Version JPT/6

EGT2
ENGINEERING TRIPOS PART IIA
______________________________________________________________________

Thursday 29 April 2021 1.30 to 3.10

______________________________________________________________________

Module 3C5

DYNAMICS

Answer not more than three questions.

All questions carry the same number of marks.

The approximate percentage of marks allocated to each part of a question is

indicated in the right margin.

Write your candidate number not your name on the cover sheet and at the top of
each answer sheet.

STATIONERY REQUIREMENTS
Write on single-sided paper.

SPECIAL REQUIREMENTS TO BE SUPPLIED FOR THIS EXAM

CUED approved calculator allowed.
Attachment: 3C5 Dynamics and 3C6 Vibration data sheet (7 pages).
You are allowed access to the electronic version of the Engineering Data Books.

10 minutes reading time is allowed for this paper at the start of the
exam.

The time taken for scanning/uploading answers is 15 minutes.

Your script is to be uploaded as a single consolidated pdf containing

all answers.

Page 1 of 6
Version JPT/6

1 This question relates to the 3C5 laboratory experiment “Gyroscopic Phenomena”.

The frequency of nutation p for the gyro when inclined at angle  is given by

1
−
𝐶ω 𝐽1 𝐼1 2
𝑝= [1 + 𝑐𝑜𝑡 2 (𝜃) + 𝑐𝑜𝑠𝑒𝑐 2 (𝜃)]
𝐴 𝐴 𝐴

where the rotor is AAC aligned with axes i, j, k as shown in Fig. 1. The moment of inertia
of the gyro assembly about k is J1 (the moments of inertia about i and j are included in
A). The moment of inertia of the stand is I1 about the vertical K. The spin rate of the
rotor is  which is considered to be ‘fast’ and constant.

Fig. 1

(a) Sketch, for 0 < < , the variation of p with  for the case I1=A and J1=A. Identify
which of I1 and J1 is the more significant near  = 2 and explain why this is the
case. [15%]

(b) Identify the three components of the couple between the gyro assembly and the
stand and explain why one of these is zero. [15%]

(c) With the aid of suitable diagrams and the Gyroscope Equations (for fast spin) write
down equations relating the motion of the rotor, gyro assembly and stand. Use Euler
angles and . [35%]

(d) Linearize these equations to find the frequency of small vibration around  = 2
and show that your result is consistent with the expression for p given above. [35%]

Page 2 of 6
2 A cylindrical solid body of radius a and height h is wobbling on a flat horizontal
surface as shown in Fig. 2.

(a) For what value of h is the body “AAA” at the centre of mass G? [10%]

(b) During the wobbling motion, G is assumed to be at rest and the angle θ between the
body’s axis of symmetry and the vertical can be taken as constant and assumed to be
small. The angular velocity of the body is described using the reference frame in Fig. 2
as 𝜔 = 𝜔1 i + 𝜔2 j + 𝜔3 k.

(i) Use a suitable no-slip condition to show that

ℎ
𝜔3 = 𝜙̇ sin 𝜃 [30%]
2𝑎

where 𝜙̇ is the rate of rotation of the reference frame about the vertical axis 𝐾.

(ii) For the case of steady-state wobbling, use a clear diagram to show that the
couple acting on the body is
ℎ
𝑚𝑔 (2 sin 𝜃 − 𝑎 cos 𝜃) [20%]

(iii) Use the second Gyroscope equation (or otherwise) to find an expression for
the rate 𝜙̇ of steady-state wobbling and, using the result of part (a), find its value
for the case of an “AAA” cylinder. [40%]
𝐾
θ 𝐾

𝑘 θ
𝑘

𝑗
h
G G
𝑖
𝑖
2𝑎
P P

Fig. 2

Page 3 of 6 (TURN OVER

Version JPT/6

3 A rigid circular ring of radius a is attached at O to a vertical shaft and shown in

Fig. 3. The centre of the ring is at A and the fixed angle between OA and the shaft is .
The vertical shaft, and with it the ring, is forced to rotate about the vertical axis at
constant angular velocity . A small bead at P of mass m slides on the ring and its
motion is described by  the angle OAP. The acceleration due to gravity is g.

(a) Show that the kinetic energy of the bead can be expressed as

1 2
T = 2𝑚𝑎2 [𝛺 2 sin2 𝜃 + (𝜃̇ + 𝛺sin𝛽(1 − cos𝜃)) ]

and find and expression for the potential energy of the bead. [30%]

(b) Use Lagrange’s Equation to find an equation of motion for the bead. Hence show
that equilibrium solutions satisfy

[g cos𝛽 − 𝑎𝛺 2 (cos𝜃cos 2 𝛽 + sin2 𝛽)]sin𝜃 = 0 [20%]

(c) Find all possible equilibrium positions and identify the stability of two of them.
Describe how their stability depends on the rotation speed  [40%]

(d) Describe how the two regimes 2 < g cos/a and 2 > g cos/a differ. [10%]

Fig. 3

Page 4 of 6
4 (a) A wire of mass 6m and of length 12a is bent into the shape defined by points
A to G, as shown in Fig. 4(a). Relative to Cartesian axes Oxyz, the coordinates of the
points are as follows: A is (–a, a, 2a); B is (a, a, 2a); C is (a, a, 0); D is (a, –a,
0); E is (–a, –a, 0); F is (–a, –a, –2a) and G is (–a, a, –2a). Find the moments of inertia
Izz and Ixy. [50%]

(b) Two trolleys of mass 𝑀1 and 𝑀2 are connected by a spring of stiffness K, as shown
in Fig. 4(b). The motion of the system is described by two degrees of freedom, consisting
of the displacement of the left hand trolley 𝑞1 and the stretch of the spring 𝑞2 .

(i) Find expressions for the kinetic and potential energies of the system in terms
of the specified degrees of freedom. [15%]

(ii) Find expressions for the generalised momenta, and hence show that the
Hamiltonian of the system is given by

1 𝑀1 + 𝑀2 𝑀2 −1 𝑝1 1
𝐻 = 2 [𝑝1 𝑝2 ] [ ] [𝑝 ] + 2 𝐾𝑞22 [20%]
𝑀2 𝑀2 2

(iii) Derive Hamilton’s equations of motion, and explain why one of the
generalised momenta is conserved during the motion of the system. [15%]

z
A
B

y
O
E C
x
D
Fig. 4(b)
G
F

Fig. 4(a)

END OF PAPER

Page 5 of 6 (TURN OVER

Version JPT/6

THIS PAGE IS BLANK

Page 6 of 6
 Part IIA Data Sheet
Module 3C5 Dynamics
Module 3C6 Vibration

1 Dynamics in three dimensions

1.1 Axes fixed in direction

(a) Linear momentum for a general collection of particles mi :

dp
= F (e)
dt

where p = M v G , M is the total mass, v G is the velocity of the centre of mass and F (e) the
total external force applied to the system.

(b) Moment of momentum about a general point P

Q(e) = (r G − r P ) × ṗ + ḣG

= ḣP + ṙ P × p

where Q(e) is the total moment of external forces about P. Here hP and hG are the moments
of momentum about P and G respectively, so that for example
X
hP = (r i − r P ) × mi ṙ i
i

= hG + (r G − r P ) × p
where the summation is over all the mass particles making up the system.

(c) For a rigid body rotating with angular velocity ω about a fixed point P at the origin of
coordinates
Z
hP = r × (ω × r)dm = Iω

where the integral is taken over the volume of the body, and where
     
A −F −E ωx x
I = −F B −D , ω = ωy  , r = y 
−E −D C ωz z
R R R
and A = (y 2 + z 2 )dm B = (z 2 + x2 )dm C = (x2 + y 2 )dm
R R R
D = yz dm E = zx dm F = xy dm
where all integrals are taken over the volume of the body.

3C5 / 3C6 data sheet 2021 Page 1

1.2 Axes rotating with angular velocity Ω
Time derivatives of vectors must be replaced by the “rotating frame” form, so that for example

ṗ + Ω × p = F (e)

where the time derivative is evaluated in the moving reference frame.

When the rate of change of the position vector r is needed, as in 1.1(b) above, it is usually easiest
to calculate velocity components directly in the required directions of the axes. Application of
the general formula needs an extra term unless the origin of the frame is fixed.

1.3 Euler’s dynamic equations (governing the angular motion of a

rigid body)
(a) Body-fixed reference frame:

Aω̇1 − (B − C)ω2 ω3 = Q1
B ω̇2 − (C − A)ω3 ω1 = Q2
C ω̇3 − (A − B)ω1 ω2 = Q3

where A, B and C are the principal moments of inertia about P which is either at a fixed point
or at the centre of mass. The angular velocity of the body is ω = [ω1 , ω2 , ω3 ] and the moment
about P of external forces is Q = [Q1 , Q2 , Q3 ] using axes aligned with the principal axes of
inertia of the body at P.

(b) Non-body-fixed reference frame for axisymmetric bodies (the“Gyroscope equations”):

AΩ̇1 − (AΩ3 − Cω3 )Ω2 = Q1

AΩ̇2 + (AΩ3 − Cω3 )Ω1 = Q2
C ω̇3 = Q3

where A, A and C are the principal moments of inertia about P which is either at a fixed point
or at the centre of mass. The angular velocity of the body is ω = [ω1 , ω2 , ω3 ] and the moment
about P of external forces is Q = [Q1 , Q2 , Q3 ] using axes such that ω3 and Q3 are aligned
with the symmetry axis of the body. The reference frame (not fixed in the body) rotates with
angular velocity Ω = [Ω1 , Ω2 , Ω3 ] with Ω1 = ω1 and Ω2 = ω2 .

1.4 Lagrange’s equations

For a holonomic system with generalised coordinates qi

d h ∂T i ∂T ∂V
− + = Qi
dt ∂ q̇i ∂qi ∂qi
where T is the total kinetic energy, V is the total potential energy and Qi are the non-
conservative generalised forces.

3C5 / 3C6 data sheet 2021 Page 2

1.5 Hamilton’s equations
(a) Basic formulation
The generalized momenta pi and the Hamiltonian H(p, q) are defined as

∂T X
pi = , H(p, q) = pi q̇i − T + V
∂ q̇i i

where it should be noted that in the expression for the Hamiltonian the velocities q̇i (p, q) must
be expressed as a function of the generalized momenta and the generalized displacements.
Hamilton’s equations are
∂H ∂H
q˙i = , ṗi = − + Qi .
∂pi ∂qi

(b) Extension topics

The total time derivative of some function f (p, q, t) can be expressed in terms of the Poisson
bracket {f, H} in the form

df ∂f X  ∂f ∂H ∂f ∂H

= + {f, H}, {f, H} ≡ − .
dt ∂t i
∂q i ∂p i ∂p i ∂q i

Common forms of Canonical Transform for Hamilton’s equations are:

Type Generating function 1st eqn 2nd eqn Kamiltonian

∂G1 ∂G1 ∂G1
1 G1 (q, Q, t) p= P =− K=H+
∂q ∂Q ∂t
∂G2 ∂G2 ∂G2
2 G2 (q, P , t) p= Q= K=H+
∂q ∂P ∂t
∂G3 ∂G3 ∂G3
3 G3 (p, Q, t) q=− P =− K=H+
∂p ∂Q ∂t
∂G4 ∂G4 ∂G4
4 G4 (p, P , t) q=− Q= K=H+
∂p ∂P ∂t

3C5 / 3C6 data sheet 2021 Page 3

2 Vibration modes and response

Discrete Systems Continuous Systems

1. Equation of motion
The forced vibration of an N -degree-of-freedom The forced vibration of a continuous system
system with mass matrix M and stiffness ma- is determined by solving a partial differential
trix K (both symmetric and positive definite) equation: see Section 3 for examples.
is governed by:
Mÿ + Ky = f
where y is the vector of generalised displace-
ments and f is the vector of generalised forces.

2. Kinetic Energy
1
Z
1
T = ẏT Mẏ T = ẏ 2 dm
2 2
where the integral is with respect to mass (sim-
ilar to moments and products of inertia).
3. Potential Energy
1 See Section 3 for examples.
V = yT Ky
2

4. Natural frequencies and mode shapes

The natural frequencies ωn and corresponding The natural frequencies ωn and mode shapes
mode shape vectors u(n) satisfy un (x) are found by solving the appropriate dif-
ferential equation (see Section 3) and bound-
Ku(n) = ωn2 Mu(n) ary conditions, assuming harmonic time depen-
dence.

5. Orthogonality and normalisation

 Z 
(j)T (k) 0 j 6= k 0 j 6= k
u Mu = uj (x)uk (x)dm =
1 j=k 1 j=k

T 0 j 6= k
u(j) Ku(k) =
ωj2 j = k

6. General response
The general response of the system can be writ- The general response of the system can be writ-
ten as a sum of modal responses: ten as a sum of modal responses:
N
X X
y(t) = qj (t)u(j) = Uq(t) y(x, t) = qj (t)uj (x)
j=1 j

where U is a matrix whose N columns are where y(x, t) is the displacement and qj can be
the normalised eigenvectors u(j) and qj can be thought of as the ‘quantity’ of the jth mode.
thought of as the ‘quantity’ of the jth mode.

3C5 / 3C6 data sheet 2021 Page 4

 7. Modal coordinates
Modal coordinates q satisfy: Each modal amplitude qj (t) satisfies:

q̈ + diag(ωj2 ) q = Q q̈j + ωj2 qj = Qj

 

R
where y = Uq and the modal force vector where Qj = f (x, t)uj (x)dm and f (x, t) is the
Q = UT f . external applied force distribution.

8. Frequency response function

For input generalised force fj at frequency ω For force F at frequency ω applied at point x1 ,
and measured generalised displacement yk , the and displacement y measured at point x2 , the
transfer function is transfer function is
N (n) (n)
yk X uj uk y X u (x )u (x )
n 1 n 2
H(j, k, ω) = = H(x1 , x2 , ω) = =
fj ωn2 − ω 2 F ωn2 − ω 2
n=1 n

(with no damping), or (with no damping), or

N (n) (n)
yk X uj uk y X u (x )u (x )
n 1 n 2
H(j, k, ω) = ≈ H(x1 , x2 , ω) = ≈
fj ωn + 2iωωn ζn − ω 2
2 F ωn + 2iωωn ζn − ω 2
2
n=1 n

(with small damping), where the damping fac- (with small damping), where the damping fac-
tor ζn is as in the Mechanics Data Book for tor ζn is as in the Mechanics Data Book for
one-degree-of-freedom systems. one-degree-of-freedom systems.

9. Pattern of antiresonances
For a system with well-separated resonances For a system with well-separated resonances
(n) (n)
(low modal overlap), if the factor uj uk has (low modal overlap), if the factor un (x1 )un (x2 )
the same sign for two adjacent resonances then has the same sign for two adjacent resonances
the transfer function will have an antiresonance then the transfer function will have an antireso-
between the two peaks. If it has opposite sign, nance between the two peaks. If it has opposite
there will be no antiresonance. sign, there will be no anti-resonance.

10. Impulse responses

For a unit impulsive generalised force fj = δ(t), For a unit impulse applied at t = 0 at point x1 ,
the measured response yk is given by the response at point x2 is
N (n) (n)
X u u j k
X u (x )u (x )
n 1 n 2
g(j, k, t) = yk (t) = sin ωn t g(x1 , x2 , t) = sin ωn t
ωn ωn
n=1 n

for t ≥ 0 (with no damping), or for t ≥ 0 (with no damping), or

N (n) (n)
X u u j k −ωn ζn t
X u (x )u (x )
n 1 n 2
g(j, k, t) ≈ e sin ωn t g(x1 , x2 , t) ≈ e−ωn ζn t sin ωn t
ωn ωn
n=1 n

for t ≥ 0 (with small damping). for t ≥ 0 (with small damping).

3C5 / 3C6 data sheet 2021 Page 5

 11. Step response
For a unit step generalised force fj applied at For a unit step force applied at t = 0 at point
t = 0, the measured response yk is given by x1 , the response at point x2 is

N (n) (n)
X u (x )u (x ) h i
n 1 n 2
X u u h j k
i h(x1 , x2 , t) = 1−cos ωn t
h(j, k, t) = yk (t) = 1−cos ωn t ωn2
ωn2 n
n=1

for t ≥ 0 (with no damping), or for t ≥ 0 (with no damping), or

N (n) (n)
X u (x )u (x ) h i
n 1 n 2
X u u h
j k
i h(x1 , x2 , t) ≈ 1−e−ωn ζn t cos ωn t
h(j, k, t) ≈ 1−e−ωn ζn t cos ωn t ωn2
ωn2 n
n=1

for t ≥ 0 (with small damping). for t ≥ 0 (with small damping).

2.1 Rayleigh’s principle for small vibrations

The “Rayleigh quotient” for a discrete system is

V yT Ky
= T
Te y My

where y is the vector of generalised coordinates (and yT is its transpose), M is the mass matrix
and K is the stiffness matrix. The equivalent quantity for a continuous system is defined using
the energy expressions in Section 3.

If this quantity is evaluated with any vector y, the result will be

(1) ≥ the smallest squared natural frequency;
(2) ≤ the largest squared natural frequency;
(3) a good approximation to ωk2 if y is an approximation to u(k) .

V
Formally is stationary near each mode.
Te

3C5 / 3C6 data sheet 2021 Page 6

3 Governing equations for continuous systems
3.1 Transverse vibration of a stretched string
Tension P , mass per unit length m, transverse displacement y(x, t), applied lateral force f (x, t)
per unit length.

Equation of motion Potential energy Kinetic energy

∂ 2y ∂ 2y
Z  2 Z  2
1 ∂y 1 ∂y
m 2 − P 2 = f (x, t) V = P dx T = m dx
∂t ∂x 2 ∂x 2 ∂t

3.2 Torsional vibration of a circular shaft

Shear modulus G, density ρ, external radius a, internal radius b if shaft is hollow, angular
displacement θ(x, t), applied

torque τ (x, t) per unit length. The polar moment of area is given
4 4
by J = (π/2) a − b .

Equation of motion Potential energy Kinetic energy

∂ 2θ ∂ 2θ
Z  2 Z  2
1 ∂θ 1 ∂θ
ρJ 2 − GJ 2 = τ (x, t) V = GJ dx T = ρJ dx
∂t ∂x 2 ∂x 2 ∂t

3.3 Axial vibration of a rod or column

Young’s modulus E, density ρ, cross-sectional area A, axial displacement y(x, t), applied axial
force f (x, t) per unit length.

Equation of motion Potential energy Kinetic energy

∂ 2y ∂ 2y
Z  2 Z  2
1 ∂y 1 ∂y
ρA 2 − EA 2 = f (x, t) V = EA dx T = ρA dx
∂t ∂x 2 ∂x 2 ∂t

3.4 Bending vibration of an Euler beam

Young’s modulus E, density ρ, cross-sectional area A, second moment of area of cross-section
I, transverse displacement y(x, t), applied transverse force f (x, t) per unit length.

Equation of motion Potential energy Kinetic energy

∂ 2y ∂ 4y
Z  2 2 Z  2
1 ∂ y 1 ∂y
ρA 2 + EI 4 = f (x, t) V = EI dx T = ρA dx
∂t ∂x 2 ∂x2 2 ∂t

Note that values of I can be found in the Mechanics Data Book.

The first non-zero solutions for the following equations have been obtained numerically and are
provided as follows:

cos α cosh α + 1 = 0, α1 = 1.8751

cos α cosh α − 1 = 0, α1 = 4.7300
tan α − tanh α = 0, α1 = 3.9266

3C5 / 3C6 data sheet 2021 Page 7