Principles of Aeroelasticity AERO 431 - Fall 2023

Lecture 13: Classical Flutter Analysis - 2-DOF system

Instructor: ©Dr. Mojtaba Kheiri

Department of Mechanical, Industrial and Aerospace Engineering

Concordia University

November 15, 2023

Montréal, Canada
 Theory 2-DOF flutter analysis Algorithm Example

Classical Flutter Analysis

The objective of the “classical flutter” is to determine the flight
conditions that correspond to the flutter boundary (i.e. U = UF )
Flutter boundary corresponds to conditions for which (at least) one
of the modes of motion has a simple harmonic time dependency, i.e.
x = xeiωt
It is implied that all modes of motion are convergent (i.e. stable) for
less critical flight conditions (i.e. U < UF )
All modes other than the critical ones are convergent at the flutter
boundary
1.5 1.5 15

1 1 10

0.5 0.5 5

0 0 0

-0.5 -0.5 -5

-1 -1 -10

-1.5 -1.5 -15

0 0.5 1 1.5 2 0 0.5 1 1.5 2 0 0.5 1 1.5 2

Stable (U < UF ) Flutter boundary (U = UF ) Unstable (U > UF )

Ref: Hodges D.H., Pierce G.A., 2011. Introduction to Structural Dynamics and Aeroelasticity. Cambridge University Press.
2 / 11
 Theory 2-DOF flutter analysis Algorithm Example

Classical Flutter Analysis (Cont.)

According to classical flutter analysis, it is assumed that the solution

involves simple harmonic motion. The equations of motion are then
solved for the flight conditions that yield such a solution
Classical flutter analysis cannot provide any definitive measure of
flutter stability other than the location of the stability boundary
The primary strength of classical flutter analysis is that it needs only
the unsteady airloads for simple harmonic motion
Here, it is assumed that oscillatory motions are relatively small in
amplitude (compared to the characteristic length), and thus it is
sufficient to use a linear aerodynamic theory for the computation of
airloads generated as a result of simple harmonic motion

Ref: Hodges D.H., Pierce G.A., 2011. Introduction to Structural Dynamics and Aeroelasticity. Cambridge University Press.

3 / 11
 Theory 2-DOF flutter analysis Algorithm Example

Classical flutter analysis (Cont.)

Let us consider a typical section model undergoing simultaneous plunging
and pitching motions, represented by h(t) and θ(t), respectively.
Assuming that the motion is simple harmonic, we can write
h(t) = heiωt , θ(t) = θeiωt , (1)
where ω is the frequency of motion.
Since a linear aerodynamic theory is used, the lift and pitching moment
resulting from simple harmonic motions h(t) and θ(t) will also be simple
harmonic with frequency ω; thus,
L(t) = Leiωt , M(t) = Meiωt . (2)
The amplitudes of these airloads can be obtained as complex linear
functions of h and θ as
 
h
L = −πρ∞ b3 ω 2 ℓ ℓh (k, M∞ ) + ℓθ (k, M∞ )θ) ,
b
 
4 2 h
M = πρ∞ b ω ℓ mh (k, M∞ ) + mθ (k, M∞ )θ) . (3)
b
4 / 11
 Theory 2-DOF flutter analysis Algorithm Example

Classical flutter analysis (Cont.)

As seen from expressions given in (3), the aerodynamic coefficients ℓh ,
ℓθ , mh and mθ are functions of k and M∞ which reflect the unsteadiness
and compressibility effects of the flow, respectively; k normally has a
value between 0 and 1 for conventional flight vehicles.
An example is the expressions from Theodorsen’s unsteady aerodynamic
theory:

2iC(k)
ℓh = 1 − ,
k
i 1 2C(k)
ℓθ = a + [1 + 2( − a)C(k)] + ,
k 2 k2
2i( 12 + a)C(k)
mh = a − ,
k
1 ( 1 − a)[1 − 2( 21 + a)C(k)]i 2( 12 + a)C(k)
mθ = a2 + − 2 + , (4)
8 k k2
where C(k) is called the Theodorsen’s function.
5 / 11
 Theory 2-DOF flutter analysis Algorithm Example

Two-degree-of-freedom flutter analysis

The above figure shows the two-degree-of-freedom (plunge-pitch) representation

of a typical section model. The pitch angle is represented by θ and the plunge
by h. The plunge and pitch motions are restrained by a translational and a
torsional spring of the stiffness kh and kθ , respectively.
The equations of motion for the system shown above are written as

mḧ + mbxθ θ̈ + kh h = −L, IP θ̈ + mbxθ ḧ + kθ θ = Mea , (5)

Ref: Hodges D.H., Pierce G.A., 2011. Introduction to Structural Dynamics and Aeroelasticity. Cambridge University Press.

6 / 11
 Theory 2-DOF flutter analysis Algorithm Example

Two-degree-of-freedom flutter analysis (Cont.)

where Ip is the mass moment of inertia about the reference point, and
Mea = M1/4 + b( 12 + a)L is the aerodynamic pitching moment about the
reference point; also, ( ˙ ) is the time derivative.
We assume that the motion is simple harmonic, i.e. h = heiωt , θ = θeiωt
(we assume that we are at the flutter boundary); the corresponding lift
and moment can be written as L = Leiωt and Mea = Meiωt . Thus,
equations (5) may be re-written as the following algebraic equations:

− ω 2 mh − ω 2 mbxθ θ + mωh2 h = −L,

− ω 2 mbxθ h − ω 2 IP θ + IP ωθ2 θ = M. (6)

Using expressions (3), equations (6) can be simplified to:

Ref: Hodges D.H., Pierce G.A., 2011. Introduction to Structural Dynamics and Aeroelasticity. Cambridge University Press.

7 / 11
 Theory 2-DOF flutter analysis Algorithm Example

Algebraic equations
n m ωh 2 o h n mx o
θ
[1 − ( ) ] + ℓh (k, M∞ ) + + ℓθ (k, M ∞ ) θ = 0,
πρ∞ b2 ℓ ω b πρ∞ b2 ℓ
n mx oh n I ωθ 2 o
θ P
+ m h (k, M ∞ ) + [1 − ( ) ] + m θ (k, M∞ ) θ = 0.
πρ∞ b2 ℓ b πρ∞ b4 ℓ ω
(7)
Equation (7) can be simplified by defining the following dimensionless
parameters:
m IP ωh
µ= 2
, r2 = 2
, σ= , (8)
πρ∞ b ℓ mb ωθ
where µ is the mass ratio, r the mass radius of gyration about P, and σ
the frequency ratio; thus,
n ωθ 2 oh
µ[1 − σ 2 () ] + ℓh + (µxθ + ℓθ )θ = 0,
ω b
h n ωθ o
(µxθ + mh ) + µr2 [1 − ( )2 ] + mθ θ = 0. (9)
b ω
8 / 11
 Theory 2-DOF flutter analysis Algorithm Example

The flutter determinant

The algebraic equations presented in (9) can have non-trivial solution
only if the determinant of the coefficients of h/b and θ, which is called
the “flutter determinant,” is zero.

µ(1 − σ 2 λ) + ℓh µxθ + ℓθ
2 = 0, (10)
µxθ + mh µr (1 − λ) + mθ

where λ = (ωθ /ω)2 .

The flutter determinant may be written as a quadratic equation of λ, i.e.
Aλ2 + Bλ + C = 0, where A, B and C are generally functions of k and
M∞ . This means that for any combination of k and M∞ , two roots,
which are generally complex, are found from the equation.

Tip: At the flutter boundary, motion is simple harmonic, which

necessitates that ω (and thus λ) to be a real number since h = heiωt and
θ = θeiωt .
9 / 11
 Theory 2-DOF flutter analysis Algorithm Example

The solution process

Assuming an incompressible flow problem, i.e. M∞ → 0, the solution
procedure for a system with known values of µ, r, σ and xθ is as follows:
Find the value of k at which the imaginary part of one of the two
roots for λ becomes zero, i.e. kF – there are two ways to do this:
(1) find λ1 (k) and λ2 (k) analytically using the discriminant (or ∆)
of the quadratic equation and let their imaginary parts to zero and
use, e.g. ‘fzero’ function of MATLAB, to find kF , and (2) input a
range of values for k and observe where the imaginary part for one
of the roots becomes (close to) zero
Use the root
√ for which λ is real, i.e. λF , and find ωθ /ωF from
ωθ /ωF = λF .
Find the dimensionless flutter speed from VF = (ωF /ωθ )/kF .
Knowing b and ωθ , find the dimensional flutter speed from
UF = bωθ VF .
What is the solution procedure if M∞ is not negligible?
10 / 11
 Theory 2-DOF flutter analysis Algorithm Example

Example

Using the classical flutter analysis, compute the dimensionless flutter

speed and frequency for an incompressible, two-degree-of-freedom flutter
problem in which a = −0.2, e = −0.1, µ = 20, r2 = 0.24 and σ = 0.4.
For the aerodynamic coefficients, use those of Theodorsen’s theory with
C(k) approximated as
2
0.01365 + 0.2808ik − k2
C(k) = . (11)
0.01365 + 0.3455ik − k2
Solution:
The aerodynamic coefficients from Theodorsen’s theory were given in
equation (4).
Try the MATLAB program.

Ref: Hodges D.H., Pierce G.A., 2011. Introduction to Structural Dynamics and Aeroelasticity. Cambridge University Press.

11 / 11