Chapter 2

Particles and Cylindrical Polar Coordinates

T OPICS
Here, we discuss the cylindrical polar coordinate system and how it is used in particle mechanics. This coordinate system and its associated basis vectors {er , e , Ez }
are vital to understand and practice.
It is a mistake to waste your time memorizing formulae here. Instead, focus on
understanding the material. You will repeat it countless times which will naturally
develop the ability to derive the results from scratch.

2.1 The Cylindrical Polar Coordinate System

Consider the pendulum system shown in Figure 2.1. Here, a particle of mass m is
attached by an inextensible string of length L to a fixed point O. Assuming that the
string remains taut, then the distance from O to the particle remains constant: ||r|| =
Ey
g

Tension in string
Ex
mg

t
Fig. 2.1 An example of the motion of a pendulum. The behavior of the tension T in the string
during this motion is also shown.

O.M. OReilly, Engineering Dynamics: A Primer, DOI 10.1007/978-1-4419-6360-4 2,

c Springer Science+Business Media, LLC 2010


17

18

2 Particles and Cylindrical Polar Coordinates

Ey
er

Ex
Fig. 2.2 The unit vectors er and e .

L. This system is a prototypical example of a situation where a polar coordinate

system can be effectively used.
To define a cylindrical polar coordinate system {r, , z}, we start with a Cartesian
coordinate system {x, y, z} for the three-dimensional space E3 . Using these coordinates, we define r, , and z as
y

,
z = z.
= tan1
r = x2 + y2 ,
x
The coordinate r 0. Apart from the points {x, y, z} = {0, 0, z} , given r, , and z,
we can uniquely determine x, y, and z:
x = r cos( ),

y = r sin( ),

z = z.

Here, is taken to be positive in the counterclockwise direction.

If we now consider the position vector r of a point in this space, we have, as
always,
r = xEx + yEy + zEz .
We can write this position vector using cylindrical polar coordinates by substituting
for x and y in terms of r and :
r = r cos( )Ex + r sin( )Ey + zEz .
Before we use this representation to establish expressions for the velocity and acceleration vectors, it is convenient to introduce the unit vectors er and e :


Ex
er
cos( ) sin( ) 0
e = sin( ) cos( ) 0 Ey .
0
0 1
Ez
Ez
Two of these vectors are shown in Figure 2.2.

2.2 Velocity and Acceleration Vectors

19

Note that {er , e , Ez } are orthonormal and form a right-handed basis1 for E3 . You
should also be able to see that


Ex
cos( ) sin( ) 0
er
Ey = sin( ) cos( ) 0 e .
Ez
Ez
0
0
1
Because er = er ( ) and e = e ( ), these vectors change as changes:
der
= sin( )Ex + cos( )Ey = e ,
d
de
= cos( )Ex sin( )Ey = er .
d
It is crucial to note that is measured positive in the counterclockwise direction.
Returning to the position vector r, it follows that
r = xEx + yEy + zEz = r cos( )Ex + r sin( )Ey +zEz = rer + zEz .



=rer

Furthermore, because {er , e , Ez } is a basis, we then have, for any vector b, that
b = br er + b e + bzEz = bx Ex + by Ey + bz Ez .
It should be clear that br = b er , b = b e , and bz = b Ez .

2.2 Velocity and Acceleration Vectors

Consider a particle moving in space: r = r(t). We recall that
r = rer + zEz = xEx + yEy + zEz .
As the particle is in motion, its coordinates are functions of time: x = x(t), y = y(t),
r = r(t), = (t), and z = z(t). To calculate the (absolute) velocity vector v of the
particle, we differentiate r(t):
v=

dr dr
der dz
= er + r
+ Ez .
dt
dt
dt
dt

Now, using the chain rule, e r = der /d = e . Also,

de
= er ,
d

der
= e .
d

Details of these results are discussed in Section A.4 of Appendix A.

20

2 Particles and Cylindrical Polar Coordinates

It follows that
d
dz
dr
er + r e + Ez
dt
dt
dt
dx
dy
dz
Ex + Ey + Ez .
=
dt
dt
dt

v=

To calculate the (absolute) acceleration vector a, we differentiate v with respect

to time:




d dr
dv
d
d2z
d
=
er +
a=
r e + 2 Ez .
dt
dt dt
dt
dt
dt
Using the chain rule to determine the time derivatives of the vectors er and e ,
and after collecting terms in the expressions for a, the final form of the results is
obtained:

2 
2

d
d
dr d
d 2z
d2r

r
+
r
+
2
+
Ez
e
a=
e
r

dt 2
dt
dt 2
dt dt
dt 2
=

d 2x
d2y
d 2z
E
+
E
+
Ez .
x
y
dt 2
dt 2
dt 2

We have also included the representations for the velocity and acceleration vectors
in Cartesian coordinates to emphasize the fact that the values of these vectors do not
depend on the coordinate system used.

2.2.1 Common Errors

In my experience, the most common error with using cylindrical polar coordinates
is to write r = rer + e + zEz . This is not true. Another mistake is to differentiate
er and e incorrectly with respect to time. Last, but not least, many people presume
that all of the results presented here apply when is taken to be positive in the
clockwise direction. Alas, this is not the case.

2.3 Kinetics of a Particle

Consider a particle of mass m. Let F denote the resultant external force acting on
the particle, and let G = mv be the linear momentum of the particle. Eulers first law
(which is also known as Newtons second law or the balance of linear momentum)
postulates that
dG
= ma.
F=
dt

2.4 The Planar Pendulum

21

With respect to a Cartesian basis F = ma is equivalent to three scalar equations:

Fx = max = mx,

Fy = may = my,

Fz = maz = mz,

where F = Fx Ex + Fy Ey + Fz Ez and a = ax Ex + ay Ey + az Ez .
With respect to a cylindrical polar coordinate system, the single vector equation
F = ma is equivalent to three scalar equations:

2 
d
d2r
(F = ma) er : Fr = m
r
,
2
dt
dt
2

d
dr d
(F = ma) e : F = m r 2 + 2
,
dt
dt dt
(F = ma) Ez : Fz = m

d 2z
.
dt 2

Finally, we recall for emphasis the relations

er = cos( )Ex + sin( )Ey ,
e = sin( )Ex + cos( )Ey ,
Ex = cos( )er sin( )e ,
Ey = sin( )er + cos( )e .
You will use these relations countless times in an undergraduate engineering dynamics course.

2.4 The Planar Pendulum

The planar pendulum is a classical problem in mechanics. As shown in Figure 2.3,
a particle of mass m is suspended from a fixed point O either by an inextensible
massless string or rigid massless rod of length L. The particle is free to move on a
plane (z = 0), and during its motion a vertical gravitational force mgEy acts on the
particle.
We ask the following questions: what are the equations governing the motion
of the particle and what is the tension in the string or rod? The answers to these
questions are used to construct the motion of the particle and the plot of tension as
a function of time that were shown in Figure 2.1 at the beginning of this chapter.

22

2 Particles and Cylindrical Polar Coordinates

T er

Ey
g

Ex

NEz

mgEy

Massless rod or string of length L

Particle of mass m
Fig. 2.3 The planar pendulum and the free-body diagram of the particle of mass m.

2.4.1 Kinematics
We begin by establishing some kinematical results. We note that r = Ler . Differentiating with respect to t, and noting that L is constant, gives us the velocity v.
Similarly, we obtain a from v:
d
der
= L e ,
dt
dt
2
d
d2
d de
d2
= L 2 e L
er .
a = L 2 e + L
dt
dt dt
dt
dt
v=L

Alternatively, one can get these results by substituting r = L and z = 0 in the general
expressions recorded in Section 2.2. I do not recommend this approach inasmuch as
it emphasizes memorization.

2.4.2 Forces
Next, as shown in Figure 2.3, we draw a free-body diagram. There is a tension force
T er and a normal force NEz acting on a particle. The role of the tension force is
to ensure that the distance of the particle from the origin is L and the normal force
ensures that there is no motion in the direction of Ez . These two forces are known
as constraint forces. They are indeterminate (we need to use F = ma to determine
them). One should also note that the gravitational force has the representations
mgEy = mg sin( )er mg cos( )e .

2.5 Summary

23

2.4.3 Balance Law

The third step is to write down the balance of linear momentum (F = ma):
T er + NEz mgEy = mL e mL 2 er .
We obtain three scalar equations from this vector equation:
mL = mg cos( ),

T = mL 2 mg sin( ),

N = 0.

2.4.4 Analysis
The first of these equations is a second-order differential equation for (t):
mL = mg cos( ).
Given the initial conditions (t0 ) and (t0 ), one can solve this equation and determine the motion of the particle. Next, the second equation gives the tension T in the
string or rod once (t) is known:


T er = mL 2 mg sin( ) er .
A representative example of the behavior of T during a motion of the pendulum is
shown in Figure 2.1. This figure was constructed by first numerically solving the
ordinary differential equations for (t) and then computing the corresponding T (t).
For a string, it is normally assumed that T > 0, and for some motions of the string
it is possible that this assumption is violated. In this case, the particle behaves as if
it were free to move on the plane and r = L. Regardless, the normal force NEz is
zero in this problem.

2.5 Summary
In this chapter, the cylindrical polar coordinate system {r, , z} was introduced. To
assist with certain expressions, the vectors er = cos( )Ex + sin( )Ey and e =
sin( )Ex + cos( )Ey were introduced. It was also shown that the position vector of a particle has the representations

r = rer + zEz = x2 + y2er + zEz
= r cos( )Ex + r sin( )Ey + zEz
= xEx + yEy + zEz .

24

2 Particles and Cylindrical Polar Coordinates

By differentiating r with respect to time, the velocity and acceleration vectors were
obtained. These vectors have the representations
dr
=
dt
=
dv
a=
=
dt
=
v=

rer + r e + zEz
xE
x + yE
y + zEz ,




r r 2 er + r + 2r e + zEz
xE
x + yE
y + zEz .

To establish these results, the chain rule and the important identities e r = e and
e = er were used.
Using a cylindrical polar coordinate system, F = ma can be written as three scalar
equations:


Fr = m r r 2 ,


F = m r + 2r ,
Fz = mz.
These equations were illustrated using the example of the planar pendulum.

2.6 Exercises
The following short exercises are intended to assist you in reviewing the present
chapter.
2.1.

2.2.

2.3.

2.4.

2.5.

Using Figure 2.2, verify that er = cos( )Ex +sin( )Ey and e = sin( )Ex +
cos( )Ey . Then, by considering cases where er lies in the second, third, and
fourth quadrants, verify that these definitions are valid for all values of .
Starting from the definitions er = cos( )Ex +sin( )Ey and e = sin( )Ex +
cos( )Ey , show that e r = e and e = er . In addition, verify that
Ex = cos( )er sin( )e and Ey = sin( )er + cos( )e .
Calculate the velocity vectors of particles whose position vectors are 10er and
5er + tEz , where = t. Why do all of these particles move with constant
speed ||v|| yet have a nonzero acceleration?
The position vector of a particle of mass m that is placed at the end of a
rotating telescoping rod is r = 6ter , where = 10t + 5 (radians). Calculate
the velocity and acceleration vectors of the particle, and determine the force F
needed to sustain the motion of the particle. What is the force that the particle
exerts on the telescoping rod?
In solving a problem, one person uses cylindrical polar coordinates whereas
another uses Cartesian coordinates. To check that their answers are identical,
they need to examine the relationship between the Cartesian and cylindrical

2.6 Exercises

25

polar components of a certain vector, say b = br er + b e . To this end, show

that
bx = b Ex = br cos( ) b sin( ),
2.6.

Consider the projectile problem discussed in Section 1.6 of Chapter 1. Using

a cylindrical polar coordinate system, show that the equations governing the
motion of the particle are
mr mr 2 = mg sin( ),

2.7.

2.8.

by = b Ey = br sin( ) + b cos( ).

mr + 2mr = mg cos( ),

mz = 0.

Notice that, in contrast to using Cartesian coordinates to determine the governing equations, solving these differential equations is nontrivial.
Consider a spherical bead of mass m and radius R that is placed inside a long
cylindrical tube. The inner radius of the tube is R, and the tube is pivoted so
that it rotates in a horizontal plane. Furthermore, the contact between the tube
and the bead is smooth. Here, the bead is modeled as a particle of mass m.
Now suppose that the tube is whirled at a constant angular speed (radians
per second). The whirling motion of the tube is such that the velocity vector
of the bead is v = rer + re . Show that the equation governing the motion
of the bead is
r 2 r = 0,
and the force exerted by the tube on the particle is mgEz + 2mr e .
Consider the case where the bead is initially at rest relative to the whirling
tube at a location r0 = L. Using the solution to the differential equation r
2 r = 0 recorded in Section A.5.3 of Appendix A, show that, unless L = 0,
the bead discussed in the previous exercise will eventually exit the whirling
tube.

http://www.springer.com/978-1-4419-6359-8