:
Three-Dimensional Wings
G. Dimitriadis
Introduction to Aeroelasticity

Wings are 3D
E All the methods described until now
concern 2D wing sections
E These results must now be extended to
3D wings because all wings are 3D
E There are two methods for 3D wing
aeroelasticity:
Strip theory
Panel methods
Introduction to Aeroelasticity

Strip theory
E Strip theory breaks
the wing into
spanwise small strips
E The instantaneous lift
and moment acting on
each strip are given
by the 2D sectional lift
and moment theories
(quasi-steady,
unsteady etc)
Introduction to Aeroelasticity

dy
S
y

Panel methods

Introduction to Aeroelasticity

Wake Panels
i,j+1
i+1,j+1

y
i,j
i+1,j

E The wing is
replaced by its
camber surface.
E The surface
itself is replaced
by panels of
mathematical
singularities,
solutions of
Laplaces
equation

0
x
c

Hancock Model
E A simple 3D wing model is used to
introduce 3D aeroelasticity
!

1*+,-

A rigid flat plate of span s,

chord c and thickness t,
suspended through an axis
xf by two torsional springs,
one in roll (K) and one in
pitch (K).

!!"$
!!"#

"
$

0
!"'

!"&

The wing has two degrees of

freedom, roll () and pitch
().

).

!"%
!"#
!"#
!

!"$
!

/*+,!!"#

!!"$
!!"#
)*+,-

Introduction to Aeroelasticity

Hancock model assumptions

E The plate thickness is very small
compared to its other dimensions
E The wing is infinitely rigid (in other words it
does not flex or change shape)
E The displacement angles and are
always small
E The z-position of any point on the wing is

z = y + x x f
Introduction to Aeroelasticity

Equations of motion
E As with the 2D pitch plunge wing, the
equations of motion are derived using energy
considerations.
E The kinetic energy of a small mass element
dm of the wing is given by
2
1
1 2
dT = z
dm = dm y
+ x x f 

2
2

))

E The total kinetic energy of the wing is:

m
2 2
2
2 
2






T=
2s + 3s c 2x f + 2 c 3x f c + 3x f
12

Introduction to Aeroelasticity

) )

Structural equations
E The potential energy of the wing is simply
1
1
2
V = K + K 2
2
2

E The full structural equations of motion are

then:
$ I
&I
%

I ' *

- $ K
+ 

. + &
)
I ( , / % 0

0 ' * - * M1 )+ . = + .
K ( , / , M 2 /

I = ms2 /3, I = m c 2x f s /4, I = m c 2 3x f c + 3x 2f /3

Introduction to Aeroelasticity

Strip theory
E The quasi-steady or unsteady
approximations for the lift and moment
around the flexural axis are applied to
infinitesimal strips of wing
E The lift and moment on these strips are
integrated over the entire span of the wing
E The result is a quasi-steady pseudo-3D lift
and moment acting on the Hancock wing
s

M1 = yl( y )dy
0

M 2 = mx f ( y )dy
0

Introduction to Aeroelasticity

Quasi-steady strip theory

E Denote, = and h=y. Then
l

mxf

E Carrying out the strip theory integrations

will yield the total moments around the y=0
and x=xf axes.
Introduction to Aeroelasticity

3D Quasi-steady equations
of motion
E The full 3D quasi-steady equations of
motion are given by

E They can be solved as usual

Introduction to Aeroelasticity

Natural frequencies and

damping ratios

Introduction to Aeroelasticity

Theodorsen function
aerodynamics

E Again, Theodorsen function aerodynamics

can (unsteady frequency domain) can be
implemented directly using strip theory:
l

mxf

Introduction to Aeroelasticity

Flutter determinant
E The flutter determinant for the Hancock
model is given by

E And is solved in exactly the same was as

for the 2D pitch-plunge model.
Introduction to Aeroelasticity

p-k solution

Introduction to Aeroelasticity

Comparison of flutter speeds

Wagner and
Theodorsen
solutions are
identical.
Quasi-steady
solution is the
most conservative
Ignore the other
solutions

Introduction to Aeroelasticity

Comparison of 3D and strip

theory, static case
The 3D lift distribution
is completely different
to the strip theory
result!

Introduction to Aeroelasticity

Vortex lattice aerodynamics

E Strip theory is a very gross approximation
that is only exact when the wings aspect
ratio is infinite.
E It is approximately correct when the aspect
ratio is very large
E It becomes completely unsatisfactory at
moderate and small aspect ratios (less
than 10).
Introduction to Aeroelasticity

Vortex lattice method

Introduction to Aeroelasticity

w
v
i,j+1
u

i+1,j+1

ni,j

E The basis of the

VLM is the
division of the
wing planform to
panels on which
lie vortex rings,
usually called a
vortex lattice
E A vortex ring is a
rectangle made
up of four straight
line vortex
segments.

i,j
y
ri,j

i+1,j

Characteristics of a
vortex ring

E The vector nij is a unit vector normal to the

ring and positioned at its midpoint (the
intersection of the two diagonals - also
termed the collocation point)
E The vorticity, , is constant over all the
segments.
E Each segment is inducing a velocity [u v w]
at a general point P.
E In the case where the point P lies on a vortex
ring segment, the velocity induced is 0.
Introduction to Aeroelasticity

Panelling up and solving

E The process of dividing a wing planform into

panels
E The wing can be swept, tapered and twisted.
E It cannot have thickness
E The wake must also be panelled up.
E The object of the VLM is to calculated the
values of the vorticities on each wing panel
at each instant in time
E The vorticity of the wake panels does not
change in time. Only the vorticities on the
wing panels are unknowns
Introduction to Aeroelasticity

Calculating forces
E Once the vorticities on the wing panels are
known, the lift and moment acting on the
wing can be calculated
E These are calculated from the pressure
difference acting on each panel
E Summing the pressure differences of the
entire wing yields the total forces and
moments
Introduction to Aeroelasticity

Panels for static wing

Even if the wing is
not moving, the
wake must still be
modelled because
it describes the
downwash induced
on the wings
surface and,
hence, the induced
drag.

Introduction to Aeroelasticity

Unsteady wake panels

E At each time
instance a new
wake panel is shed
in the wake.
E The previous wake
panels are
propagated
downstream at the
local flow airspeeds
Introduction to Aeroelasticity

Wake shapes
Wake shape behind a
rectangular wing that
underwent an
impulsive start from
rest.
The aspect ratio of the
wing is 4 and the
angle of attack 5
degrees.

Introduction to Aeroelasticity

Effect of Aspect Ratio on lift

coefficient
Lift coefficient variation
with time for an
impulsively started
rectangular wing of
varying Aspect Ratio. It
can be clearly seen that
the 3D results approach
Wagners function (2D
result) as the Aspect
Ratio increases.

Introduction to Aeroelasticity

Unsteady lift and drag

Unsteady lift and drag
coefficients for a birdlike wing performing
roll oscillations at
10Hz with amplitude 2
degrees.

Drag coefficient
Introduction to Aeroelasticity

Lift coefficient

Industrial use
E Unsteady wakes are beautiful but expensive
to calculate.
E For practical purposes, a fixed wake is used
with unsteady vorticity, just like Theodorsens
method.
E The wake propagates at the free stream
airspeed and in the free stream direction.
E Only a short length of wake is simulated (a
few chord-lengths).
E The result is a linearized aerodynamic model.
Introduction to Aeroelasticity

Aerodynamic influence
coefficient matrices

E 3D aerodynamic calculations can be further speeded up

by calculating everything in terms of the mode shapes of
the structure.
E This treatment allows the expression of the aerodynamic
forces as modal aerodynamic forces, written in terms of
aerodynamic influence coefficients matrices.
E These are square matrices with dimensions equal to the
number of retained modes. They also depend on
response frequency.
E Therefore, the complete aeroelastic system can be written
as a set of linear ODEs with frequency-dependent
matrices, to be solved using the p-k method.
Introduction to Aeroelasticity

Commercial packages
E There are two major commercial packages
that can calculate 3D unsteady
aerodynamics using panel methods:
MSC.Nastran
ZAERO (ZONA Technology)

E They can both deal with complete,

although idealized geometries

Introduction to Aeroelasticity

ZAERO AFA example

E The examples manual of ZAERO features
an Advanced Fighter Aircraft model.
Aerodynamic model

Structural model
Introduction to Aeroelasticity

BAH Example
E Bisplinghoff, Ashley and Halfman wing

E FEM with 12 nodes and 72 dof

Introduction to Aeroelasticity

First 5 modes of BAH wing

Introduction to Aeroelasticity

GTA Example
E Here is a very simple aeroelastic model for
a Generic Transport Aircraft

Finite element model: Bar elements

with 678 degrees of freedom
Introduction to Aeroelasticity

Aerodynamic model: 2500 doublet

lattice panels

Flutter plots
for GTA

First 7 flexible modes.

Clear flutter mechanism
between first and third
mode (first wing bending
and aileron deflection)

Introduction to Aeroelasticity

Time domain plots

for the GTA

V<VF
Introduction to Aeroelasticity

V=VF

Supersonic Transport
E The SST is a
proposal for the
replacement of
the Concorde.
E The aeroelastic
model is a halfmodel
E The
aerodynamics
contains the
wing and a
rectangle for the
wall
Introduction to Aeroelasticity

Flutter plots
for SST

First 9 flexible modes.

Clear flutter mechanism
between first and third
mode.

Introduction to Aeroelasticity