3 dB Beamwidth .5 power .707 voltage Peak power to first null Antenna Radiation Pattern 22 NN Azimuth and Elevation Beamwidths BW BW G • 4B0A 82 Where 0 ' Efficiency A ' Physical aperture area 8 ' wavelength another is: G ' X 0 BWN BW2 Where BW2 and N are the elev & az beamwidths in degrees. For approximating an antenna pattern with: (1) A rectangle; X'41253,0typical'0.7 (2) An ellipsoid; X'52525,0typical'0.55 BW ' 708 d D(22,NN) '' 10 Log 4BBP (22,NN) mmmmPin(22,NN) Sin 22 d22 dNN 0 < NN ## 360EE 0 < 22 ## 180EE [1] 3-1.1 Figure 1. Antenna Aperture ANTENNA INTRODUCTION / BASICS Rules of Thumb: 1. The Gain of an antenna with losses is given by: 2. Gain of rectangular X-Band Aperture G = 1.4 LW Where: Length (L) and Width (W) are in cm 3. Gain of Circular X-Band Aperture G = d 0 Where: d = antenna diameter in cm 2 0 = aperture efficiency 4. Gain of an isotropic antenna radiating in a uniform spherical pattern is one (0 dB). 5. Antenna with a 20 degree beamwidth has a 20 dB gain. 6. 3 dB beamwidth is approximately equal to the angle from the peak of the power to the first null (see figure at right). 7. Parabolic Antenna Beamwidth: Where: BW = antenna beamwidth; 8 = wavelength; d = antenna diameter. The antenna equations which follow relate to Figure 1 as a typical antenna. In Figure 1, BW is the azimuth beamwidth and N BW is the elevation beamwidth. Beamwidth is normally measured 2 at the half-power or -3 dB point of the main lobe unless otherwise specified. See Glossary. The gain or directivity of an antenna is the ratio of the radiation intensity in a given direction to the radiation intensity averaged over all directions. Quite often directivity and gain are used interchangeably. The difference is that directivity neglects antenna losses such as dielectric, resistance, polarization, and VSWR losses. Since these losses in most classes of antennas are usually quite small, the directivity and gain will be approximately equal (disregarding unwanted pattern characteristics). Normalizing a radiation pattern by the integrated total power yields the directivity of the antenna. This concept in shown in equation form by: (a) SPHERE (Isotropic source) P = 4 B R2 G = 0 dB P D P = 4 B R 2 G = +3 dB 2 P D P = 4 B R 2 G = +43 dB 18334 P D P = 4 B R2 G = +6 dB 4 P D (b) HEMISPHERE (c) QUARTER SPHERE (d) 1.5E SEGMENT in in in in -3 dB Beamwidth IDEAL ANTENNA PATTERN REAL ANTENNA PATTERN ( measured at the 0.5 power or 0.707 voltage points) 3D Views 2D Views Elliptical Model Rectangular Model G ' Area of Sphere Area of Antenna pattern G • 4B BWNazBW2el or 4B N2(radians) where:BWNaz ' Azmith beamwidth in radians BW2el ' Elevation beamwidth in radians 3-1.2 Figure 2. Antenna Gain Figure 3. Antenna Beamwidth Where D(2,N) is the directivity in dB, and the radiation pattern power in a specific direction is P (2,N), which is d normalized by the total integrated radiated power. Another important concept is that when the angle in which the radiation is constrained is reduced, the directive gain goes up. For example, using an isotropic radiating source, the gain would be 0 dB by definition (Figure 2(a)) and the power density (P ) at any given point would be the d power in (P ) divided by the surface area in of the imaginary sphere at a distance R from the source. If the spacial angle was decreased to one hemisphere (Figure 2(b)), the power radiated, P , would be the same in but the area would be half as much, so the gain would double to 3 dB. Likewise if the angle is a quarter sphere, (Figure 2(c)), the gain would be 6 dB. Figure 2(d) shows a pencil beam. The gain is independent of actual power output and radius (distance) at which measurements are taken. Real antennas are different, however, and do not have an ideal radiation distribution. Energy varies with angular displacement and losses occur due to sidelobes. However, if we can measure the pattern, and determine the beamwidth we can use two (or more) ideal antenna models to approximate a real antenna pattern as shown in Figure 3. Assuming the antenna pattern is uniform, the gain is equal to the area of the isotropic sphere (4Br ) 2 divided by the sector (cross section) area. [2] It can be shown that: [3] From this point, two different models are presented: (1) Approximating an antenna pattern using an elliptical area, and (2) Approximating an antenna pattern using a rectangular area. r 2 N b a Where 2=BW2, and N= BWN G ' 16 sinN sin2 – 16 N 2 (radians) ' 16 N 2 360E 360E 2B 2B ' 52525 N 2 (degrees) or 52525 BWN BW2 (degrees) a b r 2 N Where 2=BW2, and N= BWN G ' 4 B sinN sin2 ' 4 B N 2 (radians) ' 4 B N 2 360E 360E 2B 2B ' 41253 N 2 (degrees) or 41253 BWN BW2 (degrees) G ' Area of Sphere Area of Antenna pattern ' (4Br 2) 4 B r 2 sin2 sinN ' 16 sin2 sinN G ' Area of Sphere Area of Antenna pattern ' 4Br 2 r 2 sin2 sinN ' 4B sin2 sinN Gmax(dB) ' 10 Log 41253 BWN BW2 with BWN and BW2 in degrees 3-1.3 [4] [5] Approximating the antenna pattern as an elliptical area: Area of ellipse = B a b = B[ (r sin 2)/2 ][ (r sin N)/2 ]= (B r sin 2 sin N)/4 2 For small angles, sin N = N in radians, so: The second term in the equation above is very close to equation [3]. For a very directional radar dish with a beamwidth of 1E and an average efficiency of 55%: Ideally: G = 52525, or in dB form: 10 log G =10 log 52525 = 47.2 dB With efficiency taken into account, G = 0.55(52525) = 28888, or in log form: 10 log G = 44.6 dB Approximating the antenna pattern as a rectangular area: a = r sin 2 , b = r sin N, area = ab = r sin 2 sin N 2 For small angles, sin N = N in radians, so: The second term in the equation above is identical to equation [3]. Converting to dB, [6] For a very directional radar dish with a beamwidth of 1E and an average efficiency of 70%: Ideally (in dB form): 10 log G =10 log 41253 = 46.2 dB. With efficiency taken into account, G = 0.7(41253) = 28877, or in log form: 10 log G = 44.6 dB 22 88/2 88 22 A B C L ANTENNA BORESIGHT 3-1.4 Figure 4. . Antenna Sector Size vs Gain Figure 5. Directional Gain vs Wavelength Comparison between elliptical and rectangular areas for antenna pattern models By using the rectangular model there is a direct correlation between the development of gain in equation [5] and the ideal gain of equation [3]. The elliptical model has about one dB difference from the ideal calculation, but will yield the same real antenna gain when appropriate efficiencies are assumed. The upper plot of Figure 4 shows the gain for an ideal antenna pattern using the elliptical model. The middle plot shows the gain for an ideal antenna using the rectangular model. The lower plot of Figure 4 shows the gain of a typical real antenna (rectangular model using an efficiency of 70%or elliptical model using an efficiency of 47%). Gain as a function of 8: When 2 = 0, each wave source in Figure 5 is in phase with one another and a maximum is produced in that direction. Conversely, nulls to either side of the main lobe will occur when the waves radiating from the antenna cancel each other. The first null occurs when there is a phase difference of 8/2 in the wave fronts emanating from the aperture. To aid in visualizing what happens, consider each point in the antenna aperture, from A to C in Figure 5, as a point source of a spherical wave front. If viewed from infinity, the electromagnetic waves from each point interfere with each other, and when, for a particular direction, 2 in Figure 5, each wave source has a corresponding point that is one-half wavelength out of phase, a null is produced in that direction due to destructive interference. In Figure 5, the wave emanating from point A is out of phase with the wave from point B by one-half of a wavelength. Hence, they cancel. Similarly, a point just to the right of point A cancels with a point just to the right of point B, and so on across the entire aperture. Therefore, the first null in the radiation pattern is given by: Sin 2 = 8/L and, in radians, 2 = 8/L (for small angles) [7] G • 4B BWN BW2 (radians) ' 4BLazLel 82 ' 4BAe 82 3-1.5 As the angle off boresight is increased beyond the first null, the intensity of the radiation pattern rises then falls, until the second null is reached. This corresponds to a phase difference of two wavelengths between the left and right edges of the aperture. In this case, the argument proceeds as before, except now the aperture is divided into four segments (point A canceling with a point halfway between A and B, and so on). The angle 2 is the angle from the center (maximum) of the radiation pattern to the first null. The null-to-null beam width is 22. Generally, we are interested in the half-power (3 dB) beamwidth. It turns out that this beamwidth is approximately one-half of the null-to-null beamwidth, so that: BW . (½)(22) = 8/L [8] 3 dB Therefore, beamwidth is a function of the antenna dimension “L” and the wavelength of the signal. It can be expressed as follows: Note: for circular antennas, L in the following equations = diameter Bw = 8/L and BW = 8/L [9] N(az) Az eff 2(el) El eff Substituting the two variations of equation [9] into equation [3] and since L times L = A (effective capture area Az eff El eff e of the antenna), we have: [10] Note: Equation is approximate since aperture efficiency isn’t included as is done later in equation [12]. The efficiency (discussed later) will reduce the gain by a factor of 30-50%, i.e. real gain = .5 to .7 times theoretical gain. Unity Gain Antenna. If a square antenna is visualized and G=1, A = 8 / 4B. When a dimension is greater than 0.28 8 (~¼8 ) it is known as e 2 an electrically large antenna, and the antenna will have a gain greater than one (positive gain when expressed in dB). Conversely, when the dimension is less than 0.28 8 (~¼8 )(an electrically small antenna), the gain will be less than one (negative gain when expressed in dB). Therefore, a unity gain antenna can be approximated by an aperture that is ¼8 by ¼8. Beamwidth as a Function of Aperture Length It can be seen from Figure 5, that the wider the antenna aperture (L), the narrower the beamwidth will be for the same 8. Therefore, if you have a rectangular shaped horn antenna, the radiation pattern from the wider side will be narrower than the radiation pattern from the narrow side. APERTURE EFFICIENCY, 00 The Antenna Efficiency, 0, is a factor which includes all reductions from the maximum gain. 0 can be expressed as a percentage, or in dB. Several types of "loss" must be accounted for in the efficiency, 0: (1) Illumination efficiency which is the ratio of the directivity of the antenna to the directivity of a uniformly illuminated antenna of the same aperture size, (2) Phase error loss or loss due to the fact that the aperture is not a uniform phase surface, (3) Spillover loss (Reflector Antennas) which reflects the energy spilling beyond the edge of the reflector into the back lobes of the antenna, (4) Mismatch (VSWR) loss, derived from the reflection at the feed port due to impedance mismatch (especially important for low frequency antennas), and (5) RF losses between the antenna and the antenna feed port or measurement point. Ae ' 0 A ' 82G 4B G ' 4B0A 82 Where 0 ' Aperture Efficiency A ' Physical aperture area 8 ' wavelength 3-1.6 [11] The aperture efficiency, 0 , is also known as the illumination factor, and includes items (1) and (2) above; it does not result a in any loss of power radiated but affects the gain and pattern. It is nominally 0.6-0.8 for a planer array and 0.13 to 0.8 with a nominal value of 0.5 for a parabolic antenna, however 0 can vary significantly. Other antennas include the spiral (.002-.5), the horn (.002-.8), the double ridge horn (.005-.93), and the conical log spiral (.0017-1.0). Items (3), (4), and (5) above represent RF or power losses which can be measured. The efficiency varies and generally gets lower with wider bandwidths. Also note that the gain equation is optimized for small angles - see derivation of wavelength portion of equation of equation [7]. This explains why efficiency also gets lower for wider beamwidth antennas. EFFECTIVE CAPTURE AREA Effective capture area (A ) is the product of the physical aperture area (A) and the aperture efficiency (0) or: e GAIN AS A FUNCTION OF APERTURE EFFICIENCY The Gain of an antenna with losses is given by: [12] Note that the gain is proportional to the aperture area and inversely proportional to the square of the wavelength. For example, if the frequency is doubled, (half the wavelength), the aperture could be decreased four times to maintain the same gain. BEAM FACTOR Antenna size and beamwidth are also related by the beam factor defined by: Beam Factor = (D/8)@(Beamwidth) where D = antenna dimension in wavelengths. The beam factor is approximately invariant with antenna size, but does vary with type of antenna aperture illumination or taper. The beam factor typically varies from 50-70E. APERTURE ILLUMINATION (TAPER) The aperture illumination or illumination taper is the variation in amplitude across the aperture. This variation can have several effects on the antenna performance: (1) reduction in gain, (2) reduced (lower) sidelobes in most cases, and (3) increased antenna beamwidth and beam factor. Tapered illumination occurs naturally in reflector antennas due to the feed radiation pattern and the variation in distance from the feed to different portions of the reflector. Phase can also vary across the aperture which also affects the gain, efficiency, and beamwidth. CIRCULAR ANTENNA GAIN Solving equation [12] in dB, for a circular antenna with area BD /4, we have: 2 10 Log G = 20 Log (D/8) + 10 Log (0) + 9.94 dB ; where D = diameter [13] This data is depicted in the nomograph of Figure 6. For example, a six foot diameter antenna operating at 9 GHz would have approximately 44.7 dB of gain as shown by the dashed line drawn on Figure 6. This gain is for an antenna 100% efficient, and would be 41.7 dB for a typical parabolic antenna (50% efficient). An example of a typical antenna (with losses) showing the variation of gain with frequency is depicted in Figure 7, and the variation of gain with antenna diameter in Figure 8. The circle on the curves in Figure 7 and 8 correspond to the Figure 6 example and yields 42 dB of gain for the 6 ft dish at 9 GHz. +10 dBm Signal L1 6 dBi gain each Point A L2 3-1.7 Figure 6. Antenna Gain Nomograph Example Problem: If the two antennas in the drawing are “welded” together, how much power will be measured at point A? (Line loss L = L = 0.5, and 10log L or L = 3 dB) 1 2 1 2 Multiple choice: A. 16 dBm b. 28 dBm c. 4 dBm d. 10 dBm e. < 4 dBm Answer: The antennas do not act as they normally would since the antennas are operating in the near field. They act as inefficient coupling devices resulting in some loss of signal. In addition, since there are no active components, you cannot end up with more power than you started with. The correct answer is “e. < 4 dBm.” 10 dBm - 3 dB - small loss -3 dB = 4 dBm - small loss If the antennas were separated by 5 ft and were in the far field, the antenna gain could be used with space loss formulas to calculate (at 5 GHz): 10 dBm - 3 dB + 6 dB - 50 dB (space loss) + 6 dB -3 dB = -34 dBm (a much smaller signal). 50 45 40 35 30 25 20 2 4 6 8 10 12 14 16 18 FREQUENCY (GHz) EXAMPLE ONLY 50 45 40 35 30 25 20 2 4 6 8 10 12 14 16 18 DIAMETER (Feet) EXAMPLE ONLY 3-1.8 Figure 7. Gain of a Typical 6 Foot Dish Antenna (With Losses) Figure 8. Gain of a Typical Dish at 9 GHz (With Losses)