Why S - Parameter at High frequencies?
 

I am in a process of understanding the importance of S-Parameter at
high frequencies, please answer my following questions:

1. Why Measuring a wave (voltage or current) is more easier than
volage and current? ( Why measuring S-Parameter is more easier than
other parameters?)

2. How the voltage and current waves are measured?

Measuring voltage and current are difficult at high frequncies, that
is why S-Parameter is more useful at high frequncies. Is the same
thing is true for simulation?

Regards,
Ilam

Why S - Parameter at High frequencies?
 

ilam wrote:
I am in a process of understanding the importance of S-Parameter at
high frequencies, please answer my following questions:

1. Why Measuring a wave (voltage or current) is more easier than
volage and current? ( Why measuring S-Parameter is more easier than
other parameters?)

The reflection coefficients are measurable physical entities
and there is a well-defined procedure for measuring them.
Virtual reflection coefficients, such as the ones used by
hams, are not used with S-Parameters. That's an advantage.
The voltages are normalized to the characteristic impedance
of the transmission line and everything obeys the rules of
the wave reflection model.

There are certainly other methods of analysis, h-parameters,
Z-parameters, etc.
--
73, Cecil http://www.w5dxp.com

Why S - Parameter at High frequencies?
 

"Cecil Moore" wrote in message
et...
ilam wrote:
I am in a process of understanding the importance of S-Parameter at
high frequencies, please answer my following questions:

1. Why Measuring a wave (voltage or current) is more easier than
volage and current? ( Why measuring S-Parameter is more easier than
other parameters?)

The reflection coefficients are measurable physical entities
and there is a well-defined procedure for measuring them.
Virtual reflection coefficients, such as the ones used by
hams, are not used with S-Parameters. That's an advantage.
The voltages are normalized to the characteristic impedance
of the transmission line and everything obeys the rules of
the wave reflection model.

There are certainly other methods of analysis, h-parameters,
Z-parameters, etc.
--
73, Cecil http://www.w5dxp.com

The problem with some of the other parameters is that
short and open circuits are required to obtain the data.
This can cause instability in active devices.

Frank

Why S - Parameter at High frequencies?
 

Frank wrote:
The problem with some of the other parameters is that
short and open circuits are required to obtain the data.
This can cause instability in active devices.

And that, in a nutshell, is why Keith's analysis
is invalid for real-world ham transmitters. In
a real-world transmitter, incident reflected
waves do not flow through the source like they
do in the model in the human mind.
--
73, Cecil http://www.w5dxp.com

Why S - Parameter at High frequencies?
 

On Apr 10, 4:21 am, "ilam" wrote:
I am in a process of understanding the importance of S-Parameter at
high frequencies, please answer my following questions:

1. Why Measuring a wave (voltage or current) is more easier than
volage and current? ( Why measuring S-Parameter is more easier than
other parameters?)

2. How the voltage and current waves are measured?

Measuring voltage and current are difficult at high frequncies, that
is why S-Parameter is more useful at high frequncies. Is the same
thing is true for simulation?

Regards,
Ilam

Hewlett-Packard were one of the early proponents of using S-parameters
for calculations as well as for measurements. HP had a vested
interest, in that they were one of the first companies to make
accurate instruments for measuring high frequencies, up into the
microwaves. Part of that effort was customer education, and there are
some good applications notes from HP about S-parameters, and about RF
measurements in general. You should be able to find many of them on
the internet, with a bit of searching. Some of them deal with
accuracy, and in them you should find answers to your questions.
You'll find a great deal more than just those answers.

Measuring voltage or current at audio frequencies is easy enough. But
how do you measure a current at 10GHz without disturbing and changing
it? How do you accurately measure a voltage at that frequency, when
the tiniest of capacitances will change the reading? But if instead
you want to measure the standing wave ratio on an accurately-made
piece of transmission line, that can be done relatively easily, and
especially you can determine when the standing wave ratio goes to
zero: when there is no variation in voltage along the length of the
line. Then too, you can build a bridge circuit to enable measurements
at a particular impedance. 50 ohms is convenient, though it's
somewhat arbitrary. If you pick too high or too low an impedance, you
run into the same difficulties as if you try to measure a voltage or a
current, but at impedances between perhaps 20 ohms and 100 ohms, in a
coaxial-line environment, you can build loads--resistors--that
accurately terminate the line, and you can build a Wheatstone bridge
circuit, essentially, that will tell you if you have a load that's the
impedance of the bridge arms. The bridge is the most sensitive at
balance, when all the arms are the same impedance.

Also note that while it's very difficult to measure current or voltage
accurately at high frequencies, if you can build a resistive load,
then measuring power accurately is possible. When you are first
starting, you can measure power by determining the temperature rise of
a load resistor, which you made by trimming things carefully so that
there were no standing waves on the precision line feeding that
resistor. You can calibrate that quite accurately by feeding it DC
power, to determine the temperature rise per watt. If you learn to
make very tiny resistors and measure very smal temperature
differences, you can make a thermal bridge that can detect power at a
low level accurately.

Let us know if you have trouble finding the HP applications
notes...look for S-parameter notes, and look for notes on making
accurate measurements with (vector) network analyzers. Also, do a
search for information on the history of RF and microwave
measurements.

For simulation, it is simply convenient to work in the same domain as
you make measurements. As you note, there are other parameter sets
that give you equivalent information, and with modern computers, there
is effectively no loss of accuracy if you deal in any other set of
parameters used for 2-port (or N-port) networks. But do note, please,
that these are linear parameters, and they won't give you information
about nonlinearities, either ones you wanted as in frequency mixers or
ones you didn't want, that cause unwanted distortion.

Cheers,
Tom

Why S - Parameter at High frequencies?
 

K7ITM wrote:
....
...and there are
some good applications notes from HP about S-parameters, and about RF
measurements in general. You should be able to find many of them on
the internet, with a bit of searching.

Of course you'll have to search at Agilent.com since 'HP' no longer
exists with regards to the the test and measurement world, but to get
you started,

'S-Parameters... Circuit Analysis and Design' a/k/a AN-95, an oldie but
a goodie,
http://cp.literature.agilent.com/lit.../5952-0918.pdf

Enjoy!
- Galen, W8LNA

Why S - Parameter at High frequencies?
 

Cecil Moore wrote in news:vyMSh.13914$JZ3.5605
@newssvr13.news.prodigy.net:

Virtual reflection coefficients, such as the ones used by

What is a "Virtual reflection coefficients"?

Owen

Why S - Parameter at High frequencies?
 

Owen Duffy wrote:
Cecil Moore wrote in news:vyMSh.13914$JZ3.5605
@newssvr13.news.prodigy.net:

Virtual reflection coefficients, such as the ones used by

What is a "Virtual reflection coefficients"?

SQRT(Pref/Pfor) which often differs from the
physical reflection coefficient (Z02-Z01)/(Z02+Z01)

Here's an example:

100W---50 ohm line--+--1/2WL 300 ohm line--50 ohm load

The virtual reflection coefficient looking into point
'+' from the source side is SQRT(Pref/Pfor) = 0

The physical reflection coefficient looking into point
'+' from the source side is (300-50)/(300+50) = 0.7143

The virtual reflection coefficient is an effect. The
physical reflection coefficient is associated with
the causes of reflections. The S-Parameter reflection
and transmission coefficients are physical. There
are *always* reflections at a physical impedance
discontinuity but they are often zeroed out by
wave cancellation.
--
73, Cecil http://www.w5dxp.com

Why S - Parameter at High frequencies?
 

Cecil Moore wrote in news:02USh.7679$u03.2956
@newssvr21.news.prodigy.net:

Owen Duffy wrote:
Cecil Moore wrote in news:vyMSh.13914$JZ3.5605
@newssvr13.news.prodigy.net:

Virtual reflection coefficients, such as the ones used by

What is a "Virtual reflection coefficients"?

SQRT(Pref/Pfor) which often differs from the
physical reflection coefficient (Z02-Z01)/(Z02+Z01)

So, what you are referring to with the term "virtual reflection
coefficient" is the magnitude the reflection coefficient (rho). or |(Z02-
Z01)/(Z02+Z01)| (Gamma).

Did you get that usage from an optics text?

I haven't seen "virtual reflection coefficient" used to have that
meaning. Nor have I seen hams commonly, as you suggest, using rho in the
way in which you could use S parameters or Gamma.

I did note that in a Google search, more than half the relevant hits on
the first page were your writings, so it is clear that you use the term.

Owen

Why S - Parameter at High frequencies?
 

Owen Duffy wrote:
I haven't seen "virtual reflection coefficient" used to have that
meaning. Nor have I seen hams commonly, as you suggest, using rho in the
way in which you could use S parameters or Gamma.

I am handicapped by being away from my reference books
but I believe that RF engineers use rho and Gamma
interchangeably along with other symbols like RC.

The difference between a virtual reflection coefficient
and a physical reflection coefficient has nothing to
do with what it is named. No matter what is the value
of the physical reflection coefficient, e.g. s11, the
virtual reflection coefficient at a Z0-match is zero
because the reflected voltage, current, and power is
zero.
--
73, Cecil http://www.w5dxp.com

Why S - Parameter at High frequencies?
 

Owen Duffy wrote:
So, what you are referring to with the term "virtual reflection
coefficient" is the magnitude the reflection coefficient (rho). or |(Z02-
Z01)/(Z02+Z01)| (Gamma).

I believe the following web page is indicative of
how Gamma has essentially been replaced by rho.

http://www.ac6la.com/stss.html

The ARRL Antenna Book implies that rho is used
by amateur radio operators while Gamma is still
used in some professional circles.
--
73, Cecil http://www.w5dxp.com

Why S - Parameter at High frequencies?
 

The ARRL Antenna Book implies that rho is used
by amateur radio operators while Gamma is still
used in some professional circles.
--

As far as I'm concerned, Gamma is the complex reflection coefficient;
rho is its magnitude.

Why S - Parameter at High frequencies?
 

Wes wrote:
The ARRL Antenna Book implies that rho is used
by amateur radio operators while Gamma is still
used in some professional circles.

As far as I'm concerned, Gamma is the complex reflection coefficient;
rho is its magnitude.

That works as long as everyone else understands what you
are talking about. Here's a quote from:

http://www.ac6la.com/stss.html

"The SWR meter is only concerned with the magnitude of the
reflection coefficient, but rho is a complex quantity having
both a magnitude and angle. If rho were to be measured at
various points along a transmission line, the magnitude would
be fairly constant but the angle would change depending on
the electrical length of the line at the point of measurement."

And from the "ARRL Antenna Book":

"In some professional literature, Gamma is used in place of rho
to represent the reflection coefficient."

Wes, you must be a professional. :-)
--
73, Cecil http://www.w5dxp.com

Why S - Parameter at High frequencies?
 

On Apr 12, 6:39 am, "Wes" wrote:

As far as I'm concerned, Gamma is the complex reflection coefficient;
rho is its magnitude.

As far as I'm concerned, S11 = (Z-Zo)/(Z+Zo), and is commonly known as
the (port 1) reflection coefficient, and Zo = 50+j0 unless otherwise
specified -- and |S11| is the magnitude of that reflection
coefficient. If I see rho without a definition, I only have its
context to go by, and in some circles that's pretty weak and often
inaccurate. For other ports of a multiport network, of course, S22,
S33, and so forth serve.

Almost all I need to know about a TEM line can be expressed by the set
of 2-port S parameters versus frequency for the line. Normal coaxial
and open-wire lines come about as close as anything I work with to
being true linear systems. However, the two-port model covers only
the differential TEM propagation on the line, not the line versus
ground: it's not useful for analyzing the "antenna" currents on a
line. It also doesn't tell me whether lost power is lost to heating
or to radiation, and it doesn't tell me about radiation received by
the line. The S parameter set for the line can be referenced to 50
ohms (or any other useful impedance), independent of the impedance of
the line. With the S parameter set, I can determine power loss, power
transmission, images, ... all the usual things. Of course, it's not
the only way to characterize the line, but it's complete and accurate
to the extent the line really does behave linearly and that only the
TEM propagation is important. The line doesn't even have to be
uniform; it can be exponentially expanding, or have ripples in its
impedance. If the line isn't uniform, the S parameters won't be able
to tell you power dissipation versus distance along the line, but
they'll still give you net power dissipation under specified
conditions of excitation and load.

Cheers,
Tom

Why S - Parameter at High frequencies?
 

"Wes" wrote in news:1176385196.714553.293960
@y5g2000hsa.googlegroups.com:



The ARRL Antenna Book implies that rho is used
by amateur radio operators while Gamma is still
used in some professional circles.
--

As far as I'm concerned, Gamma is the complex reflection coefficient;
rho is its magnitude.

Hi Wes,

My preferred usage is the same, and for the reason that Gamma and rho can
be written directly in html as the greek symbols, and in ascii text as
Gamma and rho. The notation that rho is the magnitude of the complex
reflection coefficient and that the complex reflection coefficient is
written as rho with a bar over it is not conveniently and consistently
expressed in a short notation in ascii only media.

If it is declared (as I did in my post) then the usage is fair. The
sidetrack onto "standard" notation was just another diversionary tactic
from YKW.

Owen

Glossary:
YKW - You Know Who

Why S - Parameter at High frequencies?
 

This topic has been discussed in this newsgroup several times in the
past. In my posting of June 17, 2005 in the thread " Calculating VSWR
from rho and rho from VSWR" you'll find a summary of what I found in use
by authors of 16 different textbooks. As you'll see, if you look it up
at http://groups.google.com, there's no single standard usage.

Personally, my preference is the same as Wes' and Owen's - Gamma for
complex reflection coefficient, rho for its magnitude. But usage is so
varied that it's necessary to explicitly show what the symbols mean
unless it's obvious from the context.

Roy Lewallen, W7EL

Owen Duffy wrote:
"Wes" wrote in news:1176385196.714553.293960
@y5g2000hsa.googlegroups.com:

The ARRL Antenna Book implies that rho is used
by amateur radio operators while Gamma is still
used in some professional circles.
--
As far as I'm concerned, Gamma is the complex reflection coefficient;
rho is its magnitude.

Hi Wes,

My preferred usage is the same, and for the reason that Gamma and rho can
be written directly in html as the greek symbols, and in ascii text as
Gamma and rho. The notation that rho is the magnitude of the complex
reflection coefficient and that the complex reflection coefficient is
written as rho with a bar over it is not conveniently and consistently
expressed in a short notation in ascii only media.

If it is declared (as I did in my post) then the usage is fair. The
sidetrack onto "standard" notation was just another diversionary tactic
from YKW.

Owen

Glossary:
YKW - You Know Who

Why S - Parameter at High frequencies?
 

On Apr 12, 2:04 pm, Roy Lewallen wrote:
This topic has been discussed in this newsgroup several times in the
past. In my posting of June 17, 2005 in the thread " Calculating VSWR
from rho and rho from VSWR" you'll find a summary of what I found in use
by authors of 16 different textbooks. As you'll see, if you look it up
athttp://groups.google.com, there's no single standard usage.

Personally, my preference is the same as Wes' and Owen's - Gamma for
complex reflection coefficient, rho for its magnitude. But usage is so
varied that it's necessary to explicitly show what the symbols mean
unless it's obvious from the context.


In my response in the thread "Question about reflection/transmission
coefficients", April 28, 2003 I had only 10 different references.

Why S - Parameter at High frequencies?
 

Wes wrote:
On Apr 12, 2:04 pm, Roy Lewallen wrote:
This topic has been discussed in this newsgroup several times in the
past. In my posting of June 17, 2005 in the thread " Calculating VSWR
from rho and rho from VSWR" you'll find a summary of what I found in use
by authors of 16 different textbooks. As you'll see, if you look it up
athttp://groups.google.com, there's no single standard usage.

Personally, my preference is the same as Wes' and Owen's - Gamma for
complex reflection coefficient, rho for its magnitude. But usage is so
varied that it's necessary to explicitly show what the symbols mean
unless it's obvious from the context.


In my response in the thread "Question about reflection/transmission
coefficients", April 28, 2003 I had only 10 different references.

Well, then, it's your turn again, Wes, and right on the biannual
schedule. Should it be 22 references (arithmetic increase) or 26
(exponential) this time?

Roy Lewallen, W7EL

Why S - Parameter at High frequencies?
 

On Apr 10, 4:21 am, "ilam" wrote:
I am in a process of understanding the importance of S-Parameter at
high frequencies, please answer my following questions:

1. Why Measuring a wave (voltage or current) is more easier than
volage and current? ( Why measuring S-Parameter is more easier than
other parameters?)

2. How the voltage and current waves are measured?

Measuring voltage and current are difficult at high frequncies, that
is why S-Parameter is more useful at high frequncies. Is the same
thing is true for simulation?

Regards,
Ilam

Search the agilent website for the original application note on S-
parameters.. it provides a great explanation of why S-parameters are
used.

You've correctly assessed the situation.. at microwave frequencies,
it's much easier to measure forward and reflected signals (and
calibrate for non-ideal directional couplers, etc.) than to measure
voltage and/or current.

Why S - Parameter at High frequencies?
 

Cecil Moore wrote:
Owen Duffy wrote:

So, what you are referring to with the term "virtual reflection
coefficient" is the magnitude the reflection coefficient (rho). or |(Z02-
Z01)/(Z02+Z01)| (Gamma).


I believe the following web page is indicative of
how Gamma has essentially been replaced by rho.

http://www.ac6la.com/stss.html

The ARRL Antenna Book implies that rho is used
by amateur radio operators while Gamma is still
used in some professional circles.

"still used"... I don't know that expressing reflection coefficient as
gamma is an archaic notation. You see it a lot in performance
specifications for RF test equipment like network analyzers. Where the
magnitude is meant (and Gamma isn't real) you'd see it written with
vertical bars on either side ||.

For instance, in the latest Agilent PNA manuals, they use capital Gamma.

here's an interesting interactive smith chart
http://www-ece.eng.uab.edu/JCaldwel/...art/SChart.htm


lower case gamma is used for the complex propagation constant in
transmission lines.

Why S - Parameter at High frequencies?
 

Cecil Moore wrote:

And from the "ARRL Antenna Book":

"In some professional literature, Gamma is used in place of rho
to represent the reflection coefficient."

I would say in *most* professional literature...

Why S - Parameter at High frequencies?
 

Jim Lux wrote:
Cecil Moore wrote:
"In some professional literature, Gamma is used in place of rho
to represent the reflection coefficient."

I would say in *most* professional literature...

The above quote from the ARRL Antenna Book is from
the 1988 edition.
--
73, Cecil http://www.w5dxp.com

Why S - Parameter at High frequencies?
 

Cecil Moore wrote:
Jim Lux wrote:

Cecil Moore wrote:

"In some professional literature, Gamma is used in place of rho
to represent the reflection coefficient."


I would say in *most* professional literature...


The above quote from the ARRL Antenna Book is from
the 1988 edition.

Hmm.. yes, well fashion does change in the professional literature over
time, and your citation is almost 20 years old. However, while cleaning
out a lab downstairs at work, I ran across a reprint of:
Kerns, David M., "Definitions of v,i, Z, Y, a, b, Gamma, and S", Proc.
IEEE, v55 n6, June 1967, pp892-900

no rhos to be found.

In another useful paper, Dybdal,R.B., Ott,R.H., "Coherent RF Error
Statistics", IEEE Trans MTT, vMTT-34,n12, Dec 86
(anyone doing precision measurements on RF systems with mismatches
should read this paper, if you want to get a handle on measurement
uncertainties..)
Gamma is used for the reflection coefficient, and rho is used for, of
all things, the phase of the reflection.

That's not to say that there's not a paper around from the 70s that uses
rho.. and certainly, if you go back to the 30s and 40s, one might find
rho (as a magnitude of the reflection coefficient) more commonly used,
reflecting (no pun intended) the use of instruments that didn't measure
phase. Precision phase measurement of reflections is a relatively
recent (last 30 years?) phenomena..

So, I'll revise and say in *most current* professional literature...