Curtain array

Curtain array at international shortwave broadcasting station, Moosbrunn, Austria. It consists of 4 columns of horizontal wire dipoles, suspended in front of a wire screen. The vertical parallel wire feedlines to each column of dipoles are visible. The entire antenna is mounted on a rotating truss structure, allowing it to be pointed in different directions.

Antenna G1, a curtain array, at Hörby shortwave station operated by Radio Sweden. It consists of 16 horizontal wire dipoles in a 4x4 array, suspended in front of a wire screen. Each of the 4 columns of dipoles is fed by a separate open-wire transmission line, which can be seen exiting at an angle from the center of each column. The diagonal wires in the foreground are guy wires. The CCIR designation for this type of antenna (below) is HR 4/4/0.5

Curtain arrays are a class of large multielement directional wire radio transmitting antennas, used in the shortwave radio bands.^[1] They are a type of reflective array antenna, consisting of multiple wire dipole antennas, suspended in a plane, often in front of a "curtain" reflector made of a flat vertical screen of many long parallel wires.^[1] These are suspended by support wires strung between pairs of tall steel towers, up to 300 feet (90 m) high.^[1] They are used for long-distance skywave (or skip) transmission; they transmit a beam of radio waves at a shallow angle into the sky, which is reflected by the ionosphere back to Earth beyond the horizon. Curtain antennas are mostly used by international short wave radio stations to broadcast to large areas at transcontinental distances.^[1]

Because of their powerful directional characteristics, curtain arrays are often used in government propaganda radio stations to beam propaganda broadcasts over national borders into other nations. For example, curtain arrays were used by Radio Free Europe and Radio Liberty to broadcast into Eastern Europe.

History

Curtain arrays were originally developed during the 1920s and 1930s when there was a lot of experimentation with long distance shortwave broadcasting. The underlying concept was to achieve improvements in gain and/or directionality over the simple dipole antenna.

In the early 1920s, Guglielmo Marconi, pioneer of radio, commissioned his assistant Charles Samuel Franklin to carry out a large scale study into the transmission characteristics of short wavelength waves and to determine their suitability for long distance transmissions. Franklin invented the first curtain array aerial system in 1924, known as the 'Franklin' or 'English' system.^[2][3]

Other early curtain arrays included the Bruce array patented by Edmond Bruce in 1927,^[4] and the Sterba curtain, patented by Ernest J. Sterba in 1929.^[5] The Bruce array produces a vertically-polarised signal, Sterba arrays (and the later HRS antennas) produce a horizontally-polarised signal.

The Sterba array was used by Bell Labs and others during the 1930s and 1940s. The Sterba curtain is however a narrowband design and is only steerable by mechanical means. However, as far back as the mid-1930s, Radio Netherlands was using a rotatable HRS antenna for global coverage. Since the 1950s the HRS design has become more or less the standard for long distance high power shortwave broadcasting (> 1000 km).

Curtain arrays were used in some of the first radar systems, such as Britain's Chain Home network. During the Cold War, large curtain arrays were used by the Voice of America, Radio Free Europe, and Radio Liberty, and analogous Western European organizations, to beam propaganda broadcasts into communist countries, which censored Western media.

Description

Curtain arrays at Radio Free Europe transmitter site, Biblis, Germany

The driven elements are usually half wave dipoles, fed in phase, mounted in a plane 1/4 wavelength in front of the reflector plane.^[1] The reflector wires are oriented parallel to the dipoles. The dipoles may be vertical, radiating in vertical polarization, but are most often horizontal, because horizontally polarized waves are less absorbed by earth reflections.^[1] The lowest row of dipoles are mounted more than 1/2 wavelength above the ground, to prevent ground reflections from interfering with the radiation pattern.^[1] This allows most of the radiation to be concentrated in a narrow main lobe aimed a few degrees above the horizon, which is ideal for skywave transmission.^[1] A curtain array may have a gain of 20 dB greater than a simple dipole antenna.^[1] Because of the strict phase requirements, earlier curtain arrays had a narrow bandwidth, but modern curtain arrays can be built with a bandwidth of up to 2:1, allowing them to cover several shortwave bands.^[1]

Rather than feeding each dipole at its center, which requires a "tree" transmission line structure with complicated impedance matching, multiple dipoles are often connected in series to make an elaborate folded dipole structure which can be fed at a single point.

In order to allow the beam to be steered, sometimes the entire array is suspended by cantilever arms from a single large tower which can be rotated. Alternately, some modern versions are constructed as phased arrays in which the beam can be steered electronically, without moving the antenna. Each dipole or group of dipoles is fed through an electronically adjustable phase shifter, implemented either by passive networks of capacitors and inductors which can be switched in and out, or by separate output RF amplifiers. Adding a constant phase shift between adjacent horizontal dipoles allows the direction of the beam to be rotated in azimuth by a limited angle.

Nomenclature

Former Radio France Internationale (RFI) Issoudun Relay station feeders and curtain arrays.

Since 1984 the CCIR has created a standardised nomenclature for describing curtain antennas, consisting of 1 to 4 letters followed by three numbers:

• First symbol - Indicates the orientation of the dipoles in the array:
  • "H" indicates the dipoles are oriented horizontally, so the antenna radiates horizontally polarized radio waves.
  • "V" indicates the dipoles are oriented vertically, so the antenna radiates vertically polarized radio waves.
• Second symbol (if present) - Indicates whether the antenna has a reflector. If it is missing, the antenna lacks a reflector, so the dipole array will radiate its energy in two beams in both directions perpendicular to its plane, 180° apart.
  • "R" indicates that there is a simple (passive) reflector on one side of the array, so the antenna radiates a single beam.
  • "RR" indicates that the array has some kind of "reversible reflector", so the direction of the beam can be switched 180°. Very few of this type have ever been built. RCI Sackville in Canada may have 2 HRRS type antennas—perhaps the only ones in North America.
• Third symbol (if present) - "S" indicates that the array is steerable.

Following the letters come three numbers "x/y/z". "x" and "y" specifies the dimensions of the rectangular array of dipoles, while "z" gives the height above the ground of the bottom of the array:

• "x" (an integer) is the number of horizontal rows of dipoles.
• "y" (an integer) is the number of vertical columns of dipoles.
• "z" (a decimal fraction) is the height above ground in wavelengths of the lowest row of dipoles in the array.

For example, a "HRS 4/5/0.5" curtain antenna has a rectangular array of 20 dipoles, 4 dipoles high and 5 dipoles wide, with the lowest row being half a wavelength off the ground, and a flat reflector behind it, and the direction of the beam can be steered. An HRS 4/4/0.5 steerable antenna with 16 dipoles is one of the standard types of array seen at shortwave broadcast stations worldwide.

Simulated radiation pattern of a 15.1 MHz HR 6/4/1 curtain antenna (24 horizontal dipoles organized in 6 rows of 4 elements each, in front of a reflector), driven by a 500 kW transmitter. The transmitter is located in Seattle and the pattern covers Central America and parts of South America, showing the long distances achieved with this antenna. The main lobe of the pattern is flanked by two sidelobes, which appear curved due to the global map projection.

References