Towertalk
[Top] [All Lists]

[TowerTalk] Rebutal to Steve Best's "Initial State" Mismatch Arguments

To: <towertalk@contesting.com>
Subject: [TowerTalk] Rebutal to Steve Best's "Initial State" Mismatch Arguments
From: jreid@aloha.net (Jim Reid)
Date: Tue, 25 Aug 1998 08:35:51 -1000

-----Original Message-----
From: sbest@cushcraft.com <sbest@cushcraft.com>
To: kn6di@groupone.net <kn6di@groupone.net>; towertalk@contesting.com
<towertalk@contesting.com>
To: <towertalk@contesting.com>
Date: Tuesday, August 25, 1998 3:01 AM
Subject: Re[2]: [TowerTalk] Mismatch Loss and Tuners


Steve Best continues to write his stuff which has NO MEANING in
the HF spectrum world!!!!

For example:

>     In plain English:
>
>     One of the objectives in the transmit system is to eliminate
>     reflections arriving at the front end of the transmit circuit. This
>     can be accomplished in many ways.  The most common are to use an
>     antenna tuner or to use an antenna with a feedpoint impedance equal to
>     the characteristic impedance of the transmission line.

Reid comment:  [Which can only be done,  exactly,  at ONE single frequency
.]

>     With a matched antenna, all power arriving at the antenna will be
>     delivered to the antenna with no reflections.  The significant source
>     of loss in the system will be that which occurs as a result of cable
>     attenuation between the transmitter and the antenna.

>     To properly analyze a tuner we must first describe the initial state
>     of the tuner and then develop into the steady state of the tuner.  To
>     do a full analysis, voltages must be used in all calculations.  Steady
>     state power levels can be determined once the steady state voltages
>     are known.

Ok,  here we  go to develpe the initial state,  and an assume steady
state condition on a typical low loss coax transmission line,  conneted
between tuner and the antenna terminals.  Line will be 100 feet in length.
And we will use only voltage.

I have written to Steve numerous times pointing out the silliness of his
concept of the "initial state of the tuner"

Here is his definition of same:

>     In the initial state (the state prior to reflections arriving at the
>     tuner output from the antenna), the tuner will not present a 1.0:1
>     VSWR at its input.  Because of this, there will be an initial
>     reflected voltage present at the tuner input.

And when does the first reflection arrive back at the tuner?
On 100 feet of typical coax,  say low loss RG213, which has
a propagation velocity of 66% of the velocity in space,  the time
for an RF signal to travel out the line length,  and then reflect
back to the input end is 30 nanoseconds for the round trip.

What does that number mean,  in plain English??

Consider an RF circuit to responding  to such a short time pulse,that is
for the voltage of the reflected wave to reach about 90%
of it's eventual "initial value",  the bandwidth of response
required for the circuits to be "aware" of this initial voltage
would have to be 2700  MHz !!!   That is why even radar signals
at microwave frequencies between 1 and 10 GHz are not even
this brief; radar pulses below 10 GHz are in microseconds.

Now look again at Steve's following lines:

>     Therefore, the initial
>     power delivered through the tuner to the antenna will be less than the
>     total power delivered to the tuner from the transmitter.  The voltage
>     (power) delivered through the tuner will then arrive at the antenna
>     where some will be reflected (based upon antenna VSWR).  This
>     reflected voltage will arrive back at the tuner output.  At the tuner
>     output, another reflected voltage will be created based upon the tuner
>     VSWR.  Note that the tuner output VSWR will not be infinite therefore,
>     ALL of the voltage will not be re-reflected.  This process will
>     continue until the steady state condition is reached.

And how many of these multiple round trips will be made before
they have faded away to even less meaninglessness?  Well, lets
be generous,  and assume 20 round trips,  which is way more
than will actually occur.  So now we are talking about 20 times
30 nanoseconds to come to Steve's "steady state condition".
So we will assume it takes 600 nanoseconds to reach steady
state,  or 0.6 microseconds.  Now what bandwidth of circuit
is required to "be aware",  that is for all of this phenomena
to make ANY DIFFERECNCE RF ELECTRICALLY AT ALL.

The answer:  600 KHz.  How many of us have HF equipment
which has  that simultaneous,  instantaneous bandwidth?
Of course,  the answer is none,  in our amateur radio shacks.
The electrons in our recievers and transmitters can not
move in time to respond to the problem Steve insisits
is important.  It is not physically,  electrically possible for
the phenomena about which Steve writes to do anything
to,  within,  or about our HF signals -- it is all over and done
with before the transmitter knows anything about it, and
makes no difference at all to transmitting efficiency,  that
is the problem of reflections ONLY.  They are too brief
and fast for this "initial condition" reflection build
up transient phenomena  to do anything at HF.

It is similar to worrying about where on the RF cycle
you key your rig:  sure at one moment the RF cycle
voltage might be "0";  so at that instant,  nothing is
radiated from the antenna!  So.  per Steve,  that
would represent an inefficient engineering problem
which must be solved.  Actually,  the rise time of a
30 wpm CW "dit" is several milliseconds in order
that we not produce a key click transient;  and,  as
I have mentioned before,  the entire "dit" signal
will last about 40 milliseconds.  This is
thousands of times longer in time duration of any of the
"initial condition" states which concern Steve,  and which
he feels will reduce our antenna's radiating efficiency.

Steve goes on:

>In the steady
>     state condition, the multiple reflections will create a steady state
>     voltage at the tuner output.  This voltage will be delivered back to
>     the tuner input through the tuner components.  The voltage delivered
>     back to the tuner input will be the negative of the initial reflected
>     voltage at the tuner input causing a complete cancellation of the
>     initial reflection. Hence, the steady state 1.0:1 VSWR.
>
>     The difference between this discussion and that of Walt Maxwell's is
>     that Walt does not start by considering the initial state of the
>     tuner.

I hope  what I have tried to explain above,  helps all who might
have read this far,  see why Walt has no need
to consider the "initial state";  it is meaningless to HF rigs!!


>     In any event, even considering the transmission line losses, the
>     matched antenna will radiate more steady state power than the
>     mismatched antenna and tuner.

Well,  lets see how much more.  In 100 feet of 213 at 14 MHz,  my wire
tables show a loss of just less than 1 dB per 100 feet.  So lets call it
1 dB of loss.

In the first case,  perfect match between rig,  line,  and antenna,  the
loss is 1 dB of our power  converted to heat in the line;  or our
l voltage ratios at the antenna have dropped to about 89% of the
magnitude generated and input to the line at the rig input end.

Now,  lets suppose the line was mismatched to the anenna,  such
that the tuner has had to "tune out" a 3:1 impedance mismatch,
or a 3:1 VSWR on the line.  So,  how much of the 89% of the
voltage left at the antenna after the first trip up the line,  is going
to be reflected because of this mismatch?  Answer, 50%,
or half of the remaining 89%,  or 44 1/2 % will return back
down the line,  to suffer another drop of 89% due to the
trip back down to the tuner,  where it arrives with an
amplitude of 89% or 44 1/2% or,  now at 39% of its initial
voltage amplitude.  The tuner returns it back up the line,
and it arrives 15 nanoseconds later at the antenna terminals
with now an amplitude of 89% of 39%,  or  35%.  So after just
one reflection round trip,  we are going to add 50% more
voltage to the antenna input terminals 45 nanosecond after
the inital 89% of our transmitters output voltage got there.
This adds to the radiated energy going on,  as the key is
still down on our 30 wpm "dit",  in fact we have not even
reached the full ampliftude of the "dit" as yet because
of the several millisecond rise time required in our
HF transmitter.

So now we are radiating not just 89% of our initial signal,  but
actually,  now 89% plus 35% more,  and this just 45 nanoseconds
after key down.  So now our integrated output signal voltage
amplitude is even greater than we thought it would be because
of the hearing loss in the coax transmission line!!!

Hey,  maybe Walt Maxwell's "reflection gain" really does happen,
when we add in the benefit of addative reflection energy,  that is
power plus some time to add more power!!

And so it continues,  while the key is down,  we are not loosing
"much",  that is a net tiny fraction of a dB to multiple reflection
heating loss in our transmission line.

Steve goes on:

>Remember that with the tuner, the
>     reflections from the antenna must travel the transmission line twice
>     before arriving back at the antenna.

Yes,  and this to have turned into a benefit,  doesn't it,  when we
look at all the actual details of what is going on;  and  considering
the real time factors.

More from Steve:

>     Which "works" better - matched antenna or mismatched antenna/tuner?
>     The answer to this question requires a much more involved analysis.
>     In order to "work", the transmit system only needs to provide a
>     useable signal level at the receive site.  In many cases, both will
>     "work" quite well.

I have tried,  in plain English,  to give a "more involved analysis"
I hope it is understood.  You decide which system you believe
works better in any useful way in the HF amateur frequency bands.

73,  Jim, KH7M
On the Garden Island of Kauai



--
FAQ on WWW:               http://www.contesting.com/towertalkfaq.html
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm


<Prev in Thread] Current Thread [Next in Thread>