"Reg Edwards" wrote in message ...

When the line is not lossless, ie., it has appreciable
attenuation in dB per 1/4-wavelength, then the ratio is
'distorted' and has a phase angle. So negative values
of indicated SWR can be expected at some values of |
Vmax | / | Vmin |


What are you talking about? If it have losses, and they
are dissipative losses, the amplitude of the voltage will
decrease due to voltage drops. That would be moving AWAY
from having a greater reflected voltage than an incident one.

But, that's impossible anyways with a passive network.

The concept of Negative SWRs is rubbish.

SWR is calculated from the square of | rho |. As I've
said before, immediately | rho | is squared, half the
information it contains is junked. Any
discussion/argument about power waves following
rho-squared on a lossy (a real ) line is meaningless
piffle.

Anybody who writes books about power waves, selling
them to make a living, is obtaining money under false
pretences.

On the other hand we should be kind to otherwise
unemployed Ph.D's. They too have wive's and kid's to
clothe, feed and provide a roof over their heads.
That's life!
---
Reg.

Remind me not to be YOUR book when it comes out!



The ratio Pref/Pfwd is directly related to the ratio [rho].

Pref/Pfwd = [rho]**2 Absolute value brackets are a must!


Consider that after the absolute value brackets, the phase information
is gone. But since we are going to a ratio of average (RMS)
values OR peak values of power, it doesn't matter.

In other words, if you use V**2/R, the "V" can be either peak or
RMS, it doesn't matter, because it is a ratio. And of course, the "R"
doesn't matter either. And of course, the phase information is gone
with
the absolute value brackets.

If you agree that the Pref/Pfwd ratio cannot be greater than 1
for a passive network, then neither can the [Vref/Vfwd]= rho be
greater
than 1 either.

Some people wanna rewrite some books here.


Slick

wrote in message ...

And yes, |rho| can be greater than unity for a passive load.

...Keith


Absolute Rubbish.. Could you produce a passive circuit that
will reflect a greater voltage than what you feed it? I'd
LOVE to see that...


The ratio Pref/Pfwd is directly related to the ratio [rho].

Pref/Pfwd = [rho]**2 Absolute value brackets are a must!


Consider that after the absolute value brackets, the phase information
is gone. But since we are going to a ratio of average (RMS)
values OR peak values of power, it doesn't matter.

In other words, if you use V**2/R, the "V" can be either peak or
RMS, it doesn't matter, because it is a ratio. And of course, the "R"
doesn't matter either. And of course, the phase information is gone
with
the absolute value brackets.

If you agree that the Pref/Pfwd ratio cannot be greater than 1
for a passive network, then neither can the [Vref/Vfwd]= rho be
greater
than 1 either.

Some people wanna rewrite some books here.


Slick

"David Robbins" wrote in message ...
sorry, no scanner here.

how do you get rho1? please give me the Zo and Zl to try out, i have been
playing for a while with the basic equations and haven't found a case where
either formulat gives rho1.

and of course if |rho|=1 then swr can never be negative.

I think Reg put it best:

"Dear Dr Slick, it's very easy.

Take a real, long telephone line with Zo = 300 - j250 ohms at 1000 Hz.

Load it with a real resistor of 10 ohms in series with a real
inductance of
40 millihenrys.

The inductance has a reactance of 250 ohms at 1000 Hz.

If you agree with the following formula,

Magnitude of Reflection Coefficient of the load, ZL, relative to line
impedance

= ( ZL - Zo ) / ( ZL + Zo ) = 1.865 which exceeds unity,

and has an angle of -59.9 degrees.

The resulting standing waves may also be calculated.

Are you happy now ?"
---
Reg, G4FGQ


If it were not for Reg pointing out this example, i wouldn't have
researched and corrected my original, "purely real" Zo post with the
more general conjugate Zo formula.

And i researched it because i knew that you cannot have a R.C.
greater than one for a passive network (you can only have a R.C.
greater than one for an active network, which would be a "return gain"
instead of a "return loss"), so i knew that when Zo is complex, my
original post must have been wrong.


Intelligent people can be close-minded, that is for certainly, in
which case, their intelligence is blunted.



Slick

On 3 Sep 2003 23:16:39 -0700, (Dr. Slick) wrote:


I think Reg put it best:

Hi OM,

If you are going to quote him as an authority supporting you, you
should at least accept his offer of a bridge to settle this hash
shouldn't you? You can't run far on one legged stilts.

73's
Richard Clark, KB7QHC

Richard Harrison wrote:

Keith wrote:
"I suppose one might phrase it as "There is no such thing as a power
reflection coefficient" when it is not physically meaningful to separate
the total power as the sum of the incident and reflected power so the
equatiomn:
[rho] = sq. rt. (Pref / Pfwd) has no meaning."

We don`t have a choice of options on a menu to select or reject from.
Reality is whatever it is and we accept it and describe it as best we
can.

Terman says on page 97 of his 1955 edition:
"{rho} = (SWR-1) / SWR + 1."

Power varies as the equare of the voltage, because when you increase the
volts you also automatically increase the amps (Ohm`s law). Thus, Terman
has a subscript at the bottom of page 97 which is relevant:
"The definition of standing-wave ratio is sometimes called voltage
standing-wave ratio (VSWR) to distinguish it from the standing-wave
ratio expressed as a power ratio which is (Emax / Emin) squared."

In my long rxperience, I`ve found it`s never profitable to argue with
Terman. He is as close to infallible as any wrirter I`ve ever read.

Terman may be infallible, but I often find it unwise to trust his
interpreters.

The mention of SWR strongly implies lossless lines since VSWR varies
along a lossy line. Perhaps in prose previous to the equation above
he has limited his discussion to the lossless case. Quotes out of
context must be interpreted with great care.

....Keith

"Dr. Slick" wrote:

wrote in message ...
Cecil Moore wrote:

wrote:
And yes, |rho| can be greater than unity for a passive load.

But the power reflection coefficient cannot be greater than 1.0
which is what the argument is all about.

Which is entirely consistent with my previous statement:
It follows that when rho is greater than unity, it is not 'physically
meaningful to separate the total power as the sum of the incident and
reflected power' so the equation
|rho| = Sqrt(Pref/Pfwd)
has no meaning.


It certainly does, because the ratio Pref/Pfwd is directly related
to the
ratio [rho]. Consider that after the absolute value brackets, the
phase information is gone. But since we are going to a ratio of
average (RMS)
values OR peak values of power, it doesn't matter.

Are you gonna re-write some books?

Don't think I need to. But many need to read their books with more care.

The root question is:
Do Prev and Pfwd have physical meaning?
or
Are Prev and Pfwd just poorly named quantities which are useful
in certain (common) circumstances?

Once you settle on the latter definition, your difficulties will
disappear and you will be free to experience rho through its full
range of values.

The proof that Prev and Pfwd are not, in general, physical things
has been offerred on many recent threads.

Although useful, it is unfortunate that Pnet does equal Pfwd-Prev
for many common situations since people are tempted to generalize.
The belief then becomes so ingrained, that when examples are
presented which demonstrate the lack of generality, the
example is rejected, or the equations modified, rather than
examining the incorrect beliefs which lead to the difficulties.

....Keith

"Dr. Slick" wrote:

wrote in message ...

And yes, |rho| can be greater than unity for a passive load.

...Keith

Absolute Rubbish.. Could you produce a passive circuit that
will reflect a greater voltage than what you feed it? I'd
LOVE to see that...

Several examples have been presented, but rather than accepting
them, you changed the definition of rho. Perhaps you could build
one of these circuits to determine if modifying the definition
of rho was appropriate.

The ratio Pref/Pfwd is directly related to the ratio [rho].

Pref/Pfwd = [rho]**2 Absolute value brackets are a must!

Consider that after the absolute value brackets, the phase information
is gone. But since we are going to a ratio of average (RMS)
values OR peak values of power, it doesn't matter.

In other words, if you use V**2/R, the "V" can be either peak or
RMS, it doesn't matter, because it is a ratio. And of course, the "R"
doesn't matter either. And of course, the phase information is gone
with
the absolute value brackets.

If you agree that the Pref/Pfwd ratio cannot be greater than 1

Which I haven't since Pref and Pfwd are just computed numbers and
the result for some circuits is that Pref/Pfwd is greater than 1.
Of course, Pnet is not equal to Pfwd-Pref in these circumstances
so there is no violation of basic physics. It is just that the
computation of Pfwd and Pref does not really produce real powers
(though, again unfortunately, the dimension of the quantity produced
is power).

for a passive network, then neither can the [Vref/Vfwd]= rho be
greater
than 1 either.

....Keith

Richard Harrison wrote:


In my long rxperience, I`ve found it`s never profitable to argue with
Terman. He is as close to infallible as any wrirter I`ve ever read.


All of the handy-dandy transmission line formulas
that we have been using for many years apply
specifically to lossless lines.

A line with loss has a complex value of Z0. If the
imaginary part of Z0 is more than a few percent of
the real part we should use different methods.

One famous example:

Pload = Pforward - Preflected

is one that has to be treated with suspicion if
the line has appreciable loss (complex Z0).

Another is :

SWR = [1+|rho|]/[1-|rho|]

At high values of rho close to 1.0, SWR becomes a
totally useless concept. This is true regardless
of which formula for rho that we use.

We use the Smith chart outer circle to plot
lengths of transmission line, for example stubs
and matching transformers. We assume these lines
taken by themselves are lossless and have infinite
SWR (the outer circle of the Smith chart is the
"locus" of infinite SWR).

If we know the matched loss of a particular coax
(dB per 100 ft) it is far better to use a math
program and calculate everything, if the matched
loss is not negligible. The computer is much more
revealing than the Smith chart when line loss is
significant.

Bill W0IYH

Dr. Slick wrote:
If you agree that the Pref/Pfwd ratio cannot be greater than 1
for a passive network, then neither can the [Vref/Vfwd]= rho be
greater than 1 either.

But apparently rho=(Z2-Z1)/(Z2+Z1) can be greater than unity. So
those are not the same reflection coefficients. A physical rho
and an image rho are quite often different values. s11 is a physical
rho. Vref/Vfwd is an image rho.
--
73, Cecil http://www.qsl.net/w5dxp



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

Dr. Slick wrote:
If you agree that the Pref/Pfwd ratio cannot be greater than 1
for a passive network, then neither can the [Vref/Vfwd]= rho be
greater
than 1 either.

Sqrt(Pref/Pfwd) cannot be greater than one. (Z2-Z1)/(Z2+Z1) can be
greater than one. Both are defined as 'rho' but they are not
always equal. (Z2-Z1)/(Z2+Z1) is a physical reflection coefficient.
Sqrt(Pref/Pfwd) is an image reflection coefficient.
--
73, Cecil http://www.qsl.net/w5dxp



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