Variational Calculus in Physics

(a) Functionals. Basic idea of functionals. Extremization of action as a basic principle in mechanics. Lagrangian
formulation. Euler’s equations of motion for simple systems harmonics oscillators, simple pendulum, spherical
pendulum, coupled oscillators. Cyclic coordinates symmetries and conservation laws. Legendre transformations and
the Hamiltonian formulation of mechanics. Canonical equations of motion. Application to simple systems.

Calculus of variation primarily deals with finding maximum and minimum values of an
integral involving a certain function [𝑠𝑎𝑦 𝑦(𝑥)] to find the extremum path 𝑦 = 𝑦(𝑥) followed by
a system.
The problem is to find 𝑦 = 𝑦(𝑥) which makes a certain integral,
𝑥2
𝐼 = ∫ 𝑓(𝑥, 𝑦, 𝑦 ′ ) 𝑑𝑥 𝑠𝑡𝑎𝑡𝑖𝑜𝑛𝑎𝑟𝑦.
𝑥1

The integral 𝐼 is called functional, the function 𝑦(𝑥) which makes the 𝐼 stationary is called
extremal.
Q: Explain the terms functional, extremal with examples.
Example 1: Let us find the shortest distance between two given points (𝑥1 , 𝑦1 ) 𝑎𝑛𝑑 (𝑥2 , 𝑦2 ) the
elementary arc length,

𝑑𝑦 2 1
𝑑𝑠 = √𝑑𝑥 2 + 𝑑𝑦 2 = √1 + ( ) = (1 + 𝑦 ′2 )2
𝑑𝑥

Hence the path length,

𝑥2 𝑥2 1
𝑆 = ∫ 𝑑𝑠 = ∫ (1 + 𝑦 ′ )2 𝑑𝑥
𝑥1 𝑥1

But there may be numbers of different paths connecting the points.

Thus, for minimum or maximum (extremum) path length 𝑆, the integral must be minimum.
𝑥2 𝑥2 1
𝛿𝑆 = 𝛿𝑙 = 𝛿 ∫ 𝑑𝑠 = 𝛿 ∫ (1 + 𝑦 ′2 )2 𝑑𝑥 = 0
𝑥1 𝑥1
1
Here, 𝐼 = 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑎𝑙; (1 + 𝑦 ′2 )2 is the extremal.

Example 2 :
Let us design a bend path connecting two points (𝑥1 , 𝑦1 ) 𝑎𝑛𝑑 (𝑥2 , 𝑦2 ) such that a bead will take
the minimum time to slide down.
𝑑𝑠
Let the velocity at any time 𝑡 is 𝑣 =
𝑑𝑡
Hence,
1
𝑑𝑠 (1 + 𝑦 ′2 )2
𝑑𝑡 = = 𝑑𝑥
𝑣 𝑣
For the path along which time taken is the minimum the integral
1
𝑡2 𝑥2 (1
+ 𝑦 ′2 )2
∫ 𝑑𝑡 = ∫ 𝑑𝑥
𝑡1 𝑥1 𝑣

Must be extremum.
Hence,
1
𝑡2 𝑥2 (1
+ 𝑦 ′2 )2
𝛿 ∫ 𝑑𝑡 = 𝛿 ∫ 𝑑𝑥 = 0
𝑡1 𝑥1 𝑣

Here,
1
𝑥2 (1
+ 𝑦 ′2 )2
∫ 𝑑𝑥 𝑖𝑠 𝑓𝑢𝑛𝑐𝑡𝑖𝑜𝑛𝑎𝑙 ;
𝑥1 𝑣
1
(1 + 𝑦 ′2 )2
𝑖𝑠 𝑒𝑥𝑡𝑟𝑒𝑚𝑎𝑙.
𝑣
Q: Obtain Euler’s equation of motion. Hence extend it to Euler-Lagrange equation of
motion.
𝑑𝑦
We consider a function 𝑓(𝑥, 𝑦, 𝑦 ′ ) [𝑤ℎ𝑒𝑟𝑒 𝑦 ′ = ] . The value
𝑑𝑥
of the integral
𝑥2
𝐼 = ∫ 𝑓(𝑥, 𝑦, 𝑦 ′ ) … … … … … … … (1)
𝑥1

Depends on the function 𝑓, the function 𝑦 = 𝑦(𝑥), and the two

end pints 𝑥1 , 𝑥2 . Fig. 1 shows an example of various paths
between two points.

The basic problem is the calculus of variations if finding a function 𝑦 = 𝑦(𝑥) such that the
integral of the function 𝑓(𝑥, 𝑦, 𝑦 ′ ) has a stationary value on a given interval.
We consider a parameter 𝛼 such that 𝑦(𝛼, 𝑥) will give all possible paths between 𝑥1 , 𝑥2 as,
𝑦(𝛼, 𝑥) = 𝑦(𝑥) + 𝛼𝜂(𝑥) … … … … … . (2)
𝛼 is a parameter independent of 𝑥 and 𝜂(𝑥) is a differential function. Clearly when 𝛼 = 0,
𝑦(0, 𝑥) = 𝑦(𝑥) , is the original (extremum) path.
 𝜕𝑥
ℎ𝑒𝑟𝑒, 𝑖) =0
𝜕𝛼
𝑖𝑖) 𝜂(𝑥1 ) = 𝜂(𝑥2 ) = 0.
For this family of curve, we design the integral,
𝑥2
𝐼 = ∫ 𝑓[𝑥, 𝑦(𝛼, 𝑥), 𝑦 ′ (𝛼, 𝑥)] 𝑑𝑥
𝑥1
𝑥2
′ (𝛼)
𝜕𝑓 𝜕𝑥 𝜕𝑓 𝜕𝑦 𝜕𝑓 𝜕𝑦 ′
ℎ𝑒𝑛𝑐𝑒, 𝐼 =∫ [ + + ] 𝑑𝑥
𝑥1 𝜕𝑥 𝜕𝛼 𝜕𝑦 𝜕𝛼 𝜕𝑦 ′ 𝜕𝛼

𝑥2
𝜕𝑓 𝜕𝑓
=∫ [𝜂 (𝑥) + 𝜂′ (𝑥) ] 𝑑𝑥
𝑥1 𝜕𝑦 𝜕𝑦′

𝑥2 𝑥2
𝜕𝑓 𝜕𝑓
= ∫ 𝜂(𝑥) 𝑑𝑥 + ∫ 𝜂′ (𝑥) 𝑑𝑥
𝑥1 𝜕𝑦 𝑥1 𝜕𝑦 ′

𝑥2
𝜕𝑓 𝜕𝑓 𝑥2 𝑥2
𝜕 𝜕𝑓
= ∫ 𝜂(𝑥) 𝑑𝑥 + [𝜂 (𝑥) ′ ] − ∫ 𝜂(𝑥) ( ) 𝑑𝑥
𝑥1 𝜕𝑦 𝜕𝑦 𝑥 𝑥1 𝜕𝑥 𝜕𝑦
1

𝑥2
𝜕𝑓 𝜕 𝜕𝑓
=∫ [ − ( )] 𝜂(𝑥) 𝑑𝑥
𝑥1 𝜕𝑦 𝜕𝑥 𝜕𝑦 ′

Since, 𝜂(𝑥1 ) = 𝜂(𝑥2 ) = 0,

For the path 𝑦 = 𝑦(𝑥) = 𝑦(0, 𝑥) to be extremum we set,
𝜕𝐼
𝐼 ′ (0) = ( ) =0
𝜕𝛼 𝛼=0
For arbitrary function 𝜂(𝑥) ≠ 0 , we must have the integrand,
𝜕𝑓 𝜕 𝜕𝑓
− ( )=0
𝜕𝑦 𝜕𝑥 𝜕𝑦 ′
i.e.
𝜕 𝜕𝑓 𝜕𝑓
( ′) − =0
𝜕𝑥 𝜕𝑦 𝜕𝑦
This is known as the Euler equation.

The δ notation:
The result of calculus of variation are very often expressed in term of a compact 𝛿 −
𝑛𝑜𝑡𝑎𝑡𝑖𝑜𝑛 as,

𝑥2
𝜕𝐼 ′ (𝛼)
𝜕𝑓 𝜕 𝜕𝑓
𝛿𝐼 = 𝑑𝛼 = 𝐼 𝑑𝛼 = [ ∫ − ( )] 𝜂(𝑥)𝑑𝑥 𝑑𝛼
𝜕𝛼 𝑥1 𝜕𝑦 𝜕𝑥 𝜕𝑦 ′
𝑥2
𝜕𝑓 𝜕 𝜕𝑓 𝜕𝑦
=∫ [ − ( )] 𝑑𝛼 𝑑𝑥
𝑥1 𝜕𝑦 𝜕𝑥 𝜕𝑦 ′ 𝜕𝛼
𝑥2
𝜕𝑓 𝜕 𝜕𝑓
=∫ [ − ( )] 𝛿𝑦 𝑑𝑥
𝑥1 𝜕𝑦 𝜕𝑥 𝜕𝑦 ′

Since, 𝛿𝑦 is arbitrary, we can get 𝛿𝐼 = 0 only if integrand vanishes i.e.

𝜕𝑓 𝜕 𝜕𝑓
− ( )=0
𝜕𝑦 𝜕𝑥 𝜕𝑦 ′
Euler Lagrange equation:
The result can be generalised to the case where 𝑓 is a function of many independent variables
𝑦𝑗 , and their derivatives 𝑦𝑗̇ then,
𝑥2
𝛿𝐼 = 𝛿 ∫ 𝑓 (𝑥, 𝑦1 𝑦2 𝑦3 … … 𝑦1̇ 𝑦2̇ 𝑦3̇ … ) 𝑑𝑥
𝑥1

Thus,
𝜕 𝜕𝑓 𝜕𝑓
( ′) − =0
𝜕𝑥 𝜕𝑦𝑗 𝜕𝑦𝑗

This is known as Euler-Lagrange equation.

Problem 1: Show that the shortest distance between two points is a straight
line.
The elementary length of the line connecting the points 𝐴(𝑥1 , 𝑦1 ) and 𝐵(𝑥2 , 𝑦2 )

𝑑𝑦 2 1
𝑑𝑠 = √𝑑𝑥 2 + 𝑑𝑦 2 = √1 + ′2
𝑑𝑥 = (1 + 𝑦 )2 𝑑𝑥
𝑑𝑥 2
𝑥2 1
′2 2
𝑆 = ∫ (1 + 𝑦 ) 𝑑𝑥
𝑥1
 1
For 𝑆 to be minimum the function 𝑓 = (1 + 𝑦 ′ )2 should satisfy the Euler
Lagrange’s equation,
𝜕 𝜕𝑓 𝜕𝑓
( ′) − =0
𝜕𝑥 𝜕𝑦 𝜕𝑦
1
′2 )2
𝑓 = (1 + 𝑦
Hence,
𝜕𝑓
=0
𝜕𝑦
1
𝜕𝑓 1 −
′2 2
′
= (1 + 𝑦 ) 2𝑦 ′
𝜕𝑦 2
1
= 𝑦 ′ (1 + 𝑦 ′2 )−2
𝜕 𝜕𝑓 𝜕 ′(
1
′2 )−2
( )= (𝑦 1 + 𝑦 )
𝜕𝑥 𝜕𝑦 ′ 𝜕𝑥
Thus,
1
𝜕 ′
−
′2 2
(𝑦 (1 + 𝑦 ) ) = 0
𝜕𝑥
1
−2
′ ′2
𝑜𝑟, 𝑦 (1 + 𝑦 ) =𝑘
𝑦 ′2
𝑜𝑟, ′2
= 𝑘2
1+𝑦

′2
𝑘2
𝑜𝑟, 𝑦 =
1 − 𝑘2
Hence,

𝑘2
𝑦 =√
′
= 𝑚 (𝑠𝑎𝑦 𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡)
1 − 𝑘2

Hence,
𝑦 = 𝑚𝑥 + 𝑐
It represents a straight line.
Problem 2:
THE BRANCHISTOCHORNE PROBLEM: -
In this problem, we find a curve joining two
points along which a particle falling from rest under
the influence of gravity travels from higher to the
lower point in the minimum time.
Suppose 𝑣 is the speed of the particle along the
curve, then in traversing 𝑑𝑠 portion of the curve,
𝑑𝑠
time spent will be . So that total time taken by the
𝑣
particle in moving form highest point 1 to the lowest
point 2 will be:
2
𝑑𝑠
𝑡12 = ∫
1 𝑣
Suppose vertical distance of fall up to the point 2 be 𝑥, then from the conservation
principle for the energy of the particle, we find that
1
𝑚𝑣 2 = 𝑚𝑔𝑥
2
𝑜𝑟, 𝑣 = √2𝑔𝑥
Then,
2 2
√𝑑𝑥 2 + 𝑑𝑦 2 √1 + 𝑦 2
𝑡12 = ∫ =∫ 𝑑𝑥
1 √2𝑔𝑥 1 √2𝑔𝑥
2
𝑡12 = ∫ 𝑓 𝑑𝑥
1
1
′2 2
1+ 𝑦
𝑓=( )
2𝑔𝑥
For 𝑡12 to be minimum, the equation
 𝑑 𝑑𝑓
( )=0
𝑑𝑥 𝑑𝑦 ′
𝑑𝑓 1 𝑦′
=
𝑑𝑦 ′ √2𝑔𝑥
√1 + 𝑦 ′ 2

So that the above equation becomes,

𝑑 𝑦′
( )=0
𝑑𝑥 √2𝑔𝑥 √1 + 𝑦 ′2

𝑦′
𝑜𝑟, = 𝐶,
𝑥√1 + 𝑦 ′2
Where 𝐶 is a constant. On rearranging the above equation, we find
𝑦 ′2
= 𝑥 (1 + 𝑦 ′2 )
𝐶
1
𝑜𝑟, 𝑦 ′2 ( − 𝑥) = 𝑥
𝐶
2 𝑥
𝑜𝑟, 𝑦 ′ ( − 𝑥 2 ) = 𝑥 2
𝐶
𝑥
𝑜𝑟, 𝑦′ = 1
𝑥 2
(𝐶 − 𝑥 2 )

Let us put
1
= 2𝑎,
𝐶
So that on integration,
−1
𝑥 1
𝑦 = 𝑎 cos (1 − ) − (2𝑎𝑥 − 𝑥 )2 + 𝐶 ′
2
𝑎
Where 𝐶 ′ is the new constant of integration.
Obviously if 𝐶 ′ be zero then 𝑦 will be zero for 𝑥 to be zero. In such a case,
𝑥 1
𝑦 = 𝑎 cos−1 (1 − ) − (2𝑎𝑥 − 𝑥 2 )2
𝑎
Which represent an inverted cycloid with its base along 𝑦 − 𝑎𝑥𝑖𝑠 and cusp at the
origin.
Q: Deduction of Newton’s second law of motion from Hamilton’s principle: -
Suppose a particle of mass 𝑚 at the position 𝑟 = 𝑟(𝑥, 𝑦, 𝑧) is moving under the
action of a field of force 𝐹. The kinetic energy of the particle is
1
𝑇= 𝑚 |𝑟̇ 2 |
2
And the variation of work done is
𝛿𝑊 = 𝐹 ⋅ 𝛿𝑟 = −𝛿𝑉
Now, from Hamiltonian’s variational principle
𝑡2 𝑡2
0 = 𝛿 ∫ (𝑇 − 𝑉) 𝑑𝑡 = ∫ (𝛿𝑇 − 𝛿𝑉) 𝑑𝑡
𝑡1 𝑡1
𝑡2
= ∫ (𝑚𝑟⃗̇ ⋅ 𝛿𝑟⃗̇ + 𝐹⃗ ⋅ 𝛿𝑟⃗) 𝑑𝑡
𝑡1
𝑡2 𝑡2
= ∫ 𝑚𝑟⃗̇ ⋅ 𝛿𝑟⃗̇ 𝑑𝑡 + ∫ 𝐹⃗ ⋅ 𝛿𝑟⃗ 𝑑𝑡
𝑡1 𝑡1

Integrating the first term on RHS by parts, we get

𝑡2 𝑡2
𝑡
0 = 𝑚𝑟⃗̇ ⋅ 𝛿𝑟⃗|𝑡21 − ∫ 𝑚𝑟⃗̈ ⋅ 𝛿𝑟⃗ 𝑑𝑡 + ∫ 𝐹⃗ ⋅ 𝛿𝑟⃗ 𝑑𝑡
𝑡1 𝑡1
𝑡2
= − ∫ (𝑚𝑟⃗̈ − 𝐹⃗ ) ⋅ 𝛿𝑟⃗ 𝑑𝑡
𝑡1

Because 𝛿𝑟⃗ vanishes at end points. The above equation is true for every virtual
displacement 𝛿𝑟 and hence integrand must vanish.
i.e.
𝑚𝑟̈ = 𝐹
Which is the Newton’s second law of motion.
Hamilton’s Principle of Extremization of action:
The principle of least action – or, more accurately, the principle of stationary
action- is a variational principle that, when applied to the action of a mechanical
system, can be used to obtain the equations of motion for that system.
[The principle can be used to rederive Newtonian, Lagrange and Hamiltonian equations of
motion, and even general relativity (see Eistein-Hibert action). The physicist Paul Dirac and
after him Julian Schwinger and Richard Feynman, demonstrated how this principle can also be
used in quantum calculations. It was historically called “least” because its solution requires
finding the path that has the least value.]

Q: Give the statement of principle of least action.

The motion of the system from 𝑡1 to 𝑡2 is such that the line integral of 𝐿 = 𝑇 − 𝑉 is
an extremum for the path of motion.
i.e.
𝑡2
∫ 𝐿 𝑑𝑡 = 0
𝑡1

Q: Derive Lagrange’s equation of motion from Hamilton’s principle.

According to Hamilton’s variational principle, motion of the conservative system
from the 𝑡1 to 𝑡2 is such that the line integral,
𝑡2
∫ 𝐿 𝑑𝑡
𝑡1

is extremum. i.e.
𝑡2
𝛿 ∫ 𝐿 𝑑𝑡 = 0
𝑡1

Considering all possible paths of motion in the configuration space (with general
coordinates 𝑞𝑗 , ) and label each with a value of parameter 𝛼 we have,
𝑡2
𝐼 (𝛼 ) = ∫ 𝐿 [𝑞𝑗 (𝑡, 𝛼 ), 𝑞̇ 𝑗 (𝑡, 𝛼 ), 𝑡] 𝑑𝑡
𝑡1

Hence,
 𝑡2
𝜕𝐼 (𝛼 ) 𝜕𝐿 𝜕𝑞𝑗 𝜕𝐿 𝜕𝑞𝑗̇ 𝜕𝐿 𝜕𝑡
= ∫ ∑[ + + ] 𝑑𝑡
𝜕𝛼 𝑡1 𝜕𝑞𝑗 𝜕𝛼 𝜕𝑞̇ 𝑗 𝜕𝛼 𝜕𝑡 𝜕𝛼
𝑗
𝜕𝑡
Ass there is no variation in time along any path i.e. = 0;
𝜕𝛼
𝑡2 𝑡2
𝜕𝐼 (𝛼 ) 𝜕𝐿 𝜕𝑞𝑗 𝜕𝐿 𝜕𝑞𝑗̇
∴ =∫ ∑ 𝑑𝑡 + ∫ ∑ 𝑑𝑡
𝜕𝛼 𝑡1 𝜕𝑞𝑗 𝜕𝛼 𝑡1 𝜕𝑞 ̇
𝑗̇ 𝜕𝛼
𝑗 𝑗

Integrating by parts the second term,

𝑡2
𝑡2 𝑡2
𝜕𝐼 (𝛼 ) 𝜕𝐿 𝜕𝑞𝑗 𝜕𝐿 𝜕𝑞𝑗 𝑑 𝜕𝐿 𝜕𝑞𝑗
=∫ ∑ 𝑑𝑡 + [∑ ] −∫ ∑ ( ) 𝑑𝑡
𝜕𝛼 𝑡1 𝜕𝑞𝑗 𝜕𝛼 𝜕𝑞𝑗̇ 𝜕𝛼 𝑡1 𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝛼
𝑗 𝑗 𝑗
𝑡1

Again, there must be no variation at the end points i.e.

𝜕𝑞𝑗 𝜕𝑞𝑗
(𝑡1 ) = (𝑡 ) = 0
𝜕𝛼 𝜕𝛼 2
𝑡2
𝜕𝐼 (𝛼 ) 𝜕𝐿 𝑑 𝜕𝐿 𝜕𝑞𝑗
= ∫ ∑[ − ( )] 𝑑𝑡
𝜕𝛼 𝑡1 𝜕𝑞𝑗 𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝛼
𝑗

Hence, multiplying by 𝑑𝛼,

𝑡2
𝜕𝐼 (𝛼 ) 𝜕𝐿 𝑑 𝜕𝐿 𝜕𝑞𝑗
𝑑𝛼 = ∫ ∑ [ − ( )] 𝑑𝛼𝑑𝑡
𝜕𝛼 𝑡1 𝜕𝑞 𝑗 𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝛼
𝑗
𝑡2
𝜕𝐿 𝑑 𝜕𝐿
𝑖. 𝑒. 𝛿𝐼 (𝛼 ) = ∫ ∑ [ − ( )] 𝛿𝑞𝑗 𝑑𝑡
𝑡1 𝜕𝑞 𝑗 𝑑𝑡 𝜕𝑞̇ 𝑗
𝑗

According to the Hamilton’s principle 𝛿𝐼 = 0; ℎ𝑒𝑟𝑒 𝑞𝑗 are independent

coordinates.
Thus, coefficients of each 𝑞𝑗 must be zero.
𝜕𝐿 𝑑 𝜕𝐿
− ( )=0
𝜕𝑞𝑗 𝑑𝑡 𝜕𝑞̇ 𝑗
𝑑 𝜕𝐿 𝜕𝐿
𝑜𝑟, ( )− =0
𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝑞𝑗
Which is Lagrange’s equation of motion for the conservative system.
 FEW BASIC IDEAS IN LAGRANGIAN FORMULATION
Constraints: -
A motion that cannot proceed in any arbitrarily manner, is called constrained
motion. The conditions which restricts the motion of the system are called
constraints.
e.g. (1) Let 𝑅 be the fixed radius of the container containing gas molecules. Gas
molecules are constraints by the walls and can move only inside the container. So,
the constraint relations are,
|⃗𝑟⃗𝑖 | ≤ 𝑅

e.g. (2) If we consider the motion of a simple pendulum, the bob (𝑥, 𝑦) is restricted
to move in 𝑥, 𝑦 plane and its distance 𝑙 is to remain fixed from the point of
suspension. So, the constrains relation are,
𝑧=0
𝑥 2 + 𝑦 2 = 𝑙2
Constraint Force: -
A constrained motion realized by forces that arises wherever the body in motion, in
in contact with constraining curves or surfaces. These forces are called forces of
constraints or constraint.
Generalised Co-ordinates :-
Degrees of Freedom :- The number of independent ways in which a mechanical system
can move without violating any constraint is called the number of degrees of freedom of
the system.
Def. : The minimum number of independent variables require to specify the position
and configuration of a dynamical system.
Generalised Co-ordinates :-
For a system of N particles, free from constraints we require a total 3N independent Co-
ordinates to describe its configuration completely.
Let, there are now K constraints of the type,
 𝑓𝑠 (𝑟⃗⃗⃗⃗⃗
1 , ⃗⃗⃗⃗,
𝑟2 ⋯ ⋯ ⃗⃗⃗⃗⃗,
𝑟𝑁 𝑡) = 0 𝑠 = 1,2,3, ⋯ ⋯ 𝐾
acting on the system.
Thus the system has only 3𝑁 − 𝑘 independent Co- ordinates or degrees of freedom.
These 3𝑁 − 𝑘 independent Co-ordinates represented by the variables,
𝑞1 , 𝑞2 , 𝑞3 , ⋯ ⋯ 𝑞3𝑁−𝑘
These are called generalized Co-ordinates. The Co-ordinates ⃗⃗⃗⃗⃗
𝑟1 , ⃗⃗⃗⃗,
𝑟2 ⋯ ⋯ ⃗⃗⃗⃗⃗
𝑟𝑁 in terms of
these Co-ordinates Can be expressed as,
𝑟1 (𝑞1 , 𝑞2 , ⋯ ⋯ 𝑞𝑛 )
𝑟1 = ⃗⃗⃗⃗
⃗⃗⃗⃗
𝑟2 1 , 𝑞2 , ⋯ ⋯ 𝑞𝑛 )
𝑟2 = ⃗⃗⃗⃗(𝑞
⃗⃗⃗⃗
⋮
𝑟𝑁 = ⃗⃗⃗⃗⃗(𝑞
⃗⃗⃗⃗⃗ 𝑟𝑁 1 , 𝑞2 , ⋯ ⋯ 𝑞𝑛 ) Where 𝑛 = 3𝑁 − 𝑘
Generalised Co-ordinates are not unique. Depending on Problem , it is Convenient to
select some Co-ordinates of dimensions of angle , energy, or may be combination of angles
and Co-ordinates.
Lagrange’s equations in term of the generalized coordinate,
𝑑 𝜕𝐿 𝜕𝐿
( )− = 0 ; 𝑊ℎ𝑒𝑟𝑒 𝐿 = 𝑇 − 𝑉 ⋯ ⋯ (12)
𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝑞𝑗
L is called Lagrangian of the system.
Note: In Newtonian approach, the focus is on focus and vector quantities like
acceleration, momentum etc. In Lagrangian approach the focus is on the Scalar energies
like K.E, P.E. of the system and no vector quantity by directly enters into it. Scalars are
easier to deal with and thus this method simplifies the problems.
• Kinetic Energy : To express K.E. in the generalized Co-ordinates system we consider
generalized Co-ordinates 𝑞𝑗 (𝑗 = 1,2,3, … . . 𝑛) where,

⃗𝑟⃗𝑖 = ⃗𝑟⃗(𝑞
𝑖 𝑗 , 𝑡)

𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖
∴ ⃗𝑟⃗𝑖̇ = ∑ 𝑞𝑗̇ +
𝜕𝑞𝑗 𝜕𝑡
1
𝐾. 𝐸. 𝑇 = ∑ 𝑚𝑖 𝑣𝑖2
2
𝑖

1 𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖

= ∑ 𝑚𝑖 (∑ 𝑞𝑗̇ + ) . (∑ 𝑞𝑘̇ + )
2 𝜕𝑞𝑗 𝜕𝑡 𝜕𝑞𝑘 𝜕𝑡
𝑖 𝑗 𝑘
 1 𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖 1 𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖 1 𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖
= ∑ ∑ ∑ 𝑚𝑖 . 𝑞𝑗̇ 𝑞𝑘̇ + ∑ 𝑚𝑖 (2 ∑ . 𝑞𝑗̇ ) + ∑ 𝑚𝑖 ( . )
2 𝜕𝑞𝑗 𝜕𝑞𝑘 2 𝜕𝑞𝑗 𝜕𝑡 2 𝜕𝑡 𝜕𝑡
𝑖 𝑗 𝑘 𝑖 𝑗 𝑖

Hence,

𝑇(𝑞̇ , 𝑞, 𝑡) = ∑ ∑ 𝑎𝑗𝑘 𝑞𝑗̇ 𝑞𝑘̇ + ∑ 𝑏𝑗 𝑞𝑗̇ + 𝑐

𝑗 𝑘 𝑗

1 𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖
𝑤ℎ𝑒𝑟𝑒, 𝑎𝑗𝑘 = ∑ 𝑚𝑖 .
2 𝜕𝑞𝑗 𝜕𝑞𝑘
𝑖
𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖
𝑏𝑗 = ∑ 𝑚𝑖 .
𝜕𝑞𝑗 𝜕𝑡
𝑖
2
1 𝜕𝑟⃗⃗𝑖
𝑐 = ∑ 𝑚𝑖 ( )
2 𝜕𝑡
𝑖

Thus K.E.is made up of three theorem,

(1) One term is independent of generalized velocity.
(2) One term is linear in velocity.
(3) One term is quadratic in velocity.
𝜕𝑟𝑖
If time does not appear explicitly in the transformation eqns. Then = 0 thus 𝑏 = 𝑐 = 0.
𝜕𝑡

∴ 𝑇 = ∑ ∑ 𝑎𝑗𝑘 𝑞𝑗̇ 𝑞𝑘̇

𝑗 𝑘

i.e. kinetic energy is a homogeneous quadratic function of generalized velocity.

• Generalized momentum : The partial derivative of the Lagrangian w.r.t. the
generalized velocity 𝑞𝑗̇ is Called generalized momentum 𝑝𝑗 corresponding to Co-
ordinates 𝑞𝑗 i.e.

𝜕𝐿
𝑝𝑖 =
𝜕𝑞𝑗̇

• Cyclic or ignorable Co-ordinates : If the Lagrangian of a system is not the function of

a given Co-ordinate 𝑞𝑗 , then the Co-ordinates 𝑞𝑘 is said to be Cyclic or ignorable Co-
ordinates.

i.e.
𝜕𝐿
=0
𝜕𝑞𝑘

To Prove that if any Co-ordinate is cyclic then the corresponding momentum is

conserved:-
If the generalized Co-ordinates 𝑞𝑗 is Cyclic, then we have,
𝜕𝐿
=0
𝜕𝑞𝑗
So from Lagrange’s eqn.
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝑞̇ 𝜕𝑞
𝑑 𝜕𝐿
𝑂𝑟, ( )=0
𝑑𝑡 𝜕𝑞𝑗̇
𝑑
𝑂𝑟, (𝑝 ) = 0
𝑑𝑡 𝑗
𝑖. 𝑒. 𝑝𝑗 = 𝐶𝑜𝑛𝑠𝑡. 𝑜𝑟 𝐶𝑜𝑛𝑠𝑒𝑟𝑣𝑒𝑑
i.e. generalized momentum conjugate to a Cyclic Co-ordinate is Conserved during the
motion.

Conservation Laws and Symmetry Properties :

1. Homogeneity of Space and Conservation of linear momentum :-
Homogeneity in space means the mechanical properties of a closed System remain
unchanged by any parallel displacement of the entire system in space.
i.e. 𝛿𝐿 = 0 for infinitesimal displacement of the system as ⃗𝑟⃗𝑖 → ⃗𝑟⃗𝑖 + 𝛿𝑟⃗⃗𝑖 → ⃗𝑟⃗𝑖 + 𝜖

𝜕𝐿 𝜕𝐿 𝜕𝐿
∴ 𝛿𝐿 = ∑ ∙ 𝛿𝑟⃗⃗𝑖 = ∑ 𝜖⃗ . = 𝜖⃗. (∑ )
𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖 𝜕𝑟⃗⃗𝑖
𝑖 𝑖 𝑖

As 𝜖 is a constant vector of infinitesimally small in magnitude.

We require 𝛿𝐿 = 0 for any arbitrarily 𝜖
𝜕𝐿
∴∑ =0
𝜕𝑟𝑖
𝑖

With is condition Lagrangian Condition becomes

𝑑 𝜕𝐿
∑ ( )=0
𝑑𝑡 𝜕𝑟𝑖̇
𝑑
𝑂𝑟, ∑ 𝑝 =0
⃗⃗⃗⃗
𝑑𝑡 𝑖

𝑂𝑟, ∑ ⃗⃗⃗⃗
𝑝𝑖 = 𝐶𝑜𝑛𝑠𝑡.
Thus the total linear momentum of any system is conserved due to homogeneity of space.
2. Isotropy of space and Conservation of angular momentum
Due to arbitrary of space, the Lagrangian of system remains
unchanged under arbitrary rotation.
Let us Consider an arbitrary infinitesimal rotation of the system
about 𝑛̂(𝑍 𝑎𝑥𝑖𝑠). Due to an Small rotation 𝜕𝜃 change in position
Vector, |𝛿𝑟⃗| = 𝐴𝐵𝛿𝜃 = 𝑟𝑖 sin 𝜑 𝜕𝜃 ∴ 𝛿𝑟⃗ = 𝜕𝜃𝑛̂ × ⃗𝑟⃗𝑖

𝑇𝑎𝑘𝑖𝑛𝑔 𝐿 = 𝐿(𝑟⃗⃗⃗⃗𝑖 , ⃗𝑟⃗𝑗̇ )

Change in 𝐿
𝜕𝐿 𝜕𝐿
𝛿𝐿 = ∑ ( ∙ 𝛿𝑟⃗⃗𝑖 + ∙ 𝛿𝑟⃗⃗𝑖̇ )
𝜕𝑟⃗⃗𝑖 ̇
𝜕𝑟⃗⃗𝑖
𝑖

⃗⃗⃗⃗𝑖̇ ∙ 𝛿𝑟⃗⃗𝑖 + ⃗⃗⃗⃗

= ∑(𝑝 𝑝𝑖 ∙ 𝛿𝑟⃗⃗𝑖̇ )
𝑖
𝑑
= ∑(𝑝
⃗⃗⃗⃗𝑖 ∙ 𝛿𝑟⃗⃗)
𝑖
𝑑𝑡
𝑖
𝑑
= ∑[𝑝
⃗⃗⃗⃗.𝑖 (𝜕𝜃𝑛̂ × ⃗𝑟⃗)]
𝑖
𝑑𝑡
𝑖
𝑑
= ∑[𝑝
⃗⃗⃗⃗𝑖 ∙ (𝜕𝜃𝑛̂ × ⃗𝑟⃗)]
𝑖
𝑑𝑡
𝑖
𝑑
= (∑ 𝜕𝜃𝑛̂ ∙ (𝑟⃗⃗𝑖 × ⃗⃗⃗⃗)
𝑝𝑖 )
𝑑𝑡
𝑖
𝑑
= 𝜕𝜃𝑛̂ ∙ (∑(𝑟⃗⃗𝑖 × ⃗⃗⃗⃗)
𝑝𝑖 )
𝑑𝑡
𝑖

Since for arbitrary 𝜕𝜃, 𝛿𝐿 = 0,

𝑑
∴ ∑ ⃗𝑟⃗𝑖 × ⃗⃗⃗⃗
𝑝𝑖 = 0
𝑑𝑡
𝑖

𝑑
𝑂𝑟. ⃗⃗𝑖 ) = 0
∑(𝛺
𝑑𝑡
𝑖

⃗⃗ = 𝐶𝑜𝑛𝑠𝑡.
∴ 𝛺 ⃗⃗ = ∑ ⃗𝑟⃗𝑖 × ⃗⃗⃗⃗)
(𝛺 𝑝𝑖

3. Homogeneity of time and Conservation of Energy :-

Homogeneity of time implies that Lagrangian of a system does not depend explicitly on
time t.
 𝜕𝐿
𝑖. 𝑒. =0
𝜕𝑡
Total time derivative of Lagrangian, 𝐿 = 𝐿(𝑞𝑖 , 𝑞𝑖̇ )
𝑑𝐿 𝜕𝐿 𝜕𝐿
=∑ 𝑞𝑖̇ + 𝑞̈
𝑑𝑡 𝜕𝑞𝑖 𝜕𝑞𝑖̇ 𝑖
𝑖
𝑑𝐿 𝜕𝐿 𝑑 𝜕𝐿 𝜕𝐿
= ∑ 𝑝̇ 𝑖 𝑞𝑖̇ + 𝑝𝑖 𝑞𝑖̈ [𝐴𝑠 [𝑝𝑖 = ]& ( ) = ]
𝑑𝑡 𝜕𝑞𝑖̇ 𝑑𝑡 𝜕𝑞𝑖̇ 𝜕𝑞𝑖
𝑖
𝑑
=∑ (𝑝 𝑞̇ )
𝑑𝑡 𝑖 𝑖
𝑑
∴ [∑ 𝑝𝑖 𝑞𝑖̇ − 𝐿] = 0
𝑑𝑡
𝑖

∴ ∑ 𝑝𝑖 𝑞̇ 𝑖 − 𝐿 = 𝐶𝑜𝑛𝑠𝑡. ⋯ ⋯ (1)

We denote this constant quantity as H , called Hamiltonian of the system i.e.

𝐻 = ∑ 𝑝𝑖 𝑞𝑖̇ − 𝐿
𝑖

Let us show now it is related with total mechanical energy.

IF the transformation equations do not contain time explicitly , 𝑇 will be a
homogeneous quadratic function of generalised velocities and by Euler’s theorem,
𝜕𝑇
∑ = 2𝑇
𝜕𝑞𝑖̇
𝑖

Thus, 𝐻 = ∑ 𝑝𝑖 𝑞𝑖̇ − 𝐿
𝑖

𝜕𝐿
=∑ 𝑞̇ − 𝐿
𝜕𝑞𝑖̇ 𝑖
𝑖
𝜕(𝑇 − 𝑉)
=∑ 𝑞𝑖̇ − 𝐿 [𝑉 ≠ 𝑉(𝑞𝑗̇ )]
𝜕𝑞𝑖̇
𝑖
𝜕𝑇
=∑ 𝑞̇ − 𝐿
𝜕𝑞𝑖̇ 𝑖
𝑖
= 2𝑇 − 𝑇 + 𝑉
= 𝑇 + 𝑉 = 𝐸(𝑇𝑜𝑡𝑎𝑙)
It shows the Conservation of total Energy. So Summing up , the laws of Conservation of
linear momentum , angular momentum and total energy are an immediate Consequences
of the Symmetry properties of Space and time.
An interesting point to be noted that the pairs (𝑟⃗, 𝑝⃗) ; (𝜃, 𝐿⃗⃗); (𝑡, 𝐸) involved in the
Conservation laws, play important roll in uncertainly principle in quantum mechanics.
Hamiltonian mechanics:-
In Lagrangian formulation , generalized Co-ordinates (𝑞𝑖 , 𝑠) and generalized velocities
(𝑞𝑖̇ , 𝑠) are used as independent Co-ordinate to formulate dynamical problems which
results in 2nd order linear differential equations. In Hamiltonian formulation , generalized
momenta (𝑞𝑖 , 𝑝𝑖 ) are used as basic variables to formulate problems. It is based on the
Hamiltonian function 𝐻(𝑝, 𝑞. 𝑡) of the system.
Resulting 1st order linear differential equations are easier to deal with. It appears not
only powerful but also proves to be Convenient, particularly to deal with problems of
modern Physics.

Hamilton’s equation of motion from the Hamiltonian of the system :-

The Hamiltonian , in general , is a function of generalized momenta (𝑝𝑖 ) generalized Co-

ordinate (𝑞𝑗 ) and time (𝑡).

𝑖. 𝑒. 𝐻 = 𝐻(𝑞𝑗 , 𝑝𝑗 , 𝑡)
𝜕𝐻 𝜕𝐻 𝜕𝐻
∴ 𝑑𝐻 = ∑ 𝑑𝑞𝑗 + ∑ 𝑑𝑝𝑗 + ⋯ ⋯ ⋯ (1)
𝜕𝑞𝑗 𝜕𝑝𝑗 𝜕𝑡
𝑗 𝑗

From definition, 𝐻 = ∑𝑖 𝑝𝑖 𝑞𝑗̇ − 𝐿

Where 𝐿(𝑞1 , 𝑞2 , … , 𝑞𝑛 , 𝑝1 , 𝑝2 , … 𝑝𝑛 , 𝑡) is the Lagrangian of the system.

∴ 𝐻𝑒𝑟𝑒 𝑑𝐻 = ∑ 𝑞𝑗̇ 𝑑𝑝𝑗 + ∑ 𝑝𝑗 𝑑𝑞𝑗̇ − 𝑑𝐿

𝑗 𝑗

𝜕𝐿 𝜕𝐿 𝜕𝐿
= ∑ 𝑞𝑗̇ 𝑑𝑝𝑗 + ∑ 𝑝𝑗 𝑑𝑞𝑗̇ − ∑ 𝑑𝑞𝑗 − ∑ 𝑑𝑞𝑗̇ − 𝑑𝑡
𝜕𝑞𝑗 𝜕𝑞𝑗̇ 𝜕𝑡
𝑗 𝑗 𝑗

𝑑 𝜕𝐿 𝜕𝐿
= ∑ 𝑞𝑗̇ 𝑑𝑝𝑗 + ∑ 𝑝𝑗 𝑑𝑞𝑗̇ − ∑ ( ) 𝑑𝑞𝑗 − ∑ 𝑝𝑗 𝑑𝑞𝑗̇ − 𝑑𝑡
𝑑𝑡 𝜕𝑞𝑗̇ 𝜕𝑡
𝑗 𝑗 𝑗 𝑗

𝑑 𝜕𝐿 𝜕𝐿 𝜕𝐿
Where We put , ( ) = 𝜕𝑞 & = 𝑝𝑗 from
𝑑𝑡 𝜕𝑞 ̇ 𝑗 𝑗 𝜕𝑞𝑗̇

Lagrangian formulation.
𝜕𝐿
∴ 𝑑𝐻 = ∑ 𝑞𝑗̇ 𝑑𝑝𝑗 − ∑ 𝑝𝑗̇ 𝑑𝑞𝑗 − 𝑑𝑡 ⋯ ⋯ ⋯ (2)
𝜕𝑡
𝑗 𝑗
Comparing (1) & (2)
𝜕𝐻
𝑞𝑗̇ = ⋯ ⋯ ⋯ (3)
𝜕𝑝𝑗
𝜕𝐻
−𝑝𝑗̇ = ⋯ ⋯ ⋯ (4)
𝜕𝑞𝑘
𝜕𝐿 𝜕𝐻
& − =
𝜕𝑡 𝜕𝑡
Eqns. (3) & (4) are called Hamiltonian equation or Hamiltonian Canonical equation of
motion.
𝜕𝐿
• If a Co-ordinate 𝑞𝑘 is cyclic in Lagrangian, then = 0 which gives,
𝜕𝑞𝑘

𝑑 𝜕𝐿
( )=0
𝑑𝑡 𝜕𝑞𝑘̇
𝑑
𝑜𝑟, 𝑝 =0 𝑂𝑟, 𝑝𝑘 = 𝑐𝑜𝑛𝑠𝑡.
𝑑𝑡 𝑘
𝜕𝐻
From Hamiltonian equation of motion, = 𝑝𝑘̇ = 0 ; 𝐻 is independent of 𝑞𝑘
𝜕𝑞𝑘

i.e. a cyclic Co-ordinates in lagrangian will also be Cyclic in Hamiltonian.

Prove that if Lagrangian of a system is not explicitly dependent on time, then Hamiltonian
is conservative.

Hamiltonian is a function of generalized Co-ordinates (𝑞𝑗 ) generalized momentum (𝑝𝑖 )

and time (𝑡)
∴ 𝐻 = 𝐻(𝑞1 , 𝑞2 , … , 𝑝1 , 𝑝2 , … , 𝑡)
𝑑𝐻 𝜕𝐻 𝜕𝐻 𝜕𝐻
=∑ 𝑞𝑗̇ + ∑ 𝑝𝑗̇ +
𝑑𝑡 𝜕𝑞𝑗 𝜕𝑝𝑗 𝜕𝑡
𝑗 𝑗
𝜕𝐻
= − ∑ 𝑝̇𝑗 𝑞𝑗̇ + ∑ 𝑞𝑗̇ 𝑝𝑗̇ +
𝜕𝑡
𝑗 𝑗
𝜕𝐻
=
𝜕𝑡
𝜕𝐻 𝜕𝐿
But we have, =
𝜕𝑡 𝜕𝑡

𝑑𝐻 𝜕𝐻
= =0 𝑂𝑟, 𝐻 𝑖𝑠 𝑐𝑜𝑛𝑠𝑒𝑟𝑣𝑎𝑡𝑖𝑣𝑒.
𝑑𝑡 𝜕𝑡
Legendre Transformations:
This is a mathematical technique used to change the basis from one set of variables to
other. Let 𝑓(𝑥, 𝑦) is a function of two variables 𝑥, 𝑦. Let 𝑔 be another function of (𝑥, 𝑣)
such that
𝑔(𝑥, 𝑣) = 𝑦𝑣 − 𝑓(𝑥, 𝑦)
Hence, we can have,

𝜕𝑔 𝜕𝑓 𝜕𝑓
(= 0) = 𝑣 − 𝑖. 𝑒 𝑣 =
𝜕𝑦 𝜕𝑦 𝜕𝑦
And,
𝜕𝑔 𝜕𝑓 𝜕𝑔
=𝑦− (= 0) 𝑔𝑖𝑣𝑒𝑠 𝑦 =
𝜕𝑣 𝜕𝑣 𝜕𝑣

Similarly ,
From the Lagrangian,
𝐿 = 𝐿(𝑞𝑗 , 𝑞̇ 𝑗 , 𝑡)
Using Legendre transformation, we define new function of 𝑞𝑗 , 𝑝𝑗 such that,

𝐻(𝑝𝑗 , 𝑞𝑗 , 𝑡) = ∑ 𝑝𝑗 𝑞̇ 𝑗 − 𝐿(𝑞𝑗 , 𝑞̇ 𝑗 , 𝑡) … . (1)

𝑗
Now differentiating both sides with respect to 𝑝𝑗 and 𝑞̇ 𝑗 separately,
𝜕𝐻
𝑞̇ 𝑗 = … … … … … … (𝑖𝑖)
𝜕𝑝𝑗
And,
𝜕𝐿
𝑝𝑗 = … … … … … … … . (𝑖𝑖𝑖)
𝜕𝑞̇ 𝑗
Now differentiating both sides with respect to time,
𝑑 𝜕𝐿 𝜕𝐿
𝑝̇𝑗 = ( )= … … (𝑖𝑣)
𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝑞𝑗
(using Lagrange equation of motion).
Now from

𝐻(𝑝𝑗 , 𝑞𝑗 , 𝑡) = ∑ 𝑝𝑗 𝑞̇ 𝑗 − 𝐿(𝑞𝑗 , 𝑞̇ 𝑗 , 𝑡)
𝑗

differentiating both sides with respect to 𝑞𝑗

𝜕𝐻 𝜕𝐿
=−
𝜕𝑞𝑗 𝜕𝑞𝑗
From equation (𝑖𝑣),
𝜕𝐻
𝑝̇𝑗 = − … … … (𝑣)
𝜕𝑞𝑗
Equations (𝑖𝑖) and (𝑣) are known as the Hamilton’s canonical equation of motion.

CLASSICAL MECHANICS (Problems)

30.

Basic
1. What is the advantage of Lagrangian formulation over Newtonian formulation.
𝑨𝒏𝒔. −Newton formulation is based on the concept of force while the Lagrangian
formulation is based on the Concept of energy. Since energy is a Scalar quantity ,
Lagrangian equation remain invariant under Co-ordinate transformation. In some
Problem particularly when there are forces of Constraints , all the forces may not be
known , but K.E. and P.E. may be obtained.

2. Prove that the momentum Conjugate to a cyclic Co-ordinate is Conserved.(2011)

𝑨𝒏𝒔. −If the Co-ordinate is cyclic means Lagrangian independent of that Co-ordinate
From Lagrangian equation of motion
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝑞𝑗
 𝜕𝐿
If 𝑞 is cyclic then =0
𝜕𝑞𝑗

𝑑 𝜕𝐿
∴ ( )=0
𝑑𝑡 𝜕𝑞̇ 𝑗
𝜕𝐿
∴ = 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡 ; 𝑝𝑗 = 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡
𝜕𝑞̇ 𝑗
∴ Canonically conjugate momentum w.r.t. Cyclic Co-ordinates is Conserved.

3. Show that Lagrange’s equation of motion remains unchanged in form if you add to
the Lagrangian a total time derivative of any function of generalized Co-ordinates
and time.
𝑨𝒏𝒔. − Let us Consider 𝐿 be the Lagrangian of a system
𝑑 𝜕𝐿 𝜕𝐿
∴ ( )− = 0 ⋯ ⋯ (1) (𝑞𝑗 = 1,2,3 … . 𝑛)
𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝑞𝑗
𝑑𝐹
∴ 𝐿′ = 𝐿 + (𝑞 , 𝑞 , 𝑞 … . . 𝑞𝑛 , 𝑡)
𝑑𝑡 1 2 3
We are required to prove that,
𝑑 𝜕𝐿′ 𝜕𝐿′
( )− =0
𝑑𝑡 𝜕𝑞̇ 𝑗 𝜕𝑞𝑗
Now, 𝐹 = 𝐹(𝑞1 , 𝑞2 , … . 𝑞𝑛 , 𝑡)
𝑛
𝑑𝐹 𝜕𝐹 𝜕𝐹
=∑ 𝑞𝑗̇ +
𝑑𝑡 𝜕𝑞𝑗 𝜕𝑡
𝑗=1
𝑛
𝑑𝐹 𝜕𝐹 𝜕𝐹
∴ 𝐿′ = 𝐿 + =𝐿+∑ 𝑞𝑗̇ +
𝑑𝑡 𝜕𝑞𝑗 𝜕𝑡
𝑗=1

𝑛
𝑑 𝜕𝐿′ 𝑑 𝜕 𝜕𝐹 𝜕𝐹
∴ ( )= { (𝐿 + ∑ 𝑞𝑗̇ + )}
𝑑𝑡 𝜕𝑞𝑗̇ 𝑑𝑡 𝜕𝑞𝑗̇ 𝜕𝑞𝑗 𝜕𝑡
𝑗𝑘=1

𝑑 𝜕𝐿 𝜕𝐹 𝜕 𝜕𝐹
= { + + ( )}
𝑑𝑡 𝜕𝑞𝑗̇ 𝜕𝑞𝑗 𝜕𝑞𝑗̇ 𝜕𝑡
𝑑 𝜕𝐿 𝜕𝐹 𝜕 𝜕𝐹
= { + 𝜕𝑞 + 𝜕𝑡 (𝜕𝑞 ̇ )}
𝑑𝑡 𝜕𝑞 ̇ 𝑗 𝑗 𝑗

𝑑 𝜕𝐿 𝜕𝐹 𝜕𝐹
= { + 𝜕𝑞 } , 𝑆𝑖𝑛𝑐𝑒 =0
𝑑𝑡 𝜕𝑞 ̇ 𝑗 𝑗 𝜕𝑞𝑗̇
 𝑑 𝜕𝐿 𝑑 𝜕𝐹
= ( ) + 𝑑𝑡 (𝜕𝑞 )
𝑑𝑡 𝜕𝑞 ̇ 𝑗 𝑗

𝜕𝐿 𝑑 𝜕𝐹
=
𝜕𝑞𝑗
+ ( ) 𝑢𝑠𝑖𝑛𝑔 (1)
𝑑𝑡 𝜕𝑞 𝑗

𝜕𝐿 𝑑 𝜕𝐹 𝜕 𝜕𝐹
= + ∑𝑛𝑘=1 (𝜕𝑞 ) 𝑞𝑘̇ + 𝜕𝑡 (𝜕𝑞 )
𝜕𝑞𝑗 𝑑𝑞𝑘 𝑗 𝑗

𝜕 𝜕𝐹 𝜕𝐹
= [𝐿 + ∑𝑛𝑘=1 𝑞𝑘̇ + ]
𝜕𝑞𝑗 𝜕𝑞𝑘 𝜕𝑡

𝜕 𝜕𝐹 𝜕𝐹
= [𝐿 + ∑𝑛𝑘=1 𝑞𝑘̇ + ]
𝜕𝑞𝑗 𝜕𝑞𝑘 𝜕𝑡

𝜕 𝜕𝐹 𝜕𝐿′
= [𝐿 + ] =
𝜕𝑞𝑗 𝜕𝑡 𝜕𝑞𝑗
𝑑 𝜕𝐿′ 𝜕𝐿′
( )− =0
𝑑𝑡 𝜕𝑞𝑗̇ 𝜕𝑞𝑗
𝑑𝐹
Thus 𝐿′ = 𝐿 + satisfies the Lagrange’s equation .
𝑑𝑡

4. Prove that , the lagrangian of a system is not explicity dependent on time , then
Hamiltonian is Conserved.(2011,09)
𝑨𝒏𝒔, −We know, Hamiltonian is a function of generalised momenta 𝑝𝑘 and time , i.e.
𝐻 = 𝐻(𝑞1 , 𝑞2 … . . 𝑞𝑘 , . . 𝑞𝑛 ; 𝑝1 , 𝑝2 , … 𝑝𝑘 … 𝑝𝑛 , 𝑡)
𝑑𝐻 𝜕𝐻 𝜕𝐻 𝜕𝐻
∴ =∑ 𝑞̇ 𝑘 + ∑ 𝑝̇𝑘 +
𝑑𝑡 𝜕𝑞𝑘 𝜕𝑝𝑘 𝜕𝑡
𝑘 𝑘

𝑑𝐻 𝜕𝐻
∴ = − ∑ 𝑝̇ 𝑘 𝑞̇ 𝑘 + ∑ 𝑞̇ 𝑘 𝑝̇ 𝑘 + (𝑈𝑠𝑖𝑛𝑔 𝑐𝑎𝑛𝑜𝑛𝑖𝑐𝑎𝑙 𝑒𝑞𝑢𝑎𝑡𝑖𝑜𝑛)
𝑑𝑡 𝜕𝑡
𝑘 𝑘

𝑑𝐻 𝜕𝐻
=
𝑑𝑡 𝜕𝑡
𝜕𝐻 𝜕𝐿
𝐵𝑢𝑡 =−
𝜕𝑡 𝜕𝑡
𝑑𝐻 𝜕𝐿
∴ =−
𝑑𝑡 𝜕𝑡
If 𝐿 is not an explicit function of time,
𝜕𝐿 𝑑𝐻
i.e. = 0, We have = 0 ; ∴ 𝐻 = 𝐶𝑜𝑛𝑠𝑡.
𝜕𝑡 𝑑𝑡

This shows if 𝐿 is not an explicit function of time, The Hamiltonian .It is Constant of
motion i.e. Conserved.
 5. For the Hamiltonian = 𝒑𝟏 𝒒𝟏 − 𝒑𝟐 𝒒𝟐 − 𝒂𝒒𝟐𝟏 − 𝒃𝒒𝟐𝟐 , Solve the Hamilton’s equation of
𝒑 +𝒃𝒒
motion and Prove that 𝟐 𝒒 𝟐 = 𝑪𝒐𝒏𝒔𝒕𝒂𝒏𝒕 and 𝒒𝟏 𝒒𝟐 = 𝑪𝒐𝒏𝒔𝒕. 𝒂, 𝒃 are Constants.
𝟏
𝒑𝟏 , 𝒑𝟐 , 𝒒𝟏 , 𝒒𝟐 are generalised Co-ordinates.(2005)
𝑨𝒏𝒔. − 𝐻 = 𝑝1 𝑞1 − 𝑝2 𝑞2 − 𝑎𝑞12 − 𝑏𝑞22
𝜕𝐻
∴ 𝑞̇ 1 = = 𝑞1 ⋯ ⋯ (𝑖)
𝜕𝑝1
𝜕𝐻
& 𝑞̇ 2 = = −𝑞2 ⋯ ⋯ (𝑖𝑖)
𝜕𝑝2
𝜕𝐻
∴ 𝑝̇1 = − = −(𝑝1 − 2𝑎𝑞1 ) ⋯ ⋯ (𝑖𝑖𝑖)
𝜕𝑞1
𝜕𝐻
& 𝑝̇2 = − = −(−𝑝2 − 2𝑏𝑞2 ) ⋯ ⋯ (𝑖𝑣)
𝜕𝑞2
From equation (𝑖) & (𝑖𝑖)
𝑞̇ 1 = 𝑞1 & 𝑞2 = 𝑞̇ 2
Multiplied both sides we have
−𝑞̇ 1 𝑞2 = 𝑞1 𝑞̇ 2
∴ 𝑞1 𝑞̇ 2 + 𝑞̇ 1 𝑞2 = 0
𝑑
∴ (𝑞 𝑞 ) = 0
𝑑𝑡 1 2
∴ 𝑞1 𝑞2 = 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡(𝐶, 𝑆𝑎𝑦)
(𝑝2 + 𝑏𝑞2 ) (𝑝2 + 𝑏𝑞2 )𝑞2 1
𝐿𝑒𝑡 𝑥 = = = (𝑝2 + 𝑏𝑞2 )𝑞2
𝑞1 𝑞1 𝑞2 𝐶
1
∴𝑥= (𝑝2 𝑞2 + 𝑏𝑞22 )
𝐶
𝑑𝑥 1
= (𝑝̇ 𝑞 + 𝑝2 𝑞̇ 2 + 2𝑏𝑞2 𝑞̇ 2 )
𝑑𝑡 𝐶 2 2
1
= [−(−𝑝2 − 2𝑏𝑞2 )𝑞2 + 𝑝2 (−𝑞2 ) − 2𝑏𝑞22 ]
𝐶

=0
∴ 𝑥 = 𝐶𝑜𝑛𝑠𝑡.
𝑝2 + 𝑏𝑞2
∴ = 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡.
𝑞1
BASIC HAMILTONIAL PROBLEMS

6. For a Simple Pendulum , Write down the Hamiltanian , and obtain Hamiltonian’s
equation of motion. (2011)

𝑨𝒏𝒔. −Let us consider 𝜃 be the angular displacement of the simple pendulum

From the equilibrium position and 𝑙 be the length of the pendulum and 𝑚 be
the mass of the bob.
1
Kinetic Energy of the bob 𝑇 = 𝑚𝑙 2 𝜃 2
2

Potential Energy 𝑉 = 𝑚𝑔𝑙(1 − cos 𝜃)

1
Lagrangian 𝐿 = 𝑇 − 𝑉 = 𝑚𝑙 2 𝜃 2 − 𝑚𝑔𝑙(1 −
2
cos 𝜃)
𝜕𝐿
∴ 𝑃𝜃 = = 𝑚𝑙 2 𝜃̇ ⋯ ⋯ (1)
𝜕𝜃̇
𝑃𝜃
𝜃̇ =
𝑚𝑙 2
Now Hamiltonian,

𝐻 = ∑ 𝑝𝑘 𝑞̇ 𝑘 − 𝐿
𝑘

𝑃𝜃 1 𝑃𝜃 2
= 𝑝𝜃 . 2
− [ 𝑚𝑙 2 ( ) − 𝑚𝑔𝑙(1 − cos 𝜃)]
𝑚𝑙 2 𝑚𝑙 2

𝑃𝜃2 𝑃𝜃2
= 2− + 𝑚𝑔𝑙(1 − 𝑐𝑜𝑠𝜃)
𝑚𝑙 2𝑚𝑙 2
𝑃𝜃2
= + 𝑚𝑔𝑙(1 − 𝑐𝑜𝑠𝜃)
2𝑚𝑙 2

Hence Hamiltonian’s equation are

𝜕𝐻 𝑝𝜃
∴ 𝜃̇ = = 2
𝜕𝑝𝜃 𝑚𝑙
𝜕𝐻
𝑎𝑛𝑑 𝑝̇𝜃 = − = −𝑚𝑔𝑙 sin 𝜃
𝜕𝜃
𝑜𝑟, 𝑚𝑙 2 𝜃̈ + 𝑚𝑔𝑙 sin 𝜃 = 0 (𝑃𝑢𝑡𝑡𝑖𝑛𝑔 𝑡ℎ𝑒 𝑣𝑎𝑙𝑢𝑒 𝑜𝑓 𝑝̇ 𝜃 )
 𝑔
∴ 𝜃̈ + 𝑠𝑖𝑛𝜃 = 0
𝑙
Nor very small limit of oscillation 𝑠𝑖𝑛𝜃 ≈ 𝜃
𝑔
Hence, 𝜃̈ + 𝜃 = 0
𝑙

This is the equation of motion.

7. For a spherical pendulum find the equation of motion.

A spherical pendulum is essentially a simple pendulum which is free to move through
the entire space about the point of suspension. The bob of such a pendulum moves on the
surface of a sphere whose radius is equal to the length of the pendulum.
It is obvious that spherical polar coordinates are most suitable to locate the position of
the bob of the pendulum. Moreover, since length is a contant, only 𝜃 and 𝜙 are need be
specified.

The velocity of the bob is given by

𝑣⃗ = 𝑟̇ 𝑟̂ + 𝑟𝜃̇𝜃̂ + 𝑟𝑠𝑖𝑛 𝜃𝜙̂
= 𝑙𝜃̇𝜃̂ + 𝑙𝑠𝑖𝑛 𝜃𝜙̂
In spherical polar Co-ordinates system K.E. of the particle-
1
𝑇 = 𝑚{𝑙 2 𝜃̇ 2 + 𝑙 2 sin2 𝜃 𝜙̇ 2 }
2
1
𝐿 = 𝑇 − 𝑉 = 𝑚{𝑙 2 𝜃̇ 2 + 𝑙 2 sin2 𝜃 𝜙̇ 2 } − (−𝑚𝑔𝑙𝑐𝑜𝑠𝜃)
2
 1
= 𝑚{𝑙 2 𝜃̇ 2 + 𝑙 2 sin2 𝜃 𝜙̇ 2 } + 𝑚𝑔𝑙𝑐𝑜𝑠𝜃
2
Now
𝜕𝐿
= 𝑚𝑙 2 𝜃̇
𝜕𝜃̇
𝑑 𝜕𝐿
( ) = 𝑚𝑙 2 𝜃̈
𝑑𝑡 𝜕𝜃̇
𝜕𝐿
= 𝑚𝑙 2 𝑠𝑖𝑛𝜃 cos 𝜃 𝜙̇ 2 − 𝑚𝑔𝑙 𝑠𝑖𝑛𝜃
𝜕𝜃
Hence the equation of motion in 𝜃,

𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃
𝑚𝑙 2 𝜃̈ − 𝑚𝑙 2 𝑠𝑖𝑛𝜃 cos 𝜃 𝜙̇ 2 + 𝑚𝑔𝑙 𝑠𝑖𝑛𝜃 = 0 … … … (1)

𝜕𝐿
= 𝑚𝑙 2 sin2 𝜃 𝜙̇
𝜕𝜙̇
𝑑 𝜕𝐿
( ) = 𝑚𝑙 2 sin2 𝜃 𝜙̈
𝑑𝑡 𝜕𝜃̇
𝜕𝐿
= 𝑚𝑙 2 sin2 𝜃 𝜙̇
𝜕𝜙̇

𝑑 𝜕𝐿
( ) = 𝑚𝑙 2 sin2 𝜃 𝜙̈
𝑑𝑡 𝜕𝜙̇

𝜕𝐿
=0
𝜕𝜙

Hence the equation of motion in 𝜙,

𝑑 𝜕𝐿
( )=0
𝑑𝑡 𝜕𝜙̇
Thus ,
𝑝𝜙 = 𝑚𝑙 2 sin2 𝜃 𝜙̇ = 𝐾(𝑐𝑜𝑛𝑠𝑡𝑎𝑛𝑡)

From equation (1),

 𝑝𝜙 2
𝑚𝑙 2 𝜃̈ − 𝑚𝑙 2 𝑠𝑖𝑛𝜃 cos 𝜃 ( 2 2 ) + 𝑚𝑔𝑙 𝑠𝑖𝑛𝜃 = 0
𝑚𝑙 sin 𝜃
𝑝𝜙2
𝑚𝑙 𝜃̈ −
2
+ 𝑚𝑔𝑙 𝑠𝑖𝑛𝜃 = 0
𝑚𝑙 2 sin3 𝜃

𝟏 𝟏
8. The Lagrangian of a particle is given by 𝑳 = 𝟐 𝒎𝒙̇ 𝟐 − 𝟐 𝒌𝒙𝟐 . Find out the Conjugate
momentum and the Hamiltonian. (2008)
1 1
𝑨𝒏𝒔. − 𝐿 = 𝑚𝑥̇ 2 − 𝑘𝑥 2
2 2

∴ Generalized Co-ordinates 𝑥
∴ Generalized momentum Corresponding to this Co-ordinates
𝜕𝐿 𝜕 1 1
𝑝𝑥 = = ( 𝑚𝑥̇ 2 − 𝑘𝑥 2 ) = 𝑚𝑥̇
𝜕𝑥̇ 𝜕𝑥̇ 2 2
𝑝𝑥
𝑥̇ =
𝑚
Hamiltonian 𝐻 = ∑𝑘 𝑝𝑘 𝑞̇ 𝑘 − 𝐿
1 1
= 𝑝𝑘 𝑥̇ − 𝑚𝑥̇ 2 + 𝑘𝑥 2
2 2
𝑝𝑘 1 𝑝𝑥2 1 2 𝑝𝑥
∴ 𝐻 = 𝑝𝑘 . − 𝑚 2 + 𝑘𝑥 [𝑆𝑖𝑛𝑐𝑒, 𝑥̇ = ]
𝑚 2 𝑚 2 𝑚
𝑝𝑥2 𝑝𝑥2 1 2
= − + 𝑘𝑥
𝑚 2𝑚 2
𝑝𝑥2 1 2
= + 𝑘𝑥
2𝑚 2
9. Describe the Hamiltonian and Hamilton’s equation for an ideal Spring-mass
arrangement.
1
𝐴𝑛𝑠. − K.E. of the System 𝑇 = 𝑚𝑥̇ 2
2
P.E. of the System (𝑉) = ∫ 𝐹. 𝑑𝑥

𝑉 = − ∫ 𝐹. 𝑑𝑥

= − ∫(−𝑘𝑥)𝑑𝑥
1
= 𝑘𝑥 2
2

∴ 𝐻 = ∑ 𝑝𝑘 𝑞̇ 𝑘 − 𝐿
𝑘
= 𝑝̇𝑥 𝑥̇
1 1
− [ 𝑚𝑥̇ 2 − 𝑘𝑥 2 ]
2 2
1 1
𝐿 = 𝑚𝑥̇ 2 − 𝑘𝑥 2
2 2
𝜕𝐿 ̇ 𝑝𝑥
𝑝𝑥 = = 𝑚𝑥̇ = 𝑥 =
𝜕𝑥̇ 𝑚
1 1
𝐻 = 𝑚𝑥̇ 2 − 𝑚𝑥̇ 2 + 𝑘𝑥 2
2 2
𝑝𝑥2 1 2
∴𝐻= + 𝑘𝑥
2𝑚 2
Hamilton’s equation are
𝜕𝐻 𝑝𝑥
𝑥̇ = = = 𝑝𝑥 = 𝑚𝑥̇ = 𝑝̇𝑥 = 𝑚𝑥̈
𝜕𝑝𝑥 𝑚
𝜕𝐻
& 𝑝̇𝑥 = − = −𝑘𝑥 ⋯ ⋯ (𝑖)
𝜕𝑥
∴ 𝑚𝑥̈ + 𝑘𝑥 = 0 [𝑆𝑖𝑛𝑐𝑒 𝑝𝑥 = 𝑚𝑥̇ ]
𝑘
∴ 𝑥̈ + 𝑥 = 0 ⋯ ⋯ (𝑖𝑖)
𝑚
This is the equation of motion of Spring mass system.

𝒑𝟐 𝟏
10. The Hamiltonian of the system is given by 𝑯 = 𝟐𝒎 + 𝟐 𝒌𝒙𝟐 + 𝜶𝒙𝟒 obtain the
equation of motion.
𝑝2 1
𝑨𝒏𝒔. − 𝐻 = + 𝑘𝑥 2 + 𝛼𝑥 4
2𝑚 2
 𝜕𝐻 𝑝
∴ 𝑥̇ = = ⋯ ⋯ (𝑖)
𝜕𝑝 𝑚
𝜕𝐻
𝐴𝑛𝑑 𝑝̇ = − = −𝑘𝑥 − 4𝛼𝑥 3
𝜕𝑥
∴ 𝑚𝑥̈ = −𝑘𝑥 − 𝛼𝑥 3
𝑘 4𝛼 3
∴ 𝑥̈ + 𝑥+ 𝑥 =0
𝑚 𝑚

BASIC LAGRANGIAN

11. Set up the Lagrangian for a Simple Pendulum and obtain the equation describing
the motion.
𝑨𝒏𝒔. − Let us consider 𝜃 be the angular displacement of the simple pendulum
From the equilibrium position and 𝑙 be the length of the pendulum and 𝑚 be
the mass of the bob.
1
Kinetic Energy of the bob 𝑇 = 𝑚𝑙 2 𝜃 2
2

Potential Energy 𝑉 = 𝑚𝑔𝑙(1 − cos 𝜃)

Lagrangian
1
𝐿 = 𝑇 − 𝑉 = 𝑚𝑙 2 𝜃 2 − 𝑚𝑔𝑙(1 − cos 𝜃)
2
So, The equation of motion is-
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃
𝑑
𝑂𝑟, (𝑚𝑙 2 𝜃̇) + 𝑚𝑔𝑙 sin 𝜃 = 0
𝑑𝑡
𝑂𝑟, 𝑚𝑙 2 𝜃̈ + 𝑚𝑔𝑙 sin 𝜃 = 0
𝑔
𝑂𝑟, 𝜃̈ + sin 𝜃 = 0
𝑙
This represents the equation of motion of a simple pendulum.
For small angular displacement sin 𝜃 ≈ 𝜃
𝑔
∴ 𝜃̈ + 𝜃=0
𝑙
 12. Find out the equation of motion of a projectile –
1
𝑇 = 𝑚(𝑥̇ 2 + 𝑦̇ 2 )
2

𝑉 = 𝑚𝑔𝑦
𝐿 =𝑇−𝑉
1 1
= 𝑚𝑥̇ 2 + 𝑚𝑦̇ 2 − 𝑚𝑔𝑦
2 2

Equation in motion in 𝑥 −
𝑑 𝜕𝐿 𝜕𝐿
( ) − 𝜕𝑥 = 0
𝑑𝑡 𝜕𝑥̇
𝑑
𝑂𝑟, (𝑚𝑥̇ ) = 0
𝑑𝑡

∴ 𝑥̈ = 0

13. Describe the Lagrangian equation for an ideal Spring-mass arrangement.

1
𝐴𝑛𝑠. − K.E. of the System 𝑇 = 𝑚𝑥̇ 2
2

P.E. of the System (𝑉) = ∫ 𝐹. 𝑑𝑥

𝑉 = − ∫ 𝐹. 𝑑𝑥

= − ∫(−𝑘𝑥)𝑑𝑥
1
= 𝑘𝑥 2
2
1 1
𝐿 = 𝑚𝑥̇ 2 − 𝑘𝑥 2
2 2
Now,
𝜕𝐿
= 𝑚𝑥̇
𝜕𝑥̇
𝑑 𝜕𝐿
( ) = 𝑚𝑥̈
𝑑𝑡 𝜕𝑥̇
𝜕𝐿
= −𝑘𝑥
𝜕𝑥
Equation in motion in 𝑥 −,
 𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝑥̇ 𝜕𝑥
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝑥̇ 𝜕𝑥
∴ 𝑚𝑥̈ + 𝑘𝑥 = 0
𝑘
∴ 𝑥̈ + 𝑥 = 0 ⋯ ⋯ (𝑖𝑖)
𝑚
This is the equation of motion of Spring mass system.
Additional problems
14. Construct equation of motion of a Simple pendulum whose point of Suspension
moves vertical according to the law, 𝒚𝟏 = 𝒂 𝒄𝒐𝒔 𝝎𝒕. The mass of the bob is 𝒎 and
length is 𝒍. (2005)

𝑨𝒏𝒔. − Let us, Consider the instantaneous Co-ordinates of 𝑚 are (𝑥, 𝑦1 )

∴ 𝑥1 = 𝑙 sin 𝜃
𝑦1 = 𝑎 cos 𝜔𝑡 + 𝑙 cos 𝜃
𝑥̇ 1 = 𝑙 cos 𝜃 𝜃̇
𝑦̇1 = −𝑎𝜔 sin 𝜔𝑡 − 𝑙 sin 𝜃𝜃̇
Kinetic energy of the bob
1
∴ 𝑇 = 𝑚(𝑥̇ 12 + 𝑦̇ 2 )
2

1
= 𝑚(𝑙 2 cos 2 𝜃𝜃̇ 2 + 𝑎2 𝜔2 sin2 𝜔𝑡
2
+ 𝑙 2 sin2 𝜃𝜃̇ 2
+ 2𝑎𝜔𝑙 sin 𝜔𝑡 sin 𝜃𝜃̇)
The potential energy of the bob
𝑉 = −𝑚𝑔𝑦1 = −𝑚𝑔(𝑎 cos 𝜔𝑡 + 𝑙 cos 𝜃)
The Lagrangian of the pendulum
𝐿=𝑇−𝑉
1
= 𝑚[𝑙 2 cos 2 𝜃𝜃̇ 2 + 𝑎2 𝜔2 sin2 𝜔𝑡 + 𝑙 2 sin2 𝜃𝜃̇ 2 + 2𝑎𝜔𝑙 sin 𝜔𝑡 sin 𝜃𝜃̇]
2
+ 𝑚𝑔(𝑎 cos 𝜔𝑡 + 𝑙 cos 𝜃)
1
= 𝑚[𝑙 2 𝜃̇ 2 + 𝑎2 𝜔2 sin2 𝜔𝑡 + 2𝑎𝜔𝑙 sin 𝜔𝑡 sin 𝜃𝜃 + 𝑚𝑔(𝑎 coṡ 𝜔𝑡 + 𝑙 cos 𝜃)]
2
The Lagrangian equation in 𝜃 Co-ordinates
 𝑑 𝜕𝐿 𝜕𝐿
( )− = 0 ⋯ ⋯ (𝑖)
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃
𝜕𝐿
= 𝑚𝑙 2 𝜃̇ + 𝑚𝑎𝜔𝑙 sin 𝜔𝑡 sin 𝜃
𝜕𝜃̇
𝑑 𝜕𝐿
( ) = 𝑚𝑙 2 𝜃̈ + 𝑚𝑎𝜔𝑙 sin 𝜔𝑡 cos 𝜃𝜃̇ + 𝑚𝑎𝜔2 𝑙 cos 𝜔𝑡 sin 𝜃
𝑑𝑡 𝜕𝜃̇
𝜕𝐿 1
And = 𝑚[2𝑎𝜔𝑙 sin 𝜔𝑡 cos 𝜃𝜃̇] + 𝑚𝑔𝑙(− sin 𝜃)
𝜕𝜃 2

= 𝑚𝑎𝜔𝑙 sin 𝜔𝑡 cos 𝜃 𝜃̇ − 𝑚𝑔𝑙 sin 𝜃

Putting this values in eq.(𝑖) we have,
𝑚𝑙 2 𝜃̈ + 𝑚𝑎𝜔𝑙 sin 𝜔𝑡 cos 𝜃𝜃̇ + 𝑚𝑎𝜔2 𝑙 cos 𝜔𝑡 sin 𝜃 − 𝑚𝑎𝜔𝑙 sin 𝜔𝑡 cos 𝜃 𝜃̇ + 𝑚𝑔𝑙 sin 𝜃 = 0
𝑂𝑟, 𝑚𝑙 2 𝜃̈ + 𝑚𝑎𝜔2 𝑙 cos 𝜔𝑡 sin 𝜃 + 𝑚𝑔𝑙 sin 𝜃 = 0
𝑎𝜔2 𝑔
𝑂𝑟, 𝜃̈ + cos 𝜔𝑡 sin 𝜃 + sin 𝜃 = 0
𝑙 𝑙
𝑎𝜔2 𝑔
𝑂𝑟, 𝜃̈ + ( cos 𝜔𝑡 + ) 𝜃 = 0
𝑙 𝑙

15. What is the Lagrangian of a free particle in cylindrical polar Co-ordinates? Find the
Hamiltonian.(2006)
𝑨𝒏𝒔. − If 𝑚 be the mass of the free particle.
1
Then 𝐿 = 𝑚(𝑟̇ 2 + 𝑟 2 𝜃̇ 2 + 𝑧̇ 2 ) − 𝑉(𝑟, 𝜃, 𝑧)
2

But for free particle 𝑉(𝑟, 𝜃, 𝑧) = 0

1
∴ 𝐿 = 𝑚(𝑟̇ 2 + 𝑟 2 𝜃̇ 2 + 𝑧̇ 2 )
2
∴ 𝐻𝑎𝑚𝑖𝑙𝑡𝑜𝑛𝑖𝑎𝑛 ,

𝐻 = ∑ 𝑝𝑘 𝑞̇ 𝑘 − 𝐿 , 𝑊ℎ𝑒𝑟𝑒 𝐾 = 𝑟, 𝜃, 𝑧
𝑘

1
∴ 𝐻 = 𝑝𝑟 𝑟̇ + 𝑝𝜃 𝜃̇ + 𝑝𝑧 𝑧̇ − 𝑚(𝑟̇ 2 + 𝑟 2 𝜃̇ 2 + 𝑧̇ 2 )
2
𝜕𝐿 𝑝𝑟
𝑁𝑜𝑤, 𝑝𝑟 = = 𝑚𝑟̇ = 𝑟̇ = ;
𝜕𝑟̇ 𝑚
𝜕𝐿 𝑝𝜃
𝑝𝜃 = = 𝑚𝑟 2 𝜃̇ = 𝜃̇ =
𝜕𝜃̇ 𝑚𝑟 2
𝜕𝐿 𝑝𝑧
𝑎𝑛𝑑 𝑝𝑧 = = 𝑚𝑧̇ = 𝑧̇ =
𝜕𝑧̇ 𝑚
 𝑝𝑟 𝑝𝜃 𝑝𝑧 1 𝑝𝑟2 2
𝑝𝜃2 𝑝𝑧2
∴ 𝐻 = 𝑝𝑟 . + 𝑝𝜃 . + 𝑝𝑧 . − 𝑚 ( 2 + 𝑟 . 2 4 + 2 )
𝑚 𝑚𝑟 2 𝑚 2 𝑚 𝑚 𝑟 𝑚
1 2 1 𝑝𝜃2 1 2
∴𝐻= 𝑝𝑟 + + 𝑝
2𝑚 2𝑚 𝑟 2 2𝑚 𝑧
1 𝑝𝜃2
= (𝑝𝑟 + 2 + 𝑝𝑧2 )
2
2𝑚 𝑟
16. A particle is moving under a Potential 𝑽(𝒓, 𝜽, 𝝋) where 𝒓, 𝜽, 𝝋 are Spherical polar
Co-ordinates of the Particle at any instant. Obtain Hamilton’s equation of
motion.(2003)
𝐴𝑛𝑠. −In spherical polar Co-ordinates system K.E. of the particle-
1
𝑇 = 𝑚{𝑟̇ 2 + 𝑟 2 𝜃̇ 2 + 𝑟 2 sin2 𝜃 𝜑̇ 2 }
2
1
𝐿 = 𝑇 − 𝑉 = 𝑚{𝑟̇ 2 + 𝑟 2 𝜃̇ 2 + 𝑟 2 sin2 𝜃 𝜑̇ 2 } − 𝑉(𝑟, 𝜃, 𝜑)
2
𝜕𝐿
∴ 𝑝𝑟 = = 𝑚𝑟̇
𝜕𝑟
𝜕𝐿
𝑝𝜃 = = 𝑚𝑟 2 𝜃̇
𝜕𝜃̇
𝜕𝐿
𝑝𝜑 = = 𝑚𝑟 2 sin2 𝜃𝜑̇
𝜕𝜑̇
Hamiltonian of the particle
𝐻 =𝑇+𝑉
𝑚 𝑝𝑟2 2
𝑝𝜃2 2 2
𝑝𝜑2
= { 2 + 𝑟 . 2 4 + 𝑟 sin 𝜃 2 4 4 } + 𝑉(𝑟, 𝜃, 𝜑)
2 𝑚 𝑚 𝑟 𝑚 𝑟 sin 𝜃
1 2
𝑝𝜃2 𝑝𝜑2
= {𝑝 + + } + 𝑉(𝑟, 𝜃, 𝜑)
2𝑚 𝑟 𝑟 2 𝑟 2 sin2 𝜑
1 2
𝑝𝜃2 𝑝𝜑2
𝐻= [𝑝 + + ] + 𝑉(𝑟, 𝜃, 𝜑)
2𝑚 𝑟 𝑟 2 𝑟 2 sin2 𝜃
Hamiltonian equation are
𝜕𝐻 𝑝𝑟
∴ 𝑟̇ = =
𝜕𝑝𝑟 𝑚
2
𝜕𝐻 1 2𝑝2 2 𝑝𝜑 𝜕𝑉
And 𝑝̇𝑟 = −
𝜕𝑟
= ( 𝜃 + 𝑟 3 sin2 𝜃) − 𝜕𝑟
2𝑚 𝑟 3

𝜕𝐻 𝑝𝜃
∴ 𝜃̇ = =
𝜕𝑝𝜃 𝑚𝑟 2
 𝜕𝐻 𝑝𝜑2 2 cos 𝜃 𝜕𝑉
& 𝑝̇ 𝜃 = − = −
𝜕𝜃 2𝑚𝑟 2 sin3 𝜃 𝜕𝜃
𝑝𝜑2 cos 𝜃 𝜕𝑉
= −
𝑚𝑟 2 sin3 𝜃 𝜕𝜃
𝜕𝐻 𝑝𝜑
∴ 𝜑̇ = =
𝜕𝑝𝑞 𝑚𝑟 2 sin2 𝜃
𝜕𝐻 𝜕𝑉
𝐴𝑛𝑑 𝑝̇𝜑 = − =−
𝜕𝜑 𝜕𝜑

17. A particle of mass 𝒎 is constrained to move on the surface of a cylinder. The

particle is subjected to a force directed towards the origin( midpoint of the
cylinder) and is proportional to the distance of the particle from the origin.
Construct the Lagrangian , Hamiltonian and Hamilton’s equation of motion. (2002)
𝐴𝑛𝑠. − Force acting on the particle

𝐹⃗ = −𝑘𝑟⃗
So Potential energy
1
𝑉 = − ∫ 𝐹⃗ . 𝑑𝑟⃗ = 𝑟 2 𝑘
2
1
= (𝑥 2 + 𝑦 2 + 𝑧 2 )
2
𝑁𝑜𝑤 𝑥 = 𝜌 cos 𝜑 , 𝑦 = 𝜌 sin 𝜑
1 1
𝑉 = 𝑘(𝜌2 + 𝑧 2 ) = 𝑘(𝑎2 + 𝑧 2 )
2 2

1
P.E. = 𝑘(𝑎2 + 𝑧 2 )
2
1
K.E. 𝑇 = 𝑚(𝑥̇ 2 + 𝑦̇ 2 + 𝑧̇ 2 )
2

∴ 𝑥 = 𝑎 cos 𝜑 , 𝑦 = 𝑎 sin 𝜑
∴ 𝑥̇ 2 + 𝑦̇ 2 = 𝑎2 sin2 𝜑 . 𝜑̇ 2 + 𝑎2 cos 2 𝜑 𝜑̇ 2 = 𝑎2 𝜑̇ 2
1
∴ 𝑇 = 𝑚(𝑎2 𝜑̇ 2 + 𝑧̇ 2 )
2
Lagrangian-
1 1
𝐿 = 𝑇 − 𝑉 = 𝑚(𝑎2 𝜑̇ 2 + 𝑧̇ 2 ) − 𝑘(𝑎2 + 𝑧̇ 2 )
2 2
Hamiltonian
 1 1
𝐻 = 𝑇 + 𝑉 = 𝑚(𝑎2 𝜑̇ 2 + 𝑧̇ 2 ) + 𝑘(𝑎2 + 𝑧 2 )
2 2
Now,
𝜕𝐿 𝑝𝜑
𝑝𝜑 = = 𝑚𝑎2 𝜑̇ = 𝜑̇ =
𝜕𝜑̇ 𝑚𝑎2
𝜕𝐿 𝑝𝑧
𝑝𝑧 = = 𝑚𝑧̇ = 𝑧̇ =
𝜕𝑧̇ 𝑚
1 2
𝑝𝜑2 𝑝𝑧2 1
∴ 𝐻 = 𝑚 [𝑎 2 4
+ 2
] + 𝑘(𝑎2 + 𝑧 2 )
2 𝑚 𝑎 𝑚 2
1 𝑝𝜑2 𝑝𝑧2 1
= 2
+ + 𝑘(𝑎2 + 𝑧 2 )
2 𝑚𝑎 2𝑚 2
Hence the Hamilton’s equation are-
𝜕𝐻 𝑝𝑧
𝑧̇ = = 𝑂𝑟, = 𝑚𝑧̇
𝜕𝑝𝑧 𝑚
𝜕𝐻
& 𝑝̇𝑧 = − = −𝑘𝑧
𝜕𝑧
𝜕𝐻 𝑝𝜑
∴ 𝜑̇ = = 2
𝑂𝑟, 𝑝𝜑 = 𝑚𝑎2 𝜑̇
𝜕𝑝𝜑 𝑚𝑎
𝜕𝐻
& 𝑝̇𝜑 = − = 0 𝑂𝑟, 𝑝𝜑 = 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡.
𝜕𝜑

18. A simple pendulum of mass 𝒎 whose point of support moves uniformly on a

vertical circle with constant angular frequency . Set up the Lagrangian and find the
equation of motion.

𝑨𝒏𝒔. − 𝑥 = 𝑎 cos 𝜔𝑡 + 𝑙 sin 𝜃

𝑥̇ = −𝑎𝜔 sin 𝜔𝑡 + 𝑙 cos 𝜃 𝜃̇
𝑦 = −𝑎 sin 𝜔𝑡 + 𝑙 cos 𝜃
∴ 𝑦̇ = −𝑎𝜔 cos 𝜔𝑡 − 𝑙 sin 𝜃 𝜃̇
1
𝑇 = 𝑚(𝑥̇ 2 + 𝑦̇ 2 )
2
1
= 𝑚[𝑎2 𝜔2 sin2 𝜔𝑡 + 𝑙 2 cos 2 𝜃 𝜃̇ 2
2
− 2𝑎𝜔𝑙 sin 𝜔𝑡 cos 𝜃𝜃̇ + 𝑎2 𝜔2 cos 2 𝜔𝑡
+ 𝑙 2 sin2 𝜃 𝜃̇ 2 + 2𝑎𝜔𝑙 sin 𝜃 cos 𝜔𝑡𝜃̇]
 1
= 𝑚[𝑎2 𝜔2 + 𝑙 2 𝜃̇ 2 − 2𝑎𝜔𝑙(sin 𝜔𝑡 cos 𝜃 − sin 𝜃 cos 𝜔𝑡)𝜃̈]
2
𝑉 = −𝑚𝑔𝑦
= +𝑚𝑔𝑎 sin 𝜔𝑡 − 𝑚𝑔𝑙 cos 𝜃
∴𝐿 =𝑇−𝑉
Equation of motion →
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃
𝑑
𝑂𝑟, [𝑚𝑙 2 𝜃̇ − 𝑎𝜔𝑙𝑚(sin 𝜔𝑡 cos 𝜃 − sin 𝜃 cos 𝜔𝑡)]
𝑑𝑡
− [𝑚𝑎𝜔𝑙 sin 𝜔𝑡 sin 𝜃𝜃̇ + 𝑚𝑎𝜔𝑙 cos 𝜃 cos 𝜔𝑡𝜃̇ + 𝑚𝑔𝑙 sin 𝜃] = 0
𝑑
𝑂𝑟, [𝑚𝑙 2 𝜃̇ − 𝑚𝑎𝜔𝑙 sin(𝜔𝑡 − 𝜃)] − 𝑚𝑎𝜔𝑙 cos(𝜔𝑡 − 𝜃)𝜃̇ − 𝑚𝑔𝑙 sin 𝜃 = 0
𝑑𝑡
𝑂𝑟, 𝑚𝑙 2 𝜃̈ − 𝑚𝑎𝜔2 𝑙 cos(𝜔𝑡 − 𝜃)
+ 𝑚𝑎𝜔𝑙 cos(𝜔𝑡 − 𝜃)𝜃̇ − 𝑚𝑎𝜔𝑙 cos(𝜔𝑡 − 𝜃)𝜃̇ − 𝑚𝑔𝑙 sin 𝜃 = 0
𝑂𝑟, 𝑙 2 𝜃̈ − 𝑎𝜔2 𝑙 cos(𝜔𝑡 − 𝜃) − 𝑔𝑙 sin 𝜃 = 0
𝑂𝑟, 𝑙𝜃̈ − 𝑎𝜔2 cos(𝜔𝑡 − 𝜃) − 𝑔 sin 𝜃 = 0

19. A simple pendulum of mass 𝒎 whose point of support moves uniformly on a

horizontal circle. With angular frequency 𝝎. Find equation of motion.
1
𝑨𝒏𝒔. − 𝑇 = 𝑚(𝑥 2 𝜑̇ 2 )
2
1
= 𝑚𝑙 2 sin2 𝜃 𝜑̇ 2
2

𝑉 = −𝑚𝑔𝑙 cos 𝜃
𝐿 =𝑇−𝑉
1
= 𝑚𝑙 2 sin2 𝜃 𝜑̇ 2 + −𝑚𝑔𝑙 cos 𝜃
2

∴ Lagrangian equation of motion in 𝜃 −

𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃
𝑑
𝑂𝑟, (0) − 𝑚𝑙 2 sin 𝜃 cos 𝜃 𝜑̇ 2 + 𝑚𝑔𝑙 sin 𝜃 = 0
𝑑𝑡
𝑂𝑟, 𝑙 sin 𝜃 cos 𝜃 𝜑̇ 2 − 𝑔 sin 𝜃 = 0
𝑔
∴ 𝜑̇ 2 − = 0 ⋯ ⋯ ⋯ (1)
𝑙 cos 𝜃
Equation in motion in 𝜑 −
𝑑
(𝑚𝑙 2 sin2 𝜃 𝜑̇ ) = 0
𝑑𝑡
𝑂𝑟, 𝑚𝑙 2 sin2 𝜃 𝜑̈ = 0 [𝑆𝑖𝑛𝑐𝑒 𝜃 =
𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡 , 𝜃̇ = 0]
𝑂𝑟, 𝜑̈ = 0
𝑔
∴ 𝜑̇ = 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡 = 𝜔 𝑎𝑛𝑑 𝜔 = 𝜑̇ = √ 𝐹𝑟𝑜𝑚 (1)
𝑙 cos 𝜃
∴ 𝐼 = 𝑚𝐾 2 , 𝐾 → 𝑅𝑎𝑑𝑖𝑜𝑢𝑠 𝑜𝑓 𝑔𝑦𝑟𝑎𝑡𝑖𝑜𝑛
𝑥 = 𝑙 sin 𝜃 ; 𝑦 = 𝑙 cos 𝜃
𝑥̇ = 𝑙 cos 𝜃 𝜃̇ ; 𝑦̇ = −𝑙 sin 𝜃 𝜃̇
1
∴ 𝑇 = 𝑚(𝑥̇ 2 + 𝑦̇ 2 )
2
1
= 𝑚(𝑙 2 cos 2 𝜃 𝜃̇ 2 + 𝑙 2 sin2 𝜃 𝜃̇ 2 )
2

1
= 𝑚𝑙 2 𝜃̇ 2
2
1
= 𝐼𝜃̇ 2
2
∴ 𝑉 = −𝑚𝑔𝑦
= −𝑚𝑔𝑙 cos 𝜃
1
∴ 𝐿 = 𝑇 − 𝑉 = 𝑚𝑙 2 𝜃̇ 2 + 𝑚𝑔𝑙 cos 𝜃
2
Lagrangian equation of motion in 𝜃 −
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃
𝑑
𝑂𝑟, (𝑚𝑙 2 𝜃̇) + 𝑚𝑔𝑙 sin 𝜃𝜃̇ = 0
𝑑𝑡
∴ 𝐼𝜃̈ + 𝑚𝑔𝑙 sin 𝜃 = 0

20. Write down the lagrangian of a body moving in Central force field. Hence, Show that
angular momentum is conserved for the system. (2012)
1
𝑨𝒏𝒔. − 𝐿 = 𝑇 − 𝑉(𝑟) = 𝑚(𝑟 2 + 𝑟 2 𝜃 2̇ ) − 𝑉(𝑟)
2

Here L is not a function of 𝜃.

 𝜕𝐿
∴ =0
𝜕𝜃
Let 𝑝𝜃 be the corresponding momentum.

𝜕𝐿
∴ 𝑝𝜃 = = 𝑚𝑟 2 𝜃̇
𝜕𝜃̇
is the angular momentum
Now,
𝑑 𝜕𝐿 𝜕𝐿
𝑝𝜃̇ = ( )= = 0 & 𝑝𝜃 = 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃

𝒌
21. A particle is moving under the influence of a central force ⃗𝑭⃗ = − 𝒓𝟐 𝒓̂. Set up the
Lagrangian and hence derive the equation of motion.
𝑶𝒓, A particle of mass 𝑚 is moving in a plane under an inversa square law attractive
force. Set up the Lagrangian and hence obtain the equation describing its motion.
𝑨𝒏𝒔. − Let (𝑟, 𝜃) be the plane polar Co-ordinates of the particle of mass 𝑚. The Kinetic
1
energy of the particle is given by 𝑇 = 𝑚(𝑟̇ 2 + 𝑟 2 𝜃 2̇ )
2

The potential energy 𝑉 is given by-

𝑟 𝑟
𝐾 𝐾
𝑉 = − ∫ 𝐹. 𝑑𝑟 = + ∫ 𝑑𝑟 = −
𝑟2 𝑟
∞ ∞

The Lagrangian is given by,

1 𝑘
𝐿 = 𝑇 − 𝑉 = 𝑚(𝑟̇ 2 + 𝑟 2 𝜃̇ 2 ) +
2 𝑟
𝜕𝐿 𝜕𝐿 𝑘
∴ = 𝑚𝑟̇ ; = 𝑚𝑟𝜃̇ 2 − 2
𝜕𝑟̇ 𝜕𝑟 𝑟
𝜕𝐿 𝜕𝐿
= 𝑚𝑟 2 𝜃̇ ; =0
𝜕𝜃̇ 𝜕𝜃
The Lagrangian equation (or equation of motion) in the variable 𝑟 is given by,
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝑟̇ 𝜕𝑟
𝑑 𝑘
𝑂𝑟, (𝑚𝑟̇ ) − (𝑚𝑟𝜃̇ 2 − 2 ) = 0
𝑑𝑡 𝑟
𝑘
∴ 𝑚𝑟̈ − 𝑚𝑟𝜃̇ 2 + =0
𝑟2
The Lagrangian equation in the variable 𝜃 is given by,
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝜃̇ 𝜕𝜃
𝑑
(𝑚𝑟 2 𝜃̇) = 0 ⋯ ⋯ (1)
𝑑𝑡
∴ 𝑚𝑟 2 𝜃̈ + 2𝑚𝑟𝑟̇ 𝜃̇ = 0
From (1) 𝑚𝑟 2 𝜃̇ = 𝐶𝑜𝑛𝑠𝑡𝑎𝑛𝑡
Thus the angular momentum of the particle is constant.

22. Two masses 𝒎𝟏 & 𝒎𝟐 are tied at the end of a light inextensible . String of length 𝒍
which passes round a light frictionless pully. Find the Lagrangian of the system &
hence its equation of motion.
𝑨𝒏𝒔. − The K.E. of the system is
1 1
𝑇 = 𝑚1 𝑥̇ 12 + 𝑚2 𝑥̇ 22
2 2

∴ 𝑥1 + 𝑥2 = 𝑙 = 𝐿𝑒𝑛𝑔𝑡ℎ 𝑜𝑓 𝑡ℎ𝑒 𝑠𝑡𝑟𝑖𝑛𝑔

𝑥̇ 1 = −𝑥̇ 2
∴ 𝑥̈ 1 = −𝑥̈ 2
1
𝑇 = (𝑚1 + 𝑚2 )𝑥̇ 12
2
Potential energy of the system is –
𝑉 = −𝑚1 𝑔𝑥1 − 𝑚2 𝑔𝑥2 = −𝑚1 𝑔𝑥1 − 𝑚2 𝑔(𝑙 − 𝑥1 )
= −(𝑚1 − 𝑚2 )𝑔𝑥1 − 𝑚2 𝑔𝑙
Hence the Lagrangian-
1
𝐿 = 𝑇 − 𝑉 = (𝑚1 + 𝑚2 )𝑥̇ 12 + (𝑚1 − 𝑚2 )𝑔𝑥1 + 𝑚2 𝑔𝑙
2
The equation of motion of the system is given by
𝑑 𝜕𝐿 𝜕𝐿
( )− =0
𝑑𝑡 𝜕𝑥̇ 1 𝜕𝑥1
𝑑
{(𝑚1 + 𝑚2 )𝑥̇ 1 } − (𝑚1 − 𝑚2 )𝑔 = 0
𝑑𝑡
𝑂𝑟, (𝑚1 + 𝑚2 )𝑥̈ 1 − (𝑚1 − 𝑚2 )𝑔 = 0
 𝑚1 − 𝑚2
∴ 𝑥̈ 1 = 𝑔
𝑚1 + 𝑚2
𝑚1 −𝑚2 𝑚1 −𝑚2
So the acceleration of 𝑚1 is 𝑔 and that of 𝑚2 is − 𝑔
𝑚1 +𝑚2 𝑚1 +𝑚2