Hit Counter

7. Agent Work and Field Energy Flux.

In this section the following requirement of energy conservation is tested: the work expended per cycle by a driving agent, to force one or more charges to oscillate, must equate to the field energy flux per cycle time through a surface that encloses the oscillation site. The expended work per cycle is denoted as ‘W’, and the energy flux per cycle time is denoted as ‘ERad’. The objective is thus to demonstrate, over a range of frequencies etc., that

. (7_1)

The point charge field solutions are used to compute ERad. Since these solutions are based on Maxwell’s equations, they are presumably rigorous and the article accordingly tests the formulas used to compute W. Only non-relativistic cases (where the maximum speed is <<c) are considered. The relativistic formulas for W will be considered in one or more later sections.

Readers who wish to run the Basic programs (or their equivalent) on their own computers are advised that the computation of retarded time, and values derived from it, seems to work well only when the field evaluation point is at least a half wavelength from the source charge. Very small differences between the current time and the retarded time often do not produce accurate results, given the limited number of significant digits in most floating point processors. (Floating point processors compute products of very large and very small values accurately, but information is lost when two such numbers are added.)

In order for any charge to move periodically, a driving agent must theoretically counteract the charge’s inertial and radiation reaction forces. Furthermore, if two or more charges move periodically and are close enough to influence one another, then theoretically the driving agent must also counteract the interactive forces.

Symmetry can be exploited over the enclosing surface by limiting the motion of the charge(s) to the x-axis. In the multiple charge case the charges need not be equal, nor need their oscillation amplitudes, frequencies and/or phases be equal. The only constraints are that (a) they never occupy the same x-position simultaneously, and (b) their oscillation frequencies be integer multiples of some arbitrary, base frequency wo. Denoting a charge’s position at time t=0 as x(0), the kinematic equation is therefore:

(7_2)

The power expended by the inertial reaction force (and by the agent’s counteraction to it) integrates to zero over a cycle time. Hence the inertial reaction force plays no role in the energy flux per cycle through an enclosing surface. The electromagnetic mass is accordingly of no consequence, and we may stipulate that the charges are point charges. (The radiation reaction force of a point charge is the same as that of a spherical shell of charge, etc.) The non-relativistic formula for the radiation reaction force is:

. (7_3)

At the other charge’s position the point charge field solutions can be used to compute qi’s Ex field at time t, and hence the interactive force on that other charge.

Denoting the net interactive force acting on charge qi as Fi(Interact), the power expended by the driving agent on charge qi is

. (7_4)

Thus Wi, the net work per cycle expended on qi, is

. (7_5)

And of course W, the work per cycle expended on both charges, is

. (7_6)

An important rule is that, whereas the E and B field vectors of the individual charges are additive, the Poynting vectors are not. In brief, the electric and magnetic fields of the individual charges must be summed before computing the Poynting vector at any point on the enclosing surface.

For motion constrained to the x-axis, the surface of choice is spherical and centered on the origin. At any time epoch t, the net E and B field vectors for both charges need then be computed only at point (R,q) in the xy-plane, where 0<q<p is the angle between the positive x-axis and a radial line from the origin to the spherical surface.

The general approach is to compute the component of S (the Poynting vector) normal to the spherical surface at point (R,q), and to multiply that normal component by the area of the narrow strip of width RDq and concentric to the x-axis. Multiplication of this result by Dt (the interval between time epochs) then gives the energy flux through that particular strip in the time interval Dt. And summing the results for all the sphere’s strips, and over all the Dt’s comprising a period of oscillation, yields the net energy flux per cycle through the entire enclosing spherical surface.

Two Runs are made, one for a lone charge and one for a pair of charges. In the lone charge case the agent theoretically need only counteract that charge’s radiation reaction force. In the two-charge case the agent must counteract the radiation reaction force, plus the interactive force experienced in the other charge’s electric field.

The values for the oscillation amplitude(s), etc., are specified in Tables 7_1 and 7_2 (for Runs 1 and 2 respectively), along with the computed values of ERad and W. The 1-charge program can easily be altered, and the 2_charge program can readily be modified to deal with more than two charges. Note how W = ERad in every case.

Table 7_1

Frequency(Hz)

Amplitude(m)

Erad

W

1E6

.0477

3.9470E-4

3.9470E-4

5E6

4.77E-3

4.9337E-4

4.9337E-4

8E5

9.947E-4

8.7832E-8

8.7832E-8

5E6

9.543E-3

1.9735E-3

1.9735E-3

1E6

7.958E-4

1.0979E-7

1.0979E-7

Single Charge

 

Table 7_2

freq1(Hz)

freq2(Hz)

q1(coul)

q2(coul)

Amp1(m)

Amp2(m)

phi1(degrees)

phi2(degrees)

W

Erad

1E6

1E6

1E-3

1E-3

.048

.048

0

0

8.9822E-4

8.9822E-4

2E6

1E6

1E-3

1E-3

.024

.048

0

0

1.9735E-3

1.9735E-3

1E7

5E6

1E-3

1E-3

2.386E-3

1.592E-4

90

30

1.9470E-3

1.9470E-3

8E5

8E5

1E-3

1E-3

9.947E-4

9.947E-4

180

45

1.5854E-7

1.5854E-7

1E6

2E6

.1

1.5

.048

3.979E-4

180

0

4.9351

4.9351

Two Charges