what happens to reflected energy ?
 

On Jun 25, 6:47*pm, Keith Dysart wrote:
Now this I agree with. Superpose volts, current and fields to your
heart's content. Just don't attempt it for power.

Nobody except you, using instantaneous power, has attempted to
superpose power. The Power-Density/Interference equation does NOT
superpose power because it contains the interference term. If the
interference term were omitted than it would be an attempt to
superpose power, but I have never omitted the interference term except
when it was zero.

You should study the design of the generator described previously and
repeated below, for your convenience:
Can you make your redistribution explanation work for this one?

Here's what your schematic looks like. How about a web page graphic
instead?

http://www.w5dxp.com/Keith.JPG
--
73, Cecil, w5dxp.com

what happens to reflected energy ?
 

On Jun 25, 7:07*pm, Keith Dysart wrote:
Using superposition, when you add Vrev2(tau) to Vfor1(rho) you get
zero. With zero voltage comes 0 energy transfer.

Completing your above sentence: With zero voltage comes 0 energy
transfer *in the direction of travel of the original waves that were
superposed*. Assuming that you believe in the conservation of energy
principle, what happened to the energy in the two component voltage
waves necessary for their existence before they cancel each other? If
they didn't contain any energy, they would be zero but we know they
are not zero, i.e. they are 35.7 volts each. That original wave energy
is redistributed and *transfered* in the opposite direction, the only
other direction available in a transmission line.

One cannot argue with a forked tongue that the superposed waves never
existed in the first place because that would violate the laws of
physics and superposition.

Do you really need rho^2 to understand what goes on in a transmission
line?

Not using rho^2 is why you are so confused. If you actually cared
where the energy goes, you would be forced to use rho^2 or at least
multiply the superposition component voltages and currents to obtain
the power in the superposition component wavefronts.

In the earlier example, the impedance discontinuity has a physical
voltage reflection coefficient of 0.7143 and a physical power
reflection coefficient of 0.51. If you consider the steady-state power
conditions, you will calculate a virtual power reflection coefficient
of 0.0 and a virtual voltage reflection coefficient of 0.0. Which
reflection coefficient is correct? Obviously, physical trumps virtual
every time.

The 50v source voltage reflected at the 0.7143 reflection coefficient
is 35.7 volts and it exists in a 50 ohm environment. Simple math
yields the power = (35.7)^2/50 = 25.5 watts. Where did the energy in
that 25.5 watt EM wave go? One can obtain the same value by
calculating the current: 1a(0.7143) = 0.7143. Power = 35.7(0.7143) =
25.5 watts.

So you can get by without using rho^2 but to determine where the
energy is going, one needs to at least multiply the EM traveling-wave
voltage by the EM traveling-wave current (or calculate the ExH
Poynting vectors).

In fact, this would be a good application for your instantaneous power
calculations. Where is the energy going that is in the instantaneous
power being reflected by the impedance discontinuity?
--
73, Cecil, w5dxp.com

what happens to reflected energy ?
 

On Jun 25, 3:27*pm, lu6etj wrote:
Sorry. Cecil, I do not catch you (final numeric example), would you
mind give to me a more explanatory/explicit answer?

I previously had a senior moment and changed contexts in the middle of
a posting and I apologize for any confusion. Would you enlighten me as
to the area of the discussion that you don't catch? Do you understand
physical reflection and transmission coefficients and their effect on
voltage, current, and power?
--
73, Cecil, w5dxp.com

what happens to reflected energy ?
 

On Jun 26, 7:41*am, K1TTT wrote:
On Jun 26, 12:22*am, Keith Dysart wrote:





On Jun 25, 7:30*am, K1TTT wrote:

On Jun 25, 7:46*am, lu6etj wrote:

On 24 jun, 17:54, K1TTT wrote:

On Jun 24, 3:25*pm, Cecil Moore wrote:

On Jun 24, 9:20*am, lu6etj wrote:

Oh, I'm so sorry Cecil, I should have written "However I can not
visualize a simple PHYSICAL mechanism/example to generate
such system in a TL". Anyway, your additional info it is very useful to
me. Thanks.

The physical mechanism is the Z01==Z02 impedance discontinuity with
its associated reflection coefficient, rho. We can see that reflection
on a TDR so it is indeed a PHYSICAL mechanism.
--
73, Cecil, w5dxp.com

don't forget the OTHER physical mechanism that is necessary,
superposition... the ability to add voltages, currents, and fields in
linear circuits and media.

I mentioned same comment in another post. We use superposition
principle in two different contexts. Superposition theorem in circuit
theory, and wave superposition. Wave (traveling) superposition deals
with f(t,x,y,z) and usually with puntual magnitudes, E, H, D, B, etc)
while circuit theory deals with a subset f(t) phenomena and with
integrated magnitudes (V, I). Sometimes that becomes a confused
issue :)

Miguel

NO, superposition is always the same. *it is the linear addition of
currents or fields in a linear media. *it works the same for circuits
as for em waves.

the big problem are the people who confuse the formulas for adding
powers with adding fields or currents/voltages and forget the phase
terms.

the other big problem is keith who seems to want to separate his waves
into separate time and space variables and leaves out the requirement
that wave functions must be dependent on both space AND time.
basically any solution to the wave equations derived from maxwell's
laws must be of the form f(t-x/v). *this leads him to the erroneous
conclusions he gets from trying to compare his batteries to wave
propagation. *this is the same problem people have with standing
waves, they have separate dependence on t and x, so they can't travel
and can't transport energy.- Hide quoted text -

I see that the stress induced by considering DC waves is causing you
to misinterpret my writings.

May I suggest an alternate exploration for you. Assuming that you
accept TDR and know how to use Reflection Coefficients to compute
voltage and current reflections, then consider what happens
when a rectangular pulse is launched from a matched generator in
to a transmission line. For simple reflection coefficients like
0, 1, and -1 compute the reflected pulse. For both the forward
and reflected direction compute the voltage and current on the line
before the pulse arrives, as it passes and after it has passed.

Compute the energy in the pulse, and how long a distance it
occupies on the transmission line. Compute the power as the
pulse is passing.

Be sure you know what happens to the pulse when it re-enters
the generator. For simplicity, assume a generator constructed
using the Thevenin circuit.

Make sure all the results are in agreement; especially, the
energy delived by the source and the energy dissipated in the
various resistors.

Now make the pulse longer and longer... until it looks like
a step function. And do the computations again.

Determine if the results agree with those I previously
presented for the DC example.

...Keith

PS: Barring errors, they will.

why would i want to do all that work? *

It would be an opportunity for you to deepen your understanding of the
behaviour of transmission lines.

there is no way that my answers will agree with your misconceptions. *

I am not convinced. You have not yet found any errors in my
expositions,
so if you do not make any errors, I expect we will agree on the
outcome,
though perhaps not on the interpretation, for you disagree when I say
"do not assign TOO much reality to the energy in reflected waves.

You seem to want your reflected waves to always transport energy, but
are unhappy that this leads to a line that was originally excited with
a step function having energy flowing in both directions even though
the current is zero all along the line.

Cecil simply sidesteps these little inconveniences by refusing to
consider anything other than sinusoidal RF excitation and by
refusing to consider any time based analysis. Such is not the path
to understanding, deep or otherwise.

you'll just come up with an even uglier generator to try to make it fit.

My generators are pretty simple. So far I have only used 3: Thevenin,
Norton, and one with an interesting constant input power
characteristic.

oh, and by the way, your fancy 2 generator and 2 resistor 'constant
power' source isn't what you think it is. *go back to basic circuits
101 and you will find that any linear network like that can be reduced
to either a simple one source one impedance norton or thevenin
equivalent. *

You have confused a bit, models with implementation. As I said in the
original: "Consider a generator constructed as below". I am not
using an equivalent circuit, but a construction. Only when dealing
with the actual construction is it valid to examine the internal
energy flows. An "equivalent" circuit is equivalent for external
behaviour but not necessarily for internal, so I avoid them when
examining the internals.

....Keith

what happens to reflected energy ?
 

On Jun 27, 5:49*pm, Keith Dysart wrote:
On Jun 26, 7:41*am, K1TTT wrote:



On Jun 26, 12:22*am, Keith Dysart wrote:

On Jun 25, 7:30*am, K1TTT wrote:

On Jun 25, 7:46*am, lu6etj wrote:

On 24 jun, 17:54, K1TTT wrote:

On Jun 24, 3:25*pm, Cecil Moore wrote:

On Jun 24, 9:20*am, lu6etj wrote:

Oh, I'm so sorry Cecil, I should have written "However I can not
visualize a simple PHYSICAL mechanism/example to generate
such system in a TL". Anyway, your additional info it is very useful to
me. Thanks.

The physical mechanism is the Z01==Z02 impedance discontinuity with
its associated reflection coefficient, rho. We can see that reflection
on a TDR so it is indeed a PHYSICAL mechanism.
--
73, Cecil, w5dxp.com

don't forget the OTHER physical mechanism that is necessary,
superposition... the ability to add voltages, currents, and fields in
linear circuits and media.

I mentioned same comment in another post. We use superposition
principle in two different contexts. Superposition theorem in circuit
theory, and wave superposition. Wave (traveling) superposition deals
with f(t,x,y,z) and usually with puntual magnitudes, E, H, D, B, etc)
while circuit theory deals with a subset f(t) phenomena and with
integrated magnitudes (V, I). Sometimes that becomes a confused
issue :)

Miguel

NO, superposition is always the same. *it is the linear addition of
currents or fields in a linear media. *it works the same for circuits
as for em waves.

the big problem are the people who confuse the formulas for adding
powers with adding fields or currents/voltages and forget the phase
terms.

the other big problem is keith who seems to want to separate his waves
into separate time and space variables and leaves out the requirement
that wave functions must be dependent on both space AND time.
basically any solution to the wave equations derived from maxwell's
laws must be of the form f(t-x/v). *this leads him to the erroneous
conclusions he gets from trying to compare his batteries to wave
propagation. *this is the same problem people have with standing
waves, they have separate dependence on t and x, so they can't travel
and can't transport energy.- Hide quoted text -

I see that the stress induced by considering DC waves is causing you
to misinterpret my writings.

May I suggest an alternate exploration for you. Assuming that you
accept TDR and know how to use Reflection Coefficients to compute
voltage and current reflections, then consider what happens
when a rectangular pulse is launched from a matched generator in
to a transmission line. For simple reflection coefficients like
0, 1, and -1 compute the reflected pulse. For both the forward
and reflected direction compute the voltage and current on the line
before the pulse arrives, as it passes and after it has passed.

Compute the energy in the pulse, and how long a distance it
occupies on the transmission line. Compute the power as the
pulse is passing.

Be sure you know what happens to the pulse when it re-enters
the generator. For simplicity, assume a generator constructed
using the Thevenin circuit.

Make sure all the results are in agreement; especially, the
energy delived by the source and the energy dissipated in the
various resistors.

Now make the pulse longer and longer... until it looks like
a step function. And do the computations again.

Determine if the results agree with those I previously
presented for the DC example.

...Keith

PS: Barring errors, they will.

why would i want to do all that work? *

It would be an opportunity for you to deepen your understanding of the
behaviour of transmission lines.

there is no way that my answers will agree with your misconceptions. *

I am not convinced. You have not yet found any errors in my
expositions,
so if you do not make any errors, I expect we will agree on the
outcome,
though perhaps not on the interpretation, for you disagree when I say
"do not assign TOO much reality to the energy in reflected waves.

You seem to want your reflected waves to always transport energy, but
are unhappy that this leads to a line that was originally excited with
a step function having energy flowing in both directions even though
the current is zero all along the line.

Cecil simply sidesteps these little inconveniences by refusing to
consider anything other than sinusoidal RF excitation and by
refusing to consider any time based analysis. Such is not the path
to understanding, deep or otherwise.

you'll just come up with an even uglier generator to try to make it fit..

My generators are pretty simple. So far I have only used 3: Thevenin,
Norton, and one with an interesting constant input power
characteristic.

oh, and by the way, your fancy 2 generator and 2 resistor 'constant
power' source isn't what you think it is. *go back to basic circuits
101 and you will find that any linear network like that can be reduced
to either a simple one source one impedance norton or thevenin
equivalent. *

You have confused a bit, models with implementation. As I said in the
original: "Consider a generator constructed as below". I am not
using an equivalent circuit, but a construction. Only when dealing
with the actual construction is it valid to examine the internal
energy flows. An "equivalent" circuit is equivalent for external
behaviour but not necessarily for internal, so I avoid them when
examining the internals.

...Keith

but the equivalent points out that your statements about it sourcing
constant power is incorrect.

i have also pointed out that your statements about your 'step wave'
are obviously incorrect because you have applied assumptions that are
only valid in the sinusoidal steady state to a step function that can
never be in steady state. neither can your pulses for that matter, so
all the assumptions are worthless, you must do the complete analysis
including the summations for the infinite fourier decomposition of
your step or pulses to get the full picture... in a transient
analysis. as was pointed out if you go VERY far into the future with
a battery connected to an open circuit piece of coax there can be no
currents and therefore no waves propagating in the line. its only in
the detailed transient analysis that you haven't done where you will
see the propagating steps going back and forth.

what happens to reflected energy ?
 

On Jun 27, 12:49*pm, Keith Dysart wrote:
Cecil simply sidesteps these little inconveniences by refusing to
consider anything other than sinusoidal RF excitation and by
refusing to consider any time based analysis.

That's simply false. Using Fourier analysis, I reduce anything other
than a sinusoid to multiple sinusoidal RF excitations, perform the
sinusoidal analysis, and then use superposition to find the answer. I
also reject any example where Maxwell's equations do not work. Your
insistance that magical waves can somehow exist during DC steady-state
violates the known laws of physics. EM waves CANNOT exist during DC
steady-state because electrons are traveling at a constant velocity.
You can measure DC voltage with an AC voltmeter but that doesn't
change DC voltage to AC voltage.
--
73, Cecil, w5dxp.com

what happens to reflected energy ?
 

On Jun 26, 9:05*am, Cecil Moore wrote:
On Jun 25, 4:00*pm, Keith Dysart wrote:

That's the time domain. Variation in the instantaneous energy flow.

What you seem to be missing is that the *energy content* of power
(total joules) must be conserved but the instantaneous power (joules/
second) does not have to be conserved as you have argued numerous
times in numerous examples.

In any region, the energy flowing in (i.e. power) to the region minus
the energy flowing out (i.e. power) is equal to the additional energy
per unit time (i.e. power) being stored in the region. While not
called the "conservation of power law" it is an obvious corollary
to "conservation of energy".

The only question that needs to be
answered is: In a system designed to eliminate reflections and
interference, does all the reflected energy eventually get dissipated
in the source resistor. The answer is yes because there is nowhere
else for it to go.

The obvious alternative is that the computed energy in the reflected
wave is sometimes just a figment. Or is something else happening with
the step function example?

Not to mention that in your 1/8 wavelength example (http://
www.w5dxp.com/nointfr.htm)
you do not explain where the energy is stored so that it can be
returned at a different time.

There is no conservation of power principle and
that includes instantaneous power. So it is irrelevant what/where
instantaneous power might do/go during a single cycle.

Such declarations do permit an easy out, despite not aligning with
reality.

Now I understand that instantaneous power dictates some physical
design considerations as in waveguides. But since instantaneous power
does not fall under the conservation of energy principle, it is simply
irrelevant to the present discussion. What happens over a complete
cycle is what is relevant.

If that is the case, the whole concept of reflected energy seems
somewhat bogus. Over a whole cycle, the power delivered by the
generator is passed on towards the load. If that is all you want
to know, then there is no need at all for "reflected power".

However, in any and every case, it is energy that is conserved,
not power.

Yes. But see the related corollary above.

How many joules are in that dt
sliver of time when the instantaneous power is 100 watts? It's those
joules that must be conserved, not the instantaneous power.

Still having problems with mapping the concepts from calculus to the
real world, I see.

You didn't answer my previous question. If you measure 100 watts of
instantaneous power at 100 places within an inch of each other, does
that mean there is 10000 watts of instantaneous power in that one inch
of wire? That is the only logical conclusion based on your argument
and assertions.

No more than "If you measure 100 watts of *average* power at 100
places
within an inch of each other, does that mean there is 10000 watts of
*average* power in that one inch of wire?"

But it is a way of thinking that you like to use to distract
yourself from the really interesting results.

Any argument based on the conservation of power is
doomed to fail. Please get real.

Please study the corollary above.

Not quite
'as useless as tits on a boar hog, or as Hecht said, putting it
mildly: "of limited utility"'.

One could argue that tits on a boar hog are not completely useless
and, therefore, instantaneous energy is exactly as useless (or exactly
as useful) as tits on a boar hog. (Hint: Without the existence of the
tit gene in the male, female hogs would probably not have tits.)

So which is it? Is instantaneous energy flow a useful concept? Or is
it
not? You previously suggested an understanding of the value (when I
mentioned "real power folk"), but seem to continue to want to argue
its complete lack of usefulness.

And to stop besmirching Hecht, it seems most probable that his
comment was in the context of optics. After all, the book had that
title.

....Keith

what happens to reflected energy ?
 

On Jun 26, 9:20*am, Cecil Moore wrote:
On Jun 25, 6:47*pm, Keith Dysart wrote:

You should study the design of the generator described previously and
repeated below, for your convenience:
Can you make your redistribution explanation work for this one?

Here's what your schematic looks like. How about a web page graphic
instead?

http://www.w5dxp.com/Keith.JPG

That does present a bit of a challenge to decipher.

Try using groups.google.com to read the message.

After opening the topic, scroll to the top and click "Options" which
should be found in the same header as the Topic. Click "Fixed font".

This significantly improves readability.

....Keith

what happens to reflected energy ?
 

On Jun 26, 9:49*am, Cecil Moore wrote:
On Jun 25, 7:07*pm, Keith Dysart wrote:

Using superposition, when you add Vrev2(tau) to Vfor1(rho) you get
zero. With zero voltage comes 0 energy transfer.

Completing your above sentence: With zero voltage comes 0 energy
transfer *in the direction of travel of the original waves that were
superposed*. Assuming that you believe in the conservation of energy
principle, what happened to the energy in the two component voltage
waves necessary for their existence before they cancel each other?

The fundamental question is: "did they have energy?"

Let us express this as a hypothesis:

Hypothesis 1: The component voltage waves have energy.

Then it should follow that we can trace this energy and discover
where it goes.

At least three examples have been proposed where the energy can
not be properly traced:

Example 1: Step function applied to a transmission line. After the
line settles, a forward and reflected voltage wave
continue on the line but no energy is being transferred.
Example 2: On a line with infinite VSWR no energy crosses a
voltage minimum or maximum.
Example 3: With the 1/8 wavelength line described in
http://www.w5dxp.com/nointfr.htm the energy can not be
properly accounted for on a moment by moment basis.

Only one counter-example was needed to disprove the hypothesis,
three have been found. There may be more.

Hypothesis is disproved.

No matter how many examples are found that support the hypothesis,
the hypothesis is still disproved.

....Keith

what happens to reflected energy ?
 

On 27 jun, 11:37, Cecil Moore wrote:
On Jun 25, 3:27*pm, lu6etj wrote:

Sorry. Cecil, I do not catch you (final numeric example), would you
mind give to me a more explanatory/explicit answer?

I previously had a senior moment and changed contexts in the middle of
a posting and I apologize for any confusion. Would you enlighten me as
to the area of the discussion that you don't catch? Do you understand
physical reflection and transmission coefficients and their effect on
voltage, current, and power?
--
73, Cecil, w5dxp.com

Thanks Cecil:

Examples of you that I saw in recent weeks were about interferences
generated by a single real generator and reflections, resulting, for
example, in constructive/destructive interference responsible of
changes in energy flow direction starting from the line point where
those interferences occurs. In a nutshell: In post 127 I asked you for
an example/experiment based in TWO real generators to assimilate it to
more familiar double slit interference phenomenom (TWO coherent
sources) rendering energy redistribution inside one dimensional TL
space, two sources in tridimensional space gives maximuns and nulls on
screen (redistribution). Well, I ask you for identical example in
unidimensional space rendering a phenomenom similar as reflection (one
wave) but with interference (two waves). Sorry I do not know how
better translate my question to english words.
I not catched your answer because it does not match my question :)

Thank you very much in advance. Miguel - LU6ETJ

PS: On Google interface, inside thread's tittle = options you can
select fixed text, to correctly see ASCII drawings.