LARGE WIND TURBINE SITING HANDBOOK: TELEVISION INTERFERENCE ASSESSMENT Technical Report No. 5 Thomas B.A. Senior and Dipak L. Sengupta Radiation Laboratory Department of Electrical and Computer Engineering The University of Michigan Ann Arbor, Michigan 48109 14438-5-T = RL-2271 April 1981 Prepared for Department of Energy Wind Systems Branch Division of Solar Energy Work performed under Contract EY-76-S-02-2846.A004

NOTICE This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States nor any agency thereof, nor any of their employees, makes any warranty, expressed or implied, or assumes any legal liability or responsibility for any third party's use or the results of such use of any information, apparatus, product or process disclosed in this report, or represents that its use by such third party would not infringe privately owned rights.

ACKNOWLEDGEMENT It is a pleasure to acknowledge the suggestions made by Mr. Neil Kelley of SERI during preparation of this Handbook. -i

ABSTRACT The rotating blades of a horizontal axis wind turbine can distort the video portion of a television signal and thereby interfere with TV reception in the vicinity of the machine. The nature of this interference is described and methods are presented for calculating the zone within which the interference may be severe. Specific results are given for the MOD-OA, MOD-1 and MOD-2 machines as functions of the TV frequency.

TABLE OF CONTENTS Section 1. Introduction 2. Interference Phenomena 3. Modulation Threshold and the Interference Zone 4. Interference Zone Calculations 5. Computed Values of rl and r2 6. Effects of Directional Receiving Antennas 7. TVI Assessment Procedure 7.1 Method A 7.2 Method B 8. A Typical TVI Assessment 9. A Note on the Smooth Earth Model 10. Conclusions 11. Appendix A: Computer-Graphical Method for Calculation of the Interference Zone A.1 Theoretical Considerations A.2 Computational Procedures A.3 Specific Computations 12. Appendix B: Glossary of Symbols 13. References Page 1 2 4 10 11 14 15 15 16 20 22 22 25 25 30 34 51 53

1. Introduction The purpose of this Handbook is to show how to predict the region where a large horizontal axis wind turbine (WT) can significantly affect the reception of television (TV) signals. Such information can assist in the selection of a site for a WT facility and in the analysis of the interference under a variety of circumstances. The Handbook is a revision of one previously published [1] and differs from it in several particulars. As a result of data that have become available since that time it is apparent that a change in height at which the WT reflects TV signals can have just as much effect as a change in the blade area responsible for the reflection. Since the tower height used throughout [1] was 30 m, the curves given there are applicable only to a MOD-OA (or similar) machine. In addition, on-site and laboratory measurements of WT-produced interference have shown that the modulation threshold used to specify the interference zone increases with decreasing time delay of the secondary (scattered) signal relative to the primary. This not only reduces the interference zone radius in the forward scattering region but also permits some simplification in the method used to calculate the entire zone. Interference distances are tabulated as functions of the frequency and distance to the TV transmitter for MOD-OA, MOD-1 and MOD-2 machines. For these same machines with blades other than metallic, information is provided to enable the corresponding numbers to be obtained. -1 -

-2 - The present Handbook provides two methods for performing the assessment of television interference or TVI effects of a large horizontal axis WT after determining its interference zone: (i) Method A, sufficient for a first order estimation of the interference zone, is less rigorous but comparatively easier to apply, and (ii) Method B, suitable for more accurate. determination of the interference zone, is more detailed but somewhat difficult to use. References [2-6] contain a detailed description of the theoretical and experimental investigations that have been performed, and the reader may wish to consult these. The TV interference is intimately connected with the characteristics of radio wave propagation over the earth and in the presence of the WT. It is assumed that the waves propagate over a homogeneous smooth spherical earth having a radius 6.34 x 103 km, with relative permittivity Er = 15 and conductivity a = 0.01 siemens/m. The medium above the earth is assumed to be free space and the effects of any terrain inhomogeneity or irregularity are ignored. 2. Interference Phenomena It has been found [2] that the rotating blades of a horizontal axis WT can interfere with TV reception by producing video distortion. At a given distance from a WT the interference increases with increasing frequency and is therefore worst on the upper UHF Channels. It also decreases with increasing distance from the machine, but in the worst cases can still produce objectionable video distortion at distances up to a few kilometers. For ambient (primary) signals well

-3 - above the noise level of the receiver, there is in general no significant dependence on the ambient field strength, and no audio distortion has been observed. The interference is caused by the time-varying amplitude modulation (AM) of the received signal produced by the rotating blades. In a neighborhood of a WT the signals scattered by the blades combine with the primary signal to create a form of time-varying multipath, thereby amplitude modulating the total received signal. The amplitude modulation waveform consists of sync pulses repeating at twice the rotation frequency of the blades. The width of the pulses is inversely proportional to the electrical length of a blade. If sufficiently strong, these extraneous pulses can distort the received picture, whereas the audio information, being transmitted by frequency modulation, remains unaffected. When the blades are stationary the scattered signal may appear on the TV screen as a ghost whose position (separation) depends on the difference between the time delays suffered by the primary and scattered signals. A rotation of the blades then causes the ghost to fluctuate, and if the ghost is sufficiently strong, the resulting interference can be objectionable. In such cases the received picture displays a horizontal jitter in synchronism with the blade rotation. As the interference increases, the entire (fuzzy) picture shows a pulsed brightening, and still larger interference can disrupt the TV receiver's vertical sync, causing the picture to roll over ('flip') or even break up. This type of interference occurs when the interfering signal reaches the receiver as a result of scattering, primarily specular, off the broad faces of the blade(s). As the angle ~ (see

-4 - Fig. 1) between the WT-transmitter and WT-receiver directions increases, the separation of the ghost decreases, and a somewhat greater interference is now required to produce the same amount of video distortion. In the forward scattering region when the WT is almost in line between the transmitter and the receiver, there is virtually no difference in the times of arrival of the primary and secondary signals. The ghost is then superimposed on the undistorted picture and the video interference appears as an intensity (brightness) fluctuation of the picture in synchronism with the blade rotation. In all cases, the amount of interference depends on the strength of the scattered signal relative to the primary one, and this decreases with increasing distance from the WT. 3. Modulation Threshold and the Interference Zone For the purpose of the present discussion we use Fig. 1 which shows the locations of the distant TV transmitter (T), the phase center of the omnidirectional receiving antenna ((R), and the scattering center (B) of a blade, all assumed to lie in a horizontal plane, i.e., the plane of the paper. It is also assumed that the blade rotates in a vertical plane through M-M (see Fig. 1), and that this plane is so oriented as to direct the specularly reflected field to the receiver. The extraneous amplitude modulation index of the total field at R is EB(R) m (R(1) E (R)

-5 - R (reciever at r,f) r M T ( I I - -- ( M (distant 1V transmitter) phase center of the WT blade B Figure 1. Geometry of the WT blade scattering.

-6 - where ET(R) is the amplitude of the direct (or primary) field at R and EB(R) is the amplitude of the secondary (or scattered) field which reaches the receiver after reflection off a rotating blade of the WT. The modulation threshold m0 is defined as the largest value of the extraneous amplitude modulation index m of the received signal for which the video distortion is still judged to be acceptable. The threshold is obviously somewhat subjective, but as a result of detailed laboratory measurements [2], field tests [2,3], and controlled measurements using a small scale model [4], it has been established for all TV Channels as 0.15 for p: 0 increasing to 0.35 for f: 180 degrees as shown in Fig. 2. The threshold is largely independent of the signal strength at the receiver and, for a given WT, determines the maximum distance from the WT at which the interference can still be objectionable. On the assumption that the WT is so oriented (in azimuth) as to direct the maximum scattered signal to the receiver, the region where m > m0 is defined as the interference zone [1,7]. That portion of the zone which is produced by specular reflection off the blades is roughly a cardioid centered at the WT with its maximum pointing towards the TV transmitter, and is called the backward interference region. There is also a narrow lobe, called the forward interference region, directed away from the transmitter resulting from forward scattering off the blades. The shape of the interference zone is sketched in Fig. 3 where r1 and r2 may be identified as the backward and forward interference distances, respectively, and these generally provide the maximum distances from the WT at which objectionable interference can occur in the two regions.

-7 - 0.4 r - /. -. mo(0) 0.2 -.. O - -.>-" 0 I I I I I I I ' i I I 0 60 120 - 180 Figure 2. - - 0 ( degrees ) -;..... Chosen threshold modulation index m (p) as a function of the angular position of the receiver with respect to the WT-transmitter direction.

-8 - Forward Interference Region = 0oo R1 Distant TV Transmitter (not drawn to scale) r- = 180~ PI2 Backward Interference Region Figure 3. Sketch of the TV interference zone of a WT.

-9 - In the backward region the interference is generally independent of the ambient or primary signal strength, and the interference region is obtained as r = r(p) for which m > m = 0.15. The forward region interference is somewhat dependent on the ambient signal strength. [3], being strongest in fringe areas of reception where the signal strength is low. The forward interference region is that for which m > mo where m = 0.35 in a strong signal area, decreasing uniformly to 0.15 when the ambient signal is weak. Consequently, in fringe areas r(T) = r2 > r(O) = rl, i.e., the forward interference determines the maximum interference distance of the WT. In areas where the signal is not weak the reverse is generally true, and r1 then determines the maximum interference distance. Knowing r1 and r2, the interference zone of a WT can be approximately obtained from [1,7] r(p) = ri cos 0 < ~ J < - - (2) and TrL sin[ e sin )] ) = r2 X L - l < < <, (3) rL e X sin where Le is the equivalent length of the WT blade and x is the wavelength. Assuming now that the maximum scattered field at R (i.e., EB(R)) is directly proprotional to the equivalent scattering area Ae of each blade and inversely proportional to the distance r from the WT (Fig. 1), the two interference distances are approximately given by [7] 2A T 2A T 2A e ET(B) r E ) (4) r ( EB) Xm T 2 Xm T (4) o E (RI)

-10 - where R1 = (r1,0), R2 = (r2,T) (see Fig. 3) and ET(B) is the amplitude of the direct (ambient) field of the TV transmitter at the WT blade B. It is evident that knowledge of Ae and of the secondary to primary field ratio, i.e., EB(R)/ET(R) at R, or of the ratio of the primary (ambient) fields at B and R, i.e., ET(B)/ET(R), is necessary to determine rl and r2 and, hence, the interference zone. Theoretical determination of both of these ratios requires the use of rigorous field expressions and the elaborate numerical computation discussed in [1,2]. However, interference zone calculations become much simpler if the ambient field strength ratio is known from measurements made at the wind turbine site and at receiver locations. For a rough assessment of the television interference (TVI) associated with a WT it is sufficient to obtain r1 and r2 as a function of frequency. 4. Interference Zone Calculations The calculation of the interference zone of a WT is based on the theoretical expressions for the fields derived in [2] and the computational procedure discussed in [1]. The computer program is itself an extension of one developed [8] for calculating ground wave propagation over a homogeneous spherical earth. The transmitting and receiving antennas are assumed omnidirectional and horizontally polarized, and located at heights h1 and h2, respectively, above the earth's surface. The height of the scattering or phase center of the blade is h3. At a given frequency and for given hi, h2 and h3, the program computes the field quantities ET(R) and/or E (B) as functions

-11 - of the great circle distance (d) from the transmitter and the scattered field EB(R) as a function of the distance (r) from the WT. Though the program can handle any specified values for the relative permittivity r and conductivity a of the earth and any earth radius (a) corrected to account for atmospheric refraction, the computations were all performed for the values: a = 6.34 x 103 km, c = 15, a = 0.01 s/m. With the frequency f and distance d as inputs, the quantities B T E (R) and ET(R) are obtained at a sequence of distances from the WT along the same great circle path from the transmitter. From a visual comparison of the two sets of data, the farthest points on either side of the WT at which m = m0 are located. These specify the two distances r1 and r2. If desired, the actual interference zone can be found by solving (1) with m = m0 or estimated by using (2) and (3). As is evident, however, the process requires computation of a whole new series of field strength values for every TV transmitter and its distance from the WT. A simpler graphical method which makes use of a general set of computed data was developed in its place [1]. The details of the computer-graphical method used to determine rl, r2 and the interference zone for large wind turbines are described in Appendix A. 5. Computed Values of r, and r2 The present section gives the interference distances of MOD-OA, MOD-1 and MOD-2 wind turbines obtained numerically as discussed in the

-12 - previous sections. All calculations were carried out with hi = 300 m and h2 = 10 m for omnidirectional transmitting and receiving antennas. The basic parameters used for the three machines with metal blades were: h3 (m) Ae (m2) Le (m) MOD-OA 30 12 15 MOD-1 45 40 25 MOD-2 60 140 63 The computed interference distances r1 = r(0O) and r2 = r(180~) for the three machines are shown in Table l(a)-(c), for f = 50, 100, 200, 300, 500, 700, 900 MHz with d = 20, 40, 60, 80, 100 km. For completeness, we have also included in the table the interference distance r(90~) in the backward region. Note that the radio horizon distance for these combinations of the transmitting antenna and WT tower heights are d = 81,86 and 89 km for MOD-OA, MOD-1 and MOD-2 machines, respectively; the corresponding distance for the given transmitting and receiving antenna heights is d = 73 km. The results in Table 1 show that for omnidirectional receiving antennas r1 generally exceeds r2 within the radio horizon, but beyond the horizon, i.e., in fringe areas, r1 is less than r2. In those few cases where data for the MOD-2 machines are missing from Table 1(c), the forward scattered signal dominates out to a large distance from the WT and r2 is then many tires larger than rl.

-13 - Table 1: Computed Interference Zone Radii (a) MOD-OA d(km) f(MHz) 20 40 60 80 >100 50 0.08 0.04 0.03 0.08 0.04 0.03 0.08 0.04 0.03 0.08 0.04 0.03 0.08 0.04 0.03 100 0.14 0.08 0.06 0.16 0.08 0.07 0.16 0.09 0.07 0.16 0.09 0.07 0.16 0.09 0.07 200 0.18 0.10 0.08 0.29 0.17 0.13 0.36 0.17 0.13 0.33 0.17 0.14 0.33 0.17 0.14 300 0.20 0.11 0.09 0.43 0.24 0.20 0.50 0.26 0.21 0.50 0.26 0.23 0.49 0.26 0.23 500 0.25 0.14 0.11 0.58 0.32 0.26 0.79 0.43 0.36 0.84 0.47 0.39 0.83 0.47 0.39 700 0.34 0.19 0.15 0.56 0.31 0.25 1.1 0.61 0.52 1.1 0.70 0.59 1.1 0.69 0.59 900 0.35 0.24 0.20 0.59 0.32 0.26 1.3 0.75 0.66 1.5 0.93 0.82 1.4 0.92 0.82 In each box the three entries are, from left to right, r(O), r(90), r(180). (b) MOD-1 d(km) f(MHz) 20 40 60 80 >100 50 0.36 0.2q 0.16 0.36 0.21 0.17 0.37 0.21 0.17 0.37 0.21 0.18 0.37 0.21 0.18 100 0.60 0.32 0.26 0.70 0.41 0.34 0.72 0.44 0.35 0.71 0.44 0.36 0.71 0.44 0.36 200 0.61 0.34 0.28 1.3 0.78 0.66 1.6 0.91 0.76 1.5 0.91 0.76 1.5 0.91 0.76 300 0.66 0.36 0.29 1.8 1.1 0.93 2.2 1.4 1.2 2.3 1.4 1.3 2.3 1.4 1.3 500 0.84 0.45 0.38 1.8 1.1 0.93 3.2 2.2 2.1 3.4 2.6 2.7 3.4 2.6 2.7 700 1.1 0.63 0.55 1.8 1.1 0.93 4.0 2.9 2.7 4.6 4.0 5.5 5.4 5.2 7.7 900 1.5 0.80 0.68 1.9 1.1 0.93 4.2 3.3 4.2 4.4 4.0 5.8 5.2 5.4 11 1.5_0. 80_0.681_ 1. 9 _1.1_0.93 In each box the three entries are, from left to right, r(0), r(90), r(180). (c) MOD-2 __ d(km) f(MHz) 20 40 60 80 >100 50 0.96 0.70 0.64 1.0 0.74 0.66 1.0 0.75 0.68 1.0 0.76 0.68 1.0 0.76 0.68 100 1.5 1.1 1.0 1.9 1.4 1.2 2.0 1.5 1.4 2.0 1.5 1.4 2.0 1.5 1.4 200 1.8 1.2 1.1 3.3 2.6 2.6 3.8 3.1 3.1 3.8 3.2 3.2 3.8 3.2 3.2 300 2.0 1.3 1.2 4.2 3.5 3.6 5.2 4.6 5.0 5.4 4.9 5.6 5.4 4.9 5.8 500 2.6 1.6 1.4 4.7 3.7 4.0 7.4 7.2 10 7.8 8.6 16 7.8 8.5 18 700 3.4 2.2 1.8 5.0 3.6 4.1 8.5 8.9 18 9.7 12 * 9.7 12 * 900 4.2 2.8 2.5 5.2 3.7 4.3 8.7 9.6 * 1l 15 * 11 16 * In each box the three entries are, from left to right, r(O), r(90), r(180).

-14 - 6. Effects of Directional Receiving Antennas The interference distances given in Table 1 were obtained with an omnidirectional receiving antenna. Almost all TV reception is obtained with a directional antenna and it is therefore of interest to know its effect on the interference. For normal reception the mainbeam of the antenna is directed at the desired transmitter to receive maximum signal. Let the voltage pattern of the antenna be f(W') where p' is measured from the center of the mainbeam with f(O) = 1, and f(q') < 1 for 0' f 0. Referring to Fig. 1, it can be seen that for d >> r, (' O 180~ - p and that p' may be interpreted as the angle between the transmitter and WT directions as seen from R. It is now simple to show that when a receiving antenna having a voltage pattern f((') is used to receive the signals, the appropriate interference distances can be obtained by multiplying those given in Table 1 by the factor f ('). The effect on ri and r2 is as follows. In the backward direction p = 0 and f(p') = f(1800)which is the back-to-front ratio of the antenna. For a typical TV antenna, f(180~) = -12 dB = 1/4 which implies that all r1 values given in Table 1 are decreased by a factor of 4. However, in the forward region, 0' 0, and f(p') = 1, showing that r2 is unaffected. Thus, in contrast to the case of an omnidirectional receiving antenna, r2 now generally determines the maximum distance from the WT at which the interference is still severe.

-15 - 7. TVI Assessment Procedure The computed interference distances given in Table 1 and the computer-graphical method for determining the interference zone discussed in Appendix A can be used to obtain similar information about a WT similar to the MOD-OA, MOD-1 or MOD-2 machine located above a smooth and homogeneous spherical earth. In many cases, a rough estimate of the TVI effects in the vicinity of these machines may be obtained from a knowledge of their interference distances; however, for a better assessment of the TVI effects around a WT, it is desirable to determine its complete interference zone. In the following we provide two basic methods for the determination of a desired interference zone. 7.1 Method A. It uses approximate relationships in combination with computed interference distances to determine the interference zone. The method is less rigorous but comparatively easier to use. A step-by-step procedure to be followed for such as assessment is given below for an omnidirectional antenna: Use of Table 1: (i) For the given WT, obtain the turbine height h3 (in m), equivalent blade scattering area A' (in m2) and the equivalent blade length Le (in m2). From h3, identify the set of data (a), (b) or (c) of Table 1 to be used. (ii) In the table chosen in (i), identify the column for which d is closest to the distance in km of the desired TV transmitter. (iii) From the column chosen in (ii), select the row appropriate to the frequency f (in MHz) closest to that of the desired TV Channel

-16 - number (see Table 2) and obtain rl = r(0~), r2 = r(l80~) and, if desired, r(90~) (iv) If the equivalent scattering area Ae of the WT is different from the A used in Table 1, multiply each distance obtained in (iii) by Ae/Ae. Steps (i) through (iv) complete the determination of ri and r2 and r(90~). To obtain a rough estimate of the interference zone: (v) Use Eq. (2) with r1 obtained in (iii) to obtain the backward interference region, and use Eq. (3) with r2 obtained in (iii) to obtain the forward interference region. 7.2 Method B. It determines the interference zone with the help of the detailed numerical and graphical method described in Appendix A. Although the method is more accurate, it is more time consuming, and difficult to apply. The step-by-step procedure to be followed is as follows (see also Appendix A): Use of Table 1: (i) For the given WT, obtain the turbine height h3 (in m), equivalent blade scattering area A' (in m2) and the equivalent blade e length Le (in m2). From h, identify the set of data (a), (b) or (c) of Table 1 to be used. (ii) In the table chosen in (i), identify t-he column for which d is closest to the distance in km of the desired TV transmitter. (iii) From the column chosen in (ii), select the row appropriate to the frequency f (in MHz) closest to that of the desired TV Channel number (.see Table 2) and obtain r1 = r(0~),r2 = r(180~) and, if desired, r(90~),

-17 - Table 2: Television frequencies Channel Center Frequency (MHz) No. video audio I Channel Center Frequency (MHz) No. video I audio.... 2 3 4 5 6 7 8 9 10 11. 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 55.25 61.25 67.25 77.25 83.25 175.25 181.25 187.25 193.25 199.25 205.25 211.25 471.25 477.25 483.25 488.25 495.25 501.25 507.25 513.25 519.25 525.25 531.25 537.25 543.25 549.25 555.25 561.25 59.75 65.75 71.75 81.75 87.75 179.75 185.75 191.75 197.75 203.75 209.75 215.75 475.75 481.75 487.75 493.75 499.75 505.75 511.75 517.75 523.75 529.75 535.75 541.75 547.75 553.75 559.75 565.75 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 567.25 573.25 579.25 585.25 591.25 587.25 603.25 609.25 615.25 621.25 627.25 633.25 639.25 645.25 651.25 657.25 663.25 669.25 675.25 681.25 687.25 693.25 699.25 705.25 711.25 717.25 723.25 729.25 571.75 577.75 583.75 589.75 595.75 601.75 607.75 613.75 619.75 625.75 631.75 637.75 643.75 649.75 655.75 661.75 667.75 673.75 679.75 685.75 691.75 697.75 703.75 709.75 715.75 721.75 727.75 733.75 i - -- --

-18 - Television frequencies (continued) Channel Center Frequency (MHz) No. Video audio I I 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 i 735.25 741.25 747.25 753.25 759.25 765.25 771.25 777.25 783.25 789.25 795.25 801.25 807.25 813.25 819.25 825.25 831.25 837.25 843.25 849.25 855.25 861.25 867.25 873.25 879.25 885.25 I 739.75 745.75 751.75 757.75 763.75 769.75 775.75 781.75 787.75 793.75 799.75 805.75 811.75 817.75 823.75 829.75 835.75 841.75 847.75 853.75 859.75 865.75 871.75 877.75 883.75 889.75 I I J- -- — - -- —

-19 - (iv) If the equivalent scattering area A' of the WT is different from the A used in Table 1, multiply each distance obtained in (iii) by Ae/Ae. (v) (a) Choose X, note cos p. (b) Identify the appropriate set of data, i.e., (a), (b), or (c) in Table 3 and obtain y. Obtain y from Fig. 5 and 1 2 determine y = y1 + Y2' (c) From the given A, d and h identify Fig. 6(x) where x = a,b, e 3 or c. From the given d, identify Fig. 7(x) where x = a,b,..., or e. (d) Overlay Fig. 6(x) and 7(x) such that the y level of Fig. 6(x) is at the 0 level of 7(x) with r = 1 of Fig. 6(x) at the r = Icos f\ position of Fig. 7(x). Read off from 6(x) the value of r at which the f (MHz) curves cross, using the solid lines if p < 90~ (dashed lines if q > 90~) in Fig. 7(x). The result is r(f). (vi) If A' f Ae, multiply r(p) of (vi) by A'/Ae. Repeat (i) through (vii) for all other Channels. Steps (i) through (vi) complete the determination of rl, r2 and r(f) for omnidirectional receiving antennas. With a directional receiving antenna having a voltage polar diagram f(4') such that f(O) = 1, p' = 0 being the direction of the maximum of the main beam, (vii) multiply the r1, r2, r( ) obtained in (i) through (vi) by f(p') where p' = 180~ - 4.

-20 - 8. A Typical TVI Assessment In this section we use the TVI assessment procedure given above to determine the interference zone of the MOD-1 WT on Channel 56 whose transmitter is 75 km away. The known parameters are: video and audio carrier frequencies of Channel 56 are (from Table 2) 723.2 and 727.75 MHz; d = A = e Proceed as Method A 75 km, 40 m2, L = 25 m. e fo 1 ows: (i) Since the machine is MOD-1, use Table l(b). (ii) Choose f = 700 MHz (X = 0.43 m), d = 80 km and entE Table l(b) at (700, 80) to obtain (iii) r1 = 4.6 km, r2 = 5.5 km and, if desired, r (90~) = 4.0 km. If the equivalent blade scattering area of the given machine different from that of the MOD-1 WT (for example, if A'/A = then modify the distances obtained in (iii) as follows: (iv) ri = 4.6 - 1.15 km, r2 = =5 - 1.38 km r (90~) = 4 = 1.0 km r(900) -4.0 1.0 km er is 1/4), Assuming A' = Ae = 40 m2, obtain the approximate interference zone as TrL (v) X- 0.017 (=1.00), e = 183 e

-21 - backward interference region (Eq. (2)): r(4) = 4.6 cos 2 km, 0 179o forward interference region (Eq. (3)): r(55 sin(183 sin m) km, 179~ < < 180~ 183 sin p Method B (see Section A.3) (vi) See Appendix A for detailed calculations of r1,r2-and r(90~), and for a more accurate determination of the interference zone using the computer-graphical method. Steps (i) through (vi) complete the assessment for an omnidirectional receiving antenna. With a directional receiving antenna whose pattern f(p') is known, proceed as follows: (vii) If the back-to-front ratio f(180~) = -12 dB = 1/4, then 4.6 rl= 4 = 1.15 km, r2 = 5.5 km (as in (iii)) approximate backward interference region: r(f) = 1.15 cos 2 km, 0 < _ < 170~ approximate forward interference region: same as in (v) above.

- 22 - All interference distances in the backward region obtained in step (vi) should be multiplied by f(,1800 - 9. A Note on the Smooth Earth Model The calculation discussed above have been carried out on the assumption of a smooth and homogeneous spherical earth, and it is realized that local topography could increase or decrease the size of the interference zone obtained in Section 7. Such effects are particularly important when the wT is located in a mountainous region where the ambient field strengths at the WT and the receiving sites may differ significantly from those computed using the smooth earth model. It is therefore desirable to incorporate the effects of the local terrain and/or the hills and other dominant features of the terrain on the field strengths into the computer model. It is understood that a computer program called RAPIT (Radio Propagation over Irregular Terrain) is available [9]. 10. Conclusions Using results obtained from a computer program for ground wave propagation over a smooth and homogeneous spherical earth, an efficient graphical procedure has been developed for predicting the zones of TV interference about a large WT. The rough shapes of these zones are similar for all horizontal axis machines but their size increases with increasing WT height, blade area and TV Channel number or frequency. The interference is therefore worst on the

-23 - higher UHF Channels and increases in severity with the size of the machine. The observed interference effects depend significantly on the quality of the receiving antenna. With a poor (omnidirectional) or a wrongly oriented good (directional) antenna, any TVI effects in a given situation could be pronounced, and under similar conditions the backward interference would be dominant, implying r1 > r2. However, with a properly oriented good antenna, the backward region interference is reduced by an amount depending on the side and back lobe characteristics of the antenna and the maximum interference distance of a WT is generally determined by the forward scattering from the WT. For the MOD-OA, MOD-1 and MOD-2 machines with metal blades, the interference distances in three directions are tabulated at seven frequencies spanning the VHF and UHF TV bands for five WT-transmitter distances, and under most circumstances these parameters are sufficient to define the entire zone. At other frequencies and/or distances adequate accuracy can be obtained by linear interpolation of the tabulated values, but if, for any reason, it is felt desirable to compute the interference zone precisely, the graphs which are needed to do this are included herewith. The method is such than an accuracy of better than 0.2 km is achievable, particularly if a light table is used in conjunction with transparencies made from the graphs. The data which are presented are for the standardized heights hi = 300 m for the transmitting antenna, h2 = 10 m for the receiving antenna, and h3 = 30, 45 and 60 m, corresponding to the tower heights of the MOD-OA, MOD-1 and MOD-2 machines, respectively. For heights

-24 - which differ substantially from these it is necessary to generate a new set of graphs before proceeding with the interference calculations, and we comment that the difference between h2 and h3 has a significant effect on the level of interference observed. If, on the other hand, the blade size and/or materials alone differ from the ones discussed here, the graphs are still applicable, and it is only necessary to change the equivalent blade scattering area Ae. For blades of similar shape to those on the MOD-OA and MOD-1 machines the scattering efficiency is approximately 0.65 and 0.25 for metallic and non-metallic blades, respectively, and these values can be used to deduce the equivalent area from the projected area. The interference zone and the distances discussed here are for the worst possible situation which generally occurs when the turbine blades are so positioned in pitch as well as azimuth to direct the maximum scattered signal to the receiving antenna. Depending on the prevailing winds and the quality of the receiving antenna, some or all parts of the interference zone could suffer interference only for a fraction of the total time. Finally, we remark that the present computer method based on radio wave propagation over a smooth spherical earth requires modification to account for terrain effects.

Appendix A: Computer-Graphical Method for Calculation of the Interference Zc A.1 Theoretical Considerations. The calculation of the interference zones of a WT is based on the theoretical expressions for the fields derived in [2] and differs somewhat from the procedure used in [1]. We consider a TV transmitter (T) at a height hi (in m) above the earth radiating a horizontally polarized signal with effective power P (in kW). The signal impinges on a WT whose tower height is h3 (in m) located a distance d (in km) from the transmitter, and is also picked up by a receiving antenna (R) at a height h2 (in m) above the earth and distance r (in km) from the WT. The receiving antenna is assumed omnidirectional and in the absence of the WT the direct field at R would constitute the only signal. The rotating blade(s) of the WT scatter some of the energy incident upon them and are the source of a secondary signal which is also picked up by the receiving antenna. The ratio of the secondary and primary signals at R is the amplitude modulation index and, hence, a measure of the interference produced by the WT. The above concepts can be expressed in mathematical form as follows. The amplitude modulation index of the total field at R is m = EB (R)/ET(R) (A.1) T B where E (R) is the amplitude of the primary field at R and E (R) is the amplitude of the secondary field which reaches the receiver after

-26 - reflection off a rotating blade (B) of the WT. The latter field is itself the product of the direct field of amplitude ET(B) illuminating the WT and the field EBT(R) at the receiver of a transmitter of effective strength N located at the blade. We can therefore write EB(R) = ET(B) N EBT(R) (A.2) giving m = N E(B E(R). (A.3) E (R) In practice, the height h3 at which the blade scattering occurs is usually different from the height h2 of the receiving antenna, and the ratio ET(B)/ET(R) will then differ from unity for this reason as well as because of the distance r of the receiving antenna from the WT. It is convenient to separate these two effects by writing ET(B) = ET(B') ET(B) (A.4) E (B') where B' is an 'equivalent' blade located at the WT, but at a height h2 equal to that of the receiving antenna. On inserting this into (A.4), we have m N E (B ET(B') EBT(R). (A.5) ET(B ) E (R)

-27 - The parameter N is determined by the scattering characteristics of the blades. Laboratory and full scale tests and analyses indicate [2,3] that the blade scattering is predominantly in the specular (mirror reflection) and forward (blocking) directions, and is usually directed towards the receiving antenna when the blade is horizontal. We can therefore take the height h3 of the phase (scattering) center of the blade to be the same as that of the WT tower. It is also found that for machines such as the MOD-OA and MOD-1 having two fully controllable blades, each blade contributes individually, whereas with the MOD-2 just a portion of the single blade is responsible for the largest scattered signal observed. With both types of machine, the rotation of the blades generates a periodic impulsive form of modulation. For blades of large electrical size, the scattering can be estimated using the physical optics approximation, and the calculation can be further simplified by treating each blade as a rectangular metal plate whose equivalent (scattering) area is Ae [3]. Figure 1 shows the locations of the distant TV transmitter, the receiver and the scattering center of a blade, all assumed to lie in a horizontal plane, i.e., the plane of the paper. It is also assumed that the blade rotates in a vertical plane through M-M (see Fig. 1), and that this plane is so oriented as to direct the specularly reflected field to the receiver. The parameter N characterizing a blade as a re-radiator is then N = N n(p) 0 (A.6)

-28 - where Aef (A.7) NO = 0 4.5 x 104 /P 4.5x cos for 0 1 <H Tr - /Le n(f) =, (A.8).sinc (e sin () for f - X/L < < v_ f is the frequency (in MHz), Le is the effective length (in m) of that portion of the blade responsible for the scattering, and sinc x = sin x 7T X Note that A cos (/2 is the projection of A perpendicular to the direction of incidence. If the plane of the blade rotation is always chosen so as to direct the maximum reflected or forward scattered field to the receiver, c also specifies the angle which the line RB makes with TB and, hence, the angular position of the receiver. N then achieves its maximum value No when c = 0 and r, corresponding respectively to backscattering off the blades (R in line between T and B) and forward scattering (R on the extension of TB). We now consider the factors other than N on the right-hand side of (A.5). The ratio ET(B)/ET(B') is obviously independent of ( and r, and measures the change in primary field strength due to the differing heights of the blade (i.e., WT tower) and the receiving antenna. It

-29 - is therefore a height gain factor which is rather easily computed for any f, d, hl, h2 and h3. The next factor ET(B')/ET(R) is weakly dependent on r. It is unity when r = 0, i.e., when the receiver is at the WT tower, and remains unity if the receiver is moved away from the WT along the circumference of a circle centered on the transmitter, i.e., for (: 90~. For a fixed r the ratio is an increasing function of p, being less than unity for 0 <- < 90~ and greater than unity for > 90~, but if r is small the j dependence is relatively slight. Finally, there is the factor EBT(R). This is a rapidly decreasing function of r independent of p, and is the main source of the r dependence of the right-hand side of (A.5). As r increases, the right-hand side of (A.5) decreases, and a procedure for determining the interference zone is now apparent. For any given p, r is increased until the threshold modulation index mo (see Fig. 2) is attained. The resulting r then specifies the radius of the interference zone in that direction. If the value of r so obtained is fairly small, e.g., < 1 km, most of the c dependence is provided by the factor N, and the slight dependence that ET(B')/ET(R) provides is, in part, cancelled by the variation of the threshold modulation index mo. In this case the zone has the shape indicated by n( ), i.e., is a cardioid plus a forward spike, and a knowledge of r in just two directions, one in the forward part of the zone (i.e., ( = 180~) and one in the backward part, e.g., g = 0 or 90~, is sufficient to establish the entire zone. Unfortunately, this is not true if r is larger.

-30 - A.2 Computational Procedures. A computer program is available [2,8] for calculating ground wave propagation over a smooth homogeneous spherical earth. At any frequency f (in MHz) the program computes the field of a horizontally polarized transmitting antenna height h, (in m) above the earth as a function of the distance d (in km) to a horizontally polarized receiving antenna of height h2 (in m). The field strength obtained is typically as shown in Fig. 4 for h, = 300 m, h2 = 10 m and f = 500 MHz. The distance d to the radio horizon is d = 3.57 (h"12 + h1/2) (A.9) 1 2 and the horizon is marked in Fig. 4. Well within the horizon distance the field strength oscillates as a result of interference between the free space and ground-reflected waves and here a change in d of less than 1 km can produce a field strength variation of more than 30 dB. Since our concern is to calculate the maximum distance from a WT at which a given level of video distortion can occur, it is sufficient to ignore the oscillations and confine attention to the peak values of the field. As is evident from equation (A.3) of [8] with P = 1, the values are such that E 0.3d mV/m (A.10) independent of h, h2 and f, though the locations of the peaks do, of course, depend on all three parameters. From (A.10) it follows that throughout the interference region

-31 - 20 log E = -(10.46 + 20 log d), (A.11) and the corresponding straight line is indicated in Fig. 4. The field strength program is the basis of our method for computing the interference zones about a WT. It is assumed that the equivalent (blade) scattering area Ae and the tower height h3 are known for the WT in question, and all calculations are carried out for a transmitting antenna of height hi = 300 m and a receiving antenna of height h2 = 10 m, typical of the values encountered in practice. Since the effective radiated power P plays no role in our analysis, we henceforth set P = 1. The following computations are performed: (i) field strength as a function of distance r (in km) for a transmitter of height h3 and a receiver of height h2 = 10 m at 7 frequencies spanning 50 < f < 900 MHz, from which plots of 20 log EBT (R) vs. r are constructed; (ii) field strength of a transmitter of height hi = 300 m at a receiver of height h3 at a distance d = 20(20)100 km away at the same seven frequencies; (iii) field strength of a transmitter of height hi = 300 m at a receiver of height h2 = 10 m distance d ~ r away for the s-ame values of d and the same frequencies as above, with incremented values of r up to (say) 15 km. In all cases the actual field strengths in the interference region are replaced by those deduced from (A.11).

-32 - The second set of data in conjunction with the third for r = 0 gives ET(B)/ET(B') at seven frequencies and five WT-to-transmitter distances d. The ratio is just the height gain factor associated with the different receiver and WT tower heights. For the values of h3 of concern to us it is found that if d > 100 km the ratio is virtually independent of d and can, in fact, be obtained from an analytical expression for the height gain factor appropriate to the deep shadow region [10]. When normalized to the values for r = 0, the third set of data then yields plots of 20 log ET(R)/E (B') vs. r, representing the dB attenuation as a function of the distance r of the receiver from the WT. For d > 100 km the attenuation is also independent of d to a first approximation and can be determined analytically [10,11]. A graphical method for calculating the interference zone about a WT is as follows. From (A.5) 20 log EBT(R) = 20 log E(R) + y (A.12) E'(B') where y = Y1 + Y2 (A.13) with m (0) BT o E (B') Yl = 20 log N E (A.14) N0 ET(B) and m (q) Y2 = 20 m ()n() (A.15)

- jJ Knowing Ae for the WT blade, No can be computed (see (A.7) with P = 1) at each of the chosen frequencies. From (ii) and (iii) above with mo(0) = 0.15 (see Fig. 2), a table is then compiled showing Yi for all combinations of the five ranges and s.even frequencies as shown in Table 3. Also, from Fig. 2 and the equation (A.8) for n(p), Y2 can be computed for any X, 0 <_ < 180~. The results are independent of d and (for all practical purposes) f and the particular WT, and are plotted in Fig. 5. To compute the interference zone of the WT for some specific TV transmission, the gist of the method is as follows. Knowing the actual frequency (see Table 2 to convert TV Channel number into frequency) and the WT-transmitter distance, we use the data for the closest frequency f (in MHz) and distance d (in km). By appropriately overlaying graphs of 20 log EBT (R) and 20 log ET(R)/ET(B') vs. r, intercepts of corresponding curves are found, and these give the zone radius r(f) for any desired azimuthal location ) of the receiver with respect to the WT-transmitter direction. The procedure is illustrated in the next section and if greater accuracy is needed, results can be obtained for frequencies and distances on both sides of the actual ones, and linear interpolation employed. In computing the interference zone the effect of using a directional antenna can be taken into account as discussed in Section 6. In the expression (A.15) for Y2, divide the function m (p) by f(p') 0 where ~' is now the angle between the directions to the transmitter and the WT as seen from R (approximately 180 - e for d >> r). The

-34 - results of this operation are illustrated in [4] where the interference by the MOD-OA machine on Block Island is examined. A.3 Specific Computations. Our main concern is with the MOD-OA, MOD-1 and MOD-2 machines whose tower heights h3 will be taken as 30, 45 and 60 mn respectively. An important parameter affecting the interference is the equivalent scattering area Ae of a blade. This is related to the projected geometrical area A via the scattering p efficiency n such that Ae = nAp, and the projected areas of the MOD-OA and MOD-1 blades are 18 and 62.5 m2 respectively. From laboratory model and full scale scattering tests [3] the scattering efficiencies of these metal blades were found to be 0.67 and 0.63 respectively, implying A = 12 and 40 m2 for the two machines. For a MOD-OA blade of fiberglass construction with a minimal configuration (stage #2 of [3]) of lightning arrestor strips, the corresponding scattering efficiency is 0.27, so that A = 5 m2 approximately. These values of n and Ae are also applicable to the corresponding wooden blade. In the case of the MOD-2 machine, measurements of the scattering from a small scale model blade show that the largest scattering is provided by a central section of the blade of approximate length 63 m, and the equivalent scattering area implied by this is A = 140 m2. The relevant parameters for the three machines with metal blades are then as follows:

-35 - h3(m) Ae(m2) Le(m) MOD-OA 30 12 15 MOD-1 45 40 25 MOD-2 60 140 63 Tables 3(a) through (c) on page 38 list Yl (see (A.14)) for these machines for f = 100, 200, 300(200)900 MHz and d = 20(20)100 km. The following graphs are needed to compute the interference zones, and are presented in the figures cited. 20 log EB(R) vs r for the seven frequencies and the three tower heights listed above: Figs. 6(a) through (c). 20 log ET(R)/ET(B') vs. r for the seven frequencies and the five WT-transmitter distances d = 20(20)100 km: Figs. 7(a) through (e). For a receiver closer to the transmitter than the WT, i.e., 0 < ~ < 90~, the appropriate curves are shown as solid lines, whereas for 90~ < ) < 180~ the lines are broken. To illustrate the procedure, consider the computation of the interference zone for a MOD-1 WT on TV Channel 56 whose transmitter is 75 km away. From Table 2 the closest frequency for which curves are provided is 700 MHz, and the closest value of d is 80 km, implying Fig. 7(d). From Table 3(b) the appropriate dB level y1 is then -28.1.

-36 - 1. ( = O. Since Y2 = 0, y = Yl = -28.1. Overlap Figs. 6(b) and 7(d) positioning the -28.1 level of Fig. 6(b) at the 0 level of Fig. 7(d). Read off the value of r at which the 700 MHz curves cross, using the solid lines in Fig. 7(d). The result is r(O) = 4.6 km. 2. ( = 180~. Figure 5 gives Y2 = 7.4. Hence, y = Yl + Y2 = -28.1 + 7.4 = -20.7. Overlay Figs. 6(b) and 7(d), positioning the -20.7 level of Fig. 6(b) at the 0 level of Fig. 7(d). Read off the value of r at which the 700 MHz curves cross, using the dashed lines in Fig. 7(d). The result is r(180) = 5.5 km. 3. ( = 90~. Figure 5 gives Y2 = 5.5. Hence, y = Yl + Y2 = -28.1 + 5.5 = -22.6. Using Fig. 6(b) alone, read off the value of r at which the 700 MHz curve crosses the -22.6 level. The result is r(90) = 4.0 km. Under most circumstances the approximate shape of the interference zone is known, and the entire zone is then determined by the values of r(O), r(180) and, perhaps, r(90). However, if greater accuracy is required, r(() can be computed for any desired ( using the above-mentioned graphs. The approach is based on the fact that for ( f 0 the distance of the receiver from the transmitter is approximately d-rcos ( and not d-r. The manner in which this affects the computation of r(() for ( f 0, 90 or 180~ is illustrated by the following example.

4. q = 60~. Figure 5 gives Y2 = 2.6. Hence, y = Yl + Y2 = -28.1 + 2.6 = -25.5. Also, cos 4 = 0.5. Overlay Figs. 5(b) and 6(d) such that the -25.5 level of Fig. 6(b) is at the 0 level of Fig. 7(d) with the r = 1 of Fig. 6(b) at the r = Icos fj = 0.5 position of Fig. 7(d). Read off from Fig. 6(b) the value of r at which the 700 MHz curves cross, using the solid lines (dashed if 4 > 90~) in Fig. 7(d). The result is r(60) = 4.5 km. Three typical interference zones are shown in Figs. 8(a) (c), all computed for the MOD-1 machine with a transmitter at a d > 100 km. The first two are adequately approximated by through distance (A) r(O) cos qb/2 r(l) = r(180) sinc (-e sin f) for 0 < q < 180(1 for 180(1 - -L ) e - ) < < 80 <.. < 180~ (16) and the zone then consists of a cardioid and a forward spike as suggested by the behavior of n(f). For the third zone, however, a better approximation is (B) r(90) r(() = r(90) sin 4, L r(180) sinc (-e- sin 4) for 0 < 4 < 90~ for 90 < 4 < 180(1 - TL ) e for 180(1 - x ) < q < 180~ e (17)

-38 - Both shapes are completely specified by a knowledge of r(O), r(90) and r(180), and the ratio r(90)/r(O) can serve as the basis for the choice of shape. The ratio is 0.71 for type A and 1 for type B, and in those rare cases where it substantially exceeds unity, neither approximation is good. A precise determination of the interference zone is then desirable. The computed interference zone radii r(O), r(90) and r(180) for the MOD-OA, MOD-1 and MOD-2 machines with metal blades are shown in Table l(a) through (c) for f = 50, 100, 200, 300(200)900 MHz with d = 20(20)100 km. Except possibly at the highest frequencies for the larger d the zones are of type A, but in the case of the MOD-2 machine there are a few instances, e.g., f = 900 MHz with d > 60 km where even the type B shape does not provide an accurate approximation.

-39 - Table 3: dB level y1 (see equation A.14) (a) MOD-OA (Ae = 12 m2) (b) (c) d(km) f(MHz) 20 40 60 80 >100 50 11.7 11.5 11.5 11.3 11.5 100 6.4 5.5 5.4 5.4 5.3 200 4.0 -0.3 -1.0 -1.0 -0.9 300 3.4 -3.6 -4.5 -4.8 -4.6 500 1.3 -6.2 -8.3 -9.6 -9.5 700 -1.5 -5.9 -11.7 -12.9 -12.8 900 -3.7 -6.1 -13.7 -15.5 -15.3 10D-1 (Ae = 40m2) d(km) f(MHz) 20 40 60 80 >100 50 -2.0 -2.4 -2.5 -2.7 -2.6 100 -6.2 -8.3 -8.7 -8.7 -8.8 200 -6.5 -13.8 -15.2 -15.3 -15.3 300 -7.1 -16.7 -18.6 -19.2 -19.2 500 -9.2 -16.2 -22.8 -24.4 -24.4 700 -11.9 -16.4 -25.2 -28.1 -28.1 900 -14.1 -16.6 -26.5 -31.0 -30.9 MOD-2 (Ae = 140 m2) d(km) f(MHz) 20 40 60 80 >100 50 -14.9 -15.8 -16.0 -16.2 -16.1 100 -17.3 -21.5 -22.1 -22.3 -22.4 200 -17.4 -26.5 -28.7 -29.1 -29.1 300 -18.0 -28.2 -32.0 -33.2 -33.2 500 -20.0 -27.5 -35.7 -38.7 -38.9 700 -22.8 -27.2 -37.1 -42.6 -42.8 900 -25.0 -27.4 -36.7 -45.7 -45.9

-i!i iT 1W!,i I i! I IIll I I I III 'II iI, I 2 10 102 103 d (km) Figure 4. Field strength in dB above mv/m as a function of distance d from a horizontally polarized transmitter radiating 1 kw effective Dower: h. = 300 m. h, = 10 m. f = 500 MHz.

20 15 Yz... 5 - *1 0 0 --- 60.. 120 # (DEGREES) 180 Figure 5: dB level Y2 versus q (see equation 15).

<n en PI X t 0 W ~C~ UI gr N 0 (O (j IIi/ //II Illi I ii I I 1111-1- 1' rii i 1 i I i i I I 1 i r i I t j///t/j It i / II j/j jii i I iiiIlii ii i' I iiri it ellnI i i i 1 i i 11 i r i i ill I[lllllllliill![ i i I ii I iii iiiilliI trniI r i ill Ill-~-t--l -II-1 I-i I itf-tlfftlll lfiflllil I IIIll tllll Illllliliflil;fli 11 iiII I i 1iii I!iai I ii 11111. 1 I I i r:1:: r ii;ll I t(iII; ii! i; ii I I 1 iliiM-1-ll-l IPEfiMfiJ`EM~iiiill H illllllilllllliiliililiiillrl —t-ll-i-l C-ll —tlll illlllilllllliill illiifiljil.t.l::-FW::t:f`FT`U:I III'N1ITTCIIPPM iifJII II11[IIi iiiillii! iiii ii I-I ill ilKj 1 1 i ill ilii liilllliilliiiilliiiIliii:::. illii i I I rl - 1 i 1 Iii ii rirrul I iiII/I1)_II /I//j j/lII:II!I I i iiii Ilr i II- iili I i ' i i fI I 1 ililii!III ii iii!I ii iiti 1' i 1 i 1 I / i,, i; I i i i ir i 1,1 1111 it-t Iii i ii I it.1~ 1 i I r i I 111Ili -1-)- 1..i:i I([ i i: I I i il!r1 r ~r-t T~I-i I t;III i IiiI (ilillrI "' i I i I i i I I ii I 11 I I I I1 i: ) j I iii IIil.-l.-l-l.. tllr;llI 11!I ij ji;1- f jl. i (/ t t - i i 1; I! 0.2 1 10 100o r (km) Figure 6(a). Field strength [EBT(R)] in dB above mv/m as a function of distance r from a transmitter located at the WT blade phase center and radiating 1 Kw effective power: h2 = 10 m, h3 = 30 m (MOD-OA)

I_- Ifi r N W N A RaPrC ix>lii tulp I I I [ j K L fij1 i I I t LI-t 'K 1 I1 1' K.1 I1ijx1i i > I ~~*it'I, i 1 ( 4I I 0.2 1 10 100 r (km) Field strength [EBT(R) in dB above mv/m as a function of distance Field strength [E (R)] in dB above mv/m as a function of distance Figure 6(b). r from a transmitter located at the WT blade phase center and radiating 1 Kw effective power: h2 = 10 m, h3 = 45 m (MOD-1)

i [ I I I I I I I I I r rrfl i, I i Jl F 00 a fff l - I 1 1 1 1 S ' l I 0.2 1 10 100 r (km) -Figure 6 (c). Field strength [EBT(R)] in dB above mv/m as a function of distance r from located at the WT blade phase center and radiating 1 Kw effective power: h3 = 60 m (MOD-2) a transmitter h2 = 10 m,

ItI L J- II 1,, It_11II 1 1 _ i *I I' I I I I Lt 1 4:'Ul I 0.2 1.0 10 100 r (km) Figure 7(a). Ratio of the direct field strengths at the receiver (R) and WT blade Phase center (B) LET(R)/ET (B)] in dB as a function of the distance r from the VIT: = 0; --— ~=a d =20 km.

-46 - n Lv ----—.. --- —. - -- -- ': I - -,-+ v V7T7t1 7 - 5 -5 -4 -3 -2.... 1:L I 7t I -- -, -- i - -. -. -, L. — -- -4 i --- -- r — — I. -- . —I - --- - -... --.. I +-----: -- - -- I I r - - I - z 1 —1 --- -- 1 in- z w - z4 I I. i - - r_:-_ I. I - I - -- — I I I I. - I I - I -,, I I I, I I - I k q i i I- I I lfpl k i k i i 1 I i: i, i,!.! 1 - - - 1 i - Alo t I v -k, - - - -1 . - - I. - - L.. -- I —.w7- - I I - i - - -—, -. -. i i I,. i I I Id ,.. i ! ",; k i I. M P i i % i I. Ift. i. I! i I I::, 1 1, (/. V-1 — , -, - -, - — -1I + -,, I I,; i,:, I.. A r ///^! X.;.; I i I I i j _,- I., -o4-4-N-X%,-,'*- 4 ' ' -' -4- I 1< f -- - "I i IBM! I 1\1 i I I, I;, I7 -6.... -t.V- N t — —. - -..= -4 77:- - t= - - - -= — -;"Z;; --- I I I i zIozz~z-c 4 -3 -... L - -- E)- 4a - -q I - I. — - --- 7 --- - cii ci) U ci) (I) ci) -o -Q F-o (ci o .. ci) ci) C..) ci) S..ci) b.. (ci C..,) 0) ci) 5 -U) 0 0 ci) ~-* -F4 -4-, U ci) S..-o I. ci) 4-, 4 -0I 0 cJ 4-) ci 0 1 CI -ci. t:-. -J -— I Fci) 4-) E 0 S5 -4 -S.(co 4-P (I, 4-P 4 -0 0 4..) 4 -(co (A co co C: Co-I) Li LJ "I 4 i i AI, - -1 -! -.! I a. i 1 j! j I; I T., -i I 9 -8.. 7 -6.. 4 - I I -0 I 4 - J (f - - -r- - - - I I --- - - - (I)i -,- 1 ';4, I! I %4 -1 - - t - I - -- Mr -- 1 - LO 4 1,; Tr-, i I - o - I - (DA., i - 04- I'MI - Ii - -::z I k-. - try, I 11,,- - n i F- - I 1 4 - 7 — 1 T I - 1, I.i. -I.. - - - - 1, i 77tt t -t. -.:. I I 17 I t I I I I -I - I i - 1, r 1. 0 1 r"%i " - -- - I i,, Li --, I -- I -n I-TI 1 - L-" I. I I - I I I -o S.. CT) L..

- s> in ~. u< o> r4 o k -a 4J Lj -fk Ln u Q I-II 4 1 I I 11t IIII i IM P II 1i 1 I I I [al l 111111i iif Hill -F J -I r11 I I -i m -- - -.. -. ~ - _ - - - --— ~- -1:. -l.... -...^ ^ I, _^ ^ _. _-_...-... 1. V ^ - ~ - j... ^ ^~ __.; 1........];..... 11. ^: __.i.......... tk Al iii i li i, 0.2 1.0 10 100 r(km) Figure 7(c). Ratio of the direct field strengths at the receiver (R) and WT blade phase center (B) [ET(R)/ET(B)] in dB as a function of the distance r from the WT: - = 0; ---- = T. d = 60 km.

F j W iI H. II II I 0.2 1.0 10 100 r (kin) Figure 7(d). Ratio of the direct field strengths at the receiver (R) and WT blade phase center (B) [ET(R)/ET(B) in dB as a function of the distance r from the WAT: = 0; ---— p= Tr. d =80 km.

L I -I I I r r I i I I I rrOl I I I 1-11 m I I f 1"o f la II:t t I IIhIIW 0.2 1.0 10 100 r (km) Figure 7(e). Ratio of the direct field strengths at the receiver (R) and WT blade phase center B [ET(R)/ET(B)] in dB as a function of the distance r from the WT: - = 0;... —. = wr. d > 100 km.

0-3 IF 00 0*2 r (0) (kin) 0*1 0 ( a) 0 l, X% 1% I I'l 0 - - I I I I 3 (k m) 0 I0 r(0 r(kin) 0f 0 01 %%1 0 (b) ' 1 *%i I I I I I I I 0 0 0 ( C) NI o% I 01L C --- - I I "A 30 60 90 cfp (DEGREES) ia0 150 180 Figure 8. Computed interference zones (a9 o o ) for the MOD-1 with d > 100 km and (a) f = 50 MHz, (b) 'f= 500 MHz, (c) f = 900 MHz; - -- type A zone-. - - type B zone.

Appendix 2. Glossary of Symbols T,B,R Denote the TV transmitter, WT blade and TV receiver, respectively. a Effective radius of the earth = 6.34 x 103 km. A Equivalent scattering area (in m2) of a WT e blade. A Projected (geometrical) area (in m2) of a P WT blade. d Great circle distance (in km) between TV transmitter and the WT. E Maximum value of the ambient field at R, located within the radio horizon of T. EB(R) Amplitude of the scattered (or secondary) field. at R. BT E (R) Amplitude of the field at R of a transmitter at B. ET(B) Amplitude of the ambient field at B of the transmitter at T. ET(R) Amplitude of the ambient field at R of the transmitter at T. f TV Channel carrier frequency (in MHz). (Table 2). f(j') Horizontal plane voltage pattern of the receiving antenna. h, Height (in m) above earth of the phase center of the TV transmitting antenna. h2 Height (in m) above earth of the phase center of the receiving antenna. h3 Height (in m) above earth of the phase (scattering) center of the WT blade (can be taken as the TW tower height). Le Equivalent length (in m) of a WT blade. m Amplitude modulation index: Interfering signal amplitude/desired signal amplitude. -51 -

-52 - m Threshold value of the modulation index. o Mn (f) Threshold modulation as a function of p. N Effective strength of the fictitious transmitter located at the phase center of a WT blade. N Maximum value of N. o n(f) p-variation of the blade scattering. P Effective radiated power (in kW) of the TV transmi tter. r1 Interference distance in the backward direction. r2 Interference distance in the forward direction. r(f) Interference distance in the f-direction. ~r Dielectric constant of the earth. n (Ae/A p): scattering efficiency of a WT blade. X (300/f): wavelength (in m). Angle between lines BR and BT. q)' Angle between lines RT and RB.

-53 - References [1] T.B.A. Senior and D. L. Sengupta, "Wind Turbine Generator Siting and TV Reception Handbook," Technical Report on Contract No. EY-76-S-02-2846.AOO1, Wind Systems Branch, Department of Energy, Washington, D.C., 20545, January 1978 (COO/2846-1). [2] T.B.A. Senior, D. L. Sengupta and J. E. Ferris, "TV and FM Interference by Windmills," Final Report on Contract No. E-(11-1)-2846, Energy Research and Development Administration, Washington, D.C., 20001, February 1977 (DOE/TIC-11348). [3] D. L. Sengupta and T.B.A. Senior, "Electromagnetic Interference by Wind Turbine Generators," Final Report on Contract No. EY-76-S-02 -2846.A002, Wind Systems Branch, Department of Energy, Washington, D.C., 20545, March 1978 (TID-28828). [4] D. L. Sengupta and T.B.A. Senior, "Wind Turbine Generated Interference to Electromagnetic Systems," Final Report on Contract No. EY-76-S-07-2846.A003, Wind Systems Branch, Department of Energy, Washington, D.C., 20545, August 1979. [5] D. L. Sengupta, T.B.A. Senior and J. E. Ferris, "Television Interference Tests on Block Island, RI," Technical Report on Contract No. EY-76-S-02-2846.A004, Wind Systems Branch, Department of Energy, Washington, D.C. 20545, January 1980. [6] D. L. Sengupta, T.B.A. Senior and J. E. Ferris, "Measurements of Interference to Television Reception Caused by the MOD-1 WT at Boone, NC," Technical Report on Contract No. DE-SERI-XH-0-9263-1, Solar Energy Research Institute, Golden, CO, January 1981.

-54 - [7] D. L. Sengupta and T.B.A. Senior, "Electromagnetic Interference to Television Reception Caused by Horizontal Axis Windmills," Proc. IEEE, Vol. 67, No. 8, pp. 1133-1142, August 1979. [8] L. A. Berry, "Fortran Program for Calculating Ground Wave Propagation over Spherical Earth," National Bureau of Standards, Boulder, CO, 1968. [9] Spectrum Utilization Division, "RAPIT (Radio Propagation over Irregular Terrain)," National Telecommunication and Information Administration, Institute of Telecommunication Sciences, Boulder, CO, 80303. [10] P. A. Azriliant and N. G. Belkina, "Numerical Results of the Theory of Diffraction of Radio Waves around the Earth's Surface," Soviet Radio Press, Moscow, 1957. [11] N. A. Logan, "Numerical Investigations of Electromagnetic Scattering and Diffraction by Convex Objects," Lockheed Missiles and Space Company Report AFCRL-66-153, 1965.