Steve Nosko wrote:
From this I infer that Cecil believes that once the transient portion of
the response has concluded you are in the steady state and no more
reflections are occurring. Is this a correct re-statement of your belief,

No, no, no. That is a re-statement of the other side of the argument
from mine. I often use devil's-advocate type arguments.

The definition of steady-state by the other side is pure unvarying
sine waves with no noise and no modulation. Then the reflections
sorta disappear into a steady-state mush of standing waves. A
modulated TV signal, according to their argument, is not a steady-
state signal. Within this steady-state mush of standing waves, energy
never makes it from the load back to the match point. Apparently,
because of the uncertainty principle, reflected energy doesn't
actually exist anywhere until it is radiated or dissipated, i.e.
its probability wave collapses.

I have challenged them to produce a standing wave without a rearward-
traveling wave but nobody has been up to that challenge.
--
73, Cecil http://www.qsl.net/w5dxp




-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

"Steve Nosko" wrote in message
...

"Tam/WB2TT" wrote in message
...

"Henry Kolesnik" wrote in message
.. .
Richard
...I got
out the book REFLECTIONS II by Walt Maxwell W2DU. I'm typing
verbatim
from
page 2-2 and 23-1

I paraphrase / quote the quote:
Power reflected from a mismatch back into the PA is not absorbed
there.
He writes:
"when the pi-network tank is tuned to resonance, a virtual
short circuit to rearward traveling waves is created at the input
of the network. Consequently, instead of the reflected power
reaching the tubes of the amplifier, it is totally re-reflected
toward the load by the virtual short "

Hank, I'm sorry, but I believe this is not correct as stated. The
word
"totally", I believe is misleading and contrary to what I think all will
agree. Namely that there MUST be SOME real part (although the true amount
is disputed here) to the Zout of the PA and therefore SOME power MUST be
absorbed there. I believe the point of contention truely is just how much
is absorbed and how much is reflected...

...Also what happens in a transistor final with no pi?

AGAIN. Be careful that you DO NOT keep assuming that the full power is
the
incident power. Incident meaning forward power or that which the
transmitter is sending toward the load. I have a real corker below. Be
careful. Do not try this at home, I am a professional. (yea, I know
this'll
spawn all kinds of grief)


Tam sez:
Henry,
Here is an example of what you just said. Take a sine wave source, and
connect it to a 1/4 wave section of shorted transmission line through a
series resistor R. The reflected wave will reach this resistor 1/2 cycle
later, and will be in phase with the source. For a lossless transmission
line, there will be *0 Volts across the resistor*. There will be 0
current
through the resistor, and the reflected wave will be re reflected for
all
values of R, including R=Z0, because the reflected wave will not "know"
what
R is. You can get the same answer from knowing that the impedance
looking
into a 1/4 wave section of shorted transmission line is infinite.
Tam/WB2TT

Tam,
Although I believe you have digressed somewhat, I will follow this
path
since it contains a closely related concept. You obviously have a pretty
good handle on much of this (as others do, up to a point, and struggle to
make their mental models fit the reality - or to make other's mental
models
fit theirs)...However, there is a paradox here which appears to be the
root
of this disagreement and your example hit it right on the head. You have
some implicit assumptions here.

1) Zero volts across the resistor = (not said, but implied) the t-line
acts
like an open at the input end, therefore there is no current into the line
which results in the zero V across R. V=I^2 * R. I think everybody will
agree with this. (shorted 1/4 wave acts like an open and an open 1/4 wave
looks like a short - mantra of all, no?).

2) Resistor current = zero therefore reverse traveling wave gets reflected
toward the load end. The implication clearly is that with zero resistor
current, no energy can be flowing out that way, so it must be reflected.

HOWEVER...
Been a while since I went to this depth and interesting to do so,
though
unnecessary, I will anyway...

If the input to this stub acts like an open, there can be NO current,
thus no power entering, therefore there can be no forward wave, no
reflected
wave and no summation of waves to make the open in the first place and no
need to re-reflect the reflected wave from the resistor who (or is it
whom)
has no current. This appears to be the root cause of this problem.
Now, we can say that (and I think it has been said) that it makes no
difference how large these two waves (which cancel each other to form the
open circuit at the stub input) can be any absolute magnitude. 1 amp 10
amps 100 amps doesn't matter - they cancel. So what are they really?
There are two things to consider as you work out how you will resolve
this paradox in your mental model.
1) I believe the most important -- If the Z looking in to the stub is
high,
how can you send a large amount of power down the line. If you say it is
an
infinite Z then you have a really big problem explaining how it got there
in
the first place.
2) There is a tendency to assume that the forward power is the same power
as
when the load is not a short, but Zo.

Interesting puzzle, but I don;t need to go further. At some point you
must
'believe' something and I can comfortably stop there...for now.

Oh well...
--
Steve N, K,9;d, c. i My email has no u's.


Steve,
I think there is an analogy here to a high Q resonant parallel tuned
circuit. It takes very little to exite this, but there can be large
circulating current in the L and C.

BTW, on my shorted transmission line: I had looked at the current on the
transmission line - at the wrong end of the TL. There is current at the
shorted end, none on the source end in steady state. Didn't even know the
program was smart enough to know the difference. Seems to me that once you
have put all the energy into the TL that it can store, you don't have to
supply any more energy (?)

Tam/WB2TT

Steve Nosko wrote:
I paraphrase / quote the quote:
Power reflected from a mismatch back into the PA is not absorbed there.
He writes:
"when the pi-network tank is tuned to resonance, a virtual
short circuit to rearward traveling waves is created at the input
of the network. Consequently, instead of the reflected power
reaching the tubes of the amplifier, it is totally re-reflected
toward the load by the virtual short "

Hank, I'm sorry, but I believe this is not correct as stated. The word
"totally", I believe is misleading and contrary to what I think all will
agree. Namely that there MUST be SOME real part (although the true amount
is disputed here) to the Zout of the PA and therefore SOME power MUST be
absorbed there. I believe the point of contention truely is just how much
is absorbed and how much is reflected...

Steve, what you are missing is that the match point is not at the transmitter
terminals and reflections indeed do flow into the transmitter terminals. The
match point is between the tube finals and the input to the pi-net tuning
section. There are zero reflections between the input of the pi-net tuning
section and the tube finals. This is what Walter Maxwell was saying. The
"input of the network" that he is talking about is the wire between the
tube finals and the built-in pi-net tuning section and, indeed in a properly
tuned transmitter, there will be no reflections at that point which was
commonly known as a "Zg-match point".

All of Walt's writings assumed a Zg-match internal in the tube transmitter
at which all reflections were canceled. We hashed that out about 20 years
ago. That's when I first communicated with Walt through snailmail.
--
73, Cecil http://www.qsl.net/w5dxp



-----= Posted via Newsfeeds.Com, Uncensored Usenet News =-----
http://www.newsfeeds.com - The #1 Newsgroup Service in the World!
-----== Over 100,000 Newsgroups - 19 Different Servers! =-----

Cecil Moore wrote:

Steve Nosko wrote:
From this I infer that Cecil believes that once the transient portion of
the response has concluded you are in the steady state and no more
reflections are occurring. Is this a correct re-statement of your belief,

No, no, no. That is a re-statement of the other side of the argument
from mine.

Perhaps he thought you believed that because you're the only one who has
said it?

73, Jim AC6XG

Tam/WB2TT wrote:

SWCAD from Linear. I also built up a model of an ideal SWR meter.

Tam

I almost bought LLTC when they came out with that. Should have. Thanks
Tam.

jk

Tam/WB2TT wrote:
Seems to me that once you
have put all the energy into the TL that it can store, you don't have to
supply any more energy (?)

Tam/WB2TT

BINGO!! No more _energy_ flows - until or unless something changes.
(TV signals are of course always changing.)

73, Jim AC6XG

Jim Kelley wrote:

Cecil Moore wrote:

Steve Nosko wrote:
From this I infer that Cecil believes that once the transient portion of
the response has concluded you are in the steady state and no more
reflections are occurring. Is this a correct re-statement of your belief,

No, no, no. That is a re-statement of the other side of the argument
from mine.

Perhaps he thought you believed that because you're the only one who has
said it?

It's a very condensed verion of the reams of stuff that some posters have
posted over the past year. In particular, when I tried to introduce noise
and modulation to prove my point, I was told that noise and modulation are
not allowed during steady-state.
--
73, Cecil http://www.qsl.net/w5dxp

On Wed, 26 May 2004 11:44:00 -0500, "Steve Nosko"
wrote:

I think "superposition" is ther wrong term for this. If I could spell
reciprocity, I'd say that it is a better word.
Digression, though minor: Superposition refers typically, to the
application of two signals to a circuit and the resulting response being the
sum of the individual responses. no?

No/Yes. The "two" signals are those I described, the one from the
source as source and the one from load as source. This is the
convention of Circuit Analysis by Mesh or Nodes. Kirchhoff mandates
all voltages will sum to zero. The summing is a solution of
simultaneous equations - Superposition.

Now some meat I'd like to pursue...

...a 500 Ohm Source...you will never launch much power into a 50 Ohm
system because it would immediately hit a reflective interface at the
antenna connector.

There's a question I see here as to wheather the power was "launched" in
the first place and this may be part of the assumption (or set thereof)
which (my gut feel says) leads the logic astray.

It was an aside (a stage whisper).

This means that a 500 Ohm source, when confronted
by power going towards it from a 50 Ohm system will reflect most of
that power (but how did we get this power into the system in the first
place - Karma?) This, by design, will never happen in any transistor
ham rig built in the last quarter century.

I think this is true...BUT let's ask this same question of a 5.0 ohm
source?? How would you carry this through Richard C? I submit that THIS
source CAN launch significant power down the line.

And here we come to issues of initial conditions and their variation
to force a conclusion. Namely, the unstated condition is how much
voltage is available? SWR is the diagnosis, and absorption of power
is the issue. Launching of power also suffers from the same
condition, but it would be well enough to examine the reflection
first. Launching problems is a sidebar.

The answer is the transmitter source Z is 50 Ohms at rated power.

OK, so now I see you are of the "Zs=50 ohms" camp.

I have measured such things professionally and certified such facts as
accurate with a chain of references back to the National Bureau of
Standards. Such is my camp.

If a watt of power is chooglin' down the line toward it, that 50 Ohms is
going to dissipate into a watt worth of calories. This can be argued
with wave mechanics, or lumped circuit equivalents - doesn't matter
because it's all the same calories.

All ok with me _IF_ Zs=50. I'm not convinced however. I have to think
more about the modern broad-band transistor PA to have an strongly arguable
opinion on this concept.

It is a convention of design that is covered in many design texts.
Academe takes this (Zs=50) for granted so much so that it is rarely
discussed. Even here I have references (Academic) that make this
point explicit and I have offered them here in debate.

what do we need tuners for?

I think the answer here is that if the transmitter can take the strange
impedance it sees, then it isn't needed. Like Hanks 30L1 that, however,
couldn't take it very well.

However, this is not an answer to the question.

Furthermore, there is an implicit assumption here in your reasoning
which I believe is key and leading to some error. There is an unspoken
assumption that the power which makes it to the antenna after all the "
round-and-round until the antenna finally radiates it" is the FULL output
and I believe it AIN'T.

As I have never stated this, you have to answer for the conflict that
follows.

It is that little dribble that made it into the
line in the first place. a.k.a. the Tx is not generating the full output.
I think this assumption is causing much trouble.

This begs the question: What is the full output? A battery sitting on
the shelf is not generating its full output is a simple example, but
does not tell us what that full output may be. This is the problem of
definition by negatives.

It is not generating:
1W
nor
10V
nor
100°C
nor
1 H.P.

Was it ever capable of those generations in the first place? Under
what conditions? Saying what is "not" has no useful information here.

If we are discussing what happens to power impinging upon the source,
it should be obvious that if that power is OUT OF PHASE trouble
follows in the form of currents. If that power is IN PHASE trouble
follows in the form of potentials. Trouble is proportional to the
degree of mismatch. Trouble arrives through the model of waves,
trouble resides in the model of lumped equivalent circuits. It is the
same trouble either way in the form of heat. Choose your poison
because equipment manufacturers have built-in foldback mechanisms and
advise you to use a tuner for good reason.

Unless, of course, I have been smoking the wrong brand.

It will be available over the counter in British Columbia soon.

73's
Richard Clark, KB7QHC

Jim Kelley wrote:

Tam/WB2TT wrote:
Seems to me that once you
have put all the energy into the TL that it can store, you don't have to
supply any more energy (?)

BINGO!! No more _energy_ flows - until or unless something changes.
(TV signals are of course always changing.)

You can prove your point only by demonstrating an RF storage
battery in which RF energy stands still.

Got news for you, Jim. RF energy *always* flows because it *cannot*
stand still. It flows in one direction in a transmission line, at the
speed of light, until it encounters a physical impedance discontinuity
at which point reflections occur and some or all of the energy changes
directions, but always moving at the speed of light. ExH is the power.

Of course, you are talking about *NET* energy above. Forward
and reflected energy continue to flow unabated. RF component
energy has no choice except to flow at the speed of light. If
there doesn't exist a match point to turn the reflected energy
around, it *will* flow into the source. That you subtract the
reflected energy from the forward energy at the source output
to obtain the "generated energy", is just a math shortcut and
bears no resemblance to reality. It is a definition created by
man for the purpose of avoiding a very complicated problem.

It is akin to the problem of absolute motion in space-time.

It follows that when you have the same bank balance at the beginning
of the month and the end of the month, you tell everyone that you
had no debits or credits during that month. Again, you forgot to use
the word "NET".
--
73, Cecil http://www.qsl.net/w5dxp

Cecil Moore wrote:

[snip]

The definition of steady-state by the other side is pure unvarying
sine waves with no noise and no modulation.

[snip]

I have challenged them to produce a standing wave without a rearward-
traveling wave but nobody has been up to that challenge.

73, Cecil http://www.qsl.net/w5dxp

Cecil,

I don't plan to enter this endless argument about transmission lines,
but there are a couple of points you might want to consider.

* Steady state simply means that a phenomenon has no inherent reference
to its time of origin. One cannot observe a steady-state wave and
determine any clue about when it started. It contains no absolute time
markers. A transient, on the other hand, contains direct reference
information about when it was formed. There is no requirement that the
steady state phenomenon is a simple sine wave.

Many problems in math, science, and engineering exhibit both transient
and steady state solutions. It is not clear why there is any confusion
at all in this thread.

* Standing waves don't require the pre-existence of traveling waves. A
straightforward application of Maxwell's equations with appropriate
boundary conditions for the physical environment will lead to standing
waves directly. Check out one of your recently quoted favorite authors,
J. C. Slater, and look at some of the resonant cavity stuff. Most
definitely standing waves, but you will find lots of complex Bessel
functions and darn few traveling waves.

In virtually every transmission line situation proposed on RRAA, whether
tuned antenna feed lines or quarter-wave stubs, the transmission line is
resonant. It is possible to consider the resulting standing waves as the
sum of two traveling waves, but it is equally valid to consider the
transmission line as a simple resonator. The physically measurable
instantaneous voltages and currents are precisely the same whether one
considers oscillation from capacitive to inductive energy storage in the
line or the sum of two counter-traveling waves.

You often refer to wave-particle duality and to the use of S-parameters
instead of lumped circuit analysis. The world of science and engineering
is absolutely filled with this sort of dual description for physical
phenomena. In many cases one approach will be more convenient or more
intuitive, but that does not make the dual approach less valid.

There's more than one way to skin a cat.

73,
Gene
W4SZ