PHY 5246: Theoretical Dynamics, Fall 2015

Assignment # 1, Solutions

Problem 1
y

θ̂
cos θ
r̂ sin θ
− sin θ cos θ
r r̂
~r =

θ x

Starting from the position vector for a planar motion in polar coordinates
r = rr̂ ,
we can derive the velocity vector as
ṙ = ṙr̂ + r θ̇θ̂ ,
where we have used that
dr̂
= θ̇ θ̂ .
dt
In the same way, we can derive the acceleration vector as
r̈ = (r̈ − r θ̇2 )r̂ + (r θ̈ + 2ṙ θ̇)θ̂ ,
where we have used that
dθ̂
= −θ̇r̂ .
dt

Problem 2
Let us choose B = Bŷ (any other choice is allowed and will lead to equivalent results, although
the motion will be in different planes depending on the direction of B). Let us write the velocity
and acceleration of the pointlike charge (using cartesian coordinate) as
v = ṙ = ẋx̂ + ẏŷ + żẑ ,
a = r̈ = ẍx̂ + ÿŷ + z̈ẑ ,
and let us choose generic initial conditions r0 = r(t = 0) and v0 = v(t = 0) as follows

r0 = x0 x̂ + y0 ŷ + z0 ẑ , (1)
v0 = ṙ0 = ẋ0 x̂ + ẏ0 ŷ + ż0 ẑ .

The force acting on the pointlike charge is the Lorentz force given by (notice that there is no
electric field, E = 0)
F = qv × B = qB(ẋẑ − żx̂) . (2)
The equations of motion are then

mẍ = −qB ż , (3)

The solution of Eq. (4) is straightforward. There is no force in the ŷ direction and the motion is
the motion of a free particle, which we obtain by integrating twice with respect to time, using the
initial conditions in Eq. (1):
y(t) = ẏ0 t + y0 . (6)
The other two equations, Eqs. (3) and (5), are coupled. Differentiating Eq. (3) with respect to
time and substituting Eq. (5) into Eq. (3) we obtain
2
dẍ qB qB

= − z̈ = − ẋ ,
dt m m
which we can easily solve as a second order differential equation for vx = ẋ of the form
qB
v̈x + ω 2 vx = 0 with ω = .
m
The solution can be written as a linear combination of harmonic functions of the form

vx (t) = A sin(ωt) + B cos(ωt) , (7)

with A and B arbitrary constants to be determined imposing the initial conditions in Eq. (1).
Integrating Eq. (7) one more time with respect to time we get the solution for x(t) in the form
A B
x(t) = − cos(ωt) + sin(ωt) + C . (8)
ω ω
In complete analogy, differentiating Eq. (5) with respect to time and substituting Eq. (3) into
Eq. (5) we obtain
dz̈ qB qB 2
 
= ẍ = − ż ,
dt m m
which we can solve as a second order differential equation for vz = ż of the form
qB
v̈z + ω 2 vz = 0 with ω = .
m
obtaining
vz (t) = A′ sin(ωt) + B ′ cos(ωt) , (9)
and, upon integration over time,
A′ B′
z(t) = − cos(ωt) + sin(ωt) + C ′ . (10)
ω ω
Imposing that x(t) and z(t) satisfy Eq. (3) (or equivalently Eq. (5)) we derive a relation among
the integration constants in Eqs. (7) and (9), namely
A′ = B and B ′ = −A .
Furthermore, imposing the initial conditions in Eq. (1) we obtain
vx (t = 0) = x˙0 = B , (11)
′
vz (t = 0) = z˙0 = B = −A ,
A ż0
x(t = 0) = x0 = C − = C + ,
ω ω
A′ B ẋ0
z(t = 0) = z0 = C ′ − = C′ − = C′ − .
ω ω ω
Having determined all the arbitrary constants in Eqs. (12) and (13) in terms of the initial condi-
tions, we can finally write x(t) and z(t) as
ż0 ẋ0
x(t) − x0 = (cos(ωt) − 1) + sin(ωt) , (12)
ω ω
ẋ0 ż0
z(t) − z0 = (1 − cos(ωt)) + sin(ωt) . (13)
ω ω
From Eqs. (6), (12), and (13) we see that the pointlike charge moves in a helix with axis along
the direction of the magnetic field B (i.e. along ŷ), with radius R = (ẋ20 + ż02 )1/2 .

vy0

v~0
vx0

R
vy0 vz0

F~ ~ = B ŷ
B ~ = B ŷ
B
y z

Problem 3
If the mass m is not a function of time (m =constant) then
dv d m dT
 
F · v = ṗ · v = m ·v = v·v = ,
dt dt 2 dt
where T = mv 2 /2 is the kinetic energy of the particle.
If the mass is a function of time (m = m(t)) then

1d 1d  2  d
F · p = ṗ · p = (p · p) = m v · v = (mT ) .
2 dt 2 dt dt

Problem 4
The equations of motion of the individual particles are
(e)
ṗ1 = F1 + f21 , (14)
(e)
ṗ2 = F2 + f12 ,

and
(e)
 
l̇1 = r1 × F1 + f21 , (15)
(e)
 
l̇2 = r2 × F2 + f12 ,

Using Eq. (14) we can calculate

which corresponds to Ṗ = F(e) only if f12 = −f21 (weak law of action and reaction).
On the other hand, using Eq. (15) we find that
(e) (e)
L̈ = l̇1 + l̇2 = r1 × F1 + r2 × F2 + (r1 − r2 ) × f21 , (17)

where we have used the (already proved) weak law of action and reaction. The previous result
(e) (e)
corresponds to L̇ = N(e) = N1 + N2 only if (r1 − r2 ) × f21 = 0 i.e. only if f21 (and therefore
f12 ) are parallel to r12 = r1 − r2 (strong law of action and reaction).
Problem 5

ẑ
m1

~r1
v~0
T OP m1g φ
m2

b
k̂

θ ~r2
m1 a−
b
SIDE m2
k̂

m2 g

The motion of this system is more easily understood as the superposition of the translational
motion of the center of mass (CM) and the rotational motion of the two bobs with respect of to
the center of mass.
The center of mass is located along the rod at a distance b from m2 defined by the equation
m2 b − m1 (a − b) m1
= 0 −→ b = a . (18)
m1 + m2 m1 + m2
Since m2 > m1 the center of mass is closer to m2 , i.e. b < a/2. When the system is released,
the center of mass is moving with an arbitrary initial velocity v0 and from there on its motion is
governed by
(e)
Ṗ = F1 + Fe(e) = (m1 + m2 )g = Mg . (19)
Therefore the center of mass moves like a pointlike object of mass M = m1 + m2 with initial
velocity v0 , moving under the action of the gravitational force Mg, i.e. it falls along a parabolic
trajectory.
As far as the rotational motion about the center of mass goes, the system of the two masses
m1 and m2 starts off with some given initial angular momentum l0 about the center of mass, and
from there on their rotational motion is governed by

L̇ = N(e) = r1 × f1 + r2 × f2 (20)
= [(a − b)m1 g sin φ − bm2 g sin φ] k̂
m1 m1 m2
  
= a− a m1 g sin φ − ag sin φ k̂ = 0 ,
m1 + m2 m1 + m2
where φ is the angle between the plane of rotation and the direction of f1 (or f2 ), i.e. the vertical,
while k̂ is a unit vector in the direction orthogonal to the plane of rotation of the two masses.
Since the total angular momentum of the system with respect to the center of mass is conserved,
the rotational motion of the two masses happens in a constant plane. If the string does not bend,
i.e. if it behaves like a rigid rod, both velocities v1 and v2 can be expressed in terms of the
same angular velocity ω = θ̇, where θ is the angle of rotation of the two masses (the same if the
string does not bend). They can be written as v1 = bθ̇ĵ and v2 = −(a − b)θ̇ ĵ, where ĵ is a vector
orthogonal to the string/rod, in the plane of rotation, and pointing in the direction of rotation.
Assuming that the string between the two masses does not bend, the tension in the string (T)
can be calculated from the radial equations of motion of the two masses written in the center of
mass frame (where we use polar coordinates, and notice that the angular coordinate for the two
masses, θ, is the same). Notice that these equations in any fixed frame would look like,

m1 (r̈1 − r1 θ̇2 ) = −T − m1 g cos φ , (21)

but in the center of mass frame, which is accelerating with acceleration g, the last term is missing
(because we need to subtract an inertial force equal to the mass of the object times the acceleration
of the frame), and the system of radial equations reads,

−m1 (a − b)θ̇2 = −T , (23)

where we have used that r̈1 = r̈2 = 0, since r1 = a − b and r2 = b are constant. Solving for T we
get,
m1 m2
T = a θ̇2 , (25)
m1 + m2
where we have used that b = m1 a/(m1 + m2 ).

There is another quite interesting way to look at this problem. Consider the problem as a
two-body problem, i.e. as the motion of two objects under the action of a central force, directed
along the line that joins the two moving objects at all times (in this case, the tension of the string).
Using the appropriate set of coordinates to describe a two-body problem, i.e. making the change
of variables,
(r1 , r2 ) → (R, r)
where R is the position vector of the center of mass,
m1 r1 + m2 r2
R=
m1 + m2
and r is the relative position of the two masses,

r = r1 − r2 ,

we know that the motion of the system reduces to the motion of an object of mass equal to the
total mass of the system (M = m1 + m2 ) located at the center of mass (i.e. with position vector
R), plus the motion of an object of reduced mass µ = m1 m2 /(m1 + m2 ) and position vector r,
subject to the central force of the problem (T in our case). The equation of motion for M is then
simply,
M R̈ = Mg ,
while the equation for µ is the equation of an object moving along a circular orbit of constant
radius |r1 − r2 | = a, i.e.
m1 m2
µ a θ̇2 = T −→ T = a θ̇2 .
m1 + m2