Roy Lewallen wrote:
John Popelish wrote:

Roy Lewallen wrote:
. . .

Effective height determines how many volts you'll get from an open
circuited antenna.


Does that include an antenna that has been brought to resonance with
an appropriate capacitive load?


No. "Open circuited" means that there's nothing connected across the
feedpoint.

But I can design the coil to be self resonant or not, just by
adjusting the surface area of the wire, or the spacing. It is non
intuitive that if I peak the coil this way, and obtain more voltage,
it is a different case than if I peak the coil with cable capacitance,
or an additional capacitor. I guess I really don't comprehend the
point of this value.

(Snip excellent review of basic lossless radiator. Thank you.)

But more directly to the point, your tiny theoretical rod antenna would
have a gain of about 0.45 dB less than a half wave dipole, and its
capture area would be correspondingly smaller -- about 10%. This is
assuming you're looking in the best direction for each antenna. Because
the total radiated power or integral of the capture areas must be the
same for the two antennas, this means that the tiny antenna has to have
more gain or capture area than the dipole in some other directions. And
indeed it does -- the tiny antenna has slightly fatter lobes than the
half wave dipole.

I understand what you are saying.

(snip)
The effective height of a ferrite rod antenna is approximately:

(2 * pi * mueff * N * A) / lambda

where

mueff = effective relative permeability of the rod (mainly a
function of rod length)
N = number of turns
A = rod cross sectional area
lambda = wavelength


I can apply this formula directory to what I am experimenting with,
except that I have to approximate mueff. I am making the rod by
stacking ferrite beads, with various gaps between them. Can I
approximate mueff by taking the ratio of coil inductance with and
without the rod?


Yes. That's exactly what it is.

Well, now I can calculate the effective height of my antennas, even
though I am not sure what it has to do with height.


And, what if the rod area is not constant all along the rod? Since my
rods are assembled from pieces, I have a lot of freedom in this
direction.


That one I don't know the answer to.

I am also experimenting with designs that do not necessarily have a
small, coil, close to the rod. (My interest in the discussion of
extended coils is showing.) One of the possibilities that shows a
significant increase in tuned Q is an hour glass shaped coil (small
diameter in the center, but sweeping to a larger diameter at the
ends). I have been asked to try putting a rod through the center of a
flat spiral coil. It seems to me that, at some extreme, the above
formula will fail, because it assumes that essentially all the signal
energy exiting the coil was collected by the rod, and that the signal
the coil would collect by itself would be insignificant. But if my
coils get large enough, they become loop antennas in their own right,
and the rod, though it may have a significant length and area, is only
a part of what is happening. In other words, the mueff can get pretty
small, even though the rod has significant dimensions. I guess, what
I am asking are what assumptions about coil dimensions (that are not
explicitly referenced in the formula) are being made in the above formula?