Owen Duffy wrote:
If the distributed network model you favour is the S paramater model,
properly applied, it is in fact entirely consistent with the distributed
impedance line model because the parameters are derived from the solution
to the distributed impedance line model.

Given that the S-Parameter analysis is valid as explained in
HP's Ap Note 95-1 available from:

http://www.tm.agilent.com/data/stati...-1/an-95-1.pdf

--------Z01--------+--------Z02--------
a1-- b2--
--b1 --a2

where a1, a2, b1, and b2 are normalized voltages.

The equation for b1 is b1 = s11(a1) + s12(a2)

Given that a1 and a2 are in phase and that b1 = 0
then s11(a1) and s12(a2) would have to be of equal
magnitude and opposite phase thus making the reflected
power |b1|^2 equal to zero. s11(a1) and s12(a2) cancel
each other out. (What happens to the energy in the
canceled waves?)

What do you get when you square both sides of the equation:?

b1 = s11(a1) + s12(a2) = 0 reflected voltage

Since the square of any of those terms yields watts,
If we simplify by replacing complicated terms with symbols:

|s11*a1|^2 = P1 and |s12*a2|^2 = P2 we get:

|b1|^2 = 0 = P1 + P2 - 2*SQRT(P1*P2) = 0 reflected power

These squared (power) terms are all explained in Ap Note 95-1.
The intensity-irradiance-Poynting vector equation can be
derived from the S-Parameter equations. Good thing the S-Parameter
analysis is consistent with Hecht and Born & Wolf, huh?
--
73, Cecil, w5dxp.com