Keith Dysart wrote:
When the source impedance is the same as Z0 there
is no impedance discontinuity to produce a reflection.

Here's a quote from "Fields and Waves ..." by
Ramo and Whinnery: "... significance cannot be
automatically attached to a calculation of power
loss in the internal impedance of the equivalent
circuit."

And from "T-Lines and Networks" by Walter C. Johnson:
"Although the Thevenin equivalent produces the correct
exterior results, its internal power relations may be
quite different from those of the network it replaces.
The power dissipated in the equivalent impedance of the
Thevenin circuit is not the same as the power dissipated
in the resistance of the actual network."

Translation: Do not use a Thevenin source impedance
to try to track power dissipation within the Thevenin
equivalent box - it won't work. Your argument that
there is zero power dissipation in the source resistor
inside the Thevenin equivalent box is bogus.

Is the reflected wave reflected even without a
discontinuity?

The incident reflected wave is either reflected by the
source or it isn't. You cannot have it both ways. Since
the internal conditions inside a Thevenin equivalent
circuit are a complete unknown, either choice is
a possibility.

What you will find is that, for a real-world source,
the destructive interference on one side of the source
equals the constructive interference on the other side
of the source. The conservation of energy principle
prohibits having it any other way.
--
73, Cecil http://www.w5dxp.com