Cecil Moore wrote:
The formula for theoretical TOTAL losses in a *resonant* stub:

Total loss = 10*log{[(Z0-R)/(Z0+R)]^2}

where R is the measured resistance of the resonant stub and Z0
is the characteristic impedance of the stub material. You can
see the [(Z0-R)/(Z0+R)]^2 term is akin to a virtual rho^2 at
the mouth of the stub. Since rho^2 = Pref/Pfor, the losses in
the stub are equivalent to the losses in an equivalent resistance
equal to the measured virtual resistance at the mouth of the stub.

In other words, replace the stub with a resistor having the same
value of measured resistance as the stub, and calculate the I^2*R
losses in the resistor. That will be the same value as the total
losses in the stub.
--
73, Cecil http://www.qsl.net/w5dxp

On Sun, 28 Nov 2004 18:41:44 GMT, Richard Clark
wrote:

|On Sat, 27 Nov 2004 21:43:14 GMT, (Robert Lay
|W9DMK) wrote:
|
| I can see now that the
|Additional Losses Due to SWR really are dissipative and are unrelated
|to the "Mismatch Losses" and "Transducer Losses" defined on page 22-12
|of the ITT Handbook, 5th Ed.
|
|Hi Bob,
|
|I've let this simmer for a while, but I have to return to this because
|you've erred in interpretation of this particular page and those
|particular subjects. They are entirely caloric losses, not what you
|dismiss as the myth of mismatch loss.
|
|You need only review the math offered to observe they use the
|conventional "real" line loss and add more "real" line loss in
|proportion to the reflections at either one or two interfaces. The
|equations are quite literal to this and explicitly state:
| A0 = normal attenuation of line
|
|If you want deeper math, one source can be found in Chipman's (as
|unread as any here) "Transmission Lines."
|
|This is yet another of my references that attend to my recent, short
|thread on the nature of power determination error, and mismatched
|loads AND sources. In fact ALL of these references I've offered
|explicitly describe that the source MUST be matched for ANY of these
|equations about transmission lines bandied about to accurately offer
|true answers. The naive presumptions that Source Z is immaterial to
|the outcome of analysis is quite widespread here.
|
|Chipman offers the rigorous math that attends explicitly to the Smith
|Chart loss nomograph you reference elsewhere in this thread. If you
|lack access to this work, I can munge up the equations here for you.

Richard,

If you wouild cite the pages to which you refer, I would gladly scan
then to pdf and post them for all to reference.

Wes


|I will add, this math is for "lossless" lines, as is the implication
|of the Smith Chart nomograph; but it only requires you to add that in
|for yourself by restructuring the math to include loss. At that level
|of granularity, it won't be pretty; but you can rest assured it will
|be complete.
|
|73's
|Richard Clark, KB7QHC

On Sun, 28 Nov 2004 18:41:44 GMT, Richard Clark
wrote:

On Sat, 27 Nov 2004 21:43:14 GMT, (Robert Lay
W9DMK) wrote:

I can see now that the
Additional Losses Due to SWR really are dissipative and are unrelated
to the "Mismatch Losses" and "Transducer Losses" defined on page 22-12
of the ITT Handbook, 5th Ed.

Hi Bob,

I've let this simmer for a while, but I have to return to this because
you've erred in interpretation of this particular page and those
particular subjects. They are entirely caloric losses, not what you
dismiss as the myth of mismatch loss.

You need only review the math offered to observe they use the
conventional "real" line loss and add more "real" line loss in
proportion to the reflections at either one or two interfaces. The
equations are quite literal to this and explicitly state:
A0 = normal attenuation of line


I goofed on the part that is talking about transducer loss. I should
NOT have included the "Transducer Losses" in my statement above. The
Transducer losses do, as you say, include the normal attenuation of
the line, which is indeed a dissipative loss.

If you want deeper math, one source can be found in Chipman's (as
unread as any here) "Transmission Lines."

This is yet another of my references that attend to my recent, short
thread on the nature of power determination error, and mismatched
loads AND sources. In fact ALL of these references I've offered
explicitly describe that the source MUST be matched for ANY of these
equations about transmission lines bandied about to accurately offer
true answers. The naive presumptions that Source Z is immaterial to
the outcome of analysis is quite widespread here.

Chipman offers the rigorous math that attends explicitly to the Smith
Chart loss nomograph you reference elsewhere in this thread. If you
lack access to this work, I can munge up the equations here for you.
I will add, this math is for "lossless" lines, as is the implication
of the Smith Chart nomograph; but it only requires you to add that in
for yourself by restructuring the math to include loss. At that level
of granularity, it won't be pretty; but you can rest assured it will
be complete.

I'm not sure what you are saying about the loss nomograph on the Smith
Chart. If that's wrong too, then we're in big trouble. Everything ever
written that I have seen about the Smith Charts agrees that the actual
losses in the transmission line are indicated by the collapsing of the
circle as one traverses the transmission line. All you have to do is
read the "loss in 1 dB steps" scale to determine those losses.


Bob, W9DMK, Dahlgren, VA
http://www.qsl.net/w9dmk

On Sun, 28 Nov 2004 16:53:59 -0700, Wes Stewart
wrote:

If you wouild cite the pages to which you refer, I would gladly scan
then to pdf and post them for all to reference.

Hi Wes,

The math is on the bottom of pg. 203 which is supporting Fig. 9-26.

There is also a section entitled 8.8 Multiple reflections on ppg
174...176.

Then there is the specific math of fully specified matches at both
ends, that is at the source and the load, found in Fig. 10-7 that is
supported by discussion on ppg. 225...227.

All of this bears on discussion around and about the necessary
treatment of the Z of the Source, but I haven't supplied all the
citations within this one reference by any means.

Thanx, Wes. You needn't do all these scans. The group needs to do
their own heavy lifting to escape their naivety about source Z.

73's
Richard Clark, KB7QHC

On Mon, 29 Nov 2004 00:02:10 GMT, (Robert Lay
W9DMK) wrote:
I'm not sure what you are saying about the loss nomograph on the Smith
Chart. If that's wrong too, then we're in big trouble.

Hi Bob,

Chipman devotes an entire chapter on the Smith Chart, covering each
scale completely (I wonder why Reg thinks it would take so many months
to accumulate this information if he, as he alludes, is already aware
of Chipman's work.).

However, the nomograph discusses nothing in terms of any particular
line's dissipative loss as that is outside of purpose of the chart.
I'm afraid you are back in the mythical mismatch loss there. When Wes
puts that particular page online, then perhaps you may find a way to
fully describe your situation - including loss specific to your tests.

Chipman states quite bluntly:
"IF the transmission line of Fig. 9-26 is lossless...."

The point of my posts is that few actually read this material to
appreciate the very carefully stated premises. The problems arise
when derivative work is used as a source wherein that author presumes
that the reader is educated in the particulars that go left unsaid....
like all Sources' Z MUST match the line for any analysis to be valid.
Instead we get these trivializations of the matter that this is only a
concern inhabiting only specialty sources in specialty labs.

However, as applied to your recent work, the degree of mismatch
offered by your source is low enough to be negligible; but as you are
not aware of its degree, then neither are you aware of the scope of
your error.

73's
Richard Clark, KB7QHC

Just a few meandering thoughts -

For anyone assembling a collection of exact transmission line formulae, of
which there are dozens, some more applicable to practical problems than
others, it is best to begin with the most commonly used and the most
complicated formulae. Don't incorporate approximate formulae or you will
later wish that you hadn't.

Clear your minds of superfluous Smith Charts, standing waves, reflected
power, virtual short circuits, conjugate matches, etc. A clear understanding
of how transmission lines work is essential. You should be familiar with
complex hyperbolic functions. Only metric line dimensions should be used.

What are required are calculating procedures which accepts all possible
input data and finish with preferably a single number. In some cases, if not
needed, input data can be set to zero but the facility must exist.

You will then have designed a set of step-by-step routines as in computer
programs but which can be tediously and logically worked through with a
pocket calculator. The number of intermediate variables can be large. But
there can be only one unambiguous straight-line path through subroutines.

For example, in a large number of cases the single output quantity is
related to line loss, such as insertion loss in dB, or load power in watts,
or transmission efficiency in percent, or percent of input power lost in the
line itself.

But before this can be calculated it is essential to calculate input
impedance Rin+jXin for given attenuation in dB or nepers, given phase shift
in radians and given terminating impedance Rt+jXt. Then include generator
impedance Rg+jXg and internal generator volts. Having done this you are
half-way through.

Some intermediate results may be useful. Such as input impedance which
terminates a tuner or provides a source for a receiver.

Such intermediate results as reflection coefficient magnitudes and angles
may be explicitly available but may be of no practical use. What can you do
with them? The calculation is already complete. You might find an SWR
somewhere in there if you recognise it but who cares.

Well, you get the idea. But if you had the source codings of some of my
programs I can assure you they would not be of the slightest use. You may
just as well start at the begining. There are many ways of accomplishing the
same task. A mathematical program is a work of art as much as it is a set
of logical rules. But only the programmer can fully appreciate the beauty.

First prepare a list of proposed interrelated calculating formulae or
routines.
Then write the routines on paper.
Then test them on a computer.
Then spend the next 12 months removing the bugs.
Ditto, removing the bugs caused by the debugging operations.
Then publish them in the ARRL Handbook, 2009 edition, using a better
printer.
----
Reg.

On Mon, 29 Nov 2004 04:09:50 +0000 (UTC), "Reg Edwards"
wrote:

But only the programmer can fully appreciate the beauty.

Hi Reggie,

Nice posting.

Once, some many (few in your perspective) years ago, you once retorted
to my style with "this is not rec.radio.amateur.antenna.poetry."
There is more than the bouquet of that romantic tendency in your last
posting. However, neither of us is really surprised to find that in
the other. ;-)

73's
Richard Clark, KB7QHC

On Mon, 29 Nov 2004 00:28:22 GMT, Richard Clark
wrote:

|On Sun, 28 Nov 2004 16:53:59 -0700, Wes Stewart
|wrote:
|
|If you wouild cite the pages to which you refer, I would gladly scan
|then to pdf and post them for all to reference.
|
|Hi Wes,
|
|The math is on the bottom of pg. 203 which is supporting Fig. 9-26.
|
|There is also a section entitled 8.8 Multiple reflections on ppg
|174...176.
|
|Then there is the specific math of fully specified matches at both
|ends, that is at the source and the load, found in Fig. 10-7 that is
|supported by discussion on ppg. 225...227.
|
|All of this bears on discussion around and about the necessary
|treatment of the Z of the Source, but I haven't supplied all the
|citations within this one reference by any means.
|
|Thanx, Wes. You needn't do all these scans. The group needs to do
|their own heavy lifting to escape their naivety about source Z.

Hi Richard,

I did it anyway. [g] Hope this covers it:

http://users.triconet.org/wesandlind...rdClarkRef.pdf

Regards,

Wes

On Mon, 29 Nov 2004 10:27:37 -0700, Wes Stewart
wrote:
I did it anyway. [g] Hope this covers it:
http://users.triconet.org/wesandlind...rdClarkRef.pdf

Hi Wes,

Thanx very much. I can see one of two results from this general
availability. The readership here can:
1. Avoid it in stunned shame (the embarrassment in coming of age);
2. Accept it as a remarkable revelation (because it's on the web).

I would hope for a third response from those who could argue what
follows from these first principles, but the lazier ones would
complain of my "attitude" and hobble back to their beauty contests on
their crutches. ;-)

To quote one of my favorite authors, Raymond Chandler, when in "The
Big Sleep" Doghouse Reilly is admonished about the same defect, he
avers "I don't mind if you don't like my manners. They're pretty bad.
I grieve over them during the long winter evenings."

73's
Richard Clark, KB7QHC

I offer a third third response.

On p. 175, Chipman states:

"Equation (8.27) demonstrates explicitly that the shape of a standing
wave pattern representing |V(d)| as a function of d on a transmission
line is in no way affected by the quantities Vs, Zs and [rho]s at the
source."

And equation 8.29 on p. 176, the calculation of reflection coefficient,
contains no source-dependent terms. I'm sure that somewhere in the book,
the author derives SWR in terms of reflection coefficient.

These are the facts:

1. The SWR, positions of the standing waves, reflection coefficient seen
looking into the line, impedance seen looking into the line, and dB line
loss are independent of source impedance.
2. The actual amount of power delivered to a line for a given Thevenin
source voltage will, of course, depend on the source impedance, just as
it would if the source were directly connected to a load. Therefore, the
absolute amount of power dissipated in the load depends on source
impedance. The dB line loss, however, doesn't. Also, the length of time
the line requires to reach equilibrium after initially turning on the
source depends on the source impedance.

These can be found, explicitly stated and/or in easily interpreted
equation form, in a host of references.

I see nothing in the text Wes has kindly posted which contradicts these
facts, and I'm sure there's nothing elsewhere in the text that does.

I often have a hard time understanding Richard's postings, so it's
possible that he's not disagreeing with the statements I've made,
either. If so, I apologize for the misinterpretation.

Roy Lewallen, W7EL

Richard Clark wrote:

On Mon, 29 Nov 2004 10:27:37 -0700, Wes Stewart
wrote:

I did it anyway. [g] Hope this covers it:
http://users.triconet.org/wesandlind...rdClarkRef.pdf


Hi Wes,

Thanx very much. I can see one of two results from this general
availability. The readership here can:
1. Avoid it in stunned shame (the embarrassment in coming of age);
2. Accept it as a remarkable revelation (because it's on the web).

I would hope for a third response from those who could argue what
follows from these first principles, but the lazier ones would
complain of my "attitude" and hobble back to their beauty contests on
their crutches. ;-)

To quote one of my favorite authors, Raymond Chandler, when in "The
Big Sleep" Doghouse Reilly is admonished about the same defect, he
avers "I don't mind if you don't like my manners. They're pretty bad.
I grieve over them during the long winter evenings."

73's
Richard Clark, KB7QHC