next up previous
Next: PeriodicityQuasiperiodicity and Sensitive Up: Classical Mechanics Previous: Ordinary Differential Equations

Newton's Laws

Classical mechanics provides many examples of deterministic evolution with a continuous time. Newton gave us some general principles to find the rule of evolution in particular circumstances.

Newton's first laws asserts that in special frames of reference called inertial frames, if an object is ``left alone'', it either stays at rest or keeps moving with a constant velocity. The velocity is defined as the time derivative of the position, the time derivative of the velocity is called the acceleration. Our ability to identify inertial frames is clearly dependent on our ability to determine when an object is left alone or not. Newton's second law asserts that in inertial frames, the change of velocity is due to forces acting on the object considered. More precisely, the mass of an object times its acceleration is equal to functions of the position and velocity called forces.

We shall not discuss the existence of inertial frames or the methods to determine the mass of an object. We shall simply state that in many examples, one can in good approximation, establish that the state of the system is determined by the positions and velocities and that the rules of evolutions are given by expressing the acceleration in terms of the positions and velocities. We shall discuss a few simple examples: oscillations of a spring, simple pendulum and double pendulum.

One of the simplest example of Newtonian mechanics is an object of mass m attached at the end of a spring. Let assume that the object can only move in the vertical direction. We denote the vertical position by x. If the object is at rest, we say that x=0. If the spring is stretched, x>0. In order to specify the state of the system, we need to introduce another variable which takes into account the fact that different velocities are possible for a given position at a given time. Usually one takes the momentum tex2html_wrap_inline425 . tex2html_wrap_inline427 is the second derivative of x (called the acceleration) times the mass and is equal to the force which acts in the opposite direction to the displacement with respect to the equilibrium position. If x>0, the spring is stretched and pulls back toward the equilibrium point. In other words tex2html_wrap_inline433 where k is a positive constant depending on the nature of the spring. The ODE can be summarized as

eqnarray95

An important feature of this ODE is that it has a constant of motion called the energy

equation97

The phases curves have a constant energy and form concentric ellipses. If we take a system of units where m=1 and k=1, the equations of motion have the same form as the third case studied in the exercises on ODE.

The case of the planar pendulum (see Fig. 6) is slightly more complex. Again the mass m and the length l are given for definiteness and do not play any important role in the rest of the discussion. The angle tex2html_wrap_inline445 is expressed in radians where a complete revolution amounts to a change by tex2html_wrap_inline447 .

   figure102
Figure 6: The planar pendulum.

The state of the system is described by the angle tex2html_wrap_inline445 and another variable proportional to the angular velocity which we will define as tex2html_wrap_inline451 . The equations of motion read

eqnarray107

where g is the gravitational constant (9.8 meter/second tex2html_wrap_inline455 ). The constant of motion is

equation110

Some phase curves are shown below. One can see that there are two types of phases curves. The ones for which theta reaches a maximal value and retraces its way backward (those curves wrap around the center) and the curves for which tex2html_wrap_inline445 keeps increasing indefinitly. Note that tex2html_wrap_inline445 has to be understood modulo tex2html_wrap_inline461 (even if it keeps growing on he figure).

   figure113
Figure 7: Some phase curves of the planar pendulum.


next up previous
Next: PeriodicityQuasiperiodicity and Sensitive Up: Classical Mechanics Previous: Ordinary Differential Equations

Yannick Meurice
Wed Mar 3 01:21:43 CST 1999