Source impedance DOES affect the amount of energy moving in and sloshing
around in a transmission line. It DOESN'T affect the ratio of forward to
reflected waves, and therefore DOESN'T affect the SWR.

Once again, here's a way to see why. I'll restrict the discussion to a
lossless line for simplicity. When you first turn the source on, a
forward wave (voltage and current) travels toward the load. The source
impedance does play a role in determining the size of this wave; it can
be determined by analysis of a simple voltage divider circuit, with the
source voltage dividing between the source impedance and the line Z0. A
portion of the forward wave is reflected from the load unless the line
is perfectly matched. The fraction which is reflected has nothing to do
with the source impedance, and in fact it can easily be calculated from
only the line and load impedances. That fraction (magnitude and angle)
is known as the reflection coefficient -- you can find the formula in
any transmission line text, or derive it yourself very easily.

Take a look at the system just before the reflected wave returns to the
source. At each point along the line we have a forward wave and a
reflected wave, which vectorially add. These create standing waves and,
if the line is long enough, we can calculate the SWR directly as the
ratio of maximum to minimum voltage along the line. A little bit of
algebra will show that the SWR is determined entirely by the ratio of
forward to reflected waves -- their absolute values don't matter
(except, of course, as it affects their ratio). Given a reflection
coefficient, you can calculate the SWR.

Ok, now suppose that some fraction of the returning wave reflects from
the source and heads back toward the load. Say, X percent of it. When it
reaches the load, exactly the same fraction of it is reflected as was
the case for the original forward wave. That is, if the new forward wave
is X percent of the original, then the new reflected wave is also X
percent of the original reflected wave. If we add the new forward and
reflected waves to the original ones, and take the ratio of forward to
reverse, we find that the ratio of the new, combined forward wave to the
new, combined reflected wave is exactly the same as it was for the first
forward and reflected waves. It doesn't matter what X is -- no matter
what fraction of the reflected wave bounces off the source, the same
fraction of that new forward wave is reflected from the load. The SWR is
the same as it was for the original pair of waves. Eventually, we build
up a large number of pairs of forward and reflected waves. And the ratio
of each forward wave to its corresponding reflected wave is always the
same -- it's the reflection coefficient of the load. So when we add all
the forward waves into a single forward wave and all the reflected waves
into a single reflected wave, we get the same ratio. And that ratio
doesn't depend in any way on the source impedance or what fraction of
each returning wave is re-reflected from the source.

One of the nice things about this way of looking at it is that it's
entirely supported by the theory and equations describing transmission
line operation which engineers have used to design working systems for
the past hundred years or so.

Roy Lewallen, W7EL

Cecil Moore wrote:
Jim Kelley wrote:

It's hard to imagine how Rs (Zs) could have any effect on that ratio.


Consider a reactive load where energy can be locally exchanged between
the load reactance and the impedance looking back into the feedline.
Zs can certainly affect the impedance looking back into the feedline.