[Elecraft] balanced tuner

[Ron D'Eau Claire]

How about a normal L-type network? Connect one side to the balanced line,
and the other side via a current balun to the 50 coax. The tuner is
'RF-floating'. (You can run the coax through a toroid or use two seperate
wires to make the current balun.) 

73,
Arie PA3a

----------------------------------------------------------

I've done a number of experiments with this setup, measuring the amplitude
and phase of currents into the transmission line for a center fed doublet
antenna using balanced and unbalanced networks. The unbalanced networks were
tested to compare the effect of a balun at the input or at the output. 

They all worked FB. I came to these conclusions:

1) The tuner components must be well isolated from surrounding objects if
they are part of the balanced feed line (balun at the input of an
unsymmetrical network or an inherently balanced network) to avoid excessive
stray coupling. Inherently balanced networks can be arranged to provide
equal coupling to a box (e.g. the famous Johnson "Matchboxes"). An
unsymmetrical network needs more care (isolation is good!). 

2) There was no practical difference in the balance available from either a
"single ended" (unbalanced as in an L-network) tuner or an inherently
balanced tuner (link-coupled circuit with split stator tuning capacitor).
After all, when the system is adjusted to provide a 50-ohm non-reactive
impedance to the rig, the various coils and capacitors are "transparent" to
the RF. The RF "sees" a resistive 50-ohm load at the input to the tuner.
There is a possibility of imbalance caused by different losses in the
components. For example, coils tend to greater losses than air-dielectric
capacitors at RF, so a large coil can introduce ohmic losses that affect
balance in an unsymmetrical network. 

3) When the balun is placed at the *input* to the tuner, it must be properly
isolated. Putting it inside a metal box with proper conductors is important
in most situations (i.e. inside the tuner enclosure if a metal enclosure is
used). The point is to keep the RF flowing along the inside of the coax
shield from finding its way to the outside of the shield and back to outside
of the equipment enclosures. Remember, to RF, a shield is two isolated
conductors: one in each side. The RF needs to stay *inside* the enclosure of
the balun. Simply connecting the coax to a balun laying in the open is sure
to cause problems. This issue becomes more important at higher frequencies. 

4) A high degree of balance in a wire antenna is almost impossible to obtain
at HF. Differences in the proximity of buildings, trees and the ground to
each side of a center-fed wire effect the balance in the currents in the
feeders. Of course, unbalance can skew the pattern of a beam antenna, but
they are usually better isolated (and few beams use open wire lines these
days anyway). Pattern skew of a typical wire antenna caused by trees,
buildings or other sources of unbalance in the matching network is seldom of
concern. That's why so many installations work FB with no balun at all. It's
only needed if common mode currents cause "RF in the shack" issues. Remember
that the RF output at the coaxial connector on the rig is already inherently
balanced: the current flowing on the inside of the shield is equal and
amplitude and opposite in phase to that flowing along the center conductor.
The challenge is to preserve that balance where the coax shield comes to an
end by not allowing the RF to flow onto the outside of the shield, adding
the equipment on the operating desk, the station ground and possibly the
power lines to one side of the antenna system! 

Ron AC7AC