The four-course Low Frequency Radio Range system used for aircraft navigation (in the US and elsewhere) in the 1930s and 1940s used a pair of directional antenna (with figure-of-8 shaped radiation patterns) at 90 degrees to each other. One antenna would transmit the Morse code pattern for "A" (⋅-) and the other for "N" (-⋅), as follows:

four-course radio range

When the aircraft was on course, the signals from the "A" transmitter and the "N" transmitter would be at equal value and overlap to form a continuous tone:


If the antenna patterns are at 90° and the transmitted power is the same for the two signals then the system provides four different courses that a plane could fly (or eight, counting reciprocal courses away from the station).

Examining charts for actual installations, however, one often sees that the four courses are not at 90° angles:

Stout field instrument approach

How was this in practice achieved? I can see that by increasing the relative strength of one transmitter compared to the other the courses could be adjusted so they crossed at other than a right angle (like this, for the top diagram, supposing A is stronger than N):

\ N /
 \ /
 / \
/ N \

But how would it be arranged to have asymmetrical patterns such as the one in the Stout Field diagram above, where the opposing courses are not at 180° angles to each other?

  • 1
    \$\begingroup\$ I have a working 4 course radio range at my private airport. You can "crow's foot" a radio range by changing the RF phase of the antennas. Plus you can shift the courses by 90 degrees using a RF Goniometer. I use my radio range for real flying. \$\endgroup\$
    – user117685
    Jul 25, 2016 at 2:30
  • \$\begingroup\$ @Davidfrantz: by changing the phase of one pair with respect to the other pair, or of one antenna within a pair with respect to the other? \$\endgroup\$
    – cpcallen
    Aug 2, 2016 at 22:55

2 Answers 2


Here is the detailed explanation from Edward Laporte's Radio Antenna Engineering article on the Adcock Antenna. Briefly, the relative phase of the voltages feeding each opposite pair of antennas is altered a few degrees from the normal 180 degrees. This causes the opposing lobes of the antenna pattern to become "lopsided" with one lobe shrinking and the opposite one expanding. By controlling the relative strengths of the A and N sides this way, the on-course azimuths can be altered. Additionally, a goniometer is used to rotate each of the two patterns separately. Taken together, you can achieve something like this:

Adcock array pattern illustrating arbitrary relations between courses by combination of current phasings and goniometer position.

  • 3
    \$\begingroup\$ Link only answers are discouraged because links go down, please write a synopsis of the links when providing links \$\endgroup\$
    – Voltage Spike
    Nov 10, 2016 at 18:22
  • \$\begingroup\$ Sorry. WIll do now. \$\endgroup\$
    – Bob Denny
    Nov 18, 2016 at 21:57
  • \$\begingroup\$ Direct link to chapter on Adcock antennas is broken (has spurious trailing slash). I'd fix it but apparently edits of fewer than six characters are not allowed(!) \$\endgroup\$
    – cpcallen
    Dec 4, 2016 at 9:52

You can vary the strength of each of the four beams independently. Assuming the antennas are basic dipoles is overly simplistic.

Also, the terrain around the antenna array has an effect on the received signal strength for each beam.

  • \$\begingroup\$ I am not sure that just varying the signal power can achieve the second picture diagram. I think you can use four separate antennas and shield the back of each one in order to kill one of the lobes of the radiation diagram. \$\endgroup\$ May 14, 2016 at 12:41
  • \$\begingroup\$ The four separate antennas with shielding idea makes sense, but apparently most installations used Adcock antennas, which consist of four interconnected monpole antennas. Could some electrical or physical asymmetry between the two antennas in each pair change the propagation pattern? (Very little about the diagram in Wikipedia makes sense to me—each antenna is also connected to ground??—but then I know little about RF engineering, and hence my question.) \$\endgroup\$
    – cpcallen
    May 15, 2016 at 22:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.