Posted by
David Woolley (E.L) on
Jan 06, 2008; 12:48pm
URL: http://elecraft.85.s1.nabble.com/dipole-antenna-efficiency-tp458071p458098.html
John Magliacane wrote:
> When it comes to antenna efficiency, it is important to understand
> that when RF energy is applied to any antenna, three things will
> invariably happen:
For antennas in practical locations, at least one other thing will happen:
d) The antenna will convert part of the AC power into (near field)
electromagnetic energy which will induce currents in the ground,
building structure, wiring, water and gas pipes, etc. Much of that
energy will be converted to heat after it has lost, although some will
be re-radiated (I believe, in extreme cases, if the current is induced
in something large enough and resonant, the re-radiator can become the
actual antenna and the antenna act as a feed device, but, normally, for
low, and indoor antennas, this is where most of the energy turns to heat).
Normally, even without losses, induced currents in the ground below a,
low, horizontal dipole will tend to cancel the far field signal,
resulting in less power actually being radiated, and a lower radiation
resistance, requiring even lower losses in all components from antenna
wire back to and including the ATU.
(One way of considering the far field is that you need to create a
relatively large electric or magnetic field far enough from the antenna
that its phase lags that close to the antenna by a significant amount.)
> c) The antenna will reflect a portion of the applied AC voltage and
> current back to the transmitter as a result of an impedance mismatch
> between the antenna and that of the source.
This reflection abstraction causes a lot of confusion. It is possibly
easier to see it as simply a bad match between the transmitter source
impedance (which is usually rather different from the optimum load
impedance) and the antenna impedance, causing most of the DC input to
the transmitter to end up as heat in the output devices.
>
> Unless steel or nichrome wire is used, or electrically poor
> connections exist in the antenna structure, losses due to (b)
> will be low.
As already pointed out, skin effect means that this is not true. People
experimenting with small magnetic loops have to use large copper pipes
to keep ohmic losses manageable. (In some cases I suspect they are
still high compared to radiation resistance, but lower than the near
field loss resistance.)
> Effects of (c) can be reduced or eliminated by using intelligent,
> low-loss impedance matching techniques and low-loss feedline.
> (Technically, reflected power isn't a "loss" per se, since
> energy isn't dissipated when a reflection occurs.)
Note that devices capable of doing this for the sort of antenna being
considered in this thread are not easy to find, if they can be found at
all. For example, the KAT2 has a 10:1 SWR matching specification, but
matching the antenna discussed here, at infinite height above the
ground, needs a 250:1 range, or more. They can also have power losses.
> That leaves us with (a), the desired outcome of applying RF energy to
> an antenna. Since losses due to (b) and (c) are typically low and/or
> easily corrected, it is very difficult NOT to achieve high antenna
> system efficiency.
(b) and (c) are not typically low for the sort of antenna considered
here, although (c) isn't really achievable, anyway.
>
> Shortening the physical length of an antenna below that of a
> half-wavelength DOES NOT reduce its efficiency provided the
> necessary efforts to minimize resistive losses in the antenna
> structure and the impedance matching networks are made.
But, apart from possibly cooling everything to near absolute zero, they
cannot be made.
>
> That last statement is so important and so often misunderstood,
> it bears repeating:
I'd agree that there is a misunderstanding that is common. It arises
because people have difficulty with the idea that an antenna can have a
capture area that is a lot bigger than the antenna, and because people
don't understand that the real limitation on small antennas is power
losses. Large antennas have gains that equate to directivity, and
people try to extrapolate these down to small antennas, whereas there
are no Maxwellian reasons why a small antenna cannot be efficient, only
materials science, engineering and environmental ones.
>
> If we were to apply 100 watts to such an antenna, and we get zero
> watts reflected back, and the antenna and matching networks remain
> cool, then 100 watts of RF energy is being radiated from that one
> foot dipole -- the same as if a full-sized dipole were used.
Radiation is normally used to refer to far field radiation, which is the
only radiation useful for normal ham radio communications. The antenna
can remain cool even if all the power is going into heating up the
ground, or the re-inforcing bars in your concrete building.
>
> The penalties for using physically shortened antennas are:
>
> (a) Decreased operating bandwidth
I'm not sure that is inevitably true. My reference for normal mode
helices included them in the section on broadband antennas.
> (b) Decreased directivity
Directivity ceases to be a factor much below a half wave, and I don't
think that there is much difference between a halfwave dipole and a stub
dipole in free space.
> Dipoles have 2.14 dB "gain" over isotropic radiators. As we make
> our dipole shorter and shorter (and keep resonating and impedance
> matching it in the process), its directivity (b) approaches that
> of an isotropic radiator. If our radiating structure and impedance
Although I cannot find an exact formula at the moment, the gain will not
be asymptotic to 0dBi, but will, rather, not be too different from 0dBd.
Incidentally, I have a feeling that the 2.14dBi applies to an
idealized short dipole, rather than a half wave one.
All in all, then, Maxwell doesn't prohibit high efficiencies from very
small antennas, but engineering practicalities do.
--
David Woolley
Emails are not formal business letters, whatever businesses may want.
RFC1855 says there should be an address here, but, in a world of spam,
that is no longer good advice, as archive address hiding may not work.
_______________________________________________
Elecraft mailing list
Post to:
[hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
http://mailman.qth.net/mailman/listinfo/elecraft
Help:
http://mailman.qth.net/subscribers.htmElecraft web page:
http://www.elecraft.com