[TowerTalk] Reflections, Tuners and Antenna Forum Talks
Michael Tope
Michael Tope <w4ef@pacbell.net>
Thu, 11 May 2000 22:55:37 -0700
Jim
I have to agree with Steve's description as outlined below.
The tuner can't know ahead of time what is at the load
end of the transmission line. Otherwise, you could use an
RF switch and a long piece of coax to send information
faster than the speed of light. Until the incident wave reaches
the load termination where the standing wave forms and
then has time to travel back to the source, the transmission
line's input impedance will look like 50 ohms.Thus, when
calculating the amplitude of the incident wave, one must treat
the load impedance attached to tuner as a pure 50 ohms.
The impedance loading the output of the tuner will remain
50 ohms until such time as the leading edge of the reflected
wave reaches the tuner. At this time, the resultant impedance
loading the tuner will become the vector sum of the incident
and reflected voltages divided by the vector sum of the
incident and reflected currents.
There will indeed be a step discontinuity in the envelope of the
RF current and voltage waveforms at the output of the tuner. This
is why for example, VSWR is so important in analog TV
transmission - in these systems the round trip transit time from the
transmitter to the antenna can be on the same order as the rate
of information being transmitted (e.g. the reflections can cause
ghosting). In these systems the transient response can't be
neglected since the information (modulation) is changing faster than
time its takes for the transient response to settle to its steady state
value.
The most interesting thing about all this is that there needs to be
a consistency between the time domain view of transmission
line behavior and the steady state behavior for the steady state
theory to be valid. In other words if you think thru how the incident
wave and the subsequent reflections add up as a function of time,
they have to converge to values that are consistent with the steady
state solutions. There is an interesting duality at work here between
two domains. John Kraus has an illustration in his book
"Electromagnetics" 2nd ed. which shows a series of snapshots in time
of a continuous wave traversing a 1/4 transmission line transformer.
The drawing shows how the multiple reflections of the incident energy
add up to form the steady state solution for the 1/4 wave line (matched
condition). The following page shows the time progression for a short
pulse along the same 1/4 wave line. Kraus states "It is clear that the
transformer fails to function as a reflectionless device for a pulse as
short as indicated.........".
Mike, W4EF...........................
----- Original Message -----
From: Jim Reid <kh7m@hsa-kauai.net>
To: Steven Best <sbest@cushcraft.com>; <towertalk@contesting.com>
Sent: Thursday, May 11, 2000 1:53 PM
Subject: Re: [TowerTalk] Reflections, Tuners and Antenna Forum Talks
>
> Steve's post today contains within:
>
> > Now back to Jim's tuner experiment. I ran an analysis of
> > the set up Jim described, which is the basis for the numbers
> > presented here......... I could not exactly match the tuner
> > component configuration described by Jim and get the
> > correct input impedance.
>
> Recall that the actual Palstar tuner component "settings" are,
> and remain:
>
> Input C at about 65% maximum mesh, or presumably
> about 0.65 of the 350 pF total available C, or 228 pF.
>
> Inductor tap at 1.6 uHy of the total of 28 uHy inductance
> available.
>
> Output C, at 100% full mesh, or 350 pF in series with the
> transmission line.
>
> >This is not really relevant since I arbitrarily set the component
> > values such that the tuner's steady state input impedance
> > was 50 + j8 ohms...........The component values I selected
> > were series C at input = 115.71 pF; parallel L = 1.8 uH; and
> > series C at output = 164.22 pF. With these tuner components,
> > I calculated the steady state tuner input impedance to be
> > about 50 + j8 ohms.
>
> Steve, I just don't understand why using the acutal Palstar
> component settings would not be relevant. All your assumed
> settings, except for the inductor, appear to be quite a bit
> different from where the "real" capacitors are sitting; I assume
> the capacitors do have the 350 pF full spec. rating which the
> mfgr. of the Palstar tuner specifies. Maybe I will go back
> and redo the input Z measure; I could have messed it
> up!!
>
> Perhaps from the numbers I took today about my Bird
> slugs might aid in determination of the "real" forward/
> reflected power values? As I did these slug numbers,
> it occured to me that the 5000 watt slug was reading
> closer to the slug used to "set" each power level
> into the load; too bad I didn't use that one also in
> collecting some of the data, hi.
>
> Steve's post continues:
>
> > 1) The first point is to correctly determine the forward
> > driving voltage delivered to the transmission line. Remember
> > that when the "system" is first energized no reflections are
> > present at the tuner output. Therefore, the "load" impedance
> > at the tuner output is Zo. Therefore, the initial "input
> > impedance" at the tuner input is NOT 50 + j8 ohms, it is
> > calculated using circuit theory to be 58.1 - j51.7 ohms.
>
> Hmmmm, well, Steve, I know how fervently you hold this
> to be true. I am still having a difficult time with the idea
> of a "transient" changes in impedance because a reflection
> has not yet occured. You claim the tuner output sees only
> the transmission line Zo, and so therefore the input impedance
> of the tuner is greatly different from what it will be some a
> microsecond or so later (whatever the round trip propagation time
> may be in the feedline up to the antenna input terminal and
> back). Still seems to me it is better to use true wave behavior
> than this "introduced" transient change of nature, hi!
>
> Oh well, I must read your entire post very carefully, Steve;
> you are correct: it is long, and different from what many of
> us are accustomed to following.
>
> Anyway, still look forward to the ongoing discussion, and
> of course, to Dayton and next Friday!
>
> 73, Jim, KH7M
>
>
>
>
> --
> 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
>
--
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