42 

  C ha pter 2  Kinematics of Particles

2/ 3 P L A N E C U R V I L I N E A R M OT I O N
We now treat the motion of a particle along a curved path which lies
in a single plane. This motion is a special case of the more general three-
dimensional motion introduced in Art. 2/1 and illustrated in Fig. 2/1. If we
let the plane of motion be the x-y plane, for instance, then the coordinates
z and ␾ of Fig. 2/1 are both zero, and R becomes the same as r. As men-
tioned previously, the vast majority of the motions of points or particles
encountered in engineering practice can be represented as plane motion.
Before pursuing the description of plane curvilinear motion in any
specific set of coordinates, we will first use vector analysis to describe
the motion, since the results will be independent of any particular coor-
dinate system. What follows in this article constitutes one of the most
basic concepts in dynamics, namely, the time derivative of a vector.
Much analysis in dynamics utilizes the time rates of change of vector
quantities. You are therefore well advised to master this topic at the
outset because you will have frequent occasion to use it.
Consider now the continuous motion of a particle along a plane
curve as represented in Fig. 2 /5. At time t the particle is at position A,
which is located by the position vector r measured from some conve-
nient fixed origin O. If both the magnitude and direction of r are known
at time t, then the position of the particle is completely specified. At
time t + Δt, the particle is at A′, located by the position vector r + Δr.
We note, of course, that this combination is vector addition and not sca-
lar addition. The displacement of the particle during time Δt is the
vector Δr which represents the vector change of position and is clearly
independent of the choice of origin. If an origin were chosen at some
different location, the position vector r would be changed, but Δr would
be unchanged. The distance actually traveled by the particle as it
moves along the path from A to A′ is the scalar length Δs measured
along the path. Thus, we distinguish between the vector displacement
Δr and the scalar distance Δs.

Velocity
The average velocity of the particle between A and A′ is defined as
vav = Δr /Δt, which is a vector whose direction is that of Δr and whose
magnitude is the magnitude of Δr divided by Δt. The average speed of

Path of
particle
v′
A′

Δs Δv
Δr A′
r + Δr
A v a
v′
v
A
r A

Figure 2/5
 A rt i c l e 2 / 3     Pl a n e Curv i l i ne ar M o t io n    43

the particle between A and A′ is the scalar quotient Δs /Δt. Clearly, the
magnitude of the average velocity and the speed approach one another
as the interval Δt decreases and A and A′ become closer together.
The instantaneous velocity v of the particle is defined as the limiting
value of the average velocity as the time interval approaches zero. Thus,

Δr
v = lim
Δt
 

Δ tS0

We observe that the direction of Δr approaches that of the tangent to

the path as Δt approaches zero and, thus, the velocity v is always a vec-
tor tangent to the path.
We now extend the basic definition of the derivative of a scalar
quantity to include a vector quantity and write

dr
v= = ṙ (2/4)
dt

The derivative of a vector is itself a vector having both a magnitude

and a direction. The magnitude of v is called the speed and is the scalar

ds
v = v = = ṡ
dt

At this point we make a careful distinction between the magnitude

of the derivative and the derivative of the magnitude. The magnitude of
the derivative can be written in any one of the several ways dr /dt =
 ṙ = ṡ = v = v and represents the magnitude of the velocity, or the
speed, of the particle. On the other hand, the derivative of the magni-
tude is written dr /dt = dr /dt = ṙ, and represents the rate at which the
length of the position vector r is changing. Thus, these two derivatives
have two entirely different meanings, and we must be extremely care-
ful to distinguish between them in our thinking and in our notation.
For this and other reasons, you are urged to adopt a consistent nota-
tion for handwritten work for all vector quantities to distinguish them
from scalar quantities. For simplicity the underline v is recommended.
z
Other handwritten symbols such as v , ∼v , and v̂ are sometimes used.
With the concept of velocity as a vector established, we return to
Fig. 2/5 and denote the velocity of the particle at A by the tangent vector v
and the velocity at A′ by the tangent v′. Clearly, there is a vector change in
the velocity during the time Δt. The velocity v at A plus (vectorially) the
change Δv must equal the velocity at A′, so we can write v′ − v = Δv. In-
spection of the vector diagram shows that Δv depends both on the change
in magnitude (length) of v and on the change in direction of v. These two
changes are fundamental characteristics of the derivative of a vector.

Acceleration
The average acceleration of the particle between A and A′ is defined
as Δv /Δt, which is a vector whose direction is that of Δv. The magnitude
of this average acceleration is the magnitude of Δv divided by Δt.
44    C ha pter 2  Kinematics of Particles

The instantaneous acceleration a of the particle is defined as the

limiting value of the average acceleration as the time interval ap-
proaches zero. Thus,

Δv
a = lim
Δt
 

Δ tS0

By definition of the derivative, then, we write

dv
a= = v̇ (2/5)
dt

As the interval Δt becomes smaller and approaches zero, the direction

of the change Δv approaches that of the differential change dv and,
thus, of a. The acceleration a, then, includes the effects of both the
change in magnitude of v and the change of direction of v. It is appar-
ent, in general, that the direction of the acceleration of a particle in
curvilinear motion is neither tangent to the path nor normal to the
path. We do observe, however, that the acceleration component which is
normal to the path points toward the center of curvature of the path.

Visualization of Motion
A further approach to the visualization of acceleration is shown in
Fig. 2/6, where the position vectors to three arbitrary positions on the
path of the particle are shown for illustrative purpose. There is a velocity
vector tangent to the path corresponding to each position vector, and the
relation is v = ṙ. If these velocity vectors are now plotted from some ar-
bitrary point C, a curve, called the hodograph, is formed. The derivatives
of these velocity vectors will be the acceleration vectors a = v̇ which are
tangent to the hodograph. We see that the acceleration has the same re-
lation to the velocity as the velocity has to the position vector.
The geometric portrayal of the derivatives of the position vector r
and velocity vector v in Fig. 2 /5 can be used to describe the derivative
of any vector quantity with respect to t or with respect to any other
scalar variable. Now that we have used the definitions of velocity and
acceleration to introduce the concept of the derivative of a vector, it is
important to establish the rules for differentiating vector quantities.

v3 = r·3
Path
a 3 = v·3 a 2 = v·2
v2 = r·2
r3 a 1 = v· 1
Hodograph v3
v2
O r2
v1
r1 v1 = r·1
C

Figure 2/6
 A rt i c l e 2 / 4     Re c t a n g u l a r Coord i n a t e s (x- y )    45

These rules are the same as for the differentiation of scalar quantities,
except for the case of the cross product where the order of the terms
must be preserved. These rules are covered in Art. C /7 of Appendix C
and should be reviewed at this point.
Three different coordinate systems are commonly used for describing
the vector relationships for curvilinear motion of a particle in a plane:
rectangular coordinates, normal and tangential coordinates, and polar co-
ordinates. An important lesson to be learned from the study of these coor-
dinate systems is the proper choice of a reference system for a given
problem. This choice is usually revealed by the manner in which the mo-
tion is generated or by the form in which the data are specified. Each of
the three coordinate systems will now be developed and illustrated.

2 / 4 R E C TA N G U L A R C O O R D I N AT E S ( X - Y )
This system of coordinates is particularly useful for describing mo-
tions where the x- and y-components of acceleration are independently
generated or determined. The resulting curvilinear motion is then ob-
tained by a vector combination of the x- and y-components of the posi-
tion vector, the velocity, and the acceleration.

Vector Representation
Path
The particle path of Fig. 2 /5 is shown again in Fig. 2 /7 along with
x- and y-axes. The position vector r, the velocity v, and the acceleration y
ay
a of the particle as developed in Art. 2 /3 are represented in Fig. 2 /7 to- v
a
gether with their x- and y-components. With the aid of the unit vectors vy
i and j, we can write the vectors r, v, and a in terms of their x- and j
θ
y-components. Thus, A vx ax A

yj r
r = xi + yj
v = ṙ = ẋi + ẏj (2/6) x
xi i
a = v̇ = r̈ = ẍi + ÿj
Figure 2/7

As we differentiate with respect to time, we observe that the time de-

rivatives of the unit vectors are zero because their magnitudes and di-
rections remain constant. The scalar values of the components of v and
a are merely vx = ẋ, vy = ẏ and ax = v̇x = ẍ, ay = v̇y = ÿ. (As drawn in
Fig. 2 /7, ax is in the negative x-direction, so that ẍ would be a negative
number.)
As observed previously, the direction of the velocity is always tan-
gent to the path, and from the figure it is clear that

vy
v2 = vx 2 + vy 2  v = √vx 2 + vy 2  tan ␪ =
       

vx

a2 = ax 2 + ay 2  a = √ax 2 + ay 2
       

If the angle ␪ is measured counterclockwise from the x-axis to v for the

configuration of axes shown, then we can also observe that dy /dx =
tan ␪ = vy /vx.
46    C ha pter 2  Kinematics of Particles

If the coordinates x and y are known independently as functions of

time, x = ƒ1(t) and y = ƒ2(t), then for any value of the time we can com-
bine them to obtain r. Similarly, we combine their first derivatives ẋ
and ẏ to obtain v and their second derivatives ẍ and ÿ to obtain a. On
the other hand, if the acceleration components ax and ay are given as
functions of the time, we can integrate each one separately with re-
spect to time, once to obtain vx and vy and again to obtain x = ƒ1(t) and
y = ƒ2(t). Elimination of the time t between these last two parametric
equations gives the equation of the curved path y = ƒ(x).
From the foregoing discussion we can see that the rectangular-
coordinate representation of curvilinear motion is merely the superpo-
sition of the components of two simultaneous rectilinear motions in
the x- and y-directions. Therefore, everything covered in Art. 2 /2 on
rectilinear motion can be applied separately to the x-motion and to the
y-motion.

Projectile Motion
An important application of two-dimensional kinematic theory is
the problem of projectile motion. For a first treatment of the subject,
we neglect aerodynamic drag and the curvature and rotation of the
earth, and we assume that the altitude change is small enough so that
the acceleration due to gravity can be considered constant. With these
assumptions, rectangular coordinates are useful for the trajectory
analysis.
For the axes shown in Fig. 2 /8, the acceleration components are

ax = 0  ay = −g

Integration of these accelerations follows the results obtained previ-

ously in Art. 2 /2a for constant acceleration and yields

vx = (vx ) 0 vy = (vy ) 0 − gt
1
x = x0 + (vx ) 0 t   y = y0 + (vy ) 0 t − 2 gt2
vy 2 = (vy ) 0 2 − 2 g( y − y0 )
     

In all these expressions, the subscript zero denotes initial conditions,

frequently taken as those at launch where, for the case illustrated,

vy v
vx vx
v0
g vy
v
(vy) 0 =
v0 sin θ
θ
x
(vx ) 0 = v0 cos θ

Figure 2/8
 A rt i c l e 2 / 4     Re c t a n g u l a r Coord i n a t e s (x- y )    47

x0 = y0 = 0. Note that the quantity g is taken to be positive throughout

this text.
We can see that the x- and y-motions are independent for the sim-
ple projectile conditions under consideration. Elimination of the time t
between the x- and y-displacement equations shows the path to be par-
abolic (see Sample Problem 2 /6). If we were to introduce a drag force
which depends on the speed squared (for example), then the x- and
y-motions would be coupled (interdependent), and the trajectory would
be nonparabolic.
When the projectile motion involves large velocities and high alti-
tudes, to obtain accurate results we must account for the shape of the
projectile, the variation of g with altitude, the variation of the air den-
sity with altitude, and the rotation of the earth. These factors introduce
considerable complexity into the motion equations, and numerical inte-
gration of the acceleration equations is usually necessary.

Andrew Davidhazy

This stroboscopic photograph of a bouncing ping-pong ball suggests not

only the parabolic nature of the path, but also the fact that the speed is
lower near the apex.