Constructive interference in radiowave propagation
 

On Sat, 07 Apr 2007 17:59:03 GMT, Cecil Moore
wrote:

Between coherent and mutually incoherent is
a span of signals which they call partially
incoherent.

As Reggie would say

"BAFFLEGAB!"

Constructive interference in radiowave propagation
 

On Sat, 07 Apr 2007 21:16:06 GMT, Owen Duffy wrote:

Walter Maxwell wrote in
:

On Sat, 07 Apr 2007 05:03:51 GMT, Owen Duffy wrote:

Walter Maxwell wrote in
:

On Fri, 06 Apr 2007 23:03:42 GMT, Cecil Moore
wrote:

MRW wrote:
Any comments? Really, what I'm trying to understand here is: if
constructive interference does any good in radiowave propagation.
I was thinking that with an increase in amplitude the signal would
be able to travel a little further, but the signal received may
not be accurate in terms of the information it is conveying.

Antenna gain over isotropic is an application of
constructive interference. The constructive
interference must be balanced by an equal amount
of destructive interference elsewhere to avoid
violating the conservation of energy principle.

This is what I've been trying to persuade the 'anti's' that whenthe
radiation fields from two vertical dipoles superpose at some point
in space, where their magnitudes are equal and are 180° out of
phase, the wave cancellation resulting from destructive interference
produces a null in a predetermined direction, and thus prevents
those fields from propagating any further in that direction. At the
precise instant the null is produced, the constructive interference
following the principle of energy conservation yields an increase in
the field strength in directions away from the null direction. This
explains the concept of antenna-pattern modification, and
contradicts the notion that the two fields just plow through each
other with no effect on either.


Walt, this seems inconsistent with the approach that I believe you
seem to use in analysing waves in transmission lines where you seem to
want to not only deal with the forward and reverse waves separately
(ie to not collapse them to a resultant V/I ratio at a point), but to
deal with multiply reflected waves travelling in the forward and
reverse direction (which is only necessary in the transient state).

Owen

Owen, it appears that you've misinterpreted my approach. In developing
a condition for impedance matching, such as adding a series or shunt
stub at the proper place on a transmission line, the object has always
been to generate a new reflection at the stub point of the opposite
phase to that appearing on the line at the stub point. Thus when the
stub reflection and the load reflection superpose at the stub point,
the resulting reflection coefficients of voltage and current form
either a virtual open circuit or a virtual short circuit. These
conditions are produced because when the load impedance is greater
than Zo, the resultant reflection coefficient angles at the stub point
are 0° for voltage and 180° for current, establishing a virtual open
circuit at the stub point to rearward traveling waves. When the load
impedance is less than Zo, the resultant reflection coefficient angles
are 180° for voltage and 0° for current, establishing a virtual short
circuit at the stub point for rearward traveling waves.

Hi Walt,

I read the above, and I think I can see what you are getting at, however
I think it is flawed.

If you were to try to extend this method to explain the common two stub
tuner (where the length of the stubs is adjustable and the distance
between them is fixed), you will have to deal with a situation where the
load end stub junction does not present the "virtual o/c or s/c" you
describe, your "total re-reflector concept" and you come to need to
calculate the situation on the source side of the load end stub (possibly
by conventional methods?).

Walk your explanation around a Smith chart, and explain why, if the
principles on which your explanation are based are correct, why energy
fills a 3/4 wave hi Q coaxial resonator rather than being blocked by the
virtual s/c or o/c at the first voltage minimum or current minimum.

Someone else persuing the theme that reflected waves always travel all
the way back to the source, seems to come to a position that some kinds
of matching produce a complementary reflected wave, and that really there
are two (or more) reflected waves, its just that they have zero net
energy. Some of us would accept that if the resultant is zero, there is
no wave. Otherwise, you would see a multitude of net-zero waves all
around us to complicate every analysis.

These "new" and alternative explanations are questionable and don't seem
better than the conventional explanations of a transmission line that are
set out in just about any reputable transmission lines text. What
advantages do these explanation have, who are they targeted at? Is the
"total re-reflector" concept to appeal to a dumbed down audience who can
get their mind around a bunch of words that describe specific situations
in a simple and appealing way, but an incorrect explanation nonetheless?

I think it is a real challenge to teach people a simple explanation of
what happens without telling them convenient lies that have to be
unlearned to develop further. The "reflected wave is (always) dissipated
in the PA as heat" is an example of one of those convenient lies.

Owen

Owen,

Are you saying that those who support my explanation of stub matching are telling lies!?
I can't believe you said that.

I explained this exact matching function in great detail in October 1974 QST. It was repeated in Reflections
edition 1 in 1990, and again in edition 2 in 2001. It was also published in QEX in the Mar/Apr 1998 issue.

I had to submit my manuscripts to the engineering departments of four different divisions of RCA for approval
before being allowed to even submit them to QST to determine if they were interested in publishing them. One
of the RCA engineers who reviewed my text was Jack Young, then the chief engineer of the transmitter section
of the RCA Broadcast Engineering Division. He complimented me on my explanation of stub matching using the
reflection coefficients as the parameters. On accepting my papers for publication in QST, the former Technical
Director of the League, George Grammer, W1DF, also complimented me on the unique approach explaining stub
matching, and also for my explanation of conjugate matching. Grammer was probably the most highly educated
engineer ever gracing the ARRL.

It's been 33 years since I published my presentation using reflection coefficients, and to date no one except
some of the posters on this NG have disputed it. Are you saying that all who reviewed my writings that
appeared in QST and in Reflection were being lied to?

You suggested walking through the Smith Chart with the problem. Owen, the Smith Chart is Fig. 1 in my
presentation that proves my position is correct. Please review the QEX article I referenced above to see it.
With all due respect, Owen, I believe you have misunderstood, or perhaps misconstrued the procedure I
presented. Would you please review it again to see where you might have gone wrong?

Walt

Constructive interference in radiowave propagation
 

Walter Maxwell wrote in
:

is correct. Please review the QEX article I referenced above to see
it. With all due respect, Owen, I believe you have misunderstood, or
perhaps misconstrued the procedure I presented. Would you please
review it again to see where you might have gone wrong?

Hi Walt,

I have reviewed Chapter 3 of Reflections II which you kindly sent me. I
think it contains the deveopment to which you refer.

It seems to me that Chapter 3 depends entirely on an assumption that the
phase relationship of the current and voltage of a travelling wave is 0 deg
or 180 deg, depending on the direction. This only holds true for lossless
lines and distortionless lines, and so the "proofs" developed in the
chapter are not general proofs. For example, the proof that reflected power
is purely real and of magnitude |E-|^2/Zc is not developed for the general
case, and happens to not be correct for the general case.

Owen

Constructive interference in radiowave propagation
 

On Apr 7, 7:31 pm, Richard Clark wrote:
On Sat, 07 Apr 2007 17:59:03 GMT, Cecil Moore
wrote:

Between coherent and mutually incoherent is
a span of signals which they call partially
incoherent.

That should have been "partially coherent".

As Reggie would say "BAFFLEGAB!"

Don't blame me - blame Born and Wolf - whom a lot of people respect.
--
73, Cecil, w5dxp.com

Constructive interference in radiowave propagation
 

On Apr 7, 11:45 pm, "Cecil Moore" wrote:
On Apr 7, 7:31 pm, Richard Clark wrote:

On Sat, 07 Apr 2007 17:59:03 GMT, Cecil Moore
wrote:

Between coherent and mutually incoherent is
a span of signals which they call partially
incoherent.

That should have been "partially coherent".

As Reggie would say "BAFFLEGAB!"

Don't blame me - blame Born and Wolf - whom a lot of people respect.

Do Born and Wolf offer crisp definitions of the boundaries
between coherent, partially coherent, and mutually incoherent?

Or is it a continuum arbitrarily divided into 3 regions for
the purposes of discussion?

....Keith

Constructive interference in radiowave propagation
 

On Sun, 08 Apr 2007 03:32:48 GMT, Owen Duffy wrote:

Walter Maxwell wrote in
:

is correct. Please review the QEX article I referenced above to see
it. With all due respect, Owen, I believe you have misunderstood, or
perhaps misconstrued the procedure I presented. Would you please
review it again to see where you might have gone wrong?

Hi Walt,

I have reviewed Chapter 3 of Reflections II which you kindly sent me. I
think it contains the deveopment to which you refer.

It seems to me that Chapter 3 depends entirely on an assumption that the
phase relationship of the current and voltage of a travelling wave is 0 deg
or 180 deg, depending on the direction. This only holds true for lossless
lines and distortionless lines, and so the "proofs" developed in the
chapter are not general proofs. For example, the proof that reflected power
is purely real and of magnitude |E-|^2/Zc is not developed for the general
case, and happens to not be correct for the general case.

Owen

Owen, your statement that my writings in Reflections are flawed is shocking. The feeling I get from it is like
getting sucker punched in the stomach.

Are you so narrowly oriented academically that the difference between lossless and low-loss conditions is so
great that general principles cannot be applied to situations where real-world low-loss elements are involved?

It is generally accepted that voltage and current travel forward on a low-loss line with 0° phase difference.
In low-loss lines the effect of the small negative-reactance component in the Z0 due to the loss is routinely
disregarded as insignificant. Likewise, when voltage and current travel rearward on a low-loss line, resulting
from reflection at a mismatched termination, it is generally accepted that they travel with a 180° phase
difference, disregarding the small error caused by the insignificantly-small reactance in the Z0.

Calculations performed when disregarding the small error still yield practical results in hands-on operations.
On the other hand, if everyday practical operations required calculating with the academically-perfect
conditions of the Z0, time would be lost due to the unnecessary complications involved in the calculations.

Your stated position is that applying general principles that are academically correct only with lossless
elements to operations involving low-loss elements is flawed.

C'mon, Owen, let's get practical and rescind your impeachment of Reflections.

Walt

Constructive interference in radiowave propagation
 

On Apr 7, 5:51 pm, Gene Fuller wrote:
The beams "interfere" but they do not "interact".

Of course, you can give examples where the waves survive the
superposition. But what we are talking about is when the waves do
NOT survive the superposition.

How about wave cancellation, Gene? When two coherent waves traveling
in the same direction in the same path with equal magnitudes and
opposite
phases interact, they cease to exist in the direction of the original
travel.
Ir's senseless to argue that waves that cease to exist during the
process
of superposition have not interacted with each other, don't you think?
--
73, Cecil, w5dxp.com

Constructive interference in radiowave propagation
 

On Apr 8, 6:57 am, "Keith Dysart" wrote:
Do Born and Wolf offer crisp definitions of the boundaries
between coherent, partially coherent, and mutually incoherent?
Or is it a continuum arbitrarily divided into 3 regions for
the purposes of discussion?

I don't have Born and Wolf with me at the moment but I seem
to recall that they said that any signal that was not single
frequency could not be 100% coherent. Presumably a signal
with modulation would be partially coherent.
--
73, Cecil, w5dxp.com

Constructive interference in radiowave propagation
 

Cecil Moore wrote:
On Apr 7, 5:51 pm, Gene Fuller wrote:
The beams "interfere" but they do not "interact".

Of course, you can give examples where the waves survive the
superposition. But what we are talking about is when the waves do
NOT survive the superposition.

How about wave cancellation, Gene? When two coherent waves traveling
in the same direction in the same path with equal magnitudes and
opposite
phases interact, they cease to exist in the direction of the original
travel.
Ir's senseless to argue that waves that cease to exist during the
process
of superposition have not interacted with each other, don't you think?
--
73, Cecil, w5dxp.com


Cecil,

It is easy to give examples where the waves survive the superposition,
because they always do. It is rather strange that you are making this
argument after all the back and forth about traveling waves and standing
waves. Do we now have multiple flavors of EM waves? Some that obey
superposition and some that don't?

I must have missed class the day they went over the theory of
"cancellation". Is this another one of those convenient descriptions of
results that you keep trying to remold into fundamental physical laws?

I stand 100% behind my two messages to Walt. If you actually read them
you would note that I said for most cases it makes no difference whether
the waves interfere forever or whether they interact and "cancel". As
Owen pointed out a little while ago, we generally don't want to carry
around lots of zero components in an analysis.

The bottom line is that EM waves do not interact in free space.
Linearity and superposition could not hold if that were the case.
Maxwell's equations would need to be recast.

There are exactly enough physical laws and principles now. There is no
need to invent more on RRAA.

73,
Gene
W4SZ

Constructive interference in radiowave propagation
 

Walter Maxwell wrote in
:

Walt, I can see that you have taken my comment as personal criticism. That
was not intended, and to the extent that I may have caused that, I
apologise. In that context, it is better that I refrain from further
comment.

Regards
Owen