View Single Post
  #219   Report Post  
Old April 19th 07, 10:32 PM posted to rec.radio.amateur.antenna
Dr. Honeydew Dr. Honeydew is offline
external usenet poster
 
First recorded activity by RadioBanter: Mar 2007
Posts: 10
Default Analyzing Stub Matching with Reflection Coefficients

On Apr 19, 12:44 pm, Cecil Moore wrote:
Dr. Honeydew wrote:


In other words,
viewed from both sides, show us even one instance where the system is
not correctly analyzed with your S11--S12 equations for the Z1--Z2
interface. Show us even one instance where those equations will not
tell you exactly what happens to waves coming into that interface from
either direction, and in fact from both directions at once.


You must have me confused with someone else. I'm a supporter
of the s-parameter analysis. It's others who have called
it "Gobbledegook" (sic).
--
73, Cecil, w5dxp.com


Ah, thank you. Then it follows directly from that, that when you
wrote,
A Bird wattmeter reads 100 watts forward and 100w reflected. The
current in the source is zero. The source is not only not sourcing any
forward power, it is also not sinking any reflected power.

you were not disallowing the fact that the source, which is matched to
the line, sucks up the entire reverse wave from the line. I'm happy
to see we agree on that little point after all.

Fallacies of the, "if the source absorbs the reverse wave, it must be
dissipated as heat," school: Start with a source which is a voltage
generator, 141.4Vrms sinewave, in series with a 50 ohm resistor.
Connect it to a 50 ohm load, and you have 100 watts dissipated inside
the source and 100 watts dissipated in the load. Put a 50 ohm
lossless line between the source and the load. We now have a
70.71Vrms wave from the source to the load, and no reflected. Assume
the line is 1/4 wave long; replace the load with a short. A moment
later, steady state is reached and the dissipation in the source's 50
ohm resistor dropped from 100W to 0W. We may be tempted to say that
source did not absorb the return wave and is no longer supplying the
forward wave. But that requires complete reflection of the return
wave at the interface between the source and the line. Ah, but there
is an infinite set of conditions under which the same should be
true. What if we make the line 1/2 wave long, still shorted at the
far end? In steady state, the source is still apparently delivering
no power to the line, but now, instead of it dissipating NO power,
it's suddenly dissipating 400 watts! In the line, though, we still
see 100 watts of forward power, and 100 watts of reverse power.

Ooops. The "if the source absorbs the reverse wave, it must be
dissipated as heat" school ... better go back to school. Or to the
lab. Or SOMEwhere else, till they get it figured out. It didn't hold
water when Dr. Slick (remember him?) tried to push it on us, and it
doesn't hold water now.

More fun: the guts of my source are now a 282.8V source and the 50
ohm resistor, feeding a fairly long piece of 50 ohm transmission line
to the front panel connector. The line has 6.02dB loss. Now back to
the original situation with 1/4 wave of 50 ohm line, shorted at the
far end, connected to the generator's output connector. NOW are you
going to say that the reverse wave on the 1/4 wave line stops when it
gets back to the generator? What if I put even a couple inches of
line between the guts of the source and the front panel connector,
does the reverse wave on the external line stop when it gets to the
front panel connector?

I repeat: [In the case of a line connected to a source, with the
impedance of the two matched--not conjugate] It's obvious that the
source is sourcing the forward voltage wave, and it's sucking up the
entire reverse voltage wave from the line.

What happens to that reverse voltage wave inside the source depends on
what's in the guts of the source.

From the lab,

Bunsen