You are on page 1of 6

Marion and Thornton

Tyler Shendruk
October 1, 2010
1 Marion and Thornton Chapter 7
Hamiltons Principle - Lagrangian and Hamiltonian dynamics.
1.1 Problem 6.4
_
`

_
`
-
z
6
r

,
,
s
Figure 1: Geodesic on circular cylinder
Show that the geodesic on the surface of a right circular cylinder is a segment
of a helix.
Recall that a geodesic is the shortest path along the surface between any
two points on that surface. We know that an element of length on the surface
of a cylinder is
ds
2
= dr
2
+ r
2
d
2
+ dz
2
. (1)
On a given cylinder of radius r = R we can have variation of both and z. We
can choose either to be our independent variable.
z as independent variable Writing Eq. (1) with z as the independent vari-
able, the total length between points 1 and 2 is
s =
_
z
2
z
1
ds =
_
z
2
z
1
_
dr
2
+ r
2
d
2
+ dz
2
=
_
z
2
z
1

0
2
+ r
2
_
d
dz
_
2
+ 1dz =
_
z
2
z
1
_
1 + r
2

2
dz
=
_
z
2
z
1
f (,

; z) dz
where we have identied f (,

; z) =
_
1 + r
2

2
to be the function that
we will apply Eulers equation to inorder to get an extremum for the path
length s.
1
Apply Eulers equation to f gives
f


d
dz
f

= 0

_
1 + r
2


d
dz

_
1 + r
2

= 0
0
d
dz
_
r
2

_
1 + r
2

2
_
= 0
d
dz
_
r
2

_
1 + r
2

2
_
= 0 (2)
Therefore, since r is constant

is constant with respect to z which indi-


cates
= c
1
z + c
2
, (3)
the equation for a helix.
as independent variable If instead we write Eq. (1) with as the inde-
pendent variable, the length becomes
s =
_

2

1
ds =
_

2

r
2
+
_
dz
d
_
2
d
=
_

2

1
_
r
2
+ z

2
d =
_

2

1
f (z, z

; ) d.
Eulers equation now gives
f
z

d
d
f
z

= 0

_
r
2
+ z

2
z

d
d

_
r
2
+ z

2
z

= 0
0
d
d
_
z

_
r
2
+ z

2
_
= 0 (4)
Now z

is constant so
z = c
3
+ c
4
=
1
c
3
z c
4
= c
1
z + c
2
(5)
which is exactly the same as Eq. (3) .
Second Form Of course, one can use Eulers second form for either of these
two variables
f y

f
y

= constant for
_
f
x
= 0
_
(6)
2
where
_
f =
_
r
2
+ z

2
and x r for y z
f =
_
1 + r
2

2
and x r for y .
Lets choose z. This choice gives
c
0
= f y

f
y

=
_
r
2
+ z

2
z

_
r
2
+ z

2
=
_
r
2
+ z

2
z

_
r
2
+ z

2
c
0
_
r
2
+ z

2
= r
2
+ z

2
z

2
r
2
+ z

2
=
r
4
c
2
0
z

2
=
r
4
c
2
r
2
= c
1
. (7)
This indicates once again z

is a constant and were at the same conclusion


as previously.
And for we get
k
0
= f y

f
y

=
_
1 + r
2

_
1 + r
2

2
=
_
1 + r
2

r
2
_
1 + r
2

2
1 = k
0
_
1 + r
2

2
1
k
2
0
= 1 + r
2

2
=
1
r
2
k
2
0

1
r
2
= k
2
. (8)
1.2 Problem 7.1
What are the coordinates needed to describe a disk that is rolling on a horizontal
plane and is free to rotate about vertical axis of the plane?
The position of the disk can be adequately described by x and y but unlike
say a cylinder rolling down an incline the distance gone in either x or y can not
be related to the angle that the disk has rolled through ( in Fig. 2 ). This is
because the disk is free to rotate about z. The angle describing the spinning
(called in the gure) determines the direction in x and y.
3

-
6
x
y
z
`
_
P
PP


Figure 2: Freely rotating, rolling disk.
If R is the disk radius then the elemental path length is given by
ds = Rd = cos () dx + sin () dy
and by trigonometry in the xy-plane the rst of two dierential equations (con-
straints) is
tan () =
dy
dx
. (9)
In order to get the second dierential equation we will square the elemental
path along the at surface and state the radial form and the cartesian form:
ds
2
= (Rd)
2
= dx
2
+ dy
2
Rearrange to get the dierential equation
_
dx
d
_
2
+
_
dy
d
_
2
= R
2
(10)
which is not integrable and since there are no imaginable equations to link the
coordinates, the constraints are non-holonomic.
1.3 Problem 7.4
Consider a particle constrained to a plane that moves under force f
r
= Ar
1
which means that the potential is
U =
_

f
r
dr =
_
Ar
1
dr =
A

+ C.
If U(r = 0) = 0 then the integration constant C drops out. The kinetic energy
in cylindrical coordinates is
T =
1
2
m r
2
+
1
2
m
_
r

_
2
.
Combining these gives the Lagrangian
L = T U =
1
2
m r
2
+
1
2
m
_
r

_
2

.
4
Lagrange equation for r
L
r

d
dt
L
r
= 0
0 + mr

2
Ar
1

d
dt
[m r + 0 + 0] = 0
mr

2
Ar
1
m r = 0
Lagrange equation for
L


d
dt
L

= 0
0 + 0 + 0
d
dt
_
0 + mr
2

+ 0
_
= 0
d
dt
[L] = 0
where weve identied L = mr
2

to be the angular momentum and demon-
strated that it is conserved with time.
1.4 Problem 7.5
g

f
r

f
g
-
6
x
z
(0, 0)
,

*
r

Figure 3: Particle conned to a plane.


Again consider a particle constrained to a plane that moves under force
f
r
= Ar
1
but now also under a gravitational force f
g
= mgz. Now the
potential is
U =
_

f dr =
_
Ar
1
dr +
_
mgdz =
A

+ mgz + C
=
A

+ mgr sin + C. (11)


And again, if U(r = 0) = 0 then the integration constant C drops out. The
kinetic energy is unchanged from
T =
1
2
m r
2
+
1
2
m
_
r

_
2
. (12)
The Lagrangian is
L = T U =
1
2
m r
2
+
1
2
m
_
r

_
2

mgr sin . (13)


Lagrange equation for r
L
r

d
dt
L
r
= 0
0 + mr

2
Ar
1
mg sin
d
dt
[m r + 0 0 0] = 0
mr

2
Ar
1
mg sin m r = 0
5
r r

2
+
A
m
r
1
+ g sin = 0 . (14)
Lagrange equation for
L


d
dt
L

= 0
0 + 0 0 mgr cos
d
dt
_
0 + mr
2

0 0
_
= 0
dL
dt
= mgr cos = 0. (15)
Unlike before the angular momentum (notice its out of the plane) about
the origin is not conserved. Continuing with the Lagrange equation for
we nd
d
dt
mr
2

= mgr cos
r
2

+ 2r r

+ gr cos = 0
r

+ 2 r

+ g cos = 0 . (16)
6

You might also like