Analyzing Stub Matching with Reflection Coefficients
 

On Apr 15, 3:07 pm, Walter Maxwell wrote:
On 15 Apr 2007 14:33:40 -0700, "Jim Kelley" wrote:





On Apr 15, 6:53 am, Walter Maxwell wrote:
On Sun, 15 Apr 2007 19:23:56 +1000, Alan Peake
It is interesting to look at a single short pulse propagating along the
TL. At the stub point, the pulse must encounter a discontinuity in
impedance and therefore there will be a reflection. This can been seen
on a TDR. So there is a real reflection from a stub regardless of
whether or not it is a virtual short.
Alan
VK2ADB

I thank you for that, Alan, because, to continue, when the pulse is replaced with a sine wave, there is also a
reflection from the stub.

Hi Walt -

Begging your pardon, but don't TDR's examine the transient response of
a system, rather the steady state response?

ac6xg

You're correct, of course, Jim, but I was intuitively assuming we'd not be continuing the use of the TDR with
the sine wave signal. I'm sure my intuition wasn't communiated, sorry.

Walt- Hide quoted text -

- Show quoted text -

I guess I may have been 'intuiting' too much, myself. Since virtual
shorts and opens only appear in the steady state, I wouldn't expect
pulses to reflect off of them. I don't expect sine waves to reflect
off of them in steady state either for that matter, but that remains a
point of contention apparently.

73, Jim AC6XG

Analyzing Stub Matching with Reflection Coefficients
 

Jim Kelley wrote:
Cecil Moore wrote:

Jim Kelley wrote:
Roy is absolutely right, Cecil. Interact is a very poor choice of
terms in this discussion.

Roy did NOT say "interact" was a poor choice of terms.

That's correct. I said that interact is a poor choice of terms.

But you implied that is what Roy said just above.

chose to use it as did Hecht. Hecht says waves interact.
Roy says they don't interact.

As I said, Roy is correct.

Roy is right and Hecht is wrong??? Shirley, you jest.
Remember, Roy is the guy who stands by his use of
standing wave current to measure phase shift through
a loading coil. Phase shift in standing wave current
doesn't exist in a wire or in a coil or anywhere else.

And the funny thing is, you say that even you know of instances in
which the net fields are zero, and yet the waves propagate beyond that
point.

Where do the reflected waves go that propagate beyond that
point and are measured as zero amplitude by a Bird wattmeter?
Dr. Best said those zero energy canceled waves propagate right
into the source. What effect do waves of zero energy have? Are
you making that same stupid assertion? Watch out! Here comes
another one of those zero energy waves - good grief, look
at the size of that zero energy wave - it must be infinite.
What is infinity times zero? :-)

If you would wade through
the S-Parameter analysis with me, you would understand.

I think you just like to argue.

No, I honestly think we would pinpoint our differences. But,
of course, you would never agree to such.

If the S parameter analysis addressed where you are going wrong, then
that might be worthwhile.

Well then, let's do it and you can show me exactly where
I am going wrong. When I realize that I am wrong, I am the
first to admit it. My goal is to learn and I don't learn
anything new when I am right.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Dear Group:
Wise words from Tom and Roy.

Roy's nincompoop was not one of my students. Last week I taught about
simple op-amps and their limitations. The terms "virtual ground" and
"virtual short" (as used with ideal and almost-ideal op amps) were presented
only to facilitate my student's communication with others. Use of excess
verbiage and names is a poor idea that should be relegated to "Social
Science," some parts of which can do little other than classify. The idea
that an op-amp strives to have V- be very close to V+ is clean. Following
with the BW and slew rate limitations enhances the idea.

I first encountered the intransigence of labels in lieu of substance
when a new faculty member. My department head went ballistic when I started
a lecture on LC oscillators with the observation that they were all the same
with an LCL or CLC circuit causing a necessary phase shift and that one
could choose to connect to common any one of the tube's elements. Oh no,
said my department head, you have to show them how a Hartley differs from a
..... and he proceeded to recite the long list of named oscillators. -- I
now omit the consequences of that encounter, for children may be reading and
they are too young to learn the details of the blood sport known as academic
politics. It took many years, but in the end I won.

Names can have great utility. Simplification is, in my experience, a
necessary part of effective teaching and learning. However, I explicitly
say to my students that the art of teaching engineering resides in not
telling the student everything in one gulp. Current textbooks tend to
present everything in one gulp.

Thus we come back to the issues at hand. Things mean just what they are
defined to mean. Things are not necessarily the same as their name nor the
same as things with similar names. For once, wise and good people are
having a dialog. It seems to me to come down to this: "a virtual short" in
a transmission line is a place where waves (current, voltage) interact to
produce a small voltage and large current. A "physical short" in a
transmission line is a place where a conductor (or the like) is placed such
that the voltage is zero and the current is large. Most often, when
"physical shorts" are found they are near the ends of transmission lines.

One characteristic of a "virtual short" is that its presence or location is
dependent on frequency. Another characteristic is that signals are expected
to exist on both sides of a "virtual short." One characteristic of a
"physical short" is that it does not depend on frequency. Another
characteristic of a "physical short" is that signals exist on only one side
of the "physical short's" location.

The last seems to me to distill what is being said and what is
understood. While it may not have happened to you, every time I have
ventured to correct my brilliant and wise wife we have, after a while,
discovered that we were in agreement all along and words were attenuating
that understanding. A certain English SK would observe that an English
writer of mathematics and children's books had these ideas down a long time
ago.

73, Mac N8TT

--
J. Mc Laughlin; Michigan U.S.A.
Home:
"Roy Lewallen" wrote in message
...
K7ITM wrote:
. . .
The analogy may not be prefect, but I think it's a lot like the
usefulness of the idea of a "virtual ground" at the inverting input of
an op amp. But it's a virtual ground only under specific conditions:
strong negative feedback is active, and the non-inverting input is at
(AC, at least) ground potential. For it to be a useful concept
without too many pitfalls, the person using it has to be aware that
the conditions that make it a good approximation don't always hold.
Similarly for a "virtual short" on a line.
. . .

Let me relate a story. . .

Years ago at Tektronix, I transferred to a different group. Across the
aisle was a very bright engineer, fresh from school with a Masters or
PhD degree -- I don't recall which. I recall that his advanced education
had specialized in nonlinear control systems, very much a mathematically
complex and challenging field. His entirely academic background had been
very different from mine, so I was often enthralled by his attempts to
reconcile reality with the idealized world he had, until very recently,
occupied.

One day I found him muttering, trying this and that, until he asked for
some help with his test setup. He was driving an inverting op amp
circuit with a square wave, and he was seeing sharp pulses at the
"virtual ground" summing junction with his 'scope. He had tried moving
his probe grounds, replacing the op amp, bypassing, and everything else
he could think of, to rid his display of this obvious erroneous
artifact. The voltage at the summing junction, he explained, should
always be zero, since it's a virtual ground. Those spikes shouldn't be
there.

I tried to explain to him how a "virtual ground" was created: An input
signal initially generates a voltage at the op amp input. The op amp
responds by sending an inverted signal back to the summing junction
which adds to the initial voltage to produce very nearly zero at the
input. I explained that it can never be zero, but at best is the op amp
output voltage divided by its open loop gain. But more to the point, the
op amp isn't infinitely fast, so it takes some time for it to respond to
that initial voltage or any changes. And during that lag, the summing
junction voltage can move a great distance from zero. So the spikes are
occurring during the time it takes the op amp to respond to changes in
the input stimulus.

Well, he didn't get it. He just couldn't make the transition from the
idealized, infinitely fast and infinite gain op amps of his academic
models to the real things he had to work with. And looking back on it, I
think the basic problem was that he never really fully understood just
how that virtual ground came about even in an idealized world.

After a number more frustrating and unresolved collisions with reality,
he wisely quit and got a teaching job. I'm sure he did well in the
academic world.

Those many models we use daily to keep calculations, concepts, and
analyses manageable are just that -- models. It's imperative to
constantly be aware of the range over which those models are valid, and
alert to any situation which might make the model invalid. People
solving routine problems can, unfortunately, often get along for a long
time without realizing the limitations of their models, and can be
lulled into a belief that they're not models at all but reality. But in
the environment where I've spent most of my time, this carelessness
leads you very quickly into places which can be very difficult to get
out of.

Roy Lewallen, W7EL

Analyzing Stub Matching with Reflection Coefficients
 

On Apr 15, 7:03 pm, Cecil Moore wrote:
When I realize that I am wrong, I am the first to admit it.

Has the former ever happened?

If you are looking for an opportunity to demonstrate the
veracity of your statement, 'realize' that matching the source
impedance to the line prevents re-reflection, and then be
'the first to admit it'.

....Keith

Analyzing Stub Matching with Reflection Coefficients
 

Keith Dysart wrote:
On Apr 15, 7:03 pm, Cecil Moore wrote:
When I realize that I am wrong, I am the first to admit it.

Has the former ever happened?

It happens all the time. Then people accuse me of changing
sides in the middle of an argument. I can't win.

If you are looking for an opportunity to demonstrate the
veracity of your statement, 'realize' that matching the source
impedance to the line prevents re-reflection, and then be
'the first to admit it'.

I will admit it when you prove that "matching the source
impedance to the line" prevents reflections from the
source. You have to run bench experiments to prove
that - no 10 cent resistor hand-waving allowed.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

On Apr 15, 8:08 pm, Cecil Moore wrote:
I will admit it when you prove that "matching the source
impedance to the line" prevents reflections from the
source. You have to run bench experiments to prove
that - no 10 cent resistor hand-waving allowed.

Ahhh. A challenge. I like it.

But first, we need to agree on an experiment that you will
find convincing. We don't want any wiggle room after the
work is done, do we?

So I propose the following:

A signal generator with 50 Ohm output impedance is connected
to the left end of a length of 50 Ohm line and another signal
generator, also with 50 Ohm output impedance is connected to
the right end of the same line.

The signal generator on the left is set to frequency Fleft
and the one on the right is set to a different frequency
Fright. The line is appreciable fraction of 1 wavelength
long at Fleft and Fright.

Step 1 - Replace the signal generator on the right with a
50 Ohm terminator
Step 2 - Observe the signal at the left and right end of
the line. The signal at the right will be a delayed
and possibly reduced copy of the one on the left.
Step 3 - Replace the signal generator on the left with a
50 Ohm terminator.
Step 4 - Observe the signal at the right and left end of
the line. The signal at the left will be a delayed
and possibly reduced copy of the one on the right.
Step 6 - With both generators operating, observe the signal
at the left and right ends of the line.

MY expected result: If no reflections are occurring then the
signal at the left will be the sum of the signals observed at
the left in Step 2 and Step 4, while the signal at the right
will be the sum of the signals observed at the right in Step 2
and Step 4.

If any reflections have occurred, the reflection will modify
the signal at the generator end (for that particular frequency)
and MY expected result will not occur.

Does this cover it?

If MY expected result occurs, you will accept that 10 cent
resistors in generators will prevent re-reflections, "realize
that you are wrong and be the first to admit it."

What say ye?

....Keith

Analyzing Stub Matching with Reflection Coefficients
 

Keith Dysart wrote:
A signal generator with 50 Ohm output impedance is connected
to the left end of a length of 50 Ohm line and another signal
generator, also with 50 Ohm output impedance is connected to
the right end of the same line.

No, you miss the point. You need to prove your
assertions using an ordinary commercial amateur
radio transceiver, like an IC-706.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

On Apr 15, 10:30 pm, Cecil Moore wrote:
Keith Dysart wrote:
A signal generator with 50 Ohm output impedance is connected
to the left end of a length of 50 Ohm line and another signal
generator, also with 50 Ohm output impedance is connected to
the right end of the same line.

No, you miss the point. You need to prove your
assertions using an ordinary commercial amateur
radio transceiver, like an IC-706.

So, out of curiosity, what do you think the outcome of my
experiment would be?

Do 10 cent resistors ever work? Or is a circulator always needed
to prevent re-reflections?

....Keith

Analyzing Stub Matching with Reflection Coefficients
 

On Sun, 15 Apr 2007 18:30:59 -0500, "J. Mc Laughlin"
wrote:

One characteristic of a "virtual short" is that its presence or location is
dependent on frequency. Another characteristic is that signals are expected
to exist on both sides of a "virtual short." One characteristic of a
"physical short" is that it does not depend on frequency. Another
characteristic of a "physical short" is that signals exist on only one side
of the "physical short's" location.

Hi Mac,

It's a shame no one has offered you kudos for such a telling and
succinct observation. It stands there balanced with another book-end:

A certain English SK would observe that an English
writer of mathematics and children's books had these ideas down a long time
ago.

No doubt this is a delicious reference to how these threads approach
an agony in 8 fits: "What I tell you three times is true."

73's
Richard Clark, KB7QHC

Analyzing Stub Matching with Reflection Coefficients
 

Jim Kelley wrote:


I guess I may have been 'intuiting' too much, myself. Since virtual
shorts and opens only appear in the steady state, I wouldn't expect
pulses to reflect off of them. I don't expect sine waves to reflect
off of them in steady state either for that matter, but that remains a
point of contention apparently.

73, Jim AC6XG


Actually Jim, virtual shorts etc. act the same for pulse systems as for
CW systems. The classic case is the rotating joint in radar systems.
Alan

Analyzing Stub Matching with Reflection Coefficients
 

Begging your pardon, but don't TDR's examine the transient response of
a system, rather the steady state response?

ac6xg


You're correct, of course, Jim, but I was intuitively assuming we'd not be continuing the use of the TDR with
the sine wave signal. I'm sure my intuition wasn't communiated, sorry.

Walt

Of course, real signals aren't just a single pulse but any CW signal can
be represented as a series of pulses. One VNA I used years ago (Wiltron)
allowed you to analyse a network by driving it with pulses ,
capturing the pulse or transient response and then doing a transform to
go from the time domain to the frequency domain, thus producing the
steady-state response. Very handy for analysing antennae in a confined
space. The time window was set to exclude reflections from walls etc. so
one didn't need an anechoic chamber.
Alan

Analyzing Stub Matching with Reflection Coefficients
 

On Apr 16, 2:30 am, Alan Peake
wrote:
Jim Kelley wrote:

I guess I may have been 'intuiting' too much, myself. Since virtual
shorts and opens only appear in the steady state, I wouldn't expect
pulses to reflect off of them. I don't expect sine waves to reflect
off of them in steady state either for that matter, but that remains a
point of contention apparently.

73, Jim AC6XG

Actually Jim, virtual shorts etc. act the same for pulse systems as for
CW systems. The classic case is the rotating joint in radar systems.
Alan

But isn't that a number of cycles of RF?
Enough to reach 'steady-state'?

When I read the word 'pulse', I think rising edge, flat top,
falling edge, and the virtual short is quite different than
a real short for this signal.

....Keith

Analyzing Stub Matching with Reflection Coefficients
 

Keith Dysart wrote:
So, out of curiosity, what do you think the outcome of my
experiment would be?

With an IC-706? I don't know. Others have tried it with
varying results.

Do 10 cent resistors ever work? Or is a circulator always needed
to prevent re-reflections?

Your 10 cent resistor can be thought of as a low dB
pad of sorts. It will attenuate but not eliminate
re-reflection. Again, let me remind you of Ramo &
Whinnery's warning not to attach importance to
what is calculated to happen inside an equivalent
source. There are models available for virtually
any amplifier you might choose but I don't know
how those models handle reflections.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Alan Peake wrote:
Of course, real signals aren't just a single pulse but any CW signal can
be represented as a series of pulses.

I think within the present context, we would say a CW dot
or dash has a transient response at the leading edge and
trailing edge and achieves steady-state in the middle.
I once calculated that it takes about ~30 cycles with
rho=0.707 to get very close to steady-state.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Alan Peake wrote:
But I still can't see that a virtual short would be
different to a real short.

A virtual short is (Vfor-Vref)/(Ifor+Iref) = 0
where |Vfor| = |Vref|

For the virtual short to exist, two equal magnitude
EM waves have to be *flowing through* the virtual
short. EM waves *cannot flow through* a real short.

For real shorts:
V/I=0 is a result caused by the real short.

For virtual shorts:
A virtual short is a result caused by V/I=0.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Actually Jim, virtual shorts etc. act the same for pulse systems as for
CW systems. The classic case is the rotating joint in radar systems.
Alan


But isn't that a number of cycles of RF?
Enough to reach 'steady-state'?

When I read the word 'pulse', I think rising edge, flat top,
falling edge, and the virtual short is quite different than
a real short for this signal.

...Keith

I see your point. But as a CW signal can be thought of as series of
rectangular pulses,
then the effect of a virtual short/open/whatever, should be the same.
For a such a series,
the resultant signal at any point in the TL is of course quite different
from a single pulse. But I still can't see that a virtual short would be
different to a real short.
Alan

Analyzing Stub Matching with Reflection Coefficients
 

Cecil Moore wrote:
Alan Peake wrote:
But I still can't see that a virtual short would be different to a
real short.

I forgot to say, the following applies only to a virtual
short in the absence of a physical impedance discontinuity.
A virtual short at a physical impedance discontinuity
involves interference between forward wave components and
reflected wave components.

A virtual short is (Vfor-Vref)/(Ifor+Iref) = 0
where |Vfor| = |Vref|

For the virtual short to exist, two equal magnitude
EM waves have to be *flowing through* the virtual
short. EM waves *cannot flow through* a real short.

For real shorts:
V/I=0 is a result caused by the real short.

For virtual shorts:
A virtual short is a result caused by V/I=0.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

On Apr 15, 11:30 pm, Alan Peake
wrote:
Jim Kelley wrote:

I guess I may have been 'intuiting' too much, myself. Since virtual
shorts and opens only appear in the steady state, I wouldn't expect
pulses to reflect off of them. I don't expect sine waves to reflect
off of them in steady state either for that matter, but that remains a
point of contention apparently.

73, Jim AC6XG

Actually Jim, virtual shorts etc. act the same for pulse systems as for
CW systems.

Hi Alan -

Reflections measured by a TDR are caused by physical impedance
discontinuities. Virtual impedances are defined by the superposition
of forward and reflected voltages in the steady state. Pulsed systems
offer the ability to study the transient effects of a system by
viewing reflections caused only by changes in the characteristic
impedance of the transmission line. Since TDR doesn't use CW (not to
be confused with Morse Code) it does not operate under steady state
conditions and can therefore neither prove nor disprove the claim for
reflections from virtual impedances.

73, Jim AC6XG

Analyzing Stub Matching with Reflection Coefficients
 

Cecil Moore wrote:
Jim Kelley wrote:

Cecil Moore wrote:


Jim Kelley wrote:

Roy is absolutely right, Cecil. Interact is a very poor choice of
terms in this discussion.


Roy did NOT say "interact" was a poor choice of terms.


That's correct. I said that interact is a poor choice of terms.


But you implied that is what Roy said just above.

I observed that Roy is absolutely right, and, that 'interact' is a
very poor choice of terms in this discussion. I said it because waves
do not, according to the definition of the word, 'act upon one
another'. That of course does not mean there isn't a net effect when
they superpose. It simply means that waves do not effect other waves.
At this point I really don't expect you to understand that.

chose to use it as did Hecht. Hecht says waves interact.
Roy says they don't interact.


As I said, Roy is correct.


Roy is right and Hecht is wrong???

If Hecht actually weighed in on the subject, he would agree with Roy.
His use of the term caused you to infer something that he, I assure
you, did not intend to imply.

And the funny thing is, you say that even you know of instances in
which the net fields are zero, and yet the waves propagate beyond that
point.

Where do the reflected waves go that propagate beyond that
point and are measured as zero amplitude by a Bird wattmeter?

Take a look at the interference pattern created in space by two,
separated, coherent, point sources of light. The light waves
propagating from each point sources have absolutely no effect on each
other as they pass through one another, alternately interfering
destructively and constructively as they continue to propagate totally
unaffected by the process. It doesn't matter which direction they're
traveling; in no instance do waves destroy or act upon other waves,
totally or partially. The result of their superposition may differ
from one case to the next, but the phenomenon itself does not. But
again, at this point I don't expect you to understand this.

Dr. Best said those zero energy canceled waves propagate right
into the source.

He might have a point. But since cancelled waves convey no energy, it
doesn't really matter one way or the other, as others here have noted.

Are
you making that same stupid assertion?

All I'm trying to do is point out when you make a stupid assertion.

I think you just like to argue.


No, I honestly think we would pinpoint our differences. But,
of course, you would never agree to such.

I've already made the differences as clear as I possibly can in every
way I can think of, Cecil. That is why at this point I really don't
expect you to understand. You could, but I think it's pretty apparent
that you have too much invested in your personal theories.

73, Jim AC6XG

Analyzing Stub Matching with Reflection Coefficients
 

Walter Maxwell wrote:

On 15 Apr 2007 15:10:11 -0700, "Jim Kelley" wrote:


On Apr 15, 12:50 pm, Walter Maxwell wrote:


Seems to me that the only disagreement with my original posting is whether the condition at the stub point can
be called a 'virtual' short circuit.

Hi Walt,

Most everyone has directly expressed complete agreement with that
idea.


Here's the recurring theme:


*******Virtual impedance discontinuities do not cause
reflections.********


73, Jim AC6XG


OK Jim, if that's so, then I've got to figure out a new way to explain how antenna radiation patterns are
modified by changing the relative phase of the signals fed to multiple radiators, and by changing the spacing
between the radiators. Looks like I've had it all wrong for lo these many years. I thought I've been reading
the same references as all the other posters.

Walt

Hi Walt,

Your entire treatise is brilliant and useful with the one exception
noted clearly above. Perhaps you could cite a single one of those
references (other than Reflections of course) which directly
contradicts my simple observation of an extremely well understood
fundamental of nature.

Obviously a revision of that one circumstantial claim would have
absolutely no impact on element spacing or how waves interfere, and it
would in my view perfect the book. Once you have the currents and
fields worked out properly, they look after themselves. You don't
need to help them by inventing another mechanism for them to do their
job. Faraday, JC Maxwell and others have already worked that out to
most everyone else's satisfaction.

I think the discussion of virtual impedances and reflection
coefficients is useful as an analytical tool. But it should also
follow that the behavior being attributed to virtual entities is
likewise, virtual i.e. it behaves as though....; that the actual cause
of reflections is the real physical boundaries. That is the more
reasonable approach, Walt. IMO.

73, Jim AC6XG

Analyzing Stub Matching with Reflection Coefficients
 

On Tue, 17 Apr 2007 10:35:11 +1000, Alan Peake
wrote:

So how does one prove or disprove the existence of a real reflection
from a virtual discontinuity?

Mac already covered that in the space of four sentences:
On Sun, 15 Apr 2007 18:30:59 -0500, "J. Mc Laughlin" wrote:

One characteristic of a "virtual short" is that its presence or location is
dependent on frequency. Another characteristic is that signals are expected
to exist on both sides of a "virtual short." One characteristic of a
"physical short" is that it does not depend on frequency. Another
characteristic of a "physical short" is that signals exist on only one side
of the "physical short's" location.

73's
Richard Clark, KB7QHC

Analyzing Stub Matching with Reflection Coefficients
 

Hi Alan -

Reflections measured by a TDR are caused by physical impedance
discontinuities. Virtual impedances are defined by the superposition
of forward and reflected voltages in the steady state. Pulsed systems
offer the ability to study the transient effects of a system by
viewing reflections caused only by changes in the characteristic
impedance of the transmission line. Since TDR doesn't use CW (not to
be confused with Morse Code) it does not operate under steady state
conditions and can therefore neither prove nor disprove the claim for
reflections from virtual impedances.

73, Jim AC6XG

Yes, I had a bit of LNBF (Late Night Brain Fade) when I threw in the
rotary joint example.
What I was trying underline was that there will be a real reflection at
the point where a stub is attached - simply because it becomes a
discontinuity in the TL. A TDR can show this but of course, I agree that
a single pulse won't see the same discontinuity as a CW.
So how does one prove or disprove the existence of a real reflection
from a virtual discontinuity?
Alan

Analyzing Stub Matching with Reflection Coefficients
 

Hi Alan -

Reflections measured by a TDR are caused by physical impedance
discontinuities. Virtual impedances are defined by the superposition
of forward and reflected voltages in the steady state. Pulsed systems
offer the ability to study the transient effects of a system by
viewing reflections caused only by changes in the characteristic
impedance of the transmission line. Since TDR doesn't use CW (not to
be confused with Morse Code) it does not operate under steady state
conditions and can therefore neither prove nor disprove the claim for
reflections from virtual impedances.

73, Jim AC6XG

Hi Jim, not sure if my previous reply got through.
Yes, I have to admit to LNBF (Late Night Brain Fade) when I threw in the
rotary joint example. Of course it is CW in the pulse. I was trying to
underline that a stub puts a physical discontinuity on the TL which will
give a real reflection. But as you point out, this reflection is not the
same as the reflection from a virtual discontinuity for CW. However, if
the CW can be thought of as a series of pulses, then does that not mean
that real reflections occur and that the sum of the reflections for each
pulse looks like they have come from a virtual discontinuity?
If not, how would one go about proving or disproving the idea of
reflections from virtual discontinuities?
Alan

Analyzing Stub Matching with Reflection Coefficients
 

Alan Peake wrote:

However, if
the CW can be thought of as a series of pulses, then does that not mean
that real reflections occur and that the sum of the reflections for each
pulse looks like they have come from a virtual discontinuity?

Hi Alan -

The point of using pulses is that their width is short, and the time
between pulses is long compared to the delay times in the system. In
the case of of CW there will be standing waves all throughout the
system obscuring any possible measurement of transient response.
These pulses only reflect from physical discontinuities in the surge
impedance of the transmission line. Otherwise, TDR would be a
complete wild goose chase; a real cluster _blank_, in the vernacular
of the trade.

If not, how would one go about proving or disproving the idea of
reflections from virtual discontinuities?
Alan

Disproving the idea of reflections from virtual discontinuities would
be done, for instance, and has been suggested, by measuring the
presence of fields beyond the virtual short in a 1/4 wave stub.
Finding waves reflecting instead from the open end sure would not lend
support to the notion. The fact that the idea is inconsistent with
Maxwell's equations doesn't help either.

I don't think there is a way to prove the idea of reflections from
virtual discontinuities. But with certain specific exceptions, a
system could in other ways appear to behave as though reflections are
originating at virtual impedance discontinuities (+/- n half wavelengths).

73, Jim AC6XG

Analyzing Stub Matching with Reflection Coefficients
 

Jim Kelley wrote:
I said it because waves do
not, according to the definition of the word, 'act upon one another'.

But they can act upon one another, Jim. The Florida State web
page says so. The Melles-Groit web page says so. It says their
energy components are redistributed. How can their energy
components be redistributed if they have no effect on each
other? You really need to join me in the s-parameter analysis.

b1 = s11(a1) + s12(a2)

That's phasor math proving that components of waves a1 and
a2 have an effect on b1 and therefore on each other. Every
time two coherent waves are collinear in the same direction
in a transmission line, they have an effect on each other.
It's called interference, either constructive or destructive.

If Hecht actually weighed in on the subject, he would agree with Roy.

Good grief, Jim, now you are mind-fornicating Hecht. Hecht
would certainly not agree with your obviously false assertions.

His use of the term caused you to infer something that he, I assure
you, did not intend to imply.

Your assurance and three bucks will get me a cup of Starbucks.

Take a look at the interference pattern created in space by two,
separated, coherent, point sources of light. The light waves
propagating from each point sources have absolutely no effect on each
other as they pass through one another, alternately interfering
destructively and constructively as they continue to propagate totally
unaffected by the process.

Yes, because they are not collinear. If they don't intersect,
they also don't interfere. You can find billions of cases where
they don't interfere. That doesn't mean they don't ever interfere.

Just as illustrated on the Florida State web page, when coherent
waves are also collinear, as they are in a transmission line, they
merge into the total wave and cease to exist as separate wave
components.

b1 = s11(a1) + s12(a2)

s11(a1) and s12(a2) lose their identities and merge into b1.
If your statements were true, an s-parameter analysis wouldn't
be valid but it is. Therefore, your statements are false. That's
why you need to wade through an s-parameter analysis because
you don't understand what happens or comprehend the physics
behind it.

It doesn't matter which direction they're traveling;

On the contrary, coherent waves traveling in the same direction
in a transmission line are *collinear*. They merge and permanently
interfere with each other thus proving your strange assertions to
be false.

I've already made the differences as clear as I possibly can in every
way I can think of, Cecil.

But you are uttering assertions that are patently false. Given
two coherent waves traveling in the same direction in a Z0
transmission line, with equal magnitudes, V, and equal phases,
0 deg, what is the total magnitude? Do you even know how to do
phasor math?

V at 0 deg + V at 0 deg = ____________________________

If you need help, ask your supervisor what the answer is.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Alan Peake wrote:
So how does one prove or disprove the existence of a real reflection
from a virtual discontinuity?

One sure way would be to demonstrate a reflection from
a virtual impedance where no physical impedance discontinuity
exists. Good luck on that one.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Walter, W2DU wrote:
"OK, Jim, if that`s so, then I`ve got to figure out a new way to explain
how antenna radiation patterns are modified by changing the relative
phase of the signals fed to multiple radiators, and by changing the
spacing between the radiators."

Walter`s systen isn`t broken so it shouldn`t be fixed. Signal strength
at a point in space depends on the vector totals of its constituents.
Walter`s totals are determined by positions of the radiators and phases
of the currents in those radiators.

Obviously, where vectors are in-phase they add and where they are
out-of-phase they subtract.

The system works, that`s why the FCC endorses it.

Best regards, Richard Harrison, KB5WZI

Analyzing Stub Matching with Reflection Coefficients
 

Cecil Moore wrote:

Jim Kelley wrote:

I said it because waves do not, according to the definition of the
word, 'act upon one another'.


But they can act upon one another, Jim. The Florida State web
page says so. The Melles-Groit web page says so.

No they don't. If the waves themselves changed, then their resultant
superposition would also change. It's a completely unfounded notion,
Cecil.

It says their
energy components are redistributed.

Which is not the same as saying waves have an effect on other waves.
I said I didn't expect you to understand, and clearly you don't.

How can their energy
components be redistributed if they have no effect on each
other?

I don't know what exactly an "energy component" is, but I would assert
that it would be redistributed in the same way completely
independently of however you or I might happen to feel about it.

His use of the term caused you to infer something that he, I assure
you, did not intend to imply.

Your assurance and three bucks will get me a cup of Starbucks.

Not to mention a more realistic viewpoint.

Take a look at the interference pattern created in space by two,
separated, coherent, point sources of light. The light waves
propagating from each point sources have absolutely no effect on each
other as they pass through one another, alternately interfering
destructively and constructively as they continue to propagate totally
unaffected by the process.


Yes, because they are not collinear. If they don't intersect,
they also don't interfere. You can find billions of cases where
they don't interfere. That doesn't mean they don't ever interfere.

As I said, I don't expect you to understand, and clearly here you don't.

Just as illustrated on the Florida State web page, when coherent
waves are also collinear, as they are in a transmission line, they
merge into the total wave and cease to exist as separate wave
components.

Yes, it very effectively shows how 1 + -1 = 0. Very profound, Cecil.

ac6xg

Analyzing Stub Matching with Reflection Coefficients
 

Alan Peake wrote:

Hi Jim, not sure if my previous reply got through.
Yes, I have to admit to LNBF (Late Night Brain Fade) when I threw in the
rotary joint example. Of course it is CW in the pulse. I was trying to
underline that a stub puts a physical discontinuity on the TL which will
give a real reflection. But as you point out, this reflection is not the
same as the reflection from a virtual discontinuity for CW. However, if
the CW can be thought of as a series of pulses, then does that not mean
that real reflections occur and that the sum of the reflections for each
pulse looks like they have come from a virtual discontinuity?
If not, how would one go about proving or disproving the idea of
reflections from virtual discontinuities?
Alan

If you think of CW as a series of pulses, a "virtual short" occurs only
when an inverted reflected pulse arrives at the same point at the same
time as a non-inverted non-reflected pulse, causing the two to add to
zero. (Or, of course, more complex combinations of multiple pulses
arriving at the same point.) It isn't the same pulse which appears twice
to interfere with itself; it's different pulses of the pulse string,
sent at different times but arriving at the same point simultaneously
due to one being delayed by reflection and the other not. So you see,
the interval between those pulses is critical; if it changes, then the
location of the "virtual short" changes. This is analogous to the steady
state CW situation where the location of the "virtual short" changes
with frequency.

In theory, you could prove that reflection isn't occurring from a
"virtual discontinuity" by making an abrupt change in the excitation,
for example abruptly changing its level, then noting that the effect of
the change isn't seen back at the input until it propagates through the
"virtual discontinuity", on to physical discontinuities where reflection
actually takes place, and back. This might be difficult to do in
practice, though, except with some fairly sophisticated equipment or
very long lines because of the time intervals involved.

But let's suppose that you did somehow prove that a "virtual
discontinuity" reflects waves. Then you have to explain the mechanism by
which waves alter each other in a linear medium. Since you won't find
any such mechanism described or explained in any reputable text, you'll
have to come up with some pretty creative alternative physical laws or
derivations on your own. They would have to explain such interesting
phenomena as the diode-like nature of a "virtual discontinuity" -- that
is, why the fields interact going one way and not the other. Also,
you'll need to come up with equations which take into account the
infinite number of reflections from the "virtual discontinuities" which
occur at nearly every point along any line not terminated in its
characteristic impedance. At the end of the day, the results of those
equations have to be the same as those which assume no reflections from
"virtual discontinuities", because equations assuming no such reflection
have been in use for over a century and have so far not been found to be
in error.

Any of the proponents of "virtual discontinuity" reflections up to it?

Roy Lewallen, W7EL

Analyzing Stub Matching with Reflection Coefficients
 

Richard Harrison wrote:
Obviously, where vectors are in-phase they add and where they are
out-of-phase they subtract.

In fact, Jim Kelley's assertion that there is no interaction
between waves would result in isotropic radiation in the
far field of every antenna if one went out far enough to
measure the waves after they are propagating free of each
other. I wonder if NASA knows that?
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Cecil Moore wrote:
Richard Harrison wrote:
Obviously, where vectors are in-phase they add and where they are
out-of-phase they subtract.

In fact, Jim Kelley's assertion that there is no interaction
between waves would result in isotropic radiation in the
far field of every antenna if one went out far enough to
measure the waves after they are propagating free of each
other. I wonder if NASA knows that?

Cecil,

You don't believe in superposition, do you? It is discussed in lots of
books if you want to understand.

8-)

73,
Gene
W4SZ

Analyzing Stub Matching with Reflection Coefficients
 

Jim Kelley wrote:
Cecil Moore wrote:
But they can act upon one another, Jim. The Florida State web
page says so. The Melles-Groit web page says so.

No they don't. If the waves themselves changed, then their resultant
superposition would also change. It's a completely unfounded notion,

If what you say is true, then if we measure field strengths
far enough away from an antenna to get outside the range
of interference, then all antennas are isotropic. Why don't
you call up NASA and tell them that permanent constructive
interference doesn't exist and they might as well be
using isotropic antennas?

It says their
energy components are redistributed.

Which is not the same as saying waves have an effect on other waves. I
said I didn't expect you to understand, and clearly you don't.

Well then, please explain it to me.
Here's the s-parameter equation for wave cancellation
in the b1 direction.

b1 = s11(a1) + s12(a2) = 0

s11, a1, s12, and a2 are all real measured values. b1 is a
real measured value. All of the measured values are perfectly
consistent. Exactly how did b1 get to be 0 without s11(a1) and
s12(a2) canceling each other out?

What do *you* get when you add one volt at 0 deg to one volt
at 180 deg when they are coherent and traveling in a collinear
path in a transmission line? Assuming EM waves, a value of zero
tells us that wave cancellation has occurred. So what value do
you get?

As I said, I don't expect you to understand, and clearly here you don't.

You are a broken record, Jim, mindlessly uttering mantras to
cover up your inability to comprehend reality. I'm beginning
to understand what Roy meant by his "academic" statement.

Just as illustrated on the Florida State web page, when coherent
waves are also collinear, as they are in a transmission line, they
merge into the total wave and cease to exist as separate wave
components.

Yes, it very effectively shows how 1 + -1 = 0. Very profound, Cecil.

That, my friend, is permanent destructive interference, in
the flesh, as it were. One joule/second at 0 degrees plus
one joule/second at 180 degrees is indeed 0 joules/sec in
the direction of original travel of those two waves in a
transmission line. What happens after that is a two
joule/second reflection in the opposite direction away from
the impedance discontinuity that is causing the reflections
and permanent interference.
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Jim, AC6XC wrote:
"Virtual impedance discontinuities do not cause reflections."

Reflection is a change in direction. On a transmission line a complete
reflection is caused by a physical open or short on the line.

Current is interrupted at an open circuit. Energy in the magnetic field
at the interruption collapses generating a voltage which doubles the
line voltage at the open. This causes the current direction to reverse
(a reverse in its phase) while not changing the phase of the voltage.

Conditions necessary for the reversal in travel direction of an EM wave
are to reverse either the magnetic or electric wave`s polarization, but
not to change both.

At an actual short on a line, volts are forced to zero at the short.
Current doubles at the short, and the voltage wave reverses its
polarization. The EM wave reverses its travel direction at the short as
it did in the case of an open circuit.

When volts or amps compete against an opponent of half their magnitude,
the stronger opponent wins. So, it`s volts or amps which determine which
direction a wave travels on a transmission line at a discontinuity.

At a short or an open on a line , it is the current or voltage the
discontinuity generates which turns the wave around. The line doesn`t
care how the amps or volts came to suddenly appear at the turnaround
point. If a virtual condition can generate the energy surge or
escalation needed for a reversal in direction, it is as acceptable as a
real discontinuity, in my opinion.

Best regards, Richard Harrison, KB5WZI

Analyzing Stub Matching with Reflection Coefficients
 

Roy Lewallen wrote:
Any of the proponents of "virtual discontinuity" reflections up to it?

Would you please explain if there is any energy associated
with your "inverted reflected pulse" and your "non-inverted
non-reflected pulse" or do they exist devoid of energy?
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Alan Peake wrote:
"I was trying to underline that a stub puts a physical discontinuity on
the TL which will give you a real reflection."

Short-circuited 1/4-wave stubs have long been used as "metal insulators"
to support transmission lines. This would likely be impractical if the
stubs produced a discontinuity on the line at the operatibg frequency.
The number of stubs seems without limit also. Put enough together and
you`ve constructed rectangular waveguide.

Best regards, Richard Harrison, KB5WZI

Analyzing Stub Matching with Reflection Coefficients
 

Gene Fuller wrote:
You don't believe in superposition, do you? It is discussed in lots of
books if you want to understand.

Do you believe Jim's argument that two coherent EM
waves of equal magnitudes and opposite phases traveling
collinearly in the same direction in a transmission line
can never be canceled? If Jim is right, we can toss the
s-parameter analysis in the garbage can and join Roy in
calling it gobbledigook (sic).
--
73, Cecil http://www.w5dxp.com

Analyzing Stub Matching with Reflection Coefficients
 

Hi Alan -

The point of using pulses is that their width is short, ....


Hi Jim,
While the individual pulses are short, for them to simulate CW, they
must be right next to each other. Obviously, the pulses that cancel or
reinforce aren't the same ones, as Roy points out, and the CW which
appears to reflect from the virtual discontinuity is the sum of all
pulses, whether they are from the stub point, the other end of the stub
or the load end of the TL.

Alan

Analyzing Stub Matching with Reflection Coefficients
 

If you think of CW as a series of pulses, a "virtual short" occurs only
when an inverted reflected pulse arrives at the same point at the same
time as a non-inverted non-reflected pulse, causing the two to add to
zero. (Or, of course, more complex combinations of multiple pulses
arriving at the same point.) It isn't the same pulse which appears twice
to interfere with itself; it's different pulses of the pulse string,
sent at different times but arriving at the same point simultaneously
due to one being delayed by reflection and the other not. So you see,
the interval between those pulses is critical; if it changes, then the
location of the "virtual short" changes. This is analogous to the steady
state CW situation where the location of the "virtual short" changes
with frequency.

Hi Roy,
No disagreement with that. A single pulse will have reflections from the
open end of the stub, the load end of the TL (if it's not Zo) and the
stub attachment point. In a TDR, the first return will be from the
latter point but it won't be of the same magnitude and sign as it would
from a short. That requires all the pulses which are simulating (if you
like), the CW. It will be whatever you get when the pulse, which has
been happily travelling along a TL of Zo, meets a point where there are
now two TLs of Zo in parallel. I consider this to be a discontinuity and
will produce a real reflection albeit not, as I said, the same as that
from a virtual short etc. So, in the sense that the quarter wave stub
appears as a virtual short, I agree that the actual reflection to cause
this comes from the open end of the stub.

In theory, you could prove that reflection isn't occurring from a
"virtual discontinuity" by making an abrupt change in the excitation,
for example abruptly changing its level, then noting that the effect of
the change isn't seen back at the input until it propagates through the
"virtual discontinuity", on to physical discontinuities where reflection
actually takes place, and back. This might be difficult to do in
practice, though, except with some fairly sophisticated equipment or
very long lines because of the time intervals involved.

Yes, you'd have to have a stub of 1/4 wave plus N 1/2 waves to see this
effect. If N was large enough to see the excitation change (you could
probably use a "CW pulse" a la radar to do this) I would expect to see
some return from the stub attachment point, then the reflection from the
open end of the stub.

But let's suppose that you did somehow prove that a "virtual
discontinuity" reflects waves. Then you have to explain the mechanism by
which waves alter each other in a linear medium.

Well, I'm a believer in superposition so I'll leave that one to others :)

Analyzing Stub Matching with Reflection Coefficients
 

Alan Peake wrote:
[I wrote]
But let's suppose that you did somehow prove that a "virtual
discontinuity" reflects waves. Then you have to explain the mechanism
by which waves alter each other in a linear medium.

Well, I'm a believer in superposition so I'll leave that one to others :)

Those others have made, I'm sure, over a thousand postings so far, and
yet in them there's a complete lack of any explanation of the mechanism.

Roy Lewallen, W7EL

Analyzing Stub Matching with Reflection Coefficients
 

Jim Kelley wrote:
Cecil Moore wrote:

Jim Kelley wrote:
I said it because waves do not, according to the definition of the
word, 'act upon one another'.

But they can act upon one another, Jim. The Florida State web
page says so. The Melles-Groit web page says so.

No they don't. If the waves themselves changed, then their resultant
superposition would also change. It's a completely unfounded notion,
Cecil.

Here's an example of that "unfounded notion". Please
point out my error.

In the following example, the 100W signal generator
is equipped with a circulator load. The system is
Z0-matched during steady-state so b1 = 0 during
steady-state.

100W
SGCL--50 ohm line--x--1/2WL 291.4 ohm line--50 ohm load
a1-- b2--
--b1 --a2

b1 = s11(a1) + s12(a2) b2 = s21(a1) + s22(a2)

Let t0 be the time at which the 100W forward wave
reaches point 'x' for the first time.

Just after after t0, the source signal has split
into two parts. There are as yet, no reflections,
so a2=0. Every one of these three voltages can be
measured as real. These values remain constant
throughout steady-state.

x
a1=10----|
|----s21(a1)=5 toward the load
s11(a1)=5----|

Just after t0, b1=5. During steady-state, b1=0.

Please explain how b1 goes from 5 to 0 during the
transient build-up state without having s11(a1)
interact with anything. There, in a nutshell,
is your technical and logical contradiction.
--
73, Cecil http://www.w5dxp.com