On the Evolution of the Ideal Antenna Tuner  

(Originally published in the September/October 2005 issue of "The Canadian Amateur")

A recent article written Jack Belrose (1) and highlighted by Pat Hawker (2) has 
presented a detailed analysis of the reversible L (RL) network in antenna tuning
applications.  This reaffirmed earlier findings of the advantages of the RL 
over other designs which use more than two elements (3).  It is now proposed 
that the utility of an antenna tuner in securing a 1:1 VSWR is a function of 
the number of available configurations of L networks rather than the number of 
tuning elements (4).  Put another way, the popular T network can be improved by
replacing one of its so called input or output capacitors with a switch. This 
modification will provide four configurations with two elements instead of one 
configuration with three elements, and in accordance with our conjecture above, 
a more useful antenna system tuning unit (ASTU): the doubly reversible L (RRL) 
(5), also featured in Jack's article. 

The basic RL (Figure 1) gives us two configurations.  With the switch in 
position 'A'  the network will match impedances with resistive components lower
than that of the source. For example, if your transmitter has an output 
impedance of 50 ohms, this configuration will match a load of say 10 ohms.  In 
position 'B' impedances with higher resistive components can be matched.  It is
noteworthy that this configuration will match reactive loads with 
resistive components less than that of the source.  This has interesting 
consequences. For example if we take a vertical antenna fed against an ideal 
ground plane and make it progressively shorter, its capacitive reactance 
increases and its radiation resistance decreases.  The reactance increases at 
such a rate that the 'B' configuration will will always be able to tune the 
vertical antenna -- regardless of how short it is.  In other words it can match 
a paper clip on 160 metres.  However, Jack VE2CV observed that we should be
prepared to wind a very large inductor, accept high losses and expect little 
radiation if we intend to use a paper clip antenna on top band! I have used 
the 'B' configuration almost exclusively for over 20 years. A remote controlled
unit (Figure 2) was used at the base of a 25 metre tower with excellent 
results.  
 
By deriving the generalised design equations for the reversible "L" (6), I 
could model all L networks and concluded that although the RL could match a 
remarkable range of impedances it had its limitations. However I quickly 
discovered that many of these shortcomings could be overcome by exchanging the 
tuning elements (5).  This yields an additional two configurations that can be 
obtained with another reversing switch, to give the RRL network (figure 3) 
which I proposed in 1987.   

For those interested in experimenting with the RRL, having a large range 
of L and C is desirable. The use of individual uncoupled coils instead of a 
roler or switched inductor is preferred in the interest of avoiding high 
voltages from the 'autotransformer' effect. Switched capacitors can give a 
greater range than a single variable unit.  For a balanced aerial such as 
the G5RV or random dipole with open wire feeders, use a balun on the 
RRL input (1) and enclose the circuit in a large metal box to minimise 
stray capacitance so as to preserve the balance of the output.  A balun that
is DC grounded will guard against precipitation static discharges.

1) Belrose, J.S. "On the Quest for an Ideal Antenna Tuner" QST Vol 68,
Issue 10, October 2004.

2) Hawker P. "Technical Topics" pp34-36 Radcom, December 2004 

3) Craig J.  "The Reversible L Impedance Matching Network" 
VO News, Issue 2, August 1986 (Ev Price, Editor).  Out of print. 
Copies are maintained by the Marconi Radio Club of Newfoundland 

4) For any given range of tuning element reactances. 

5) Craig, J. "Notes on the RL Impedance Matching Network" 
VO News Issue 3 February 1987  (Ev Price, Editor).  Out of print. 
Copies are maintained by the Marconi Radio Club of Newfoundland 

6) Craig, J. VO1NA.  "A Brief Discussion of Image Operated L Networks"
Atlantic Undergraduate Physics Conference 1987, Memorial University of 
Newfoundland, St. John's Newfoundland. 

The author is most grateful to Dr J. Belrose for reviewing this 
manuscript and his insightful comments. Copies of the VO News articles 
may be found on the MRCN web site by searching for VO1MRC on the internet. 


Captions:
Figure (1)  The basic reversible "L" network.  The source is on the left 
and the load is on the right hand side. 

Figure (2) The best place for the ASTU is at the feedpoint.  This  
B configuration "L" network is used to match a 25 metre tower on
30,40,60,80 and 160 metres.  Here the network is being tuned for 
experimental work as part of the RAC/MRCN 5 MHz project. 

Figure (3) By interchanging L and C of the basic RL, we obtain the 
doubly  reversible L.  The two reversing switches can be incorporated
into a single rotary switch as suggested by VE2CV.