On 3/28/17 7:18 PM, Richard (Rick) Karlquist wrote:
On 3/28/2017 2:54 PM, jimlux wrote:
because it's not "really" an end effect or capacitance. The actual
phenomenon is that X goes through zero at a frequency where the dipole
is not exactly a half wavelength long. There's a lot of ways you can
analyze around it or conceptualize it, but they're just mental models.
Another way to think about it is that the propagation velocity down the
wire is not c, but slightly slower (because the wire has inductance and
free space capacitance, and you can calculate the prop velocity as
sqrt(L/C))
The trouble with this description is that it doesn't explain why the
third harmonic resonance of a dipole is greater that 3X the fundamental
(half wave) resonance. OTOH, the "end capacitance" concept neatly
explains it. Three times as much wire, but no additional end effects.
True enough. that's why none of the "simple" models adequately model
higher order resonances (the link to the presentation covers this
extensively). Even the antenna text books get a bit hand-wavey on the
high order resonances - they tend to focus on the first resonance, which
at least has a fairly straightforward analytical development.
And the same is true of mechanical models too, at least for "real"
systems.. that infinitely thin vibrating string or rod with no internal
friction doesn't exist.
I'm just glad that finite element methods are easy and fast to run these
days, so we don't have to try and do antenna design with analytical
equations, particular with respect to coupling between elements. With FE
techniques, you are ultimately worrying about numerical precision and
things like that, and that's something that for the most part you can
just throw computational resources at it (e.g. run in double or triple
precision, it's still fast)
OTOH, if I had to design my antennas with a slide rule and a CRC tables
book, those analytical methods would be handy.
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|