# The Army Loop (Patterson match)

The ARRL and other publications refer to the Army Loop or Patterson match.

Patterson described his antenna system at (Patterson 1967). Hams seem to call any configuration that uses only capacitors in the matching circuit a Patterson or Army loop, though they are incorrect.

The ARRL Antenna Book 21 has a nonsense circuit that cannot work.

## Another ARRL example, one that does work Above is a diagram from a much earlier ARRL and as far as I can ascertain, this is McCoy’s version the so-called ARMY Loop. (McCoy 1968) gives the middle capacitor as 500pF variable which would reduce the matching range.

Lets analyse it at 7MHz. at this frequency it is almost a tenth of a wavelength in diameter so it barely qualifies as a Small Transmitting Loop, or put another way, it a largish Small Transmitting Loop and should be a quite efficient antenna.

Lets assume that an octagonal loop of 1.5″ copper and with 5′ sides has a radiation resistance of 1.16Ω, an inductance of 11.22µH and the intrinsic Q of the inductor is 7000 (its loss resistance is about 0.7Ω) .

(Note Patterson’s loop was aluminium, and so one expects that loss resistance of the tubing alone would be about 30% higher, or about 0.9Ω.)

Further, lets assume good capacitors with Q of 2000. You can do better with vacuum caps, and you could do a lot worse with poor air dielectric caps. Above is a model of the matching circuit in RFSIM99. Above is a plot of S11 and S21.

Firstly, lets look at the half power bandwidth. Half power bandwidth is the range between points where S11 is -7dB, in this case approximately 20kHz (RFSIM99 cannot display freq below 1kHz increments), and antenna Q is therefore 7000/20=350.

Secondly, lets look at S21 for an indication of efficiency. Due to the radiation resistance being 1.16Ω, S21 for a lossless system would be 10*log(1.16/50)=-16.4dB. In this case, S21 at match is -17.4dB, so output power is 1.0dB less than a lossless match, efficiency is 79% which is pretty good for a Small Transmitting Loop, even a big one. Above is an estimate of the efficiency of such a  loop with 20kHz bandwidth using Calculate small transmitting loop gain from bandwidth measurement. The discrepancy will be due mainly to the assumed loop inductance.

That all looks pretty good and you might conclude that the Army loop is a good practical antenna.

## A smaller loop

Let’s now look at a 1m diameter circular loop of 20mm diameter copper at 7.1MHz.

Lets assume that a circular loop of 20mm copper perimeter 3.14m has a radiation resistance of 5.7mΩ, an inductance of 2.5µH and the intrinsic Q of the inductor is 3000 (its loss resistance is about 35mΩ). Above is a model of the matching circuit in RFSIM99. Note that one of the capacitors is 10,000pF, and it needs to be an adjustable high current low loss capacitor… quite a challenge. Above is a plot of S11 and S21.

Firstly, lets look at the half power bandwidth. Half power bandwidth is the range between points where S11 is -7dB, in this case approximately 8kHz (RFSIM99 cannot display freq below 1kHz increments), and antenna Q is therefore 7100/8=888.

Secondly, lets look at S21 for an indication of efficiency. Due to the radiation resistance being 5.7mΩ, S21 for a lossless system would be 10*log(0.0057/50)=-39.4dB. In this case, S21 at match is -51.8dB, so output power is 12.4dB less than a lossless match, efficiency is 5.7% which is pretty poor. Above is an estimate of the efficiency of such a  loop with 20kHz bandwidth using Calculate small transmitting loop gain from bandwidth measurement. The discrepancy will be due mainly to the assumed loop inductance.

That all looks pretty poor, low efficiency and impractical components required, and you might conclude that the Army loop is not a good practical antenna.

It all depends!

## Patterson’s tuner

Patterson’s tuner was for an octagonal loop of 40′ (12.2m) perimeter from 2.5MHz up. Above is Patterson’s tuner. Note that the middle capacitor can have value to 9000pF, much more than McCoy uses in his comparison system at (McCoy 1968).