[ Reply | Next | Previous | Up ]
From: Ken
Date: 02 Jul 1999
Time: 23:32:35
Remote Name: 207.227.238.116
Apologies for the long response. There IS a lot of confusion about the technical aspects of antennas in general. I figured I should rant on a bit...
Remy wrote:
>Something seem unclear in my mind ( at least). Many articles talk >about obtaining the same power with a large E field and a small H >field or a large H field and a small E field. Of course the product E x >H may be the same but in space, the H field is related to the E field >by the equation H = E / n where n is the intrinsic impedance of free >space and is equal to 377 ohms. This should mean that this >proportion of E and H should be respected to obtain a kind of >optimum EM waves or where am I wrong ?
For direct coupling of E and H _with_no_reactive_field_, yes, E and H must match the impedance of free space, approximately 377 ohms. Note that this is *not* the input impedance of conventional antennas.
Conventional antennas pump energy into, predominantly, H fields. This energy does not propagate; it is confined and decays at, for a small current element, 1/r^2. It is this time-varying H field that creates E fields, which in turn reciprocally regenerate H fields. This is the source of propagation in conventional antennas. It relies on a volume of space to contain a 'buffer' of time-varying reactive field with the built-in constraint of a definable minimum volume for effective conversion of reactive energy to propagating energy. E and H only become balanced *past* the reactive near field; the antenna impedance is generally different than free-space wave impedance.
CFAs apparently operate by bypassing by design the requirement for near field reactive energy, hence the requirement that the matching circuit matches the energy of each excitation to close to 377 ohms. Note that I say *close* to 377 ohms; there is likely some residual reactive field present that affects the impedance one is matching to...
>It looks like putting to much of one field or the other just create >reactive energy.
...which in turn, as I mentioned, generates fields of the other sort, and so on.
>My other interrogation is on the claim that CFA or SuperC antennas >work with high impedance so that the coils that feeds the top hat >cylinder have low loss. The capacity of this cylinder with the close >ground plate is a lot bigger than that of a quater wave antenna. >These coils have to supply a great current in order to create a high >voltage across the cylinder or ring and the ground plane. In this >way, the losses in either the phasing coil or the loading coil is not >small at all ans efficiency is certainly affected.
Small loaded antennas, which is what the Super C is, are nothing new. They are quite conventional. The H field dominance is maintained by the presence of the large capacitive loading and big currents. Now, there is nothing wrong with this design. But because of its conventional design, it is doomed to obey *in particular* (for those of you who read these things) Chu's dictum on the limits of small antennas with regard to bandwidth and efficiency. Small antennas, it is said (and until now unrefuted) can gain either efficiency or bandwidth only at the expense of the other.
Translation: small antennas can be quite efficient if properly matched, but with cruddy bandwidth; likewise, bandwidth can be large, but efficiency suffers.
Here's where the loophole may lie: This theory (regarded as law) depends on the presumption that a reactive field is required to enable a purely resistive input impedance for some frequency -- the stored electric field of a dipole balances the excess H field generated by the current distribution on the antenna.
The ideal CFA circumvents this by design, courtesy of the super- imposition of both TM and TE spherical modes simultaneously -- I think. (OK, technobabble for most out there, but translates to TWO complementary antennas working perfectly in sync.)
The upshot is that given the right loophole, very high efficiency *and* bandwidth may be possible for very small antennas. But the Super C will not do this. Instantaneous bandwidth is not advertised, and I suspect not good. Nothing new. But the CFA, if the theory holds and still works given the framework that *does* limit conventional antennas, might just turn everything upside down.
Side note: Data from NAB '99 paper suggests that it's real, and that engineers (like myself) have a lot of work to do. If Chu's BW/efficency limit is violated, then _DECADE_ instantaneous bandwidths might be possible with good efficiency in _arbitrarily_ small volumes. If so, Physics and Electromagnetics as we know it will be shaken. Hard.
Hope I didn't add to the confusion.