Indian Institute of Technology, Bombay

Department of Aerospace Engineering

AE 305 - Flight Mechanics (II)

Project Report

Dynamic Stability Analysis of Learjet 24

Group 10
Lagnesh Mahapatra (210010038)
Mayank Bajaj (210010039)
Sourabh Chouhan (210010062)
Yash Dnyaneshwar Kotkar (210010077)

April 30, 2024

Contents
1 Introduction 4

2 Objectives 4

3 Stability & Control Derivatives 5

3.1 Derivatives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
3.1.1 Longitudinal Dynamics . . . . . . . . . . . . . . . . . . . . . . . . 6
3.1.2 Lateral Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3.2 Eigenvectors & Eigenvalues . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2.1 Longitudinal Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.2.2 Lateral Dynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2.3 Polar Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8

4 Transfer Function - Longitudinal States 9

4.1 Transfer Functions for Elevator Inputs . . . . . . . . . . . . . . . . . . . . 9
4.2 Transfer Functions for Throttle Inputs . . . . . . . . . . . . . . . . . . . . 9
4.3 Changing the Stability Derivatives: Mq and Xu . . . . . . . . . . . . . . . 10
4.4 Short Term & Long Term Responses: ∆u, ∆α, ∆γ . . . . . . . . . . . . . 11

5 Transfer Function for Lateral States 12

5.1 Transfer Functions for Aileron Inputs . . . . . . . . . . . . . . . . . . . . . 12
5.2 Transfer Functions for Rudder Inputs . . . . . . . . . . . . . . . . . . . . . 12
5.3 Changing the Stability Derivatives: Lv and Nr . . . . . . . . . . . . . . . . 13
5.4 Short Term & Long Term Responses: ∆v, ∆p, ∆r, ∆ϕ . . . . . . . . . . . 14

6 Conclusions and Important Remarks 15

1
List of Figures
1 Learjet 24 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2 Polar Plots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3 Movement of Longitudinal Eigenvalues with Mq . . . . . . . . . . . . . . . 10
4 Movement of Longitudinal Eigenvalues with Xu . . . . . . . . . . . . . . . 10
5 Short Term Step Responses to 1◦ elevator deflection: (∆u, ∆α, ∆γ) . . . 11
6 Long Term Step Responses to 1◦ elevator deflection: (∆u, ∆α, ∆γ) . . . . 11
7 Movement of Lateral Eigenvalues with Lv . . . . . . . . . . . . . . . . . . . 13
8 Movement of Lateral Eigenvalues with Nr . . . . . . . . . . . . . . . . . . 13
9 Short Term Step Responses to −15◦ aileron deflection: (∆v, ∆p, ∆r ∆ϕ) . 14
10 Long Term Step Responses to −15◦ aileron deflection: (∆v, ∆p, ∆r ∆ϕ) . 14

2
List of Tables
1 Different Aircraft Parameters . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 Coefficients for Longitudinal Dynamics . . . . . . . . . . . . . . . . . . . . 6
3 Coefficients for Lateral Dynamics . . . . . . . . . . . . . . . . . . . . . . . 7
4 Pole Locations, Damping (ζ), Natural Frequency (ωn ), Time Constants (τ ) 7
5 Pole Location, Damping (ζ), Natural Frequency (ωn ), Time Constant (τ ) . 8

3
1 Introduction
Our project analyses the dynamics of the Learjet 24 aircraft. This involves studying
its longitudinal and lateral motions, stability derivatives, and control system responses.
The Learjet 24, a well-known business jet, represents a notable example of efficiency
and sophistication in aviation. Our project aims to study the mathematical models and
principles of flight mechanics that govern its behaviour. This involves understanding the
underlying dynamics governing the Learjet 24 flight. Through the calculation of eigenval-
ues and eigenvectors, we seek to identify the distinct modes governing both longitudinal
and lateral motions to learn the aircraft’s behaviour under different flight conditions. The
study extends to control theory, where we derive transfer functions for longitudinal and
lateral states. By analysing the influence of control inputs such as elevator deflection and
throttle settings, we aim to understand how these factors affect the aircraft’s stability and
controllability.

(a) Steady State Flight (b) 3 View Layout

Figure 1: Learjet 24

2 Objectives
The project aims to investigate the dynamics of Learjet 24 by using linearized small
perturbation equations and separating the longitudinal and lateral states:

1. Extract performance data available about the aircraft and create the state space
matrix for both the longitudinal and lateral dynamics

2. Obtain all the eight transfer functions for longitudinal states (u, w, q and θ) with
elevator (δe ) and throttle (δT ) as control inputs

3. Examine the influence of the stability derivatives, Cmq and Cxu , on the longitudinal
eigenvalues. Vary one stability coefficient at a time and plot the movement of the
longitudinal eigenvalues

4. Plot short-term and long-term responses of flight path angle (∆γ), speed (∆u), and
angle of attack (∆α) to a step input of 1◦ of elevator deflection

4
 5. Obtain all the eight transfer functions for lateral states (v, p, r and ϕ) with aileron
(δa ) and rudder (δT ) as control inputs
6. Examine the influence of the stability derivatives, Clβ and Cnr , on the lateral eigen-
values. Vary one stability coefficient at a time and plot the movement of the lateral
eigenvalues
7. Plot response for velocity components (v, w, u) angular velocity (p, q, r) and attitude
angles (ϕ, θ, ψ) to a step input of -15◦ of aileron deflection

3 Stability & Control Derivatives

Longitudinal Dynamics
    
∆u̇ Xu Xw 0 −g ∆u
∆ẇ  Zu Zw u0 0 ∆w
 
 =
 ∆q̇  Mu + Mẇ Zu Mw + Mẇ Zw Mq + Mẇ u0 0   ∆q 
∆θ̇ 0 0 1 0 ∆θ
 
Xδe XδT  
 Zδ e Z δ T
 ∆δ e
+ 
Mδe + Mẇ Zδe MδT + Mẇ ZδT  ∆δT
0 0

Lateral Dynamics
      
∆v̇ Yv Yp Yr − u0 g cos θ0 ∆v 0 Yδr  
 ∆ṗ   Lv Lp Lr 0    ∆p  +  Lδa Lδr  ∆δa
   
 =
 ∆ṙ  Nv Np Nr 0   ∆r  Nδa Nδr  ∆δr
∆ϕ̇ 0 1 0 0 ∆ϕ 0 0

3.1 Derivatives
All the stability and control derivatives were calculated using the data given in Pages
522-588 of [1]. The stability and control derivatives are calculated by analysing the
aerodynamic forces and moments acting on the aircraft in response to changes in its state
variables. This involves linearising the equations of motion around a trimmed condition,
typically the desired flight state such as cruise. By expressing the aerodynamic forces
and moments as linear functions of small perturbations in the aircraft’s state variables,
equations are derived to quantify how these forces and moments change with respect
to changes in parameters such as angle of attack, sideslip angle, and control surface
deflections. The trim condition was given as Max Weight Cruise Configuration.

Parameter Value Parameter Value Parameter Value

S 230 f t2 c 7.0 f t b 34.0 f t
u0 677 f t/sec q∞ 134.6 lbs/f t2 θ0 = α 2.7◦
W 13000 lbs

Table 1: Different Aircraft Parameters

5
3.1.1 Longitudinal Dynamics

QS QS
Xu = −(CDu + 2 × CD0 ) , Xw = −(CDα − CL0 ) ,
mu0 mu0
QS QS
Zu = −(CLu + 2 × CL0 ) , Zw = −(CLα − CD0 ) ,
mu0 mu0
QSc QSc QSc
Mq = Cmq c , Mu = Cmu , Mw = Cmα ,
2u0 Iy u0 Iy u0 Iy
QSc
Mẇ = Cmα̇ c 2 , Mα = u0 Mw , Mα̇ = u0 Mẇ ,
2u0 Iy
QS QSc
Zδe = Czδe , Mδe = Cmδe ,
m Iy
QSc QS QSc
Xδe = Cmδe , XTu = (CTxu + 2CTx1 ) , MTu = (CmTu + 2CmT1 ) ,
Iy mu0 Iy u0
Parameter Value Parameter Value Parameter Value
CD0 0.0216 CDu 0.104 CDα 0.30
CL0 0.13 CLu 0.40 CLα 5.84
Cm u 0.050 Cm α −0.64 Cmα̇ −6.7
Cmq −15.5 CLδe 0.46 CDδe 0
Cmδe −1.24 CTxu -0.07 CTx1 0.0335
CmT1 0 CmTu -0.003 CzTu 0

Table 2: Coefficients for Longitudinal Dynamics

    
∆u̇ -0.0167 -0.0192 0 -32.174 ∆u
∆ẇ -0.0746 -0.6634 677 0  ∆w
 
 ∆q̇  =  0.0009 -0.0105 -1.3230
  
0   ∆q 
∆θ̇ 0 0 1 0 ∆θ
−4
 
0 −3.3952 × 10  
−35.2434 0  ∆δe
+ 
−14.2726 −5.1079 × 10−5  ∆δT
0 0

3.1.2 Lateral Dynamics

QSCyβ QSbCnβ QSbClβ

Yv = , Nv = , Lv = ,
mu0 Iz u0 Ix u0
QSbCyp QSb2 Cnp QSb2 Clp
Yp = , Np = , Lp = ,
2mu0 2Iz u0 2Ix u0
QSbCyr QSb2 Cnr QSb2 Clr
Yr = , Nr = , Lr =
2mu0 2Iz u0 2Ix u0
QSCyδa QSbCnδa QSbClδa
Yδa = , N δa = , L δa =
m Iz Ix
QSCyδr QSbCnδr QSbClδr
Yδr = , Nδr = , L δr =
m Iz Ix

6
 Parameter Value Parameter Value Parameter Value
Cyβ −0.730 C yp 0 C yr 0.400
Clβ −0.110 Cl p −0.450 Clr 0.160
Cn β 0.127 Cnp −0.008 Cnr −0.200
Cyδa 0 Cyδr 0.14 Clδa 0.178
Clδr 0.019 Cnδa −0.020 Cnδr −0.074

Table 3: Coefficients for Lateral Dynamics

    
∆v̇ −0.0826 0 −676.2310 32.1383 ∆v
 ∆ṗ  −0.0061 −0.4248 0.1510 0   ∆p 
 
 = 
 ∆ṙ   0.0042 −0.0045 −0.1125 0   ∆r 
∆ϕ̇ 0 1 0 0 ∆ϕ
 
0 10.7268  
 6.6913 0.7142  ∆δa
+ 
−0.4479 −1.6572 ∆δr
0 0

3.2 Eigenvectors & Eigenvalues

The tables below give the pole locations, damping (ζ), natural frequency (ωn ) and
time constants (τ ) for the Longitudinal and Lateral dynamics.

3.2.1 Longitudinal Dynamics

Pole Damping Frequency (rad s−1 ) Time Constant (s)

−9.91 × 10−1 + 2.64i 0.351 2.82 1.01
−9.91 × 10−1 − 2.64i 0.351 2.82 1.01
−1.01 × 10−2 + 7.39 × 10−2 i 0.135 7.46 × 10−2 99.3
−1.01 × 10−2 − 7.39 × 10−2 i .135 7.46 × 10−2 99.3

Table 4: Pole Locations, Damping (ζ), Natural Frequency (ωn ), Time Constants (τ )

Based on the provided table, it’s evident that the first two eigenvalues correspond
to the Short-Period mode, while the subsequent two relate to the Phugoid mode.
The damping ratio for the Phugoid mode, which was noted at 0.135, aligns well with
theoritical results, which hover around 0.1. Conversely, the Short-Period mode exhibits
a slightly higher damping ratio, approximately 0.351. The values in the third column
represent natural frequencies. Additionally, the time constants in the fourth column
accurately adhere to the formula ζω1n .

Eigenvectors of Longitudinal State Space Matrix

 
0.0112 + 0.0196i 0.0112 − 0.0196i 0.9980 + 0.0000i 0.9980 + 0.0000i
 0.9997 + 0.0000i
 0.9997 + 0.0000i 0.0636 − 0.0019i 0.0636 + 0.0019i 

−0.0005 + 0.0039i −0.0005 − 0.0039i 0.0002 + 0.0000i 0.0002 − 0.0000i 
0.0014 − 0.0003i 0.0014 + 0.0003i −0.0002 − 0.0023i −0.0002 + 0.0023i

7
Eigenvalues of Longitudinal Dynamics
diag(−0.9914 + 2.6445i, −0.9914 − 2.6445i, −0.0101 + 0.0739i, −0.0101 − 0.0739i)

3.2.2 Lateral Dynamics

Pole Damping Frequency (rad s−1 ) Time Constant (s)

−6.17 × 10−2 + 1.69i 3.64 × 10−2 1.69 16.2
−6.17 × 10−2 − 1.69i 3.64 × 10−2 1.69 16.2
−4.95 × 10−1 1.00 4.95 × 10−1 2.02
−1.19 × 10−3 1.00 1.19 × 10−3 8.39 × 102

Table 5: Pole Location, Damping (ζ), Natural Frequency (ωn ), Time Constant (τ )

The above table shows that the first two eigenvalues is associated with the Dutch
Roll mode, while the third and fourth correspond to the Roll and Spiral modes. No-
tably, the damping ratio for the Dutch Roll mode is 0.0364, considerably lower than the
theoretical expectation of around 0.3. However, the placement of the eigenvalues aligns
with theoretical predictions, with the Roll mode exhibiting faster dynamics compared
to the Spiral mode, as expected.

Eigenvectors of Lateral State Space Matrix

 
1.0000 + 0.0000i 1.0000 + 0.0000i −0.9718 + 0.0000i 0.7815 + 0.0000i
−0.0009 + 0.0034i −0.0009 − 0.0034i −0.1045 + 0.0000i −0.0007 + 0.0000i
 
 0.0001 − 0.0025i 0.0001 + 0.0025i 0.0094 + 0.0000i 0.0295 + 0.0000i 
0.0020 + 0.0005i 0.0020 − 0.0005i 0.2110 + 0.0000i 0.6232 + 0.0000i

Eigenvalues of Lateral Dynamics

diag(−0.0617 + 1.6933i, −0.0617 − 1.6933i, −0.4952, −0.0012)

3.2.3 Polar Plots

(a) Longitudinal Dynamics (b) Lateral Dynamics

Figure 2: Polar Plots

8
4 Transfer Function - Longitudinal States
4.1 Transfer Functions for Elevator Inputs
Output u: Forward Velocity
0.6781s2 + 646s + 292.7
s4 + 2.003s3 + 8.022s2 + 0.1717s + 0.04433

Output w: Downward Velocity

−35.24s3 − 9710s2 − 161.7s − 35.29
s4 + 2.003s3 + 8.022s2 + 0.1717s + 0.04433

Output q: Pitch Rate

−14.27s3 − 9.335s2 − 0.1305s + 1.886 × 10−18
s4 + 2.003s3 + 8.022s2 + 0.1717s + 0.04433

Output θ: Pitch Angle

−14.27s2 − 9.335s − 0.1305
s4 + 2.003s3 + 8.022s2 + 0.1717s + 0.04433

4.2 Transfer Functions for Throttle Inputs

Output u: Forward Velocity
−0.0003395s3 − 0.0006744s2 − 0.000404s + 0.00109
s4 + 2.003s3 + 8.022s2 + 0.1717s + 0.04433

Output w: Downward Velocity

−0.03456s2 − 0.0007483s − 0.0001227
s4 + 2.003s3 + 8.022s2 + 0.1717s + 0.04433

Output q: Pitch Rate

−5.108 × 10−5 s3 − 3.504 × 10−5 s2 − 9.59 × 10−7 s − 1.292 × 10−23
s4 + 2.003s3 + 8.022s2 + 0.1717s + 0.04433

Output θ: Pitch Angle

−5.108 × 10−5 s2 − 3.504 × 10−5 s − 9.59 × 10−7
s4 + 2.003s3 + 8.022s2 + 0.1717s + 0.04433

9
4.3 Changing the Stability Derivatives: Mq and Xu
Instead of changing the non-dimensional derivatives, we decided to change the dimen-
sional derivatives Mq and Xu because the entire longitudinal matrix is expressed in
terms of dimensional derivatives. The values were varied in small steps around the original
values we obtained. The responses and links to all videos are provided below in Figures
3 and 4.

Figure 3: Movement of Longitudinal Eigenvalues with Mq

Video Link: Movement of Longitudinal Eigenvalues with Mq

Figure 4: Movement of Longitudinal Eigenvalues with Xu

Video Link: Movement of Longitudinal Eigenvalues with Xu

10
4.4 Short Term & Long Term Responses: ∆u, ∆α, ∆γ
Our analysis focused on the longitudinal short-term and long-term responses to
a 1◦ elevator deflection. The responses of (∆u), angle of attack (∆α), and flight
path angle (∆γ) are depicted in Figures 5 and 6. Since both pairs of eigenvalues
are complex conjugates, we naturally anticipate a superposition of two underdamped
responses. Notably, the damping is quite satisfactory, allowing the states to reach a
steady state relatively swiftly. The responses exhibit larger magnitudes in the short
term, whereas in the long term, the amplitudes decrease. Additionally, the change in ∆γ
surpasses that of ∆α, with their relationship described by ∆γ = ∆θ − ∆α.

Figure 5: Short Term Step Responses to 1◦ elevator deflection: (∆u, ∆α, ∆γ)

Figure 6: Long Term Step Responses to 1◦ elevator deflection: (∆u, ∆α, ∆γ)

11
5 Transfer Function for Lateral States
5.1 Transfer Functions for Aileron Inputs
Output v: Lateral Velocity
302.9s2 + 364.1s + 22.01
s4 + 0.6199s3 + 2.933s2 + 1.425s + 0.001695

Output p: Roll Rate

6.691s3 + 1.238s2 + 17.21s − 3.001 × 10−16
s4 + 0.6199s3 + 2.933s2 + 1.425s + 0.001695

Output r: Yaw Rate

−0.4479s3 − 0.2574s2 − 0.01821s + 0.8153
s4 + 0.6199s3 + 2.933s2 + 1.425s + 0.001695

Output ϕ: Roll Angle

6.691s2 + 1.238s + 17.21
s4 + 0.6199s3 + 2.933s2 + 1.425s + 0.001695

5.2 Transfer Functions for Rudder Inputs

Output v: Lateral Velocity

10.73s3 + 1126s2 + 501.7s − 5.461
s4 + 0.6199s3 + 2.933s2 + 1.425s + 0.001695

Output p: Roll Rate

0.7142s3 − 0.1764s2 − 4.831s − 2.231 × 10−17
s4 + 0.6199s3 + 2.933s2 + 1.425s + 0.001695

Output r: Yaw Rate

−1.657s3 − 0.799s2 − 0.03899s − 0.2289
s4 + 0.6199s3 + 2.933s2 + 1.425s + 0.001695

Output ϕ: Roll Angle

0.7142s2 − 0.1764s − 4.831
s4 + 0.6199s3 + 2.933s2 + 1.425s + 0.001695

12
5.3 Changing the Stability Derivatives: Lv and Nr
Instead of changing the non-dimensional derivatives, we decided to change the dimen-
sional derivatives Lv and Nr because the entire lateral matrix is expressed in terms of
dimensional derivatives. The values were varied in small steps around the original values
we obtained. The responses and links to all videos are provided below in Figures 7 and 8.

Figure 7: Movement of Lateral Eigenvalues with Lv

Video Link: Movement of Lateral Eigenvalues with Lv

Figure 8: Movement of Lateral Eigenvalues with Nr

Video Link: Movement of Lateral Eigenvalues with Nr

13
5.4 Short Term & Long Term Responses: ∆v, ∆p, ∆r, ∆ϕ
We’ve conducted an analysis on the dynamic response of the lateral-directional
parameters to a step input of -15° aileron deflection. Specifically, we’ve exam-
ined the variations in v, pitch rate (p), yaw rate (r), and roll angle (ϕ), as depicted
in Figures 9 and 10. Notably, we’ve omitted the step responses of the longitudinal pa-
rameters, as they remain relatively unchanged for any control surface deflection affecting
lateral-directional dynamics. We observe a tendency for divergence from the short-term
responses, whereas the long-term responses demonstrate steady-state convergence due to
the presence of two real poles. The divergence can be attributed to the low damping of
the complex poles. Typically, such low damping would lead to a second-order overdamped
response, owing to the combination of two negative real-axis poles.

Figure 9: Short Term Step Responses to −15◦ aileron deflection: (∆v, ∆p, ∆r ∆ϕ)

Figure 10: Long Term Step Responses to −15◦ aileron deflection: (∆v, ∆p, ∆r ∆ϕ)

14
6 Conclusions and Important Remarks
• The data for all coefficients were readily available in [1], facilitating the calculation
of the derivatives. Instead of Xδt , Zδt , and Mδt , we computed XTu , ZTu , and MTu .
Additionally, ZTu was assumed to be zero due to the lack of available data.

• We can clearly observe the separation of modes in the longitudinal case, and both
the damping and time constants align reasonably well with theoretical estimates.
However, the damping for the short period mode is lower compared to the typical
value of around 0.6 found in theory.

• In the lateral dynamics, distinct modes are also noticeable. However, the Dutch roll
damping is too low, only 0.0364 compared to slightly higher values in theory. Note
that here theoritical estimates also give the nearly same results as given below. But
typically it should have a bit more damping.

• The following are some important calculated values of damping using theoretical
estimates:

– Phugoid damping = 0.1402 (Very Close!!)

– Short period damping = 0.247 (Getting better!!)
– Dutch Roll damping = 0.034 (Very Close!!)

References
[1] J. Roskam, Airplane Flight Dynamics and Automatic Flight Controls, Part I, DAR-
corporation, Lawrence, KS, 1995, pp. 522–524.

[2] M. Selig, Additional models and updates, Available at

http://amber.aae.uiuc.edu/~m-selig/apasim.html.

15