"Dave Platt" wrote in message
...
In article P2G1e.110904$Ze3.11791@attbi_s51,
wrote:
By all means put me on your kill file. I asked a simple question and you
want to reply to a different question of your liking and place your
question
as one preferable to mine.
Art, in cases like this, you keep asking "simple questions" which
imply, by their very wording, a whole bunch of assumptions about how
things work which just ain't so.
The fact that you keep getting answer after answer, from a lot of
knowledgeable people, which you either don't understand or "blow past"
or that you feel evade the point of your question, ought to be saying
something to you: that there's something wrong with the questions you
ask.
No one has been able to supply the answer to my question,
Using your words ,what steers the pattern away from a circular
form from a natural circular form.
That depends on what you mean by "a natural circular form."
If you're referring to the fact that the main lobe of a dipole tends
to look circular on many of the commonly-used plots, then the pattern
isn't "circular" in any cosmic sense of the word. It's just as
correct to say that it's elliptical, or bumpy, or squashed, because
that's exactly how it will look on plots which use different circular
axes (linear, logarithmic, etc).
To try it again, though:
I thank you for that
you're asking why the pattern appears to be
compressed, as the gain increases. Fundamentally, it's due to the
fact that the antenna is sending more power out in the desired
directions (more gain), at the expense of sending less in other
directions.
Fully agreed to
This is done by creating multiple radiators, which are
offset in power and location and phase so that their individual
radiation wavefronts reinforce in the desired directions, and cancel
in the undesired directions.
Accepted as long as you can agree that a similar vector analysis with
multiple radiators can also create a non focussing pattern
When we plot the resulting RF strengths, the RF in the desired
direction is stronger (we got the gain that we want). Let's assume
that (as is common practice) we continue to plot the signal in the
strongest direction on the outer circle of the graph.
Now, one of two things will have to be true:
[1] Every other direction in the main lobe had its power "scaled up"
by the same amount... the increase in gain worked the same for all
directions within the main lobe. In this case, the shape of the
main lobe will not change at all.
O.K. this would /could be the case I am thinking of
In this case, the additional power required to achieve the
increase in gain in the main lobe will have had to come for
somewhere.
Agreed
Since it didn't come from the main lobe, it will have
had to come either from the sidelobes, or from the rear half of
the antenna's pattern.
Agreed for over all gain but not necessarilly for the lobe becomming
focussed
which is the crux of my question
There's a limit to how far you can take approach [1]. It stops
working when your sidelobes and rear half of the pattern drop to
zero... and it becomes rather ineffective some time before that,
when the largest of the side/rear lobes is maybe 10-15 dB down.
Beyond that point, there just isn't enough power left in those
backlobes to be useful.
Using a antenna computor program the main lobe at 10 degrees does not
deviate from a circle even if the F/R is more than 30 db ( note F/R vs F/B)
and this is comprised of vector addition mode as with a yagi design.
[2] The other possibility is that you didn't manage to boost the gain,
uniformly, in the entire main lobe.
Hum!
In this case, if you're still plotting the strongest signal on the
outermost circle of the graph, you'll notice that the shape of the
main lobe has changed.
No, not always, only with a yagi design in my opinion
Any direction in which the gain increase
was less than the maximum you achieved, will be closer to the
center of the circle than before. [Another way of looking at this
is that by increasing your maximum directional gain, you've
"enlarged the circle" on which you're plotting it, but that some
points didn't move outwards by the same ratio.]
Agreed thus my question
In the common case of a Yagi, when you boost the gain (say, from 10 dB
to 15 dB) there just isn't enough power available in the side and rear
lobes to make up this gain... you can't 'rob' enough directivity from
the sidelobes and rear lobe. Instead, you 'rob' the power from the
outer edges of the main lobe, and shift it in towards the center.
Now we are getting closer to my quest. How do we "rob" from the outer
edge of the main lobe is the underpinnings of my question.
You do this, most commonly, by adding additional parasitic elements,
whose location and phasing are such that their radiation reinforces
that moving in the "forward" direction, and interacts destructively
with (cancels) radiation moving outwards at an angle.
Agreed if we are adding or subtracting on a constant plane.
Could you by any chance referring to the closing vector of the vector
analysis
to consist of two vectors and where one of these vectors is the force at
right
angles to the main lobe and which deforms it. If so I am beginning to see
the light!
When you plot the resulting pattern, and scale it so that the
strongest signal is on the outer circle of the plot, you find that the
main lobe looks narrower. Part of this is due to the actual
redirection of power, and part of it is due to the fact that you've
re-scaled the graph.a
I am lost here but if we agree on my interpretation of what you said then
I am at a point where I can generate vector diagrams of different arrays
and forecast the width of the resultant lobe .Does anybody else agree
that the main lobe width can be forecast via vector analysis.
Seems like from past posts that vector analysis is not now in vogue
for electrical engineers in the U,S and only creates blank stares
when mentioned/
..
kes several wrathful deities...]
Art, if you continue to ask "simple questions", and you continue to
get back complicated and detailed answers, it really ought to convey
to you the possibility that your "simple" question is oversimplified.
Or, perhaps, that you've been given the actual (simple) answer three
or four or five times already, have rejected it, and people are trying
to explain to you why it's actually correct.
Goodbye, Art. This is/was my last attempt, I think. I doubt I'll try
again.
But David, nobody pointed to vector analysis and the particular facet
that you referred to. You are to be congratulated in pointing to a trail
of logic that could well be the direction I was looking for.
Nobody but you presented in real terms an analysis that leads to
serious consideration and I thank you very much for that.
Best regards
Art KB9MZ.....xg
g the Jade Warrior home page:
http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!