Tutorial Sheet BPT-502 Homework 1

Himanshu Sahu ([email protected])

October 13, 2020

1. Problem 1
Part 1:
Method 1 : From work-energy theorem for small displacement, we have

dT = W = F · dr
or
dT dr
=F·
dt dt
for small time interval force doesn’t vary much.
dT
=F·v
dt
Method 2: Kinetic energy given by
1 1
T = mv 2 = mv · v
2 2
taking the time derivative of both side

dT dv
= mv · = mv · a = F · v
dt dt
where we have used the following vector identity
1 d dv
(v · v) = v ·
2 dt dt

Part 2:
Here we have varying mass so that m = m(t) so we also need to consider the time derivative of mass.
Consider the left hand side of the relation,

d dm dT
(mT ) = T +m
dt dt dt
In the mean time also recall that for variable mass we have a relation
dp dv dm
=m +v
dt dt dt

d 1 dm 1 dm
(mT ) = mF · v + mv · v = m2 a · v + mv · v
dt 2 dt 2 dt
 
d dm dp
(mT ) = ma + v · (mv) = ·p=F·p
dt dt dt

1
Tutorial Sheet BPT-502 Himanshu Sahu ([email protected]) Homework 1

or
d
(mT ) = F · p
dt

Question 2.

Consider a system of n particles with position vector ri where i = 1, 2.., n and mass mi .Then by definition
center of mass
1 X
R= mi ri
M i
X
MR = mi ri
i

magnitude is just dot product of vector with itself,

X X
MR · MR = mi ri · mj rj
i j

X X
M 2 R2 = m2i ri · ri + mi mj ri · rj
i i6=j

rij represents the distance between ith and jth particles. from cosine law
2
rij = ri2 + rj2 − 2ri · rj
X 1X
M 2 R2 = m2i ri · ri + mi mj (ri2 + rj2 − rij
2
)
i
2
i6=j

Using the symmetry between index, we have

X X 1X
M 2 R2 = m2i ri · ri + mi mj ri2 − 2
mi mj rij
i
2
i6=j i6=j

X X 1X
M 2 R2 = ( mj )( m2i ri ) − 2
mi mj rij
j i
2
i6=j

the last term have no restriction because distance become zero when the index are same.

X 1X
M 2 R2 = M ( m2i ri ) − 2
mi mj rij
i
2 i,j

Question 3.

It can be deducted from the problem that the center of mass is moving around the circle, and the distance
between the masses is fixed.so there are 4 − 2 = 2 degrees of freedom. Kinetic energy of the particle in
co-ordinate system given by

1 2 1 2
T = mr + mr
2 1 2 2
transforming to center of mass co-ordinate by transformation rule given by

2
Tutorial Sheet BPT-502 Himanshu Sahu ([email protected]) Homework 1

ric = ri + R
and using r = r1 − r2
we have
T = mṘ2 + T 0
where
1 2
T0 = µṙ
2
and µ is reduced mass.
Thus finally we have
1
T = mṘ + µṙ2
2
but as Center of mass is constrained to move in circle we have no radial velocity component of center of
mass.

1 2
T = µṙ
2
There are two degree of freedom as expected.

Question 4.

The total time derivative is

d ∂ ∂ ∂ ... ∂
= + q̇ j j + q̈ j j + q j j + . . . , (1)
dt ∂t ∂q ∂ q̇ ∂ q̈

where dot denotes time differentiation. Hence the commutator between a velocity differentiation and total
time differentiation is a position differentiation
 
∂ d (1) ∂
, = . (2)
∂ q̇ j dt ∂q j

In particular,
d ∂T (2) ∂ Ṫ ∂T
= − j, (3)
dt ∂ q̇ j ∂ q̇ j ∂q

where T (t, q, q̇) is the kinetic energy. The Lagrange equations read

d ∂T ∂T (3) ∂ Ṫ ∂T
Qj = j
− j = j
−2 j, (4)
dt ∂ q̇ ∂q ∂ q̇ ∂q
where Qj is the generalized force.
A note on terminology: It is customary to only refer to Lagrange equations (4) as Euler-Lagrange equations.

Question 5.

Writing Lagrange’s equation for L :

3
Tutorial Sheet BPT-502 Himanshu Sahu ([email protected]) Homework 1

 
d ∂L ∂L
− =0
dt ∂ q̇ ∂q

Substituting
dF (q, t)
L0 = L +
dt
here we are compressing the notation using qi = q.

∂L0 ∂L0
   
d d ∂ dF ) ∂ dF
− − + =0
dt ∂ q̇ ∂q dt ∂ q̇ dt ∂q dt

To show the invariance of Langrange’s equation we need to show

 
d ∂ dF ) ∂ dF
− =0
dt ∂ q̇ dt ∂q dt
or !
d ∂ Ḟ ∂ Ḟ
=
dt ∂ q̇ ∂q
This is shown to be true
∂ Ḟ ∂F
=
∂ q̇ ∂q
d ∂ Ḟ d ∂F
=
dt ∂ q̇ dt ∂q

∂ ∂F ∂ ∂F
= + q̇
∂t ∂q ∂q ∂q
 
∂ ∂F ∂F
= + q̇
∂q ∂t ∂q

∂ Ḟ
=
∂q

Method 2: Using the least action principle, Action associated with transformed langrangian
Z t2 Z t2 Z t2
dF
δS 0 = L0 dt = Ldt + dt
t1 t1 t1 dt

δS 0 = δS + F (q2 , t2 ) − F (q1 , t1 )
Since the action defer by the constant term, both of them will lead to same Langrange’s equation.

Question 6.

Consider the Lagrangian L(qi , q̇i , t) which satisfies the Euler-Lagrange equation
 
d ∂L ∂L
− =0 (1)
dt ∂ q̇i ∂qi

4
Tutorial Sheet BPT-502 Himanshu Sahu ([email protected]) Homework 1

Consider the transformation

qi = qi (sj , t) (2)

where both index are upto n.

Such changes of variables are known as point transformations. It is relatively easy to show, by variational
calculus, that the Euler-Lagrange equation is invariant under point transformations. Here we show this
invariance explicitly via the chain rule.
First note that
∂L X  ∂L ∂qi ∂L ∂ q̇i

= + (3)
∂sj i
∂qi ∂sj ∂ q̇i ∂sj

Next,Consider that

X ∂ q̇i ∂qi
q̇i = ṡj + (4)
j
∂sj ∂t

so that

∂ q̇i ∂qi
= (5)
∂ ṡj ∂sj

Now consider that

∂L X  ∂L ∂qi ∂L ∂ q̇i

= + (6)
∂ ṡj i
∂qi ∂ ṡj ∂ q̇i ∂ ṡj

The first term is zero from 2. Thus

∂L X  ∂L ∂ q̇i 
= (7)
∂ ṡj i
∂ q̇i ∂ ṡj

hence

d ∂L X  ∂L ∂ q̇i d ∂L ∂L d ∂ q̇i

= + (8)
dt ∂ ṡj i
∂ q̇i ∂ ṡj dt ∂ q̇i ∂ q̇i dt ∂ ṡj

Putting everything together we get

  X  ∂L ∂qi 
∂L d ∂L ∂L ∂ q̇i ∂L ∂ q̇i d ∂L ∂L d ∂ q̇i
− = + − −
∂sj dt ∂ ṡj i
∂qi ∂sj ∂ q̇i ∂sj ∂ q̇i ∂ ṡj dt ∂ q̇i ∂ q̇i dt ∂ ṡj

X  ∂qi  ∂L d ∂L

∂L ∂ q̇i ∂L d ∂ q̇i

= − + −
i
∂sj ∂qi dt ∂ q̇i ∂ q̇i ∂sj ∂ q̇i dt ∂ ṡj

X  ∂L ∂ q̇i ∂L d ∂ q̇i

= −
i
∂ q̇i ∂sj ∂ q̇i dt ∂ ṡj

X ∂L ∂ q̇i 
∂L ∂ q̇i
= − =0
i
∂ q̇i ∂sj ∂ q̇i ∂sj

5
Tutorial Sheet BPT-502 Himanshu Sahu ([email protected]) Homework 1

 
∂L d ∂L
− =0
∂sj dt ∂ ṡj
Q.E.D.

Question 7.

It’s obvious to use spherical polar co-ordinate due to symmetry in spherical pendulum. Using the notation
as in popular context, we have

Figure 1: Spherical Pendulum : (Source : Wikipedia)

The Lagrangian is
1 2 2
L= mr (θ̇ + sin2 θφ̇2 ) + mgr cos θ
2
The Euler–Lagrange equations give :

d
(mr2 θ̇) − mr2 sinθ cos θφ̇2 + mgr sin θ = 0
dt
and
d
(mr2 sin2 θφ̇) = 0
dt
showing that angular momentum is conserved.

Question 8.

2 shows the double pendulum which is confined to mover in xy plane.There are two constraints and thus 2
degrees of freedom.It’s obvious to choose the angle φ1 and φ2 as generalize co-ordinate these co-ordinate are
related to cartesian co-ordinate of particle with the following relation.
x1 = 0 , y1 = l1 cos φ1
x2 = l1 sin φ1 + l2 sin φ2
and
y2 = l1 cos φ1 + l2 cos φ2
Thus the lagrangian given by

1 1
L=T −V = m1 (ẋ21 + ẏ12 ) m2 (ẋ22 + ẏ22 ) − m1 gy1 − m2 gy2
2 2

6
Tutorial Sheet BPT-502 Himanshu Sahu ([email protected]) Homework 1

Figure 2: Double Pendulum

ẏ1 = −l1 sin φ1 φ˙1 ,

ẋ2 = l1 cos φ1 φ̇1 + l2 cos φ2 φ̇2
and
ẏ2 = −l1 cos φ1 φ̇1 − l2 cos φ2 φ̇2

putting everything together into Langrangian we get

1 1
L= (m1 + m2 )l12 φ21 + m2 l22 φ22 + m2 l1 l2 φ1 φ2 cos(φ1 − φ2 ) + (m1 + m2 )gl1 cos φ1 + m2 gl2 cos φ2
2 2

Now we can write the Langrange’s equation

 
d ∂L ∂L
− =0 (9)
dt ∂ φ̇i ∂φi

∂L
= −m2 l1 l2 φ̇1 φ̇2 sin (φ1 − φ2 ) − (m2 + m1 )gl1 sin φ1
∂φ1

∂L
= (m1 + m2 )l12 φ̇1 + m2 l1 l2 φ̇2 cos (φ1 − φ2 )
∂ φ̇

Thus Equation of motion

d
((m1 + m2 )l12 φ̇1 + m2 l1 l2 φ̇2 cos (φ1 − φ2 )) = −m2 l1 l2 φ̇1 φ̇2 sin (φ1 − φ2 ) − (m2 + m1 )gl1 sin φ1
dt

(m1 + m2 )l1 φ̈1 + m2 l2 φ̈2 cos(φ1 − φ2 ) + m2 l2 φ̇22 sin(φ1 − φ2 ) + (m1 + m2 )g sin φ1 = 0

Similarly, we can derive the second differential equation and this turn out to be

l2 φ̈2 + l1 φ̈1 cos(φ1 − φ2 ) − l1 φ̇2 sin(φ1 − φ2 ) + g sin φ2 = 0