Re: NEC-LIST: Can Is as Can Do

All:

I am afraid that a typo (cap vs can) has led to looking under the
wrong cap (or can). The reference that Jack Belrose was referring to
in response to a msg I put on a while back is:

H.A. Wheeler, "The Radiansphere Around a Small Antenna," Proc IRE,
Vol. 47, pp. 1325-1331, 1959.

Harold wanted to separate out the measurement of the loss resistance
from the measurement of the total resistance Ra = Rr + Rl
     where Ra = Antenna input resistance at the terminals
                Rr = Radiation resistance
                Rl = Loss resistance

All we can measure with a bridge at the antenna terminals (on the
antenna side of any matching circuit, which has its own losses, and
which is required to get any power into the small antenna from the
transmitting source) is Ra, the real part of the antenna input
impedance. He postulated that the losses were all associated with the
near fields, which Chu had shown in 1948 to be bounded by a sphere of
radius (lambda)/2*pi. This sphere, the "radian sphere" was the subject
of Harold's paper. If you can put a conducting "can" of spherical
shape and radius of (lambda)/2*pi over the antenna, then the losses
you measure are the antenna's loss resistance (and any losses in the
can due to it not being a perfect conductor). By measuring Ra with and
without the can present, you can sort out the two components of Ra, Rr
and Rl.

Note that this was before we had NEC and other more modern
method-of-moment tools. We can now, with a NEC model that is based on
a good representation of the antenna and its surroundings compute Rr
for the small antenna structure. With a measured Ra, we can subtract
and get Rl with some accuracy. The bridge accuracy may be of the same
order as the NEC model. Harold would not have needed the "can method"
had he had available our modern computational tools.

If I get time, I will model the "can method" with NEC-4 with a
perfectly conducting can and a real can and see what I get. It would
be a nice vignette on Harold's contribution in 1959.

But one of his main points about electrically small antennas was his
observation that one cannot treat them in isolation, without
considering the matching circuit losses and any residual mismatch loss
due to an imperfect match, when considering the practicality of such
antennas in transmitting applications, like the CFA. That is why I
proposed having the manufacturer or his representative set up a CFA
and certify that it is set up properly, and then have the owner of the
transmitter attach it (however he likes) and apply power and certify
that the transmitter is operating properly. This will generate a
vertically-polarized E field out a radial from the CFA that can be
measured vs distance with a calibrated field strength meter (hopefully
in accordance with IEEE Std. 291-1991). These data can then be used to
compute the effective radiated power (ERP) of the source (transmitter,
matching circuit, and antenna combined). Some upper bound on the
transmitter power can be obtained from a measurement of the raw input
power to the transmitter, and the difference (in dB) between the ERP
(in dBW, as deduced from the field strength measurements) and the
input power (in dBW, as measured at the site) can give us some idea on
the "gain" and efficiency of the CFA.

Hope this helps clarify about the mysterious "Wheeler Cap."

George
Received on Wed Apr 21 1999 - 15:54:58 EDT