The evolute of a curve is defined as the locus of the centers of curvature of the curve. The evolute of a curve is important in the theory of the pendulum because of the property expressed by the following theorem (see [1], p. 312).

If

and

are the centers of curvature corresponding to the points

and

of a curve, the length of the arc of the evolute between

and

is equal to the difference between the radii of curvature at

and

.

The evolute of the orbit of a pendulum can be calculated in the following way (see e.g. [2], p. 79). If

is a parametric equation of the orbit of a pendulum, the unit tangent vector

of

at any

is given by:

.

The curvature of

at any

is given by the magnitude of the vector:

,

while for each

the center of curvature can be located by the vector:

.

It is interesting to note that from this definition we find that the evolute of a circle consists of a single point, the center of the circle.

To describe the motion of a pendulum, we need to derive a function

that gives the position

of the pendulum at any time

. Since the motion of a pendulum is periodic, we only need to know the period

of the pendulum and the function

for

. The function

can be derived on the basis of the principle of conservation of mechanical energy (see e.g. [1], p. 309). The calculations for this Demonstration are done numerically using standard

*Mathematica* functions.

The principle of conservation of mechanical energy requires that the kinetic energy of the pendulum at any instant be equal to the difference between its potential energy at the initial position and its potential energy at the current position. Assuming that a parametric equation of the orbit of a pendulum is

and that the endpoints of the orbit are

and

, we find:

,

where

is the acceleration of gravity.

If

gives the arc length traveled by the bob as a function of time, then

.

From this equation we find that the period

of the pendulum is

.

We also find that the time

it takes the bob to move from the initial position

to a position

is

.

Since the motion of the bob is symmetric, we only need to establish a function that gives the time

for any

only for the first half of the orbit, that is, for the first quarter of the period. By inverting this function we can obtain the functions

and

for

. Because of the symmetries in the motion of the bob, this function can be extended so that the function

gives the position

of the bob for any time

in

. Since the period of the bob can also be calculated as we indicate above, its position can be determined at any time

.

[1] G. F. Simmons,

*Calculus Gems: Brief Lives and Memorable Mathematics*, New York: McGraw-Hill, 1992.

[2] B. Davis, H. Porta, and J. Uhl,

*Calculus & Mathematica: Vector Calculus: Measuring in Two and Three Dimensions*, Reading, MA: Addison-Wesley Publishing Company, 1994.