388629-1 -F ASSESSMENT OF ELECTROMAGNETIC INTERFERENCE EFFECTS OF THE ELLENVILLE WINDFARM Final Report Dipak L. Sengupta and Joseph E. Ferris Radiation Laboratory Department of Electrical and Computer Engineering The University of Michigan Ann Arbor, Michigan 48109 Prepared for Genro Energy Systems Eastern Office 230 Delaware Avenue Delmar, New York 12054 November 1983 Work performed under Contract No. N-2 (Dated: 9-27-83) 388629-1-F = RL-2547

EXECUTIVE SUMMARY The potential interference effects of 71 wind turbines (WTs) of the proposed Ellenville Windfarm on the performance of various electromagnetic systems operating in its vicinity have been assessed theoretically. Specific non-military systems considered are: three VOR (Very High-Frequency Omni Range) systems within 35 miles of the windfarm; four radio compasses within about 30 miles of the windfarm; seven microwave links: two earth stations (ES) receiving signals from geo-stationary satellites; one cable TV (CATV) Head-end receiving the desired TV signals and available commerical TV Channels operating at Cragsmore. In addition to these systems there may be some radar, navigational and other microwave systems associated with the U.S. Military Services in the region. Since it is understood that the military outfits prefer to do their own assessment, these military systems are excluded from the present assessment. AM and FM broadcast reception outside the windfarm should not be affected significantly; within the windfarms, the reception within a few rotor diameters of the individual WTs may experience some unacceptable interference effects. These systems have also been excluded from the detailed assessment. The interference assessment has been carried out by assuming that the windfarm consists of vertical axis wind turbines (VAWTs), model F10 250, made by the Flo Wind Corporation. Windfarm interference effects to each of the systems named earlier have been assessed on the

basis of known criteria, and the assessment of such effects on specific systems are summarized below. (i) VOR Systems The VOR systems will not experience any unacceptable effects due to the windfarm. (ii) Radio Compasses The four radio compasses will not experience any significant interference produced by the windfarm. (iii) Microwave Links The performance of all of the microwave links except Links 6 and 8 will not experience any unacceptable effects due to the windfarm. Similar comments apply to the performance of the Links 6 and 8 provided that a few of the presently planned turbine sites are either modified (as recommended) or eliminated. (iv) Earth Stations The performance of the earth stationswill not experience any unacceptable effects due to the windfarm. (v) Reception of CATV Head-End The television interference (TVI) effects produced would be insignificant on all of the TV Channels of interest. (vi) TV Reception at Cragsmore Even under the assumption of non-directional receiving antennas, it appears that no unacceptable TVI effects would be produced by the wi ndfarm.

Acknowledgements The authors acknowledge with pleasure the assistance and suggestions provided by Dr. Rudolf A. Wiley of Genro Energy Systems during the collection of various data required for the assessment and preparation of the report. -i

1. Introduction The present report is concerned with an assessment of the potential effects of interference produced by the proposed Ellenville Windfarm on the performance of various electromagnetic systems operating in its vicinity. The assessment is carried out theoretically, and the specific systems considered are: (i) VHF Omnidirectional Range or VOR navigational systems and radio compass systems, (ii) microwave links, (iii) Earth Stations (ES) receiving signals from geostationary satellites, (iv) television (TV) reception, and (v) Cable TV (CATV) Head-end installations for receiving the desired TV signals. Undoubtedly, there are some AM- and FM-broadcast systems operating in the area. Reception of AM broadcast signals is usually vulnerable to various locally generated interference effects. The highest AM broadcast frequency being 1.6 MHz (x - 188 m), it is unlikely that the windfarm will produce any adverse effects unless the receiver is located within a few rotor diameters of a WT. The reception of FM broadcast signals would be even less vulnerable to such effects. For these reasons, these two systems have also been excluded from the present assessment. The interference effects of concern arise because of the time varying multipath created by a rotating wind turbine (WT) blade [1]. The primary signal is generally reflected in an almost specular (mirror-like) manner off a blade to produce a secondary (interfering) signal. The strength of the latter is proportional to the equivalent scattering area (Ae) of the blade and decreases with increasing distance from the turbine; at any given distance it also increases with increasing

-2 - frequency. If this secondary signal is sufficiently strong, it may combine with the primary signal at the receiver to produce unacceptable interference effects on the performance of the system under consideration. A key point is that because the reflection is specular, any given receiver will be affected only when the blade is suitably oriented. The nature and amount of the interference effects observed by the receiver depend on the nature of the electromagnetic system and its associated signal processing logic. It should be pointed out that the observed interference caused by the assembly of WTs in the windfarm will generally be statistical in nature [2] depending on a number of parameters. However, we shall use non-statistical analyses to estimate the effects produced by the WTs, either singly or together, on each of the electromagnetic systems mentioned earlier. Our assessment will thus pertain to the maximum effects that may occur in a given case under worst conditions. 2. Background Information Various information needed for the assessment are described in the present section. 2.1 Windfarm and Its Environment. The proposed windfarm (when fully established) will occupy a 1000 acre site approximately three miles southeast of Ellenville, NY and 30 miles west of Poughkeepsie, NY, as indicated on the road map section in Fig. 1. There is also a residential community (400 people) in Cragsmore located about three miles SE of the windfarm. The proposed windfarm site superimposed on the topographical map of the area is shown in Fig. 2 where the hexagonal sections indicate

-3 - * I I ' X1- V-. I 4 - I - Fig. 1: Road map of the Ellenville area, showing the general location of the windfarm, indicated by. (Scale: 1 inch = 10 miles).

a (0 Hi = 0 IC) o jo (D (D -I-J tj 2-V4-(~ I C iA I — A I,I, I --. 4::k I.1I I I

-5 - the regions where wind turbines (WTs) may be placed. The proposed site for the Phase I of the windfarm are the regions marked 6,7 and 11 in Fig. 2. As presently planned, during Phase I of the project 71 wind turbines will be deployed in a 300 acre area consisting of the regions indicated in Fig. 2; detailed deployment of the wind turbines for Phase I development is shown in Fig. 3 indicating the locations of 71 wind turbines. During the later phase of the project the number of WTs may be increased up to 170 deployed over a larger area. A more detailed version of the windfarm, showing the distribution of the presently planned 71 WTs, superimposed on a topographical map of the region is shown in Fig. 4 where again the hexagonal regions may contain WTs. For future reference we have indicated by CF the center of the Phase I windfarm of 71 WTs. The windfarm area is an unpopulated region with hills rising to about 2260 feet above sea level; the lowest elevation is about 2080 feet and the average is 2200 feet. It is understood that the residents of Ellenville receive TV signals through cable TV (CATV) service; only the residents of Cragsmore do not use the cable TV service. About 9.0 miles southwest from the CF and outside the farm there is a tower (Head-end) containing antennas which receive available TV signals for a CATV service. The location of the CATV antenna tower (or Head-end) is shown in Fig. 4. The points marked ES in Fig. 4 represent the locations of two satellite earth stations to be discussed later. The directional radials originating from CF in Fig. 4 refer to the directions and distances of New York, Albany, Poughkeepsie and other cities where the transmitters of TV signals originating from those cities are located.

Planned Deployment of Wind Turbines for Phase One Development (Parcels 6, 7 and 11) Fire Tower Road I Angular Range of Prevailing High Energy Winds I' I I I InterlinkinQ Roads / o 1452 Feet Fig. 3: Planned deployment of WTs for the Phase One development.

Ism- Al- Jany 4L ( 45 c~t (0 Path 6&8 WI, Me \ 'I V.. Cabo 4 \,ad -~ \ S.- Lo Path 2&3 A.".,.A. NAVN N' N, *~ 4~ A 7 ~:~) wI~ Walker / / NPhiladel phia, PA Newark, NJ S. ~>(115 ni 4 mi) New York City IL'~-~ (45 mi) N\CATV H " Fi g.4: Distribution of 71 WIT sin the proposed windfarm.

-8 - 2.2 TV Stations. It is believed that only a small community at Cragsmore (about three files from CF) directly receives the commercial TV signals available in the area; other communities in the region receive TV signals provided by a CATV organization whose Head-end is located at Wurtsboro. Table 2.1 lists the appropriate information about TV Channel signals received by the CATV system. The terrain in the windfarm area is hilly, and all of the TV Channels may not be available for direct reception at all places. Also, due to shadowing and other effects, the ambient signal levels on some of the Channels may be very weak. This may be the case at Cragsmore. 2.3 VOR Station. Throughout the country the Federal Aviation Administration (FAA) maintains VHF Omni Range (VOR) [3] ground stations which provide navigation information to aircraft in flight. From FAA maps of VOR ground stations in the area, three VOR stations have been identified within about 35 miles of the windfarm. The relevant information about the three VOR stations is given in Table 2.2. The three VOR systems listed in Table 2.2 operate at slightly different frequencies. For computational purposes we shall assume that the operating frequency of each VOR is f = 120 MHz, with wavelength x = 2.5 m. 2.4 Radio Compass. There exist four radio compasses in the region. The location of their transmitting antennas and other relevant information are given in Table 2.3. 2.5 Microwave Links. A number of microwave link paths used for point-to-point communication purposes criss-cross the windfarm area.

-9 - Table 2.1 TV Channel Signals Received by the CATV Head-end at Wurtsboro (Distance = 7.5 miles from CF, elevation = 1470 ft, antenna tower height = 250 ft) TV Channel Origin Degrees from Station Number Video Freq. (MHz) (city) North 2 55.25 New York 154~ 4 67.25 New York 154~ -- 5 77.25 New York 154~ 7 175.25 New York 154~ 9 187.25 New York 154~ WPIX 11 199.25 New York 154~ 12 205.25 Poughkeepsie 75~ WNET 13 211.25 New York 154~ WTAF 17 488.25 Philadelphia 185~ WBRE Scranton 257~ WDAN 22 519.25 Scranton 257~ WPWL 29 561.25 Philadelphia 185~ WNYE 31 573.25 New York 154~ WXTV -- Patterson 169~ WNJU -- Newark 158~ WNYL -- New York 1540

-1 0 - Table 2.2 VOR Ground Stations Near the Windfarm Designation Pawl ing Kingston Huguenot Frequency of Operation (MHz) 112.2 117.6 116.1 Direction from Windfarm East East Southwest Distance from the Center of the windfarm (miles) 34 22 27

-11 - Table 2.3 Radio Compasses Near the Windfarm Location Meier Neely Otims Frequency (MHz) 403 335 353 201 Distance from the center of the windfarm (miles) 17.21 14.25 16.64 31.46 Direction East South South West Monga

-12 -The points of origin or Head-ends of a number of such links are located near the windfarm (Fig. 4) and at a distance of about 1.15 miles from the center of the windfarm. Detailed technical information regarding the microwave links in the region was obtained from Spectrum Planning, Inc., of Richardson, TX, and is shown in Table 2.4 where the link paths are identified by numbers such as 6,8 etc. Using the data shown in Table 2.4 we have prepared a map indicating the microwave links in the windfarm region as shown in Fig. 5, where it can be seen that paths 6 and 8 are overhead and the rest have one Head-end in the windfarm area. From Fig. 5 it can be seen that the windfarm may have some impact only on the links 6 and 8. The WTs in the immediate vicinity of these two link paths may be identified in Fig. 4 and are identified in Fig. 3 as WT Nos. 62, 67, 66, 70, 69 and 71. As can be seen from Table 2.4 all links use slightly different frequency for reception and transmission and for convenience of calculation we shall assume that each link operates at a single average frequency for both. 2.6 Earth Stations. Two earth stations (ES) communicating with geo-stationary satellites are located in the vicinity of the windfarm and are shown as ES in Fig. 5. Each of the earth stations is equipped with large parabolic dish antennas (10 m and/or 15 m in diameter) which are commonly directed at the desired stationary satellite located above the equator. The present two earth station antennas would be generally directed towards the southerly direction, and their antenna beams would not cross the windfarm area. 2.7 Wind Turbines. It is understood that the wind turbine constituting the windfarm will belong to a class of vertical axis wind turbines (VAWTs) manufactured by the Flo Wind Corporation.

-13 - Table 2.4 Microwave Links in the Vicinity of the Windfarm Path 1: Common Carrier Microwave C"ALL SIG t 4 H~ STATiE & LOCATIiCN TEL1E V AT IO0N LUNG I lUDE AZIMUTH 4 DlSTh4C-E i;RnH,'3)MIT ANT TYWrrFCC, DtSlG & MANF'AC ANT G~fINN 4 i~ul'l RLL.LIVE (ANI TYP[ FCC JASXG & MA~NFAC ANI GAIN t HEIGWII i)IVRSITY Afff TYP'E FCC. DESIG 4MANIAC ANT GAIN HEIGHT EQUIPMEkNT MANUFoC MhCAUNIkLRS EMISSIUN DLSIGHT'YION F-CC L2$IG 6 '31ADY]T' XMI1 POWER h& LINK LOSS fPL-CV) S3IG.NAL I-EVFL TR0lFFIC IYPE I1)CG2:34 NYfIrL NY, WALDEN 5i6 FT 41-35-59 N 74- 7-48 W 2%.,3A DEG 13.09 MI USRiuP-3310? 6137100 CGAi:RIU[ 490. 9 v 14 60 FT K YNS? f YrL NY, ELLLNVILLE 2274 F T 41-41 - I N 74-P.1-24 Wi i1b.'6 DEC, 21.06 KM USR I OP-3J 1 07 G13'00 GA.R I L -V. I DBI 642J F T NIPPON ELECTRILC TRfIP I GD672 -i 0 i 40000 F~Y 2P T-t!0 11 &050 0 2-9.0 DE:M 0 DEl,.- 15, 1 D8Bi b72 DIGITAL 3, 0 071~5,0U 10373,UG 5.OE 1 0V95. OH li035.0E NIPPON ELECTRIC TNIG1 hA672-iOIA) /40 00 uF9Y 2-9. 0 Dfl-M 0 De~ -1j. I DItm 672 DIGITAL 11405.0B 1.1645.,PU 11325.0DC 1156b.0OB ii2'/5,00 i14145.Oi Path 2 CALL SIGN s6 OWNFIft S(,I 4 LOCATION "LEVAT 1tt LONGITUDE AZIMUTH 4 DISIANCE T'snN3"MIT ANT TYNI': FCC DESIG & MANFAC ~NT GAIN & HEIGihT RtICEIVL ANT lYPET FCC DESIG 4 MOtFflC ANI GAIN 4 HEIGHT IDIV(-RS1T[Y AN't TYfPE FCC DE?3IG 4 MAI4AC ANT GAIN 4 HEIG6HT EQLUIPM*ENT MANUFAC MANUJ' AC ILR ER S TYKf EMISSION DESIGNATION XMIT POWER 4LlNV LOC%-S;kECV 'sIGNAL LEVCL TZA~f IC TY~PE WC~GA!~5 #4YHE L NY) MoNlICELLO 15'3 FT 41-39-14 N 74-41-10 14 83.03'DEC., 17.17 MI USR 1 ~P-3J 1 07 61b9/00 (GOBRITL 49.5 DBI1 60 FT K YN59 NYTEL NY)I ELLEt4VILLE' 2274 FT 41-41 - I N 7)4-21-24 Iw 263, 2'. DEC6 C2'7.63 KI' UJSR IOP-3Ji U7 G'I37O GAiBRIfl. 47. 9 LP-I 62 FT NIPPON ELECTR<IC 'Tk P I IG672 -1 0 inA 40000F9Y 2?PT201.00500 2.9A0 DBM 0 DE:. - 15. 9 DB'M 6 72? DIGITo~L SOC0 1071~.O1? I0~S0 3.OU i07,95.OU i1035.OC1 NIPPOt" ElECTRIC TRP I IGD672-1 OJ1A 40000F9Y 2PT201 A 00AiO 29.0 DEBM 0 DP, 672 DIGITAL H145,OG 11645.0I 1 3(2-5,OG 1565, OU J vsiM5,O 0 i&5O

-14 - Path 3 GCiLL 1SIC14 6 OW4N.Lq ST0TF X- LOC-ATION L Al IT LIDE LUNI1 [UDE AZIMUTH & DISIANCE CRANSMIT ANT iYVE FCC Dt.SIG nACC A~NT GAIN 6t HEIGH~T RLLEUVE ANT 11YVE FCC, DESIG & MOPANFAC ANT GAIN & HEIGHT DIOER'SITY (iNT TYPE FCC. DESIG 6 MAINFAC AO('41HGI 6 HEIGHI EQUIrME( M(ANUFAC EMI',-,S1ON DESIGNWI IOH' F C C D LA''C;L,.3 ~TA fL I TY XMIT POWER 6 LINL LOSS RUW1V SiGN(AL LOYLL TRAFFIC TYP~E QRANS~Ml FRTL7S I O'9 5t i. uG23 MY TEL NY. h(Th1 I Ca 1.L C' 1543 FT 41-39-14 N 94-41 1.i0 w 83.03 DEG 17A7' MI IJSRIl2-P-3J I 0?/ G16900 GAE*1E K~yNS9 MYTEL NY, FLi[-NVILI F 22V 4 FT 41-41- 1 N4?41- 21-4 W 263.253 DEC?7 UsRiOP-33J107 Gi 3'OO GABR I 1 47,9 VL K' NIPPON FLFCTRIC I'kPI ILD6 72-1 Ol1A 40400H(FY 3L.0 DAMt 96 DIGI TAt. I [ 4 0-:II.1 f I t I d?1? F Y NIPPON [LEFCTRlC T;P1 I IGD6'/2 -I O it) 400 OOFYY 2'?,lT2I0 1.0 051 3L.0 DI3M -13,9 DEBM 96 DIGMITL JA9)107.V t S7 5 Dv 0?7i.0UH 0 o (00 1 124~5.ON Path 4 CALL- S IN CH ON STATEF 6 LOCATION. ELE~)nTI ON LATITUDE LONG ITUfDE AZIMUTH 6 DiSTAiNCE TktolMlf O~f TYY1'-U FCC: DLSIG & MANFAC Am:, ONI 6 HiELCIT RECTIVE ANT TYPt FCC', DI-SIG & AANFA,,C ANT GAIN & HE1tWI DkiVIRSITY AIN' TYPE FCC DESIG 6 MANFAhC ANT GAIN 6 HEIGHT EiQU I.P MINT M6NUF AC.?IANUF nCYfli~EP9S 1WYPLEHISIOON DEESICNATIUIN FCiC DF.~IlG 3ftiILITY XMIT POWER &LINE LOSS I-,ElC) `$GNAL 'LIVEL T1PeFVIC TYPE' TRAN'3MIT FREQS -11601 KYN59 NYT'LL NY, fILL[NVIt if 2274 FT 41-41- i N 74~-21 — 24 W 1i6.18 DEC, 13.09 Ml USRIOP-31 107 Gi3700 GAERIM[ 47,9 DBIt 62 FT -wcr.203' NYTEL NY, WALDEN 516 FT 41-35-59N ~96.34 DEG 2i.0f USRI OP-33107 G13700 GAPIR IF 47.9 DBI 61 NIOrON rl FCI(RIC TRPI I C1)6?2-1 0 I A 4000 iiF9Y' 31.0 DBtM -1i3. 1 DiM 96lb D I C ITAL ) FT NIPPON [LFORiIC. TiRP I1 GD672 -l 0 lA 4000 OF9Y 2P T201.60500 31,0 DBtM 0 DP...13. 1 PMi~ 96 DIGITAL J0 ) DE:, 10835.0OH

-15 - Path 5 CALL fi!U) OWHL 01RJ STe'TE: k LOCATION ELEvI)T ION L (f I'li ) E I-to 'UDE AZ)Mhu1 H d~, DISTANCE ii4iWLMUf ANT TYPIE FCC DESIC 4 MANVAC ANT uAfIN 4& HEIGH1T RECEIVti ANT TYPF' FiL'i DESI6 & hANFtAC ANT' GAiIN & fIERtT DlkYtf.1$ —ITY (IN'w TYPE FCC. DESIG t, MANFAC; Al1,T GA~IN h hEIGHT E-QL1 I PMENT MAMNUFAL1 MARU; nCKTUPCRS TYP4: EMISSION DESIGNA1TION FCC D~ESI1G STA;AiLf fY XMI'l POWL(R IJNL LOSS RL-CV SIGtNAL~ LEk1&L'TrR'FFIC TYPE TinNSMIT t:E~ 373 ME 6 3 hITM'I1.iA N4Yj JACKIF JONS 12~,6 FT 1 -1 3- 2 N 74- 4-11 t4 334.Q3 DCG 3bj,( KSV3 616 E 43 0 00 WE 397.4 Del '2 KEGbC- ATTHEA NY ELifNV!Ll F 227? 0 FT 41-41- i N 74-?21-24 W iuj,724 DlG 56.38 KM kS15676 F:43000 WE 3794 UBI 83 FT 03 Ml 1 7 Ff WLSTERN fI FCTRIC *f D-2 2 00 0F 9 2PYYOI. Ooo00 33.0 DE:M 0 DB: -27. 6 'BM 1200 CHANNEL MSG,0.0 3810.0 V 3'J9.),0 V 0, Ov 4050.0k 4i30 0V WE'STERN FL FCTR IC TD-2 2000 (F9 2PYYOI.00500 33.0 Dr:M 0 DU' -27.6 DBM i200 CHAtNNf I SL-t 370, OV 3850. 0V.i930. OV 400, Oki 40'0.OY 4i 790. Path 6 tIALL SIG1UN 6 wqt-"SThIC ~- LQLA)'ION E"LEV(- ION LUNU'i Y'UDE AZ I Ml LH & D I S Tht!C E TRAWJ"iMIT ANT 'f'fPL,' FCL: DESIC; MAF ANT GhIN H164GT RE( FIVE ANT TYPE FCC i,' ESIG & Mf4NFAC (ANI GAIN 6 f1ETCHT DIVEkITJY A4 TY'PE FCC DESIG MAtNAC; At~i GAIP &HEIGHT EQL:,iFtiEWI' MANtUFAiC h A HU f A C i 0R E RS, TYPIlE EMISSION DESICGNATION;1-T;C UCSIG t SffnfiILfTY XMIil POWER?2, LINFE L(OSS;crd:CV ~'~L~: Tk~t-'FIC IYPL: K EE8 f)rATTNEA NY, HnLIHAN HILL 524 F' 41-59-20 k 74- i- 8 W 219.72 'DEL, 27.3 M' S'15 676 B43000 WE 39.4 D8I 108 Ff KEEG6 T? TTNEA NY, E IFNVfltIf 2270 F t' 41-41 — 1 i?q 4-421 - 24 W 39,45w DEC. K S I 3506?76 I'i 3 0O 0 f W: 39., 4E'iI 44. (1 Kh WLSlER-N PI.C:TRIC T1D-2 200 00 FS,,3py0 DELi 0 N0 I15 D E) 1500 CH(WNNR ML-t 0 01 3 I/I?00k 3J7, ) OH 403C0k 'ij0AOH WVSTERN ELECTRIC VD-2 2',0 0 0 F 9 37i,0 DfIM 0 D U. -21 DB 150u CfAN1"Et M-SL:350, OH 383 0,O0i- 391 0.0OH

-16 - Path 8 GAl-L SIGN 6 Ot4JNQ STATE 6 LOCAHOI" ELEVAT ION LAI ITUDE LONGI TUDE AZIMUTH & DISTANCE TRAtqSMIT ANT TYPEL FCC EDESIG 6 MAKNfAC ANT G.AIN t NE1GIHT RECC'iVE ANT TYPE' FGC D~ES1i 6 MANFAC ANT GAIN 6 HEIGH1 DI(,!r.SJITY (IN't l'YPL FCC" DESIG 6 MANFAC ANT GAIN & HEICHT EQUIP eMENT MiNl~frAC. MANUFAC JURES EYPE. EMISSION DESIUGNATION F0CC' DEI-.3G 6 & i-Bi.T XMIT POWEI< 6 L1N L('S~c; RECV Slu-N(L LEVEL TRAt"FIC TYPE fkf*N~MlT FREQS 6?'26.9?H KEh47 NYTEL NY, HALIHA~N HILL 5 42&1 F T 41-59-22 N 74 - 1- 6 W?19.?2 1.LG p7.39 MI KS- b6?67 %-3i00 WE 43.0 bpui I (18 v K YN59 Y1IEIFL NY, EJ.LFNVIhLL 2294 FT 41 -41 - i N 74-Pl1-24 W 39.49 DEG~ 44.0OS KM KS - L~676 D:63i0 0 WE il3, (0 D~i $- 3 FT RA(YTHUCIPM XTh~ 3A 73000 OF9 2JYF 0 3 40.-0 DJPM -- 1 3.2- DV i VI DECI. 0'200 0 DD~ RAY'r HEUiN X(TR - 3 3 00 0 OiF? 2JYF 03 4 0. 0 DI:ti - 15. 2 D-Mt VI DEli 606.31OH. 0 (I;:. 0 0 ') DEi

Atp 7 w00 4.1 P at 2& Crags path 2&3n V-PNzNI tr /n.. Gln " N/AV N G N4 1;'2 1126 ok-c GN0 C. -. -,*** * * 3 K' ~.~I" 6", -- * Fig. 5: Microwave links in the vicinity of the windfarm.

-18 - Specific models considered are Flo 170, Flo 250 and Flo 500, and it is believed that the Phase I windfarm will consist of Flo 250 WTs. Relevant information about all three WTs needed for their electromagnetic interference assessment is given in Table 2.5. The blades are made of aluminum and the turbine rpm is 53. The most important parameter needed for the assessment of the electromagnetic interference caused by a wind turbine is the equivalent scattering area (Ae ) of its blade [4]. For the present WTs the appropriate Ae will be obtained by using the following [4]: Ae = w/3 (1) where w = blade width, D = rotor diameter and x = wavelength. For Flo 250, w = 0.61 m, D = 17.1 m and we obtain Ae = 2.52 /T m2 (2) We shall use Eq. (2) for the assessment of interference to all systems caused by the Flo 250 WTs. 3. Interference Assessment Procedure The interference assessment which has been carried out is analytical and, in the case of those systems which are impacted, quantitative. The procedures used are based on the analyses and

-19 - Table 2.5 Relevant Information about the Three Vertical Axis Wind Turbines WT Type Flo 170 (170 kW) Flo 250 (250 kW) Flo 500 (500 kW) Blade Width (m) 0.61 0.61 Rotor Dia. (m) 17.1 17.1 Rotor Ht. (m) 22.9 22.9 Ground Clearance (m) 5.1 5.1 Overall Ht. (m) 28 28 0.61? 25.0 37.5 5.1 42.6

-20 - techniques developed by the Radiation Laboratory during our previous studies of electromagnetic interference produced by WTs, the details of which may be found in [1,5-7]. In the present section we merely quote the basic criteria used to judge the acceptability (or unacceptability) of the interference effects produced in a given situation, and these same criteria are also used to judge the acceptability (or unacceptability) of a particular WT at a given site. The basic parameter that is used to judge the effect of WTproduced interference on an electromagnetic system is r amplitude of the interference signal caused by one WT amplitude of the desired (direct) signal (3) where the fields are computed at the receiver of the system under consideration. As mentioned in the Introduction, the interference signal is produced by scattering off the WT blade(s), and in general EB Ae r = E xd ' (4) where EBER are the amplitudes of the ambient electric fields at the WT and the receivers, respectively, X is the operating wavelength and d is the distance between the WT and the receiver. r also depends in a rather complicated manner on the ambient signal strengths at the WT and receiver locations, and on the receiving antenna characteristics [1,4]. In our previous studies we developed

-21 - approximate expressions for r under various situations. In the absence of specific information about EB and ER we shall make appropriate approximations in individual cases. Assuming that the interference effects produced by the individual machines are additive in power, the total effect produced by N WTs is then judged by the parameter rT: N 1/2 T > (rn) n=l where rn is that produced by the nth WT. In many cases we shall assume r =r..=N r, and use 1 2 rT = r. (5) In some cases only the machine(s) closest to the receiver cause most of the problem, but in other cases there can be many machines which contribute significantly to the total effect. The actual criteria (including the values of rT or r) which are used to judge the interference effects depend on the electromagnetic system under consideration, and are discussed in the following sections. 3.1 Interference to VOR and Radio Compass. In the vicinity of a VOR ground station the FAA prohibits [3] the existence of any tall scattering object which makes an angle of more than 1.50 (for metal objects) and 2.5~ (for wooden or non-metallic objects) at the phase center of the VOR antenna. It is also recommended that the amplitudes

-22 - of any reflected or scattered interfering signal relative to that of the desired signal at the receiver not exceed 20 percent. We shall use the following acceptability criterion for assessing the effect of interference on VOR performance: rT (or r) < 0.2 (or -14 dB). (6) In the absence of more detailed information about the radio compasses in the region, we shall use the acceptability criterion for such systems in the same manner (i.e., Eq. 8) as that for earth stations discussed in Section 3.3. 3.2 Interference to Microwave Link. The satisfactory performance of a microwave link system requires that there be adequate clearance between the link path, i.e., the optical line-of-sight transmission path between the two link antennas, and any nearby scattering objects. It is often required [8] that all scattering objects lie outside the first few Fresnel zones as shown in Fig. 6 and in the present case we shall use the acceptability criterion H > 3H (7) The parameter H is obtained from a knowledge of d, d and the operating wavelength. In addition to using the criterion given by Eq. (7), in some cases we have also calculated rT (or r) to estimate the magnitude of the scattered (or interfering) signal relative to the desired one.

First Fresnel.10 - - -* -.4 nen- -A '- - '-f " — ==- W * Antenna B;?Scatteriang Object Fig* 6:s Of the link 1"* object H i st antennas. outsde the first ~1 fth F1esne7 n Pesae H zone distance a the clearance of S fr he link Path.

-24 - 3.3 Interference to Earth Stations. Interference to an earth station (ES) communicating with a geo-stationary satellite has been assessed by using the Fresnel distance criterion, given by Eq. (7), used for the microwave links. We have also used the acceptability criterion rT < 0.01 (-40 dB) (8) to estimate the level of interference signal at the earth station. 3.4 Interference to Television Reception. WT interference effects to TV reception generally appear in the form of video distortion occurring at twice the rotation frequency of the blade. The dominant parameter determining the interference by a WT is the equivalent scattering area of its blade. However, at a certain distance from the WT the maximum video distortion observed depends on the state of the WT blade (i.e., pitch, plane of rotation, etc.), the ambient signal strengths at the WT and the receiver, the characteristics of the receiving antenna, and on whether the receiver is located in the forward or backward region of the WT. In the backward region the directional property of the receiving antenna may be used to discriminate against the interference effects but in the forward region this cannot be done and hence the effects may be more severe. When the blades are stationary the scattered field may appear on the TV screen as a ghost whose position (i.e., separation from the direct picture) depends on the difference between the time delays suffered by the direct and scattered signals. A rotation of the blades then causes the ghost to fluctuate, and if the ghost is sufficiently

-25 - 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 ('slip') 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 face of a blade, and is called the backward region interference. As the angle 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 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 signal at the receiver, i.e., on the modulation index of the total received signal, and the modulation threshold is defined to be the largest value of the modulation index for which the distortion is still judged to be acceptable. It can be shown f1,2,5,6] that in the case of television interference (TVI) caused by WTs, the parameter rT (or r), defined

-26 - earlier, can be interpreted as the amplitude modulation index mT (or m) suffered by the received signal due to the scattering by the rotating WT blades. Judgement of TVI effects or the video distortion observed is made on the basis of mT (or m). In the backward region for all levels of ambient signals, and in the forward region where the ambient signal is weak, interference effects are judged to be acceptable if mT (or m) < 0.15 (~ -17 dB).(9) For a receiver in the forward region where the ambient signal is strong, the corresponding criterion is mT (or m) < 0.35 (~ - 9 dB). (10) The above criteria are based on the subjective assumption [4] that the resultant video distortion is acceptable. For satisfactory performance of a CATV Head-end the requirement on the interfering signal is more severe [9] and we shall assume the following acceptability criterion: mT (or m) < 0.05 (-26 dB).(11) 4. Assessment of Interference The windfarm interference effects on various systems are quantitatively estimated in the present section. The assessment

-27 - includes the effects of 71 Flo 250 WTs which are presently planned to be installed in the windfarm. Where appropriate, information is also supplied for the effects produced by the windfarm consisting of 170 such WTs. 4.1 Interference to VOR. The interference signal ratio rT at the VOR receiver, produced by the windfarm, has been calculated for the VOR systems identified in Table 2.2. Detailed calculations of rT for specific cases are discussed in Appendix 1. rT values for the Kingston VOR system obtained for windfarms of 71 and 170 WTs are shown in Table 4.1 which indicates that for both cases the windfarm produces rT < -14 dB, i.e., any interference effects produced would be insignificant. The other VOR ground stations being farther away from the windfarm (see Table 2.2), it is unlikely that their performance would be adversely affected by the windfarm. 4.2 Interference to Radio Compasses. With the information given in Table 2.3 and using Eqs. (2) and (4) and assuming ET = ER we have calculated the appropriate r and rT values for the four radio compasses under consideration. The results are shown in Table 4.2. In all cases, the rT values are found to be less than -40 dB. 4.3 ''Interference to Microwave Links. Assessment of interference to the microwave links in the vicinity of the windfarm has been carried out on the basis of Fresnel distance criterion mentioned in Section 3.2. Details of actual calculations required for sample assessment are described in Appendix II. Among the many microwave links originating from Head-ends located near the windfarm only paths 6 and 8 pass over the windfarm (see Figs. 4 and 5). In Appendix II

-28 - Table 4.1 rT at a VOR Receiver Produced by the Windfarm rT in dB, cuased by the windfarm of 71 Flo 250 WTs 170 Flo 250 WTs Kingston VOR -62 dB -58 dB

-29 - Table 4.2 Values for the Radio Compasses Station Meier Neely Otims Monga rT (dB) 71 Flo 250 WTs -55 -54 -55 -63 170 Flo 250 WTs -51 -50 -51 -59

we have investigated the interference effects on Path 6 only; the Path 8 link operates on a higher frequency, and being oriented similar to Path 6, it is argued that the Path 6 assessment will apply to this case also. Table 4.3 lists the assessment parameters of the offending WT sites for Link Path 6. Under the criterion Ar or AH < 3H, turbines 67, 70, 69 and 71 are unacceptable at their present locations. They will be acceptable provided they are displaced from their present locations in a manner given in Table 11.2. Under this condition, all other links would be unaffected by the windfarm. 4.4 Interference to Earth Stations. As shown in Fig. 4 there exists one earth station at Ellenville at a distance of 3.16 miles (5.08 km) and at Wurtsboro at a distance of 7.5 miles (12.03 km) from the center of the windfarm. We shall assume that the earth station uses a 30 ft (10 m) diameter parabolic dish antenna at f = 4.0 GHz, i.e., x = 0.246 ft (0.075 m); at this frequency the antenna typically has a beam width of approximately 0.5 degrees and side lobe level of about -25 dB. If the interference effects are acceptable for this antenna, they would also be acceptable for a larger (49 ft or 15 m) antenna used by earth stations. We shall show the assessment for the Ellenville ES. Sample calculation: f = 4.0 GHz,x = 0.075 m, d = 5.08 km Ae = 2.52 = 0.690 m 2 i =2A/ = 3.62 x 10-3 1 ed rT for 71 machines = 3.05 x 10-2 (-30.3 dB) rT for 170 machines = 4.173 x 10-2 (-26.5 dB) Assuming antenna discrimination of -25 dB we obtain

-31 - Table 4.3 Assessment Parameters for Offending WT Sites for Flo 250 WT: Link Path 6 Site No. Ar (ft) AH (ft) 3H (ft) 62 372 35 116 67 105 60 120 66 372 56 124 70 105 74 124 69 105 69 127 71 0 94 130

rT for 71 machines = -55.3 dB rT for 170 machines = -51.5 dB In both cases, rT is less than -40 dB; therefore the windfarm would not produce unacceptable interference to the Ellenville ES. The Wurtsboro ES, being farther away from the CF, would not be affected by the windfarm. 4.5 Television Interference (TVI) Effects at the CATV Head-End. A CATV Head-end, identified as CATV in Fig. 4, is located at Wurtsboro (elevation 1470 ft) and is at a distance of 7.5 miles (12.03 km) from the center of the farm. The antenna tower height being 250 ft, the elevation of all CATV antennas is 1720 ft. It is assumed that the CATV Head-end received all TV signals listed in Table 2.1. During reception of signals from the stations listed in Table 2.1, it may be assumed that the entire windfarm is located in the backward region. We shall therefore determine the interference signals assuming all WT sites to be in the backward region. In the present case, it is reasonable to assume that the ambient TV signals at the CATV Head-end and at the WT sites are of the same order of magnitude, i.e., EB/ER = 1. For the purpose of calculation of mT it is assumed that the CATV antenna beam is directed to recieve maximum signals from the desired direction, and that the side and/or back lobe level of the antenna is at least -20 dB. Wn shall perform the assessment for TV Channels 2 (x 6 m), 22 (x = 0.5 m) and 31 (x 0.52 m). A sample calculation for Channel 2 is given below:

f = 55.25 MHz, A 6 m, d 12.03 km A = 2.52 v = 6.17 m2 (assuming the receiving antenna to be e isotropic), m = r = 2Ae/d = 1.71 x 10-4 mT for 71 Flo 250 WTs = 7Tm = 1.44 x 10-3 (-56.8 dB) mT for 170 Flo 250 WTs = / m = 2.23 x 10-3 (-53.0 dB) With -20 dB discrimination provided by the receiving antenna mT values for 71 and 170 machines are -76.8 and -73.0 dB, respectively. Calculated mT values for Channels 2, 22 and 31 applicable to the windfarm of 71 and 170 WTs are shown in Table 4.4. Under the assumption that acceptable TVI effects would occur for mT > -26 dB, the results of Table 4.4 indicate that the interference effects produced by the windfarm on the performance of the CATV Head-end would be insignificant for the Channels 2, 22, 31 and for all other Channels listed in Table 2.1. 4.5 Interference to TV Reception at Cragsmore. It is understood that the residents at Cragsmore (elevation about 1800 ft), distant 4.88 km from the center of the windfarm, receive TV signals without the help from the CATV service. In the absence of detailed knowledge of the ambient signals on the desired Channels available in the area, we shall assume that EB = ER and obtain an estimate of the mT values under the worst possible conditions, i.e., the receiving antenna is isotropic. Table 4.5 shows the mT values caused by the windfarm of 71 and 170 Flo 250 WTs for two typical Channels.

-34 - mT Values for TVI Effects at Flo 250 WTs. Table 4.4 the CATV Head-End Due to a Windfarm of (Antenna sidelobe = -20 dB) mT (dB) TV Channel No. 2 71 WTs -76.8 -66.6 -66.2 170 WTs -73.0 -62.9 -62.5 22 31 Note: TVI effects acceptable if mT < -26 dB.

-35 - Table 4.5 mT Values for TVI Effects at Cragsmore Caused by the Windfarm of Flo 250 WTs. (Receiving antenna isotropic) mT (dB) TV Channel No. 71 WTs 170 WTs 2 -49 -45 22 -39 -35

-36 - The results given in Table 4.5 when compared with the acceptability criteria given by Eqs. (9) or (10) indicate that the TVI effects at Cragsmore would be insignificant for the TV Channels 2 and 22; hence it is concluded that they would also be insignificant for all the other Channels listed in Table 2.1, assuming that they are also available in the area. 5. Conclusions The fundamental parameter required to estimate the electromagnetic interference effects of a WT is the equivalent scattering area of its blade. To the best of our knowledge, such information about the candidate WTs (i.e., Flo 170, 250 and 500) for the Ellenville Windfarm is not yet precisely known. We have obtained, only approximately, the required information by applying extrapolation laws to our present knowledge of the scattering area of the 17-m Darrieus developed by the Sandia Laboratories. Since the VAWTs developed by the Flo Wind Corporation are similar to the Darrieus, it is believed that the estimate of the scattering area used for the present assessment is valid. The TVI effects at a receiving site also depend quite strongly on the ratio of ambient signal strengths at the receiving and WT sites. In a rugged terrain like the Ellenville Windfarm it is difficult to determine these signal strengths theoretically. Although we have made approximations to these parameters based on our experience, the actual signal ratios may be different (this may be so, particularly for the assessment of TV reception at Cragsmore). For more precise TVI assessment, the desired ambient signal strengths should be measured at the receiving and WT sites.

APPENDIX I. CALCULATION OF rT FOR ASSESSMENT OF VOR INTERFERENCE It is assumed that the WTs of the farm may cause interference only if they are visible from the antenna of the VOR ground station, i.e., when the antenna and the WT(s) are within the radio line-ofsight distance. The radio line-of-sight distance (dH) between two points at heights h and h above a smooth spherical earth is 1 2 dH = f2 (F — + /F) 1 2 (I.1) where dH is expressed in miles and h, h are in feet. Identifying 1 2 h as the VOR antenna height and h as the WT height and assuming 1 2 smooth terrain between the VOR station and the WT, Eq. (I.1) can be used to determine whether the WTs in the farm would be visible from the VOR antenna. For example, in the case of Kingston VOR station h = 2200 + 2292 ft. Let the average height above the sea level of a WT in the be h = 2292 ft. Thus, from Eq. (I.1) 2 92 = farm dH = 73 miles (118 km). Under the assumption that the terrain between the Kingston VOR station (about 22 miles from the windfarm) and the windfarm is smooth, it appears that all the WTs in the farm would be visible from the VOR station. -37 -

-38 - Calculations for the Kingston VOR: rT for the Flow 250 WTs are obtained as follows: Kingston VOR: distance from the center of the windfarm d = 21.7 miles = 34.94 km. f = 120 MHz, X = 2.5 m Ae = 2.52 A m2 for one WT at a distance of 34.94 km 2Ae r - = Ad 9.14 x 10-5 (-81 dB) for 71 machines rT = /71 r = 77.02 x 10-5 (-62 dB) for 170 machines r = /70r = 119.17 x 10-5 (-58 dB). Other VORs are located at distances larger than that of the Kingston VOR (i.e., d > 21.7 miles), hence the rT values for these stations would be less than the values obtained above.

APPENDIX II. ASSESSMENT OF INTERFERENCE TO MICROWAVE LINKS We shall illustrate the assessment of windfarm interference to microwave links by describing the calculation procedure followed in a typical case. For a given WT site of elevation Hs, located at a horizontal distance 4r' from the link path of elevation h, at the location of the WT, we define the following two parameters: horizontal clearance Ar = Ar' - D/2 vertical clearance AH = h9 - hT, where D = the rotor diameter of the WT and hT = H + hWT + D/2, HWT being the total height of the WT. The acceptability criterion for the site, based on the considerations of Fresnel distance (Section 3.2), is now |AHJ or > 3H (II.1) I Ar where 1 [x(d - d )dl 1/2 H - d- (II.2) x being the wavelength and d, d as explained in Fig. 8. Figures 4 and 5 show the microwave link paths superposed on the windfarm. For each link the offending sites (generally for Ar < 3H ) are identified, and the corresponding AH, Ar and H are calculated for a given WT by using (II.1) and (II.2) -39 -

-40 - Sample Calculations for Path 6 From the data given in Table 2.4 we prepare the elevation diagram, shown in Fig. 11.1, for the link Path 6 whose one Head-end (antenna No. 1) is located near the windfarm. It is assumed that f = 4.0 GHz, X = 0.246 ft, with Flow 250 WT (h.WT = 92 ft, D/2 = 28.5 ft) at each of the offending sites (numbered according to the WT number) near Path 6, and the various parameters required for the calculation of AH and Ar are now obtained by using Figs. 4.5 and 11.1. The results are shown in Table 11.1. According to the criterion given by Eq. (11.2), of the sites considered in Table 11.1, only sites 62 and 66 are acceptable. To make the other sites acceptable according to the present cirterion the respective WTs should be displaced from their present location in a manner given in Table 11.2.

-41 - Antenna No. 1 2270+83 = 2353' AH 242' 2 Antenna No. 2 524+108 = 632' t — --- — Halihan, NY Ellenville 62 67 66 69 71 27135 mi 70 Fig. II.1 Elevation Diagrams for Link Path 6.

-42 - Flo 250 Windfarm Site No. 62 67 66 70 69 71 Ar' (ft) 400 133 400 133 133 0 Table II.1 Interference Assessment Parameters (d = 27.35 miles, f = 4.0 GHz) Ar AH di (ft) (ft) (ft) 371.5 35 6400 104.5 60 6800 371.5 56 7333 104.5 74 7333 104.5 69 7730 0 94 8133 for Link Path 6 3H1 (ft) 116 120 124 124 127 130

-43 - Table Required Displacement 11.2 of WTs for Acceptability Displace amount (ft) IWT No. 67 70 69 71 direction 25 20 25 130 West East West East or West

-44 - References 1. 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. 2. D. L. Sengupta and T.B.A. Senior, "Wind Turbine Generator Interference to Electromagnetic Systems," University of Michigan Radiation Laboratory Report 014438-3-F, August 1979. 3., "Handbook: VOR/VORTAC Siting Criteria," Federal Aviation Administration, Department of Transportation, Report 6700.11, August 7, 1968. 4. D. L. Sengupta and T.B.A. Senior, "Measurements of Television Interference Caused by a Vertical Axis Wind Machine," SERI/STR215-1881, Solar Energy Research Institute, Golden, CO 80401. 5. T.B.A. Senior and D. L. Sengupta, "Large Wind Siting Handbook: Television Interference Assessment," University of Michigan Radiation Laboratory Report 014438-5-T, April 1981. 6. D. L. Sengupta and T.B.A. Senior, "Electromagnetic Interference by Wind Turbine Generators," University of Michigan Laboratory Report 014438-2-F, March 1978. 7. 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," University of Michigan Radiation Laboratory Report 018291-1-T, January 1981. 8. Members of the Technical Staff, "Transmission Systems for Communications," Fourth Edition, Bell Telephone Laboratories, Inc., December 1971. 9. ITT, "Reference Data for Radio Engineers," Sixth Edition, Indianapolis, Indiana, p. 30-18, 1975.