Reflecting and standing waves in wires

[Reg Edwards]

Perhaps you should write a definitive essay on it=============================

Sorry, Reg. Can you repost the link? Nothing appeared here.

=============================

One line thesis repeated here -

"The SWR meter located at the output of the transmitter does NOT measure the
standing waves on the transmission line to the antenna."

If any more proof is needed, when the SWR meter used with the G5RV on rare
occasions reads 1:1 the actual SWR on the 1/2-wave feedline is at least 9:1

What a shocking state of affairs which the many scores of thousands of full
and half-size G5RV fanatics are entirely ignorant. Louis Varney kept it an
even darker secret than Saddam did and is still doing with his WMD.

I have now exposed the whole sordid mess to the light of reason. But why
don't I feel safe when walking around poorly lit alleys.

 

[Active8]

Old Wives' Tales.


The SWR meter located at the output of the transmitter does NOT measure
standing waves on the transmission line to the antenna.

I never said it does. I have one that measures forward and reverse
power and from there I can calculate the VSWR but I don't have to
because it's a crossed needles display

Most of you are fooling yourselves.


The instrument indicates only WHETHER or NOT the transmitter is loaded with
the correct design value of resistance, usually 50 ohms. Which, under
normal operating conditions, is all anybody needs to know anyway.


And the internal impedance of the transmitter has nothing to do with it.
Even the transmitter designer has no interest in what the internal impedance
is. It is NOT the designer's objective to maximise output power by matching
the internal impedance to the load.

Well I did say you can purposely mismatch a tranny to get less
power.

What *is* his main objective? I think I know what your trying to
say but I'm not sure that what I heard is what you meant.

 

[Active8]

Hello Paul,
just read carefully what Reg has written, and if you happen to
come across a 100 metre drum of coax cable try this experiment.

Tx-VSWR meter-----------long coax cable------------short/open/50

Use an old cheap VHF taxi radios or whatever you have.
Put the VSWR meter at the transmitter end of the long
coax cable and then mess about with the far end, by attaching
say a good antenna, 50 ohm load, known bad antenna,
a short circuit, open circuit, 25 ohm load.
Take note of what the VSWR meter reads in all cases.

Next, move the VSWR meter to far end of cable and try this,

Tx---------------------long coax--------------------VSWR meter-load

Put the various good loads and bad loads, short circuit,
open circuit and note what the VSWR meter reads in all
cases and compare these readings with the first experiment.

Read again what Reg has written and what you have
written about "assumed" radiation.
Hint: you get little radiation from coax cable with a dead short
at the far end.
What did the VSWR meter tell you, when it was at the TX end
and an open circuit or dead short was on the coax at the far end?

It probably told me that the return loss of the cable messed up my
reading. That's why we try to match the line to the antenna, no?
Stop der reflections from happening in der first place.

 

[Paul Burridge]

One line thesis repeated here -

"The SWR meter located at the output of the transmitter does NOT measure the
standing waves on the transmission line to the antenna."

If any more proof is needed, when the SWR meter used with the G5RV on rare
occasions reads 1:1 the actual SWR on the 1/2-wave feedline is at least 9:1

What a shocking state of affairs which the many scores of thousands of full
and half-size G5RV fanatics are entirely ignorant. Louis Varney kept it an
even darker secret than Saddam did and is still doing with his WMD.

I have now exposed the whole sordid mess to the light of reason. But why
don't I feel safe when walking around poorly lit alleys.

You're a great bloke, Reg. Take care.

 

[Reg Edwards]

The design objective is to obtain the maximum PA power output, using the
specified DC power supply, without exceeding device maximum power
dissipation, maximum voltage ratings, maximum current ratings, maximum
distortion level and maximum allowed harmonic generation, into a given load
such as 50 ohms.

After design is complete if the internal impedance could be measured it
would be found to be very much different from 50 ohms. Internal impedance
is much better calculated.

The maximum power transfer theorem, Rload = Rint, has been ignored. It
never entered the designers mind. In any case, the load should be varied to
match the generator rather than the other way about.

 

[Active8]

The design objective is to obtain the maximum PA power output, using the
specified DC power supply, without exceeding device maximum power
dissipation, maximum voltage ratings, maximum current ratings, maximum
distortion level and maximum allowed harmonic generation, into a given load
such as 50 ohms.
Yup.

After design is complete if the internal impedance could be measured it
would be found to be very much different from 50 ohms. Internal impedance
is much better calculated.

The maximum power transfer theorem, Rload = Rint, has been ignored. It
never entered the designers mind. In any case, the load should be varied to
match the generator rather than the other way about.

You mean as in making the load (50 or 75 ohms typ after matching
that bloody kludge of tubes and wires to the line) "look" like
something that conjugately matches the output active(s) and
hopefully incorporating a harmonic suppresion filter into the
matching network, right?

So in the absence of good data on the power tranny, you'd say

RL = (Vcc - Vce(sat))^2 / 2.P

and RL is what the tranny wants to see.

I wasn't trying to imply that one would try to make the active's
Zout equal to 50 ohms, in fact, I didn't even hint at that.

 

[Rick]

Within my limited knowledge on the subject, the SWR meter simply tells
you the ratio between forward and reflected power at the point in the
line the measurement is made.

Nope. The ratio you've described is Gamma, the reflection coefficient. SWR
is the ratio of the maximum/minimum voltage seen as a result of the
superposition of the forward and reverse "waves".

 

[Reg Edwards]

So in the absence of good data on the power tranny, you'd say

RL = (Vcc - Vce(sat))^2 / 2.P

and RL is what the tranny wants to see.

===================================

Exactly. And what the internal impedance of the transistor or tube is has
nothing to do with it.

( In practice the internal impedance is not constant. It changes as the load
resistance changes. It changes as the drive level of the device is changed.
It would be silly to attempt to match the load impedance to it. )

 

[Mac]

The design objective is to obtain the maximum PA power output, using the
specified DC power supply, without exceeding device maximum power
dissipation, maximum voltage ratings, maximum current ratings, maximum
distortion level and maximum allowed harmonic generation, into a given load
such as 50 ohms.

After design is complete if the internal impedance could be measured it
would be found to be very much different from 50 ohms. Internal impedance
is much better calculated.

The maximum power transfer theorem, Rload = Rint, has been ignored. It
never entered the designers mind. In any case, the load should be varied to
match the generator rather than the other way about.

In datasheets for video drivers I often see a 75-Ohm series resistor.
Naturally, they have to have a voltage gain of two to make up for the
losses in the series resistor. But the whole point of this excercise is to
make sure that any energy reflected back from the load does not get
reflected again at the source, thus potentially setting up a larger
standing wave.

Reg's statements are doubtless true for power amplifiers, but for signal
applications, it is sometimes necessary to use resistors in series with a
voltage source.

Lots of signal generators also use an actual 50-Ohm series
resistor. You can tell, because if you run them into a high-impedance
load, the amplitude is exactly double what it is with a 50-Ohm load.

Mac

 

[Active8]

In datasheets for video drivers I often see a 75-Ohm series resistor.
Naturally, they have to have a voltage gain of two to make up for the
losses in the series resistor. But the whole point of this excercise is to
make sure that any energy reflected back from the load does not get
reflected again at the source, thus potentially setting up a larger
standing wave.

Reg's statements are doubtless true for power amplifiers, but for signal
applications, it is sometimes necessary to use resistors in series with a
voltage source.

Lots of signal generators also use an actual 50-Ohm series
resistor. You can tell, because if you run them into a high-impedance
load, the amplitude is exactly double what it is with a 50-Ohm load.

Mac

Another example would be a CATV splitter which has a 150 ohm
resistor between both outputs ( || ) making it 75 per. A TDR would
have that series resistor on the output AFAIK.

 

[Russell Shaw]

=============================

Sorry, Reg. Can you repost the link? Nothing appeared here.

=============================

One line thesis repeated here -

"The SWR meter located at the output of the transmitter does NOT measure the
standing waves on the transmission line to the antenna."

Wrong (unless it's between the transmitter and a matching network).