AI4QJ wrote:
In an inductor, current lags voltage. If you connect a resitor and a
coil in parallel and apply AC, EE101 tells you that, although the phase
of the voltage across them stays the same, the current is "delayed" by
the phase angle in the inductor when compared to current in the
resistor.
No, it isn't - the phase of the current around the circuit has to stay
the same. Think of the simplest possible circuit: an AC voltage source
(of zero internal impedance) with one terminal wired to R, lumped L in
series, and directly back to the other terminal of the AC source. If the
phase of the current were delayed through L as you suggest, there would
then be a difference in phase between the two terminals of the AC
source... which is obviously not true.
It's the magnitude and phase of the voltage that varies at different
points around the circuit; but the magnitude and phase of the current
has to remain the same all the way around the loop. In more formal
terms, Kirchhoff's current law applies all around the circuit; and it
most certainly applies between the two terminals of a lumped inductance.
If this weren't so, I assure you the world we live in just wouldn't be
the same
Well, you're certainly right about that :-)
What isn't sufficiently understood is that loading by pure lumped
inductance is responsible for almost all the properties of a
physically short loaded vertical. This is easy to see for base loading,
where lumped inductance at the feedpoint is simply compensating for the
capacitive reactance of the physically short whip.
The same remains true when the loading inductance is moved further up
the whip (progressively increasing the inductance, of course, in order
to maintain a zero reactance at the feedpoint). Loading by pure lumped
inductance explains all the well-known observations: the need for more
inductance if the coil is located further up the whip; the relatively
constant magnitude and phase of the current in the part of the whip
below the loading inductance; the large step increase in voltage across
the inductance itself; and the very rapid decrease in current in the
section above the loading inductance, tapering to zero current at the
top and accompanied by a very large electric field.
Note again: all the above is explained using the classic behaviour of
lumped inductance, with ZERO difference in current magnitude and phase
between its two terminals.
Now what changes when we replace a theoretical lumped inductance with a
physically realizable coil? The main difference is that the real-life
coil occupies a significant fraction of the total physical height of the
antenna. In addition to providing inductance, the coil is now behaving
to some extent like a helically-wound section of the complete antenna.
Now it is completely reasonable to expect some 'antenna-like' behaviour
from the coil: there will be some radiation from the coil itself,
accompanied by some decrease in current with distance along the coil. We
also expect some phase shift in current between its two ends.
This 'antenna-like' behaviour in a physically long loading coil is
completely reasonable and to be expected. However, it will be difficult
to explain in detail because the 'antenna-like' properties of the coil
will be strongly influenced by the other parts of the antenna. (This
also means that measurements made on a coil in isolation will have very
limited relevance to the behaviour of the coil as part of a complete
antenna.)
There may be several different valid explanations, each looking at the
problem from a different viewpoint - no problem about that, of course.
But there will also be several INvalid explanations... so how do we tell
the difference?
All the valid explanations will have at least one thing in common. They
will ALL be able to handle the boundary condition that, when the
physical size of the coil tends towards zero, every part of its
behaviour tends towards the classic behaviour of pure lumped inductance.
In particular for this part of the discussion, the phase shift in
current between its two terminals MUST tend to zero. Now that the
loading has become pure inductance, any valid explanation must still be
able to explain all the major features of the loaded whip, as identified
above.
In other words, any explanation MUST be able to handle that boundary
condition smoothly and effortlessly, without any need for a sudden
change in its own rules and assumptions. If it cannot do that, then
logic tells us that explanation cannot be correct. For example, if at
the boundary it is still demanding a phase shift in current between the
two terminals of a lumped inductance, that cannot be correct. If it
demands that lumped inductance behaves differently in antennas than it
does in any other circuit, then again it cannot be correct.
It's not me making these rules, by the way - just recognising that they
exist, and insisting that they apply.
Boundary conditions that join up with existing knowledge are a very
useful 'logical razor' to slice and dice ideas that fail to pass those
tests. In classical logic this method is called 'reductio ad absurdum',
but that's only the Latin for something that everybody knows:
"If it don't make sense, it kain't be right."
Unfortunately, some people seem to be immune to this, or want to
negotiate waivers for their own particular ideas. Progress comes from
recognising that there are no exceptions, and being prepared to let go
of cherished ideas if it is shown that they don't work.
--
73 from Ian GM3SEK 'In Practice' columnist for RadCom (RSGB)
http://www.ifwtech.co.uk/g3sek