Stephen Martin

PHYS 210
Homework #1

Question 1:
A particle starts at rest and moves along a cycloid whose equation is
   p 
−1 a − y 2
x = ± a cos + 2ay − y
a
There is a gravitational field of strength g in the negative y direction. Obtain and solve
the equations of motion. Show that no matter where on the cycloid the particle starts out
at time t = 0, it will reach the bottom at the same time.

Solution:
We start by defining an angle θ:
 
−1 a−y
θ = cos
a
Defining x and y in terms of θ,
y =a(1 − cos θ)
x = ± a (θ + sin θ)
The potential and kinetic energies are given by
U =mgy
=mga(1 − cos θ)

1
T = m(ẋ2 + ẏ 2 )
2
1 h i
= m a2 θ̇2 (1 + cos θ)2 + a2 θ̇2 sin2 θ
2
=ma2 θ̇2 (1 + cos θ)
Continuing with the Lagrangian:
L =T − U
=ma2 θ̇2 (1 + cos θ) − mga(1 − cos θ)

∂L
= − ma2 θ̇2 sin θ − mga sin θ
∂θ
d ∂L
=2ma2 θ̈(1 + cos θ) − 2ma2 θ̇2 sin θ
dt ∂ θ̇
1
2

Thus,

2ma2 θ̈(1 + cos θ) − ma2 θ̇2 sin θ + mga sin θ = 0

Noticing that θ̈ = θ̇ ddθθ̇ :

dθ̇ h i
2ma2 θ̇ (1 + cos θ) = ma2 θ̇2 − mga sin θ
dθ

Z Z
2θ̇dθ̇ sin θdθ
=
2
θ̇ − g/a 1 + cos θ

 g
ln θ̇2 − = − ln (1 + cos θ) + C
a

C g
θ̇2 = +
1 + cos θ a
Because the particle starts at rest, we can set the boundary conditions ẋ|t=0 = 0 and ẏ = 0.
Solving for C:
g(1 + cos θ0 )
C=−
a
And our equation becomes  
2 g 1 + cos θ0
θ̇ = 1−
a 1 + cos θ
We can now integrate to solve for θ:
Z r
dθ g
q = t+D
1 − 1+cos θ0 a
1+cos θ

Z  1/2 r
1 + cos θ g
dθ = t+D
cos θ − cos θ0 a

1/2
sin2 θ
Z  r
g
dθ = t+D
(cos θ − cos θ0 )(1 − cos θ) a

Z r
sin θdθ g
p = t+D
(cos θ − cos θ0 )(1 − cos θ) a
 3

Z r
du g
− p = t+D
−u2 + (1 + cos θ0 )u − cos θ0 a

" # r
−1 −2u + 1 + cos θ0 g
sin p = t+D
(1 + cos θ0 )2 − 4 cos θ0 a

  r
−2 cos θ + 1 + cos θ0
−1 g
sin = t+D
1 − cos θ0 a
Using our boundary conditions once more:
1 − cos θ0
= sin D
1 − cos θ0

π
D=
2
Rearranging, we finally have
r 
g π
−2 cos θ + 1 + cos θ0 = (1 − cos θ0 ) sin t+
a 2

r 
1 − cos θ0 g 1 + cos θ0
cos θ = − cos t +
2 a 2
Changing back from θ to y
r 
y0 g 2a − y0
a − y = − cos t +
2 a 2

r 
y0 g y0
y= cos t +
2 a 2
At the bottom of the arc, y = 0, so r
a
t=π
g
which is independent of y0 .
4

Question 2:
A point particle of mass m is constrained to move frictionlessly on the inside of a circular
wire hoop of radius r, uniform density and mass M . The hoop is constrained to the xy-
plane, it can roll on a fixed line (the x-axis), but it does not slide, nor can it lose contact
with the x-axis.

The point particle is acted on by gravity exerting a force along the negative y-axis. At
t = 0 suppose the hoop is at rest. At this time the particle is at the top of the hoop, and
is a given velocity v0 along the x-axis.

What is the velocity vf , with respect to the fixed axis, when the particle comes to the
bottom of the hoop? Simplify your answer in the limits m/M → 0 and M/m → 0.

Solution: Using generalized coordinates x (center of hoop horizontal position) and θ

(angle between particle and top of hoop), the coordinates of the particle are given by
xm =x + r sin θ
ym =r(1 + cos θ)
We then have

U =mgym + M gr
=mgr(1 + cos θ) + M gr

1 1 1
T = m(ẋ2m + ẏm
2
) + M (ẋ2M + ẏM2
) + IωM
2 2 2
 2
1 h 2 2
i 1
2 1 2 ẋ
= m (ẋ + r cos θθ̇) + (r sin θθ̇) + M ẋ + (M r )
2 2 2 r
1 h 2 i
= m ẋ + 2ẋr cos θθ̇ + r2 θ̇2 + M ẋ2
2
The Lagrangian is
1 h i
L = m ẋ2 + 2ẋr cos θθ̇ + r2 θ̇2 + M ẋ2 − mgr(1 + cos θ) − M gr
2
The equation of motion for x is determined by
d ∂L ∂L
− =0
dt ∂ ẋ ∂x
But L does not depend on x, so ∂L/∂ ẋ = C

(m + 2M )ẋ + mr cos θθ̇ = C

 5

Solve for C using boundary conditions ẋ(0) = 0 and ẋm (0) = r cos θθ̇ = v0 :
C = mv0
(m + 2M )ẋ + mr cos θθ̇ − mv0 = 0 (1)

We now take a look at the equation for θ:

d ∂L ∂L
− =0
dt ∂ θ̇ ∂θ

mẍr cos θ − mẋr sin θθ̇ + mr2 θ̈ = − mẋr sin θθ̇ + mgr sin θ

ẍ cos θ + rθ̈ − g sin θ =0

ẍ = g tan θ − r sec θθ̈ (2)

However, equation (1) gave us

(m + 2M )ẋ − mv0 = −mr cos θθ̇

Multiplying equations (1) and (2):

(m + 2M )ẋẍ − mv0 ẍ = − mrg sin θθ̇ + mr2 θ̇θ̈

1 d d 1 d
(m + 2M ) (ẋ2 ) − mv0 ẍ =mrg (cos θ) + mr2 (θ̇2 )
2 dt dt 2 dt

Integrating with respect to t:

1 1
(m + 2M )ẋ2 − mv0 ẋ = mrg cos θ + mr2 θ̇2 + D
2 2
We use our initial conditions to solve for D:

mv02
D = −mrg −
2
Reducing our equation of motion to

m + 2M 2
ẋ − 2v0 ẋ = 2rg cos θ + r2 θ̇2 − 2rg − v02 (3)
m
6

Now, we simply look at equations (1) and (3) for the case of θ = π:

m + 2M 2
ẋ − 2v0 ẋ + v02 =r2 θ̇2 − 4rg
m

m + 2M
ẋ − v0 =rθ̇
m

This set of equations is solved by

" s  #
m m + 2M
ẋ = v0 + v02 + 2rg
m + 2M M

s  
1 m + 2M
θ̇ = v02 + 2rg
r M
(Note: there are 2 solutions to this quadratic set of equations, but the solution was chosen
so that θ̇ > 0, which is the only physical solution that makes sense)

From here, we can find ẋm |θ=π :

ẋm =ẋ − rθ̇

s  
mv0 2M m + 2M
= − v02 + 2rg
m + 2M m + 2M M

In the limiting case m/M → 0,

q
ẋm = − v02 + 4rg

which is what we would expect from a similar system with a fixed hoop. In the limiting
case of M/m → 0,

ẋm = v0

which is simply a requirement of the conservation of linear momentum.

 7

Question 3:
A double plane pendulum consists of a simple pendulum (mass m1 , length l1 ) with another
simple pendulum (mass m2 , length l2 ) suspended from m1 , both constrained to move in
the same vertical plane.

(a) Describe the configuration manifold Q of this dynamical system. Say what you
can about TQ.

(b) Write down the Lagrangian of the system in suitable coordinates.

(c) Derive Lagrange’s equations.

Solution:
(a) The configuration manifold Q lies on the surface of a torus defined by
x =(l1 + l2 cos θ2 ) cos θ1
y =(l1 + l2 cos θ2 ) sin θ1
z =l2 sin θ2
That is, the distance from the center of the “ring” to the center of the “pipe” is l1 and the
radius of the “pipe” is l2 .

The tangent bundle TQ is 4-dimensional, so it is difficult to visualize, but it simply maps

the tangent plane at each point on torus Q.

(b) The natural choice for the generalized coordinates for this system are θ1 and θ2 ,
the angles made by each pendulum with the y-axis.
x1 =l1 sin θ1
y1 = − l1 cos θ1
x2 =l1 sin θ1 + l2 sin θ2
y2 = − l1 cos θ1 − l2 cos θ2

ẋ1 =l1 cos θ1 θ̇1

ẏ1 =l1 sin θ1 θ̇1
ẋ2 =l1 cos θ1 θ̇1 + l2 cos θ2 θ̇2
ẏ2 =l1 sin θ1 θ̇1 + l2 sin θ2 θ̇2
8

The Lagrangian is then:

L =T − U

 2 
1 2 2 1 2 
= m1 l1 θ̇1 + m2 l1 cos θ1 θ̇1 + l2 cos θ2 θ̇2 + l1 sin θ1 θ̇1 + l2 sin θ2 θ̇2
2 2
+ m1 gl1 cos θ1 + m2 g (l1 cos θ1 + l2 cos θ2 )

1 1 h i
= m1 l12 θ̇12 + m2 l12 θ̇12 + l22 θ̇22 + 2l1 l2 θ̇1 θ̇2 cos(θ1 − θ2 )
2 2
+ (m1 + m2 )gl1 cos θ1 + m2 gl2 cos θ2

(c) We start with the equation for θ1 :

d ∂L ∂L
− =0
dt ∂ θ̇1 ∂θ1

(m1 + m2 )l12 θ̈1 + 2l1 l2 θ̈2 cos(θ1 − θ2 ) − 2l1 l2 θ̇2 (θ̇1 − θ̇2 ) sin(θ1 − θ2 )
+ (m1 + m2 )gl1 sin θ1 = 0

The equation for θ2 is given by

d ∂L ∂L
− =0
dt ∂ θ̇2 ∂θ2

m2 l22 θ̈2 + 2l1 l2 θ̈1 cos(θ1 − θ2 ) − 2l1 l2 θ̇1 (θ̇1 − θ̇2 ) sin(θ1 − θ2 )
+ m2 gl2 sin θ2 = 0
 9

Question 4:
A cartesian coordinate system with axes x,y,z is rotating relative to an inertial frame with
constant angular velocity ω about the z-axis. A particle of mass m moves under a force
whose potential is V (x, y, z). Set up the Lagrange equations of motion in the coordinate
system x, y, z. Show that these equations are the same as those for a particle in a fixed co-
ordinate system acted on by the force −∇V and a force derivable from a velocity-dependent
potential U , and find U .

Solution:
In the coordinates x, y, z, the Lagrangian is given by
1
L = m[ẋ2 + ẏ 2 + ż 2 ] − V (x, y, z)
2
So the equations of motion are
∂V
mẍ + =0
∂x
∂V
mÿ + =0
∂y
∂V
mz̈ + =0
∂z

The coordinate transformations are:

x0 =x cos ωt − y sin ωt
y 0 =x sin ωt + y cos ωt
z 0 =z

ẋ0 =(ẋ − ωy) cos ωt − (ẏ + ωx) sin ωt

ẏ 0 =(ẋ − ωy) sin ωt + (ẏ + ωx) cos ωt

The Lagrangian for these coordinates is

1
L = m[(ẋ − ωy)2 + (ẏ + ωx)2 + ż 2 ] − V (x, y, z)
2

1 1
= m[ẋ2 + ẏ 2 + ż 2 ] + m[ω 2 x2 + ω 2 y 2 − 2ωy ẋ + 2ωxẏ] − V (x, y, z)
2 2
10

The equations of motion are

∂V
mẍ − mω ẏ − mω 2 x − mω ẏ + =0
∂x
∂V
mÿ + mω ẋ − mω 2 y + mω ẋ + =0
∂y
∂V
mz̈ + =0
∂z

∂V
mẍ − mω 2 x − 2mω ẏ + =0
∂x
∂V
mÿ − mω 2 y + 2mω ẋ + =0
∂y
∂V
mz̈ + =0
∂z

This shows us that

∂V ∂V ∂V
F =(mω 2 x + 2mω ẏ − )x̂ + (mω 2 y − 2mω ẋ − )ŷ − ẑ
∂x ∂y ∂z

=F∗ − ∇V

We can now consider a potential energy U where

F∗ = − ∇U

∇U = − (mω 2 x + 2mω ẏ)x̂ − (mω 2 y − 2mω ẋ)ŷ

1 1
U = − mω 2 x2 − mω 2 y 2 − 2mωxẏ + 2mωy ẋ
2 2

1
U = − mω 2 (x2 + y 2 ) + 2mω(y ẋ − xẏ)
2