Mechanics

Physics 151

Lecture 3
Lagrange’s Equations
(Goldstein Chapter 1)
Hamilton’s Principle
(Chapter 2)

What We Did Last Time

! Discussed multi-particle systems
! Internal and external forces
! Laws of action and reaction

! Introduced constraints
! Generalized coordinates
! Introduced Lagrange’s Equations
! ... and didn’t do the derivation
" Let’s pick it up and start from there

Today’s Goals
! Derive Lagrange’s Eqn from Newton’s Eqn
! Use D’Alembert’s principle
! There will be a few assumptions
! Will make them clear as we go

! Introduce Hamilton’s Principle

! Equivalent to Lagrange’s Equations
! Which in turn is equivalent to Newton’s Equations

! Does not depend on coordinates by construction

! Derivation in the next lecture

1
Lagrange’s Equations Recipe

Kinetic energy
d  ∂L  ∂L
  − =0 L(q, q! , t ) ≡ T − V
dt  ∂q! j  ∂q j Potential energy
Lagrangian

! Express L = T – V in terms of generalized coordinates

{q j } , their time-derivatives {q! j }, and time t
! The potential V = V(q, t) must exist
! i.e. all forces must be conservative

Virtual Displacement
! Consider a system with constraints  r1 = r1 (q1 , q2 ,..., qn , t )
 r = r (q , q ,..., q , t )
!Ordinary coordinates ri (i = 1...N)  2 2 1 2 n

! Generalized coordinates qj (j = 1...n)  "
! Imagine moving all the particles
rN = rN ( q1 , q2 ,..., qn , t )
slightly ri → ri + δ ri q j → q j + δ q j
Virtual displacement
! Note that δri must satisfy the constraints
∂r
δ ri = ∑ i δ q j
j ∂q j
3N coordinates n coordinates
not independent independent

D’Alembert’s Principle
! From Newton’s Equation of Motion
Fi = p! i Fi − p! i = 0

! Part of the force Fi must be due to constraints

Fi = Fi( a ) + fi
“applied” force “constraint” force

! Applied force is “known” Fi( a ) = Fi( a ) (r1 , r2 ,..., ri ,..., rN , t )

! Constraint force fi (usually) does no work
! Movement is perpendicular to the force fiδ ri = 0
! Exception: friction
! Now multiply Fi( a ) + fi − p! i = 0 by δri and sum over i

2
D’Alembert’s Principle
“constraint” force is out of the game.
∑ (Fi
i
(a)
− p! i )δ ri = 0 You can forget (a)

! Force of constraints dropped out because fiδ ri = 0

! Called D’Alembert’s Principle (1743)
! Now we switch from ri to qj
∂ri ∂ri
1st term = ∑ Fi ∑ δ q j = ∑ Q jδ q j Q j ≡ ∑ Fi
i j ∂q j j i ∂q j

! Unit of Qj not always [force] Generalized force

! Qj qj is always [work]

D’Alembert’s Principle
∂ri ∂r
2nd term = ∑ p! iδ ri = ∑ p! i ∑ δ q j = ∑ mi!!ri i δ q j
i i j ∂q j i, j ∂q j

∂ri d  ∂  vi2   ∂  vi2 

! A bit of work can show !r!i →    −  
∂q j dt  ∂q! j  2   ∂q j  2 

 d  ∂T  ∂T  mvi2
= ∑   − δ q j T ≡∑

j  dt  ∂q! j  ∂q j  i 2

! D’Alembert’s Principle becomes

  d  ∂T  ∂T  
∑   dt  ∂q!  −  − Q j δ q j = 0
j    j  ∂q j  

Lagrange’s Equations
  d  ∂T  ∂T   These are free
∑   dt  ∂q!  −  − Q j δ q j = 0
j    j  ∂q j  
! Generalized coordinates qj are independent
d  ∂T  ∂T
  − = Qj Almost there!
dt  ∂q! j  ∂q j
! Assume forces are conservative Fi = −∇iV
∂ri ∂r ∂V
Q j ≡ ∑ Fi = −∑ ∇iV i = −
i ∂q j i ∂q j ∂q j
Throw this
back in

3
Lagrange’s Equations
d  ∂T  ∂ (T − V )
 − =0
dt  ∂q! j  ∂q j

∂V
! Assume that V does not depend on q! j =0
∂q! j
d  ∂L  ∂L
Finally  − =0 L = T (q j , q! j , t ) − V (q j , t )
dt  ∂q! j  ∂q j

Done!

Assumptions We Made
! Constraints are holonomic ri = ri (q1 , q2 ,..., qn , t )
! We always assume this
! Constraint forces do no work fiδ ri = 0
! Forget frictions
! Applied forces are conservative Fi = −∇ iV
Lagrange’s Eqn. itself is OK if V depends explicitly on t
!
∂V
! Potential V does not depend on q! j =0
∂q! j
Will review the last assumption later

Example: Time-Dependent
! Transformation functions may depend on t
! Generalized coordinate system may move
ri = ri (q j , t )
! E.g. coordinate system fixed to the Earth
! An example
spring constant K
natural length l
mass m on a rail

l+r
angular velocity α

4
Example: Time-Dependent
 x = (l + r ) cos α t
! Transformation functions: 
 y = (l + r ) sin α t
Kinetic energy T = { x! 2 + y! 2 } = {r! 2 + (l + r ) 2 α 2 }
m m
!
2 2
K 2
! Potential energy V = r
2

L = {r! 2 + (l + r ) 2 α 2 } − r 2
m K
2 2

d  ∂L  ∂L
Lagrange’s Equation − = mr!! − mα 2 (l + r ) + Kr = 0
dt  ∂r!  ∂r

Example: Time-Dependent
d  ∂L  ∂L
− = mr!! − mα 2 (l + r ) + Kr = 0
dt  ∂r!  ∂r
 mα 2l 
mr!! + ( K − mα 2 )  r − =0
 K − mα 2 
K − mα 2
! If K > mα2, a harmonic oscillator with ω =
m
! Center of oscillation is shifted by

! If K < mα2, moves away exponentially

! If K = mα2, velocity is constant
! Centripetal force balances with the spring force

Note on Arbitrarity
! Lagrangian is not unique for a given system
! If a Lagrangian L describes a system
dF (q, t )
L′ = L + works as well for any function F
dt
! One can prove
d  ∂  dF   ∂  dF  dF ∂F ∂F
   −   = 0 using = q! +
dt  ∂q!  dt   ∂q  dt  dt ∂q ∂t

5
Assumptions We Made
! Constraints are holonomic ri = ri (q1 , q2 ,..., qn , t )
! We always assume this
! Constraint forces do no work fiδ ri = 0
! Forget frictions
! Applied forces are conservative Fi = −∇iV
Lagrange’s Eqn. itself is OK if V depends explicitly on t
!
∂V
! Potential V does not depend on q! j =0
∂q! j
Let’s review the last assumption

Velocity-Dependent Potential
∂V ∂V
! We assumed Q j = − and = 0 so that
∂q j ∂q! j
This had to be 0
d  ∂T  ∂T d  ∂ (T − V )  ∂ (T − V )
 − = Qj  − =0
dt  ∂q! j  ∂q j dt  ∂q! j  ∂q j
! We could do the same if we had
Generalized,
∂U d  ∂U 
Qj = − +   U = U ( q j , q! j , t ) or velocity-
∂q j dt  ∂q! j  dependent
“potential”

L = T (q j , q! j , t ) − U (q j , q! j , t )

EM Force on Particle
! Lorentz force on a charged particle
Velocity-dependent.
F = q[E + ( v × B)] Can’t find a usual
! E and B fields are given by potential V
∂A
E = −∇φ − B = ∇× A Physics 15b
∂t
! Force is v-dependent " Need a v-dependent potential
U = qφ − qA ⋅ v works check

1
! Lagrangian is L = mv 2 − qφ + qA ⋅ v
2

6
Monogenic System
! If all forces in a system are derived from a generalized
potential,
∂U d  ∂U 
its called a monogenic system Qj = − +  

∂q j dt  ∂q! j 
! U is a function of q, q! , t
! Lorentz force is monogenic
! A monogenic system is conservative only if U = U (q )
∂U ∂U
! Or = =0
∂q! ∂t
! Lagrange’s Equation works on a monogenic system

Hamilton’s Principle
! We derived Lagrange’s Eqn from Newton’s Eqn using
a “differential principle”
! D’Alembert’s principle uses infinitesimal displacements
! It’s possible to do it with an “integral principle”

Hamilton’s Principle

Configuration Space
! Generalized coordinates q1,...,qn fully describe the
system’s configuration at any moment
configuration
! Imagine an n-dimensional space
space
! Each point in this space (q1,...,qn)
corresponds to one configuration of the system
! Time evolution of the system " A curve in the
configuration space
real space configuration space

7
 Action Integral
! A system is moving as q j = q j (t ) j = 1...n
! Lagrangian is L(q, q! , t ) = L(q(t ), q! (t ), t )

I = ∫ Ldt
t2
integrate t1
Action, or action integral

! Action I depends on the entire path from t1 to t2

! Choice of coordinates qj does not matter
! Action is invariant under coordinate transformation

Hamilton’s Principle
The action integral of a physical system is stationary
for the actual path
! This is equivalent to Lagrange’s Equations
! We will prove this We will also define “stationary”
! Three equivalent formulations
! Newton’s Eqn depends explicitly on x-y-z coordinates
! Lagrange’s Eqn is same for any generalized coordinates
! Hamilton’s Principle refers to no coordinates
! Everything is in the action integral

Hamilton’s Principle is more fundamental probably...

Stationary
! Consider two paths that are close to each other
! Difference is infinitesimal configuration space

! Stationary means that the t2

difference of the action integrals is
q(t )
zero to the 1st order of δq(t)
! Similar to “first derivative = 0” q(t ) + δ q(t )
δ I = ∫ L(q + δ q, q! + δ q! , t )dt − ∫ L(q, q! , t )dt = 0
t2 t2

t1 t1 t1
! Almost same as saying “minimum” δ q(t1 ) = δ q(t2 ) = 0
! It could as well be maximum

8
Infinitesimal Path Difference
! What’s δq(t)? configuration space
! It’s arbitrary … sort of t2
! It has to be zero at t1 and t2
q(t )
! It’s well-behaving Don’t worry
too much
Continuous, non-singular, q(t ) + δ q(t )
continuous 1st and 2nd derivatives
t1
! Have to shrink it to zero
! Trick: write it as δ q (t ) = αη (t )
! α is a parameter, which we’ll make " 0

! η(t) is an arbitrary well-behaving function η (t1 ) = η (t2 ) = 0

Hamilton " Lagrange

! To derive Lagrange’s Eqns from Hamilton’s Principle
t2 t2
δ I = ∫ L(q + δ q, q! + δ q! , t )dt − ∫ L(q, q! , t )dt = 0
t1 t1

Define I (α ) ≡ ∫t L(q(t ) + αη (t ), q! (t ) + αη! (t ), t )dt

t2
!
1

δI is then lim [ I (α ) − I (0) ]  ∂I 

!
  dα
α →0
 ∂α α =0
 ∂I 
! We must show that   = 0 leads to Lagrange’s Eqns
 ∂α α = 0

A bit of work. Will do it on Thursday

Summary
! Derived Lagrange’s Eqn from Newton’s Eqn
! Using D’Alembert’s Principle  Differential approach
! Assumptions we made:
! Constraints are holonomic " Generalized coordinates
! Forces of constraints do no work " No frictions
! Other forces are monogenic " Generalized potential
! Introduced Hamilton’s Principle ∂U d  ∂U 
Qj = − +  
! Integral approach ∂q j dt  ∂q! j 

! Defined the action integral and “stationary”

! Derivation in the next lecture