Walt: For somebody who missed out on Reflections II, it looks like a lot of
fine reading - many thanks!

Sven: Got delivery here in Holland today.

73,
Kjell PB3SM - SM6CPI


"Sven Lundbech" wrote in message
k...
"walt" skrev i en meddelelse
...
To order a copy of the 3rd edition of Reflections, authored by Walter
Maxwell, W2DU, go to the W2DU website at www.w2du.com, then go to page
2 and click on the following:

Order Reflections III from CQ Online BookStore

Walt, W2DU

Dear Walt,

Ever since reading your 1973 QST articles, being a second-year student at
the Polytechnics/Technical University of Copenhagen, I have referred to
your ground-breaking work for removing the myths and misunderstandings of
the (un)importance of the swr over and over again. Somewhere I still have
the ageing photocopies of all the articles. But having the opportunity to
get it all and more in one piece in Reflections III is really greast. I
have ordered my copy right away.

Later on, both in my professional life after receiving my MSc(EE) as a
radio engineer and during my 19 years as Technical Editor of 'OZ' (the
Danisk QST) I have had to weed out misconceptions about swr and
transmission lines. My early understanding from your work was a great
help. Thanks!

All the best,

vy 73
OZ7S Sven

"Kjell W. Strom" skrev i en meddelelse
...
Walt: For somebody who missed out on Reflections II, it looks like a lot
of fine reading - many thanks!

Sven: Got delivery here in Holland today.

73,
Kjell PB3SM - SM6CPI


"Sven Lundbech" wrote in message
k...
"walt" skrev i en meddelelse
...
To order a copy of the 3rd edition of Reflections, authored by Walter
Maxwell, W2DU, go to the W2DU website at www.w2du.com, then go to page
2 and click on the following:

Order Reflections III from CQ Online BookStore

Walt, W2DU

Dear Walt,

Ever since reading your 1973 QST articles, being a second-year student at
the Polytechnics/Technical University of Copenhagen, I have referred to
your ground-breaking work for removing the myths and misunderstandings of
the (un)importance of the swr over and over again. Somewhere I still have
the ageing photocopies of all the articles. But having the opportunity to
get it all and more in one piece in Reflections III is really greast. I
have ordered my copy right away.

Later on, both in my professional life after receiving my MSc(EE) as a
radio engineer and during my 19 years as Technical Editor of 'OZ' (the
Danisk QST) I have had to weed out misconceptions about swr and
transmission lines. My early understanding from your work was a great
help. Thanks!

All the best,

vy 73
OZ7S Sven


Got delivery to-day.

As mentioned earlier, most of the stuff is old hat to me - but I really look
forward to dig into the chapters concerning tx output impedance. A highly
controversial subject for decades.

vy 73
OZ7S Sven

"Sven Lundbech" wrote in
k:

....
As mentioned earlier, most of the stuff is old hat to me - but I
really look forward to dig into the chapters concerning tx output
impedance. A highly controversial subject for decades.

Here is a simple little test for the hypothesis that Zs=50+j0 that uses
equipment found in many if not most HF ham shacks.

Owen

Owen Duffy wrote in news:Xns9D81BC11E3183nonenowhere@
61.9.191.5:

"Sven Lundbech" wrote in
k:

...
As mentioned earlier, most of the stuff is old hat to me - but I
really look forward to dig into the chapters concerning tx output
impedance. A highly controversial subject for decades.

Here is a simple little test for the hypothesis that Zs=50+j0 that uses
equipment found in many if not most HF ham shacks.

Oh, the URL: http://vk1od.net/blog/?p=1028 .


Owen

On May 23, 4:31*am, Owen Duffy wrote:
Owen Duffy wrote in news:Xns9D81BC11E3183nonenowhere@
61.9.191.5:

"Sven Lundbech" wrote in
. dk:

...
As mentioned earlier, most of the stuff is old hat to me - but I
really look forward to dig into the chapters concerning tx output
impedance. A highly controversial subject for decades.

Here is a simple little test for the hypothesis that Zs=50+j0 that uses
equipment found in many if not most HF ham shacks.

Oh, the URL:http://vk1od.net/blog/?p=1028.
Owen

While the analysis of transmitter output impedance in Reflections is
flawed,
experiments (claimed to be repeatable) described in Reflections appear
to
support the conclusions of the flawed analysis.

It would be highly valuable if the results of these experiments could
be
explained in a manner that aligns with established understandings.

Such an explanation might start by describing the circuit conditions
that
result from following the manufacturer’s tuning procedures. After
all,
these usually depend on measuring currents and voltages so are only
indirectly related to power. Perhaps the resulting conditions are not
as they are usually assumed to be.

Try as I might, I have not been able to derive a mechanism to explain
the observations in Reflections. But the explanations offered in
Reflections require large chunks of linear circuit theory to be
discarded,
so this does not seem to be an appropriate path.

....Keith

On May 24, 6:30*am, Keith Dysart wrote:
Such an explanation might start by describing the circuit conditions
that result from following the manufacturer’s tuning procedures.

On an old tube transmitter, e.g. a Globe Scout, when the manufacturer
specifed a particular grid current and a particular plate current,
does that imply a particular single resistive load line for the final
tube? Why were those particular grid and load currents chosen? Maximum
efficiency? Tube life? Minimum distortion?
--
TNX & 73, Cecil, w5dxp.com

On May 24, 9:23*am, Cecil Moore wrote:
On May 24, 6:30*am, Keith Dysart wrote:

Such an explanation might start by describing the circuit conditions
that result from following the manufacturer’s tuning procedures.

On an old tube transmitter, e.g. a Globe Scout, when the manufacturer
specifed a particular grid current and a particular plate current,
does that imply a particular single resistive load line for the final
tube? Why were those particular grid and load currents chosen? Maximum
efficiency? Tube life? Minimum distortion?

Excellent questions. I have often wondered if the manufacturer's
tuning
procedures had anything to do with maximizing output power transfer,
or
were they, in fact, optimizing some other aspect.

....Keith

On May 24, 7:30*am, Keith Dysart wrote:
On May 23, 4:31*am, Owen Duffy wrote:



Owen Duffy wrote in news:Xns9D81BC11E3183nonenowhere@
61.9.191.5:

"Sven Lundbech" wrote in
. dk:

...
As mentioned earlier, most of the stuff is old hat to me - but I
really look forward to dig into the chapters concerning tx output
impedance. A highly controversial subject for decades.

Here is a simple little test for the hypothesis that Zs=50+j0 that uses
equipment found in many if not most HF ham shacks.

Oh, the URL:http://vk1od.net/blog/?p=1028.
Owen

While the analysis of transmitter output impedance in Reflections is
flawed,
experiments (claimed to be repeatable) described in Reflections appear
to
support the conclusions of the flawed analysis.

It would be highly valuable if the results of these experiments could
be
explained in a manner that aligns with established understandings.

Such an explanation might start by describing the circuit conditions
that
result from following the manufacturer’s tuning procedures. After
all,
these usually depend on measuring currents and voltages so are only
indirectly related to power. Perhaps the resulting conditions are not
as they are usually assumed to be.

Try as I might, I have not been able to derive a mechanism to explain
the observations in Reflections. But the explanations offered in
Reflections require large chunks of linear circuit theory to be
discarded,
so this does not seem to be an appropriate path.

...Keith

Keith, would you please elaborate on why you believe my analysis of
transmitter output impedance is flawed? And what is the basis for your
belief that my explanations in Reflections require large chunks of
linear circuit theory to be discarded. Could it be because you
consider the source resistance in the transmitter to be dissipative,
as in the classical generator? If so, you must be made to realize that
the source resistance of the transmitter is non-dissipative, which is
the reason that its efficiency can exceed 50%.

Or are you considering the output characteristic of the transmitter to
be non-linear? This is not the case, because the effect of energy
storage in the tank circuit isolates the non-linear input from the
output circuit, which is linear as evidenced by the almost perfect
sine wave appearing at the output of the tank.

One last question: Are you basing your dissatisfaction of Reflections
from reviewing the 2nd or 3rd edition? Chapter 19 has been expanded in
the 3rd edition, in which I presented additional proof of my position
on the subject that you should be aware of. If you haven't yet seen
the addition that appears in the 3rd ed, please let me know so that I
can send you a copy of the addition. Also include your email address
so I can send it.

Keith, you are the only person I know of who appears to have found
flaws in my presentation on this subject. Which is why I'm anxious to
know exactly why you believe my presentation is flawed.

Walt Maxwell, W2DU

On May 24, 10:55*am, walt wrote:
Keith, would you please elaborate on why you believe my analysis of
transmitter output impedance is flawed? And what is the basis for your
belief that my explanations in Reflections require large chunks of
linear circuit theory to be discarded. Could it be because you
consider the source resistance in the transmitter to be dissipative,
as in the classical generator? If so, you must be made to realize that
the source resistance of the transmitter is non-dissipative, which is
the reason that its efficiency can exceed 50%.

No problems there. There has been much confusion in this area and
anything
that reduces this confusion is beneficial.

Or are you considering the output characteristic of the transmitter to
be non-linear? This is not the case, because the effect of energy
storage in the tank circuit isolates the non-linear input from the
output circuit, which is linear as evidenced by the almost perfect
sine wave appearing at the output of the tank.

This may be the root of my disagreement. Certainly the output can be
an
arbitrarily perfect sine wave, but this simply depends on the
characteristics of the filter and not on whether the system is linear.

But the way the filter transforms the impedances is the crux of the
issue.

It is my understanding that the input impedance to a filter can be
computed by starting with the load impedance applied to the filter and
then, using the rules for series and parallel connected components,
compute the way through the filter until reaching the input and the
result is the input impedance to the filter.

Similarly, the output impedance of the filter can be computed by
starting with source impedance driving the filter, series and
paralleling
the components until reaching the output and the result is the output
impedance of the filter.

The desired impedance for the input to the filter is that impedance
which
produces the desired load on the tube. And the component values are
computed to produce this load on the tube when the correct load is
attached to the output.

For the output impedance of the filter, the question then becomes:
What
is the source impedance driving the filter? If the source is
constructed
as a Class A amplifier, then it depends on the controlling device,
and
for the simplest of circuits would be Rp of the tube. (Just for
clarity,
in this discussion Rp is the slope of the plate E/I curve with
constant
grid voltage. In an ideal tube, these lines are equidistant apart and
the
slopes are the same. Real tubes, of course, are not so well behaved,
but
this should not affect the basic discussion.)

Since the component values for the filter were chosen to provide the
optimum load to the tube, and the optimum load value has no relation
to
Rp, there is no reason to expect the filter will transform Rp to be
the conjugate of the load impedance.

For amplifiers where conduction is not for 360 degrees (Class AB, B,
C),
the controlling device is no longer time-invariant so the rules for
linear circuit analysis no longer apply. None-the-less, for example,
consider a Class AB amplifier where the tube is only cut off for 1
degree. This short cut-off would not have much affect so the analysis
for Class A would apply. As the cut-off period increases the behaviour
will diverge more and more from that of the Class A amplifier.

Simulations produce some interesting results:

Another way of measuring the source impedance is to observe the effect
on a reflected wave entering the amplifier from the load. With a
Class
C amplifier, simulation reveals that the effect on the reflected wave
depends on the phase of that wave with respect to the drive signal
applied to the tube. As the phase of the reflected wave is changed,
the reflection co-efficient experienced by the wave changes. Truly a
non-linear behaviour. Intriguingly, when the conduction angle is
exactly
180 degrees, this effect largely disappears, and the result is much as
if the source impedance of the tube was 2 times Rp, which seems to
make some sense since the tube is only conducting one-half of the
time.

One last question: Are you basing your dissatisfaction of Reflections
from reviewing the 2nd or 3rd edition? Chapter 19 has been expanded in
the 3rd edition, in which I presented additional proof of my position
on the subject that you should be aware of. If you haven't yet seen
the addition that appears in the 3rd ed, please let me know so that I
can send you a copy of the addition.

I have been reading the .pdfs at w2du.com along with correspondence
and
other writings in QST, QEX and newsgroups.

The expanded Chapter 19 at w2du.com offers more experimental evidence
that seems to support the hypothesis that the transmitter is conjugate
matched to the load after tuning,

But given, from circuit analysis, that the output impedance can not be
well defined for any but a Class A amplifier, the fascinating question
is why is there experimental evidence that agrees with the premise
that
the output impedance of a tuned transmitter is the conjugate match of
the load?

One simple example to consider which has similar behaviour is a bench
power supply that also has a constant current limiter. Set such a
power
supply to produce a voltage of 100V (more precisely a maximum voltage)
and a current limit of 2A. Apply a variable load. Maximum power will
be drawn when the load resistance is 50 ohms. Varying the resistance
on either side of 50 ohms will reduce the power which might be
misconstrued to suggest that the power supply has an output impedance
of 50 ohms, when, in fact, it has a infinite output impedance when
the load is below 50 ohms and a zero output impedance when the load
is above.

I have looked for such a simple explanation in the circuits of the
transmitters used in the experiments but was not able to find one.
So I am still puzzled by the observations.

Also include your email address so I can send it.
Keith.dot.dysart.at.gmail.com .dot. = . .at. = @

…Keith

On 25/05/2010 01:55, walt wrote:
....
Or are you considering the output characteristic of the transmitter to
be non-linear? This is not the case, because the effect of energy
storage in the tank circuit isolates the non-linear input from the
output circuit, which is linear as evidenced by the almost perfect
sine wave appearing at the output of the tank.

Well, for the purposes of application of linear circuit theory,
linearity means that V is linearly related to I, or at least dV/dI over
the operating range is substantially constant.

If the circuit is not linear in those terms, then you cannot form a
valid Thevenin equivalent circuit, and discussion of the Thevenin
equivalent series source impedance is a nonsense... it cannot be used.

Owen