&EPA
             United States
             Environmental Protection
             Agency
            Office ol Radiation Programs
            P.O. Box 96517
            Las Vegas NV 89193-8517
EPA/520/6-91/020
July 1991
             Radiation
Electric and Magnetic Fields
Near AM Broadcast Towers

-------

-------
                    GENERAL DISCLAIMER
    This document may have problems that one or more of the following disclaimer
                           statements refer to:
»  This document has been reproduced from the best copy furnished by the
   sponsoring agency.  It is being released in the interest of making
   available as much information as possible,

•  This document may contain data which exceeds the sheet parameters. It
   was furnished in this condition by the sponsoring agency and is the best
   copy available.

•  This document may contain tone-on-tone or color graphs, charts and/or
   pictures which have been reproduced in black and white.

*  The document is paginated as submitted by the original source.

«  Portions of this document are not fully legible due to the historical nature
   of some of the material.  However, it is the best reproduction available
   from the original submission.

-------

-------
                               TECHNICAL REPORT DATA
                         (Please read Instructions on the reverse before complet'
I.RiPORTNO,
 EPA 520/6-91-020
                          2.
                                                     3.
                                                          PB 92-101427
4. TITLE AND SUBTITLE
 Electric and Magnetic  Fields Near AM
 Broadcast Towers
                                                     8. REPORT DATE
                                                         July 1991  -  .,
                                                     6. PERFORMING ORGANIZATION CODE
7, AUTMOR(SJ
 Edwin- Mantiplj?
                                                     a. PERFORMING ORGANIZATION REPORT
                         'ert';^.1;,- G levels 11137. ;--Jx';
9. PERFORMING ORGANIZATION NAME A'ND ADDRESS
 U.S.  EPA-  ,- • -,' v.;;-,!. ,!•;...  -..^.. -  i.
 Office_of Radiation Programs
 Las Vegas, NV  89193-8517'
                                                     10. PROGRAM ELEMENT NO,
                                                     11, CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
 U.S." EPA ' r> " •'' '•• !:  "-'•  •'••'•• '"'-'-'
 Office of Radiation Programs
 401  M Street, SW   •
 Washington,-DC  20460	
                                                     13, TYPE OF REPORT AND PERIOD COVERED
                                                            Final  .
                                                     14, SPONSORING AGENCY CODE
IS. SUPPLEMENTARY NOTES
16, ABSTRACT
         The purpose of this  study was  .to obtain actual measurement data i
 the close-in near field of 'representative AM  boradcast  antennas  and
 compare the data  to values  predicted  by a Numerical Electromagnetic '
 Code  (NEC)  model.   Measurements of  Electric and Magnetic  fields 'were
 made along several radial directions  at distances .from  Tto' 100m from
 the transmitting  towers of  eight AM broadcast stations.   These stations
 operated at various 'frequencies, electrical heights, and  power outputs.
7.
                            KEY WORDS AND DOCUMENT ANALYSIS
                DESCRIPTORS
                                        b.lDENTIFlERS/OPEN ENDED TERMS C. COSATI Field/Group
Electric fiel'ds
magnetic fields
measurements  -
models
strenghts
broadcasts
stations	
6, DISTRIBUTION STATEMENT

      Release Unlimited
                                        1». SECURITY CLASS (This Report!
                                            Unclassified
21. NO, OF PAGES

     45
                                         2O. SECURITY CLASS (Thispage)
                                                                 22, PRICE
EPA Form 2220-1 (R«v. 4-77)   PREVIOUS EDITION is OBSOLETE  ,

-------

-------
    ELECTRIC AND MAGNETIC FIELDS
      NEAR AM BROADCAST TOWERS
          Edwin D.  Mantiply
    Office of Radiation Programs
U.S. Environmental Protection Agency
      Las Vegas, NV 89193-8517

                 and

      Robert F. Cleveland, Jr.
Office of Engineering and Technology
  Federal Communications  Commission
        Washington, DC 20554
              July 1991

-------
                         ACKNOWLEDGMENTS

This study was supported In part by the Federal Communications
Commission through an interageney agreement with the Environmen-
tal Protection Agency (RW27931344-01-4).  EPA student employees,
Toni West and Dale Green, also participated in the measurement
survey and preparation of this report.  James Hepker of Computer
Science Corporation, made the Numerical Electromagnetics Code
runs and produced the plots used in this report.  The cooperation
and assistance of the eight stations surveyed in this study is
gratefully acknowledged.
                                11

-------
                        TABLE OF-CONTENTS
Abstract.	iv
List of Figures	v
List of Tables	vi

1.  Introduction	 i

2.  Instrumentation 	 ........ 	 .   . 2
     2.1  Electric Field Instrumentation. .  	  ... 3
     2.2  Magnetic Field Instrumentation	 3

3.  Measurement Procedures.  ................. 4

4.  Antenna Modeling. .............. 	 5

5.  Results	 10
     5.1  Station A	   .11
     5.2  Station B		   .13
     5.3  Station C ............'....	15
     5.4  Station D	 17
     5.5  Station E	20
     5.6  Station F	 22
     5.7  Station G	 25
     5.8  Station H	27

6.  Discussion and Conclusions.	30
     6.1  NIC Modeling and Agreement with Measurements.  . .   .30
     6..2  Quasi-static Modeling of Close-in Field Strength.   .33
     6.3  Factors Affecting Close-in Field Strength .   .  . .   .34
     6.4  Perturbation of Fields Due to Objects Near A
          Tower Base. ...  	 ...........   .40
     6.5  Variation of Fields Along Different Radials  ...   .41
     6.6  Agreement with Values Given in FCC Bulletin 65. .   .43
     6,7  Related Studies .... 	  ........   .46

7.  Summary	47

Appendix A. Station A Graphs	  A-i
Appendix B. Station B Graphs. .......... 	  B-l
Appendix C. Station C Graphs. 	  ..........  C-l
Appendix D. Station D Graphs.	D-l
Appendix E. Station 1 Graphs	E-l
Appendix F. Station F Graphs. ...............  F-l
Appendix G. Station G Graphs. ...............  G-l
Appendix H. Station H Graphs		H-l
                               111

-------
                             ABSTRACT


Only limited data have been available that can be used to define
regions near AM broadcast towers where radiofrequeney (RF)
radiation safety standards are likely to be exceeded.  In the
past, computer models have been used to predict distances at
which various field strength levels would occur in the near field
of AM antennas.  In particular, theoretical values for electric
and magnetic fields have been determined using the Numerical
Electromagnetic Code (NEC), a computer program developed by the
Lawrence Livermore Laboratory, to calculate fields near wire
antennas of arbitrary shapes.

The purpose of this study was to obtain actual measurement data
in the close-in near field of representative AM broadcast anten-
nas and compare the data to values predicted by a NEC model.
Measurements of electric and magnetic fields were made along
several radial directions at distances from 1 to 100 m from the
transmitting towers of eight AM broadcast stations.  These
stations operated at various frequencies, electrical heights, and
power outputs.

Reasonably good agreement was obtained between measurement data
and the NEC models developed for the AM towers surveyed at
distances greater than several meters form the tower's base.  The
agreement was generally not as good closer to the tower's base.
Metal fencing or other metal objects near a tower base signi-
ficantly affect close-in fields, especially electric fields.  The
effect of these objects was not modeled.
                                IV

-------
                         LIST OF FIGURES
                                                             Page
 1.   The Physical and Numerical Model Tower	  7
 2.   Diagram of Station A Site.	12
 3.   Diagram of Station B Site	14
 4.   Diagram of station c site		16
 5.   Diagram of Station D Site	19
 6.   Diagram of Station E Site	21
 7.   Diagram of Station F Site	24
 8.   Diagram of Station G Site	26
 9.   Diagram of Station H Site	  .29
10.   Electric Field Error Analysis	  .31
11.   Magnetic Field Error Analysis	32

-------
                         LIST OF TABLES

                                                             Page

 1.   AM Broadcast Stations  Surveyed	2

 2.   Electric Field Comparison Readings	5

 3.   Modeling Information	9

 4.   Quasi-static Electric  Field Results 	  35

 5.   Quasi-static Magnetic  Field Results 	  35

 6.   Electric and Magnetic  Field Strength as a Function of
     Electrical Height .  .  	  ........  	  37

 7.   Comparison of Data From Stations According to
     Frequency and Electrical Height	38

 8.   Comparison of Measured Field Strength Between
     Stations with Different Operating Powers	39

 9.   Variation of Field Strength Values as a Function of
     Radial Direction From Single AM Towers	  42

10.   (From OST Bulletin No. 65)  Distances (in meters) at
     Which Fields From AM Stations are Predicted to Fall
     Below various Electric Field Strengths	.44

11.   Comparison of Predicted and Measured Distances at
     Which Field Strengths  are Less Than Selected Levels
     Using Numbers From OST Bulletin 65 and Data Collected
     in This Study	45
                                VI

-------
1.  INTRODUCTION

Over the past few years the Federal Communications Commission
(FCC) and the U.S. Environmental Protection Agency (EPA) have
jointly conducted several measurement surveys and other studies
of radiofrequency (RF) radiation from FCC-regulated transmitting
facilities.  These studies were performed under the terms of an
interagency agreement between the FCC and the EPA and have
involved staff from both agencies.  Broadcast stations have been
the focus of most of these studies since their relatively high
power levels increase their potential for environmental sig-
nificance.

One area of interest has been to determine the electric and
magnetic field strength in the near-field of AM broadcast towers.
The data available to define regions near AM towers where RF
safety standards may be exceeded is limited.  Computer modeling
techniques have been used in the past to predict distances at-.
which various field strengths would occur in the near-field of AM
stations.  However,  actual measurement data is needed for com-
parison with theoretical values.  Such data should help refine
prediction methods and point out potential problems involved in
modeling techniques.

The purpose of this study was to begin collecting data on actual
field strength values near AM broadcast towers.  In the immediate
vicinity of the tower, electric and magnetic field"strength
depends on a number of variables, including transmitting frequen-
cy and tower height.  This study concentrated on tower sites that
would be relatively easy to survey and model theoretically.  A
comparison of predicted field strength values versus actual
measurements would be useful in determining the accuracy of
methods presently being used for evaluating broadcast sites for
environmental RF radiation.  For example, the FCC's Bulletin No.
65 [1] uses computer modeling to determine areas that should be
restricted in order to comply with RF protection guidelines.
This bulletin generally defines "worst-case" scenarios," informa-
tion on actual field values would be helpful.

In preparation for this study, questionnaires were sent to
thirty-seven stations in the southern California area.  These
questionnaires requested the stations' cooperation and asked for
specific information about their transmitting facilities.
Southern California was selected for this study because of its
proximity to the EPA laboratory in Las Vegas and because of the
number and variety of stations in the area.  The stations that
were sent questionnaires were chosen based on their electrical
heights, frequencies, number of transmitting towers, and power
levels.  Data were needed from stations representing a range of
electrical heights and frequencies.  Single towers or relatively
simple tower arrays were .preferred.  The majority of stations
contacted operate at power levels of 5 kilowatts or less.

-------
However, absolute power was considered of secondary importance
since fields can be scaled in proportion to the square root of
power.  Over 8O% of the stations contacted responded favorably
and were considered as study subjects.  The stations ultimately
chosen were those with relatively unobstructed transmitting sites
where measurements could be made without major complications.

The study was carried out the week of August 8-12, 1988.  A total
of eight broadcast sites were surveyed.  Pertinent characteris-
tics of each station are given in Table 1 with stations iden-
tified by the letters A - H. Only station H transmitted from a
self-supporting tower.  All other active towers were guyed.
Preliminary results of this study were presented at the Eleventh
Annual Meeting of the Bioelectromagnetics Society in 1989  [2],

             TABLE  1.   AM  BROADCAST  STATIONS  SURVEYED
Station
Fre-
quency
fkHz)
Power
CkW)
Tower
Height
J meters) *
Electrical
Height
^wavelength)
No.
Trans-
mitting
Towers
  A

  B

  C
1350

1410

550
1.0

1.0
54.9

90.2
4.3 (Tower 99.1
     #1)
0.63(Tower 99.1
     #2)
0.25

0.42

0.18

0.18
1

1

'2
D
E
F
G
H
1440
1410
1070
790
1450
1.0
4.2
50.0
5.0
1.0
111.3
56.7
111.4
146.3
70.1
0.53
0.27
0,40
0,39
0.34
1
1
**1
**1
**1
 *Tower height = height of radiator above  insulator, not includ-
ing obstruction lighting
**Second  (de-tuned) tower also at site

2.  INSTRUMENTATION

At each measurement location both the electric and magnetic  field
were determined.  The electric field was measured using a broad-
band instrument.  A narrowband check of the electric field at one
location at each transmitter site was made using a fiber-optical-
ly isolated antenna and spectrum analyzer- system.  The magnetic

-------
field was measured using a: loop antenna and .field strength meter
at every location.

2.1  ELECTRIC FIELD INSTRUMENTATION

Two Instruments for Industry (IFI) model EFS-l broadband electric
field strength meters were used in this study.  These instruments
consist of a single short monopole on a conductive box.  The box
contains the readout electronics and acts as an integral part of
the antenna.  The instrument detects only the component of the
field aligned with the monopole.  For this study two orthogonal
measurements were made using the IFI instrument, one vertical and
one radial.  In a few cases tangential measurements were also
made.  The instruments were calibrated in the EPA transverse
electromagnetic (TEM) cell system using unmodulated fields [3].
The serial number 1060-E unit was used for all measurements below
300 V/m; its correction factor at 1 MHz was 0.98.  The serial
number 1059-E unit was used for all values above 300 V/m? its
correction factor at 1 MHz was 0.87.

Since these instruments use diode detectors, the measured elec-
tric field strength is expected to be greater than the actual
field strength in amplitude modulated fields.  At each location
maximum and minimum readings were taken during modulation, the
minimum readings  (corrected) are reported here.  The maximum
reading is typically 20% higher than the minimum value.  The
minimum reading will correspond more closely to the unmodulated
field strength.  By convention, transmitter powers and measured
field strengths reported for AM radio stations refer to unmodu-
lated carrier conditions.

An automated system consisting of a fiber-optically isolated
spherical dipole antenna (FOISD), spectrum analyzer, and con-
trolling computer was used to check the IFI reading at one
location at each transmitter site.  A detailed description of
this system and discussion of measurement accuracy is contained
elsewhere  [4].

2.2  MAGNETIC FIELD INSTRUMENTATION

An Eaton model 92200-3, 15" loop antenna and Potomac model FIM-
41 field strength meter were used to measure the magnetic field
at each location.  The Potomac meter had been recently calibrated
by the manufacturer  (4/25/88).  This meter was used as a mag-
netic-field standard in the laboratory to calibrate the loop
antenna in a Helmholtz coil at each AM frequency used in this
study.  During field work, the external input of the Potomac
meter was calibrated for absolute RF voltage measurement using a
synthesizer, power splitter, power meter, and 50-ohm, feed-
through resistor.  The Potomac was not used for direct magnetic-
field measurement because levels above 0.0265 A/m would exceed
the maximum meter reading.  In the field, the Potomac was used as

-------
 a  tuned  absolute  voltmeter  to.read  the voltage generated by the
 loop  antenna  In the  magnetic  field.  The  loop was oriented for a
 maximum  reading on the  Potomac meter; this direction was aligned
 with  the circumferential  magnetic field near the tower.

 3.  MEASUREMENT PROCEDURES

 A  primary objective  of  this study was to  obtain values  of elec-
 tric  and magnetic field strength near representative, and rela-
 tively simple, AM broadcast towers.  At each of the eight sites
 visited,  measurements were  made in  at least two radial  directions
 from  a transmitting  tower.  Both electric and magnetic  field
 strengths were measured at  most measurement points.

 A  non-conductive  tape-measure, 100  meters (m) in length and
 appropriately, marked, was used to define  the radials along which
 measurements  were made.  Radial distances were defined  from the
 center of a tower base.  In the area immediately adjacent to  the
 tower base, measurements  were usually made at intervals of one or
 two meters out to a  distance  of ten meters.  Field strength
 readings were then generally  made at five- or ten-meter intervals
 out to 50, 75, or 100 m.  In  some cases obstructions such as
 fences were present  near  a  tower and prevented readings from
 being taken at certain  points.

 All readings  were made  with the field probe one meter above
 ground.   For  electric field measurements, the IFI meter was set
 on top of a plexiglass  platform supported by a section  of PVC
 tubing mounted on an adjustable wooden base.  The overall height
 of this  apparatus (ground to  plexiglass platform) was one meter,
 Readings were made by moving  this apparatus along the radial  so
 that  the meter was positioned directly above the desired measure-
 ment  point.   Measurements were then made  of both the vertical and
 radial components of the  field by orienting the IFI meter with
 the monopole  pointed:   (1)  vertically upward, or  (2) parallel to
 the tape-measure  and pointed  away  from the tower.  In a few
 cases, very close to a  tower's base, measurements were  also made
 of the tangential electric-field component.  However, those
 readings were generally negligible.

 Readings were taken  by  watching the instrument needle while the
 observer stood at a  distance  of about 2-3 m away to avoid pertur-
 bation of the field  by  the  observer.  Both maximum and  minimum
:values were recorded to establish effects of signal modulation.
 Total electric field was  obtained by computing the vector sum of
 the components and multiplying by  the appropriate correction
 factor.   This usually  involved combining  only the vertical  and
 radial components because of  the minimal  contribution  from  the
 tangential component.   The  radial  component of the total field
 was only significant in the immediate vicinity of a tower base  or
 other metallic objects, such  as chain-link  fencing.  At locations
 more  distant  from the tower or from conductive objects  the

-------
vertical component predominated and constituted most of the total
field.

Magnetic field readings were made by orienting the loop antenna
until a maximum value was obtained.  This orientation was such
that the magnetic field existing circumferentially around the
tower was perpendicular to the plane of the loop.  At each
transmitter site the Potomac meter was recalibrated to read
absolute voltage at the external input.  Tuning was checked every
few measurements.  The voltage was recorded to be corrected later
using the loop calibration.

At each broadcast site an electric-field measurement was also
made using the FOISD antenna to compare with the readings made
with the IFI meter.  For each of these FOISD readings a con-
venient point was chosen where an IFI reading had already been
made, and the spherical-dipole antenna-mount was set up at that
point.  The FOISD was rotated through 360 degrees in three 120
degree increments to obtain readings in each of three orthogonal
directions (vertical, radial, and circumferential).  These
readings were made using a computer-operated spectrum analyzer
system to determine the electric field.  Table 2 shows the
results of the FOISD and IFI readings at the comparison points.

          TABLE  2.   ELECTRIC FIELD COMPARISON  READINGS

                 IFI Minimum         FOISD       Measured
                   Reading          Reading     Difference
Station              fV/a)            fV/ml          fdE)'

                                      3.8           1.5
                                      8.0           0.63
                                     18.9           0.92
                                      5.1           1.7
                                     11.7           2.6
                                     39.2           0.22
                                     19.2           1.1
                                      2.4           0.70

 using FOISD as reference

4.  ANTENNA MODELING

Theoretical .values of the near electric and magnetic field of the
AM broadcast antennas in this study were determined using the
Numerical Electromagnetic Code (NEC).  NEC is a computer program
developed by Lawrence Livermore National Laboratory which can be
used to calculate fields near wire antennas of arbitrary shape.
The program version used here is NEC2 which is contained in the
Numerical Electromagnetics Engineering Design System (NEEDS)
package [5 J.
A
B
C
D
E
F
G
H
4.5
8.6
21.0
6.2
15.8
40.2
21.7
2.6

-------
Wire antennas are modeled in NEC as a set of straight wires in
free space or above ground.  Each wire is specified by three
coordinates in space for each wire end, by a wire radius,  and by
the number of segments into which the wire is divided.  For a
given excitation voltage applied across specified segments, NEC
solves for the magnitude and phase of the current on every other
segment.  These currents are then used to calculate the near
fields of the antenna.

Many rules apply when creating a model of a real antenna.   For
example, segment length cannot be too long or short relative to
the free-space wavelength, wire radius cannot be too large
relative to the segment length, and adjacent segments cannot
change radius or length too rapidly.  An important note is that
several single-segment wires in a straight line are etpiivalent to
one wire broken into several segments.

In principle, the detailed structure of each AM tower and its
environment in this study could be modeled using NEC.  Tower
ironwork, feed point details, ground radials and ground conduc-
tivity, guy wires, conductive fences, boxes or buildings contain-
ing matching networks, and of course other towers in the system
may be important especially when very close to any of these
structures.  However, our goal was to use the field data to find
one simple modeling recipe which results in good agreement
between measured and calculated fields one meter above ground at
distances from the tower of one meter out to the edge of the
ground radial system.

The model used for guyed towers without top loading is shown in
Figure 1.  The physical tower is shown on the left and the
numerical model of the tower is represented on the right of the
figure.  For an AM radio station the entire tower is the antenna.
The transmitter is connected to the insulated tower through a
matching network.  The ground plane is enhanced by buried ground
radial wires approximately the same length as the tower height.

Guyed towers were modeled as three vertical wires having the same
radius and spacing as the three tower legs.  The base region was
approximated by extending the three wires to ground.  An excita-
tion voltage was applied across the three wire segments at the
height of the insulator.  This voltage was adjusted until the
antenna current matched the licensed value.  The length of the
wires above the excited segments was modified by up to 5.1% to
obtain an approximate match between the measured value of the
impedance listed in the station license and the impedance calcu-
lated by NEC.  The ground was assumed to be a perfectly conduct-
ing plane.  Electric and magnetic field components were calcu-
lated along a radial at one meter above ground, and the three
components of each field were combined to give the resultant
magnitude of the field vector or total field.

-------
                         Obf iruction lighl
                                       Guy Wire Insulator
                                                                                                  Oiiven Segments ai
                                                                                                  Iliii0lii ul liiii.lalm
                                                                                                    period
                                                                                                    Ground
Figure  1.   The Physical  and Numerical Model  Tower

-------
For example, the physical tower description used to create the
NEC model for.station A is as'follows.  The tower is stabilized
by insulated guy wires and supported by a grounded concrete
pedestal 31 in. (0.787 m) high.  A 9 in. (0.229 m) long
cylindrical ceramic insulator separates the tower from the
pedestal.  The steel tower is constructed from three parallel,
braced tower legs.  A horizontal cut through the tower results in
an equilateral triangle 15 in. (0.381 m) on a side.  The legs are
0.5 in.  (0.0127 m) in radius, and 180 ft (54.9 m) long.  The
tower  ags are welded to a flat plate that rests on the in-
sulator.  The other guyed towers have a section above the in-
sulator where the legs taper together.  This section is modeled
as a continuation of the parallel tower legs and is included in
the length of the tower legs.

The NEC model for station A was constructed in rectangular
coordinates from three wires extending in the.z or vertical
direction each having a radius of 0.0127 m.  A horizontal cut
through the wires results in an egxiilateral triangle similar to
one for the actual tower legs.  However, the wires in the model
extend to a perfect ground (the x, y plane at z = 0) with the
center of the triangle at the origin.  The horizontal coordinates
in meters of the wire centers are x = 0, y = 0.172; x = 0.191, y
= -0.158; and x = -0.191, y = -0.158.  Note that the fields are
calculated at z = 1 m above ground and along a radial in the x
direction for all towers.  This radial is at right angles to a
radial along y.

Each of the three vertical wires for the station A model was
divided  into 27 segments such that the total number of segments
used in the model was 81.  These segments are described starting
from the ground and going up.  The first 3 segments represent the
pedestal.  These segments are 0.279, 0.279, and 0.229 m long.
Segments adjacent to the insulator or driven segment are set to
the same length as the insulator segment, here 0.229 m.  Eight
(8) segments above the insulator increase in length such that the
ratio of length of any one segment to the next lower segment is
1.4; the bottom segment  is 0.229 m long and the top segment of
this group is 2.41 m long.  Finally, there are 15 segments, each
3.13 m in length which complete the tower.  The segments are
chosen to model the base region in some detail, avoid changing
the lengths of segments  adjacent to the driven segment or chang-
ing the  lengths of any adjacent segments abruptly, and minimize
the total number,of segments.

The measured impedance for tower A is known from the license to
be 49.3 + j 93.4 ohms.   For the above model the calculated
impedance was 42.3 + j 42.3 ohms  (initial calculated impedance).
To better match the measured  impedance, the model tower height
was revised using successive  approximations.  This process
resulted in increasing the length of the top  15 segments to 3.42
m each.  This revision increases the modeled  tower height above

                                8

-------
the insulator from 54.9 m (tower height)  to 57.7 m (adjusted
tower height), changes the calculated impedance to 54.9 + j 85.2
ohms (revised calculated impedance),  and apparently improves the
agreement between measured and calculated fields near the tower.
This process of adjusting tower height to match measured impe-
dance was not successful for all of the towers.  Adding capaci-
tance across the driven segments has been suggested as an alter-
native to adjusting tower height.

Finally, the drive voltage for tower A across each insulator
segment was adjusted to obtain a drive current of 1.51 A rms on
each of the three segments for a total current equal to the
licensed base current value of 4.52 A.

Similar information for all of the towers modeled is given in
Table 3.
Part I.
                  TABLE 3.   MODELING INFORMATION
         Pedestal  Insulator    Leg     Leg    Tower
Station   Height    Height    Spacing  Radius  Height   Total
 Code      (m)        
-------
 (325  ft)  to  three  ground anchor points at  69.2 m  (227 ft) from
 the tower base.  The  top 9.14 m  (30  ft) of each of these guy
 wires is  electrically bonded to a tower leg.  The remaining
 portions  of  the  guy wires are insulated from the tower.  The
 sections  of  guy  wire  bonded to the tower are modeled as downward
 extensions of  the  three tower legs at angles of 35 degrees to the
 tower.  In order to maintain a constant wire radius in the model,
 the guy wire sections are modeled to have  the same radius as the
 tower legs.  These modeled guy wire  sections are not changed in
 length but are translated vertically as the tower height is
 adjusted  in  the  model to match impedance.  Also, station C is the
 only  station in  the study using more than  one active tower.  A
 second tower is  operated at only 15% of the power used for the
 tower studied.   This  tower was not modeled and its effect on
 measured  fields  is only seen at locations  close to it,,

 Station H is the only self-supporting tower in the study.  The
 model was similar  to  that used for the guyed towers except the
 three driven segments correspond to  three  physical insulators.
 The tower tapers gradually from the  base up to about the mid-
 point and is then  uniform in cross-section to the top.
 The three tower  legs  were modeled such that they converge to a
 single wire  of radius 0.05 m at the  mid-point of the tower.  The
 legs  are  spaced  3.5 in apart at the tower base.

 These models were  designed to produce fields as close as possible
 to the measured  values; there was no explicit effort to generate
 field values from  the model that will be consistently greater
 than  measured  values. For this reason some NEC values are higher
 and some  lower than measured values.  In general, considering the
 number of approximations made in the modeling process, the
 agreement between  measured and predicted fields was reasonable.

 5.  RESULTS

 A considerable amount of data was collected during the course of
 this  survey.   The  results of the measurements and computer
 modeling  of  towers will be presented by individually discussing
 data  obtained  at each site.  Every transmitter site is different,
 and we found that  it  would be difficult, if not impossible, to
-find  an "ideal"  site  without any perturbing structures and with a
 "perfect" ground system.  Thus,  it is necessary to consider the
.actual layout  of each site including such  complicating factors as
 the presence of  fencing, walls, conductive objects, terrain
 obstructions,  etc.  This is especially true very  close to the
 antennas; i.e.,  within about ten meters.

 In all of the  figures that follow, measured electric or magnetic
 field values (after correction for calibration factors) are shown
 along with theoretical values obtained using the  NEC model.  The
 figures are  given  in  Appendix A through H; the appendix letter is
 the same  as  the  station code letter.  The  measured electric field

                                10

-------
values represent the total field; i.e., the resultant value
obtained from combining the vertical and radial electric-field
readings.  Both maximum and minimum electric-field values were
usually read at a given location.  The minimum readings were used
when calculating resultant electric fields in order to more
closely approximate the field due to an unmodulated carrier.
Formats for the various stations included both music and "all
talk" or "all news."   Table 1 (page 4} shows technical charac-
teristics of all the stations studied.  Note that, unless other-
wise indicated, tower height refers to height of the radiator
above the base insulator and does not include obstruction light-
ing.

5.1  STATION A

Station A transmits at a frequency of 1350 kHz from a single,
guyed tower at a power of 1000 watts.  The transmitter site is
essentially flat and dry with scrub vegetation typical of south-
ern California's central valley.  Ground conductivity in this
area is generally accepted to be about 4.0 mhos/nu  However,
ground conductivity is known to vary considerably in the valley.
The tower height is 54.9 m, and electrical height is approximate-
ly 0.25 wavelengths.  There are two chain-link fences around the
tower.  The first occurs close-in (about 1-3 meters from the
tower base) and forms a rectangle around the tower base along
with a small cinder block building, the closest wall of which is
about two meters north of the tower base.  The second fence is
about 20 m from the tower base at its closest location.  The
outer fence surrounds the tower in the shape of an equilateral
triangle with guy wires extending to just inside the vertices of
the triangle.  Both chain link fences are about 2 m high.  The
tower's ground system consists of 120 long (180 feet) equally-
spaced ground radials alternating with 120 short  (50 feet) ground
radials.  Figure 2 shows a diagram of the Station A site.

Field strength measurements were made in three directions radiat-
ing out from the tower base.  "Radial 1" extended in a westward
direction perpendicular to one side of the triangle formed by the
outer fence; readings were made out to a distance of 50 m.
"Radial 2" extended approximately toward the southeast, and
readings were made out to 100 m.  "Radial 3" extended eastwardly
toward one of the vertices of the triangular fence and in the
direction of one of the guy wires.

Figures A-l and A-2 show the results of electric-field measure-
ments made along the three radials.  Measurements made within 10
m of the tower base are plotted in Figure A-l to provide greater
resolution.  The remaining data, out to 100 m, are shown in
Figure A-2.  Predicted values obtained as a result of computer
modeling using the NEC program are also shown.  Measured values
ranged from a high of about 110 V/m  (1 m, Radial 3) to readings
of about 3-5 V/m beyond 50 m.

                                11

-------
             GATE
RADIAL 1
           CINDER
           BLOCK
          BUILDING
        1.2rfl
           t
2.4m
   .1 m
RADIAL 3
18.3m   E
   CHAIN
    LINK
   FENCE
  35.7m
Figure 2.   DIAGRAM OF STATION A SITE
               (NOT TO SCALE)
                    12

-------
The perturbing effect of the chain-link fences on field values
can be seen in both of these figures.  The fences were located at
approximately 1.2 and 19.5 m along Radial 1, at approximately 3
and 20.5 m along Radial 2, and at approximately 2 and 39 m along
Radial 3.  Readings were generally reduced near the fences.  For
example, along Radial 3 there was a large drop in field intensity
between the 1 m reading and the 3 m reading (Figure A-l).  The
reductions are not predicted by the NEC model of the tower, which
did not include a model of the fence.  Similarly, in Figure A-2
the dip in values along Radials 1 and 2 at the 20 m point corre-
lates with the position of the outer fence.

Figures A-3 and A-4 show results of magnetic field readings
around station A.  The NIC model clearly over-predicts field
values within 4-5 m of the tower base.  However, good agreement
is obtained farther out.  Measured values ranged from slightly
over 0.2 A/m to below 0.01 A/m.

Figures A-l through A-4 show that the measured field values for
both electric and magnetic field were generally lower than would
be expected based on the NEC model for distances within about 5 m
of the tower.  This overprediction is probably related, at least
partially, to the perturbing effect of the metallic fencing
surrounding the tower.  Although the electric field at the top of
a metallic fence (at 2 m height) would be expected to be en-
hanced, values measured immediately adjacent to the fences (at 1
in height) , especially on the sides away from the tower, were
generally reduced relative to nearby readings.  Farther out from
the tower there was generally good agreement between measured
field values and NEC predictions,

5.2  STATION B

Station B is a 1000 watt station transmitting from a single,
guyed tower at a frequency of 1410 kHz,  Tower height is 90.2 m
(electrical height about 0.42 wavelengths).  The station B
transmitter site is quite similar to that of Station A and is
only a few miles away.  The terrain is flat and dry, and there is
a single chain link fence around the tower.  The fence is about
3.5 m south and about 7.5m east and west of the tower base
forming a rectangle around it.  A metal building is within this
enclosed area about two meters north of the tower base.  Also
within the fenced area are two satellite dish antennas and a
metal tank.  The ground system of Station B consists of 240
alternating radials:  120, 50-foot radials interleaved with 120,
300-foot radials.  A diagram of the site is shown in Figure 3.

Measurements were made in two directions, one west and perpen-
dicular to the fence (Radial 1) and the other south and perpen-
dicular to the fence (Radial 2),  The chain-link fence is located
at approximately 7.5 m along Radial 1 and at approximately 3.5 m

                                13

-------
                CHAIN LINK
                  FENCE
             SATELLITE
             -x DISH
                SHED
               METAL
              BUILDING
                              7.5m
1.8 m A    3.5 m
GATE
       SATELLITE
         DISH
                                  TANK
                                            RADIAL 2
                              7.5m

                               RADIAL 1
    Figure 3.   DIAGRAM OF STATION B SITE
                   (NOT TO SCALE)
                        14

-------
along Radial 2.  Figures B-l and B-2  show results of electric
field measurements along these radials.  Calculated values using
the NEC model are also shown.

Good agreement between calculated and measured values for elec-
tric field strength is seen for distances beyond about 10 m. At
closer distances measured values of electric field were less than
predicted values by as much as 30 to 40 per cent.
Magnetic-field results are shown in Figures B-3 and B-4,  The NEC
model predicted magnetic fields greater than those measured at
distances less than 4 meters from the tower and predicted mag-
netic fields less than those measured at distances between 6 and
50 meters from the tower.  The maximum H-field reading was about
0.1 A/ni (at 1m).

5.3  STATION C

Station C is a directional station transmitting at 550 kHz from a
two-tower array.  Both towers are guyed, steel radiators of
uniform cross-section and heights of 99.1 m (0.18 wavelength).
The towers are series-excited and top-loaded with 30 feet of the
guy wires of each tower.  The two towers are separated by ap-
proximately 129 m.  Input powers during the measurements were
4300 watts (Tower 1) and 625 watts (Tower 2).   Only Tower 1 was
modeled.

The ground system at the site consists of 120 buried copper
radials that extend to the property line or to a common strap
between the towers.  Interspersed among these radials are 120
additional copper radials about 15 m long.  The terrain and
vegetation at the site are similar to that for Stations A and B.
In fact, all three sites are within a few miles of each other.

Steel reinforced cinder-block walls 2.6 m high and 0.19 m thick
surround each of the station C towers; also a small cinder-block
building containing the matching network adjoins each enclosure.
Tower 1 is near the center of a rectangle formed by the walls.
The outside dimensions of this rectangle are 3.25 by 3.45 m.  The
enclosure surrounding Tower 2 is similar.  An illustration of the
site is shown in Figure 4.

Measurements were made in each of three directions extending from
the base of Tower 1.  "Radial 1" extended away from Tower 1
toward Tower 2 approximately toward the northeast.  "Radial 2"
was directed to the northwest away from Tower 1 and approximately
perpendicular to Radial 1.  "Radial 3" extended away from Tower 1
in a direction opposite to that of Radial 1.

The results of measurements made along these radials are shown in
Figures C-l through C-4 in Appendix C.  In Figure C-l, an ap-
parent effect of the block enclosure can be seen by the sharp
drop in electric-field values along Radial 1 between the 1-meter

                                15

-------
        RADIAL 2
                                            BLOCK
                                            WALL
3.25 m	*~«-1.9 m
RADIAL 3
                                                      CINDER
                                                      BLOCK
                                                     BUILDING
                            BLOCK
                             WALL
         Figure 4.   DIAGRAM OF STATION C SITE
                        (NOT TO SCALE)

-------
and 2-meter measurements.  Both of these readings were made
inside the enclosure.  However, the 2-meter reading was made near
the inside corner of the wall.

Both of these measurements include contributions from significant
radial field components.  Although the vertical field component
predominated in the case of the 1-meter value, the radial
component was greater than the vertical component at the 2-meter
location.  A significant (and predominant) radial field component
was also measured along Radial l at the 4-meter point, just
outside the block wall.  However, at the 5-meter location the
vertical component again became predominant.

In both Figures C-l and C-2, predicted electric-field values were
in fairly good agreement with measured values past a distance of
about 5 m.  Closer in, predicted values were higher than measured
values (except at 1 m) for all three radial directions.

The effect of Tower 2 on electric field strength was not apparent
along Radial 1 until past the mid-point between the two towers.
A rise in field strength began to be detected at about 100 m and
field strength continued to rise (not shown in Figure C-2),
reaching a maximum value of about 27 V/m near the block wall
surrounding Tower 2 (distance of 126 m from Tower 1).  However,
as the block wall was approached the value dropped to about 12
V/m just outside the enclosure, showing, an apparent perturbation
caused by the wall.  However, inside Tower 2's enclosure, a
measurement of 152 V/m was obtained about 1 m from the tower.

Magnetic field strength results are shown in Figures C-3 and C-
4.  In Figure C-3 it can be seen that the NEC prediction for l m
is higher than the measured value along Radial 1 (about 3 A/m
versus 1.5 A/m).  This is in contrast to the 1-meter electric-
field value where the situation is reversed.  The apparent
perturbing effect of the block wall on the electric field, seen
in Figure C-l, -is not observed in Figure C-3, indicating that the
magnetic field is not significantly influenced by the enclosure.

In general, with the exception of certain locations along Radial
1, predicted and measured values for the magnetic field are in
reasonably good agreement 1  Closer to Tower 2 along Radial 1,
past the mid-point between towers, magnetic field strength began
to rise, reaching a maximum of about 1.3 A/m at a distance l m
from Tower 2 (not shown in the figures).  As with the Tower 1
enclosure, there was no apparent affect of Tower 2's block
enclosure on the magnetic-field readings.

5.4  STATION D

Station D transmits from a single, guyed, uniform cross-section
tower at 1440 kHz with 1000 watts of power.  The height of
Station D's tower is approximately 111 m  (electrical height 0.53

                                17

-------
wavelengths).  The tower sits atop a 1-meter high concrete
pedestal and 0.2-meter insulator, and is about 3.5m from the
corner of a small cinder-block transmitter building situated
approximately to the east of the tower.  A matching network in a
metal enclosure is located immediately adjacent to the tower
base.  Both the building and the tower are enclosed by a chain-
link fence approximately 2 m in height.  The ground inside the
fence is covered with loose gravel.  The terrain surrounding the
fenced in area is basically flat and dry with fairly dense scrub
brush.  The ground system consists of 120 radials, about 85 m
long, interspaced with 120 additional radials, each about 15.2 m
long.  Figure 5 is a diagram of station D§s site.

Measurements were made in three directions from the tower base.
Radials were designated as "Radial 1"  (approximately north), a
snort "Radial 2" (approximately south), and "Radial 3M (perpen-
dicular to these two and approximately west).  None of these
radials extended through the transmitter building.  The chain-
link fence is approximately 9.5 m from the tower along Radial 1,
about 5.8 m from the tower along Radial 2, and about 2.7 m away
along Radial 3.  A vertical metal pipe, approximately 0.15 m in
diameter, is located near the 4-meter point along Radial 1.

Measurement results and NEC predictions are shown in Figures D-l
through D-4 in Appendix D,  With regard to the electric field,
there was good agreement between measured and predicted values
beyond 2 m but poor agreement in the immediate vicinity of the
tower base  (0-2 m).  This was apparently due to the presence of
locally perturbing objects.  For example, the 1-meter reading
along Radial 1 was made directly next to coupling loops used for
tower lighting.

The effect of the chain-link fence on the electric field can also
be seen.  Along Radial 1, the fence was located between the 9 and
10 m, and the perturbing effect of the fence is evident in Figure
D-2 at 10 m.  The 15-meter reading was greater than the 10-meter
reading that was made just outside the fence.

The 9-meter reading incorporated a predominant radial field
component.  Significant radial field components  (i.e., at least
25% of the value of the vertical component) were also detected at
all points up to and including the 4-meter measurements.  At 3 m
and 4 m along Radial 1 (near the pipe mentioned previously) the
radial and vertical components were essentially of egual mag-
nitude.

Tangential field components were also measured at a few locations
near the tower base of Station D.  Generally, tangential field
components were negligible (under 3 V/m) except near locally
perturbing objects.  The maximum tangential component measured
was at the 1-meter location along Radial 1, where a reading of


                                18

-------
                   t
GATE    RADIAL 1
RADIAL 3
            SHORT
            METAL
            POLE
2.7m
CHAIN LINK
  FENCE
        9.5m
                       3.5m ^
                           METAL
                         ENCLOSURE
                     5.8m
                     RADIAL 2
Figure s.   DIAGRAM OF STATION D SITE
               (NOT TO SCALE)
                     19

-------
about 37 V/m. was obtained (compared to a vertical component of
780 V/m and a radial component of 280 V/m).

Magnetic field measurements for Station D are plotted in Figures
D-3 and D-4.  There was very good agreement with the predicted
values close-in to the tower as can be seen in Figure D-3.
However, Figure D-4 shows that at distances beyond about 25 m,
along Radials 1 and 3 (Radial 2 measurements only extended to 4
m) , there was an increase in magnetic field readings.  The NEC
curve also increased after about 40 m.  This result may seem
counterintuitive.  However, since these fields are determined
essentially inside the antenna system (in the reactive near
field) there is no reason why fields cannot increase with dis-
tance.  This result may be related to the tower's height being
close to one-half wavelength.

5.5  STATION E

The transmitting antenna used by Station E is a single, guyed,
uniform cross-section tower approximately 57 m tall.  The operat-
ing frequency is 1410 kHz, resulting in an electrical height of
about 0.27 wavelength.  Input power to the antenna was 4200 watts
during the measurements.  The surrounding terrain is flat and
dry, consisting for the most part of loose soil with little
vegetative growth outside of a few trees.  The ground system for
Station E consists of 120 equally spaced, buried copper radials,
each about 46-52 m long, interspaced with 120 additional radials,
each about 15.2 m in length.

Station E's tower is surrounded by a rectangular chain-link fence
with wide metal strips interwoven diagonally through the links.
The fence is about 1.5 in tall and there is about a 2-3 m clear-
ance between the tower and fence.  The tower sits atop a 0.25-
meter high concrete pedestal with an additional 0.45 m in height
provided by the insulator.  A metal tuning box, also inside the
fenced enclosure, is located about 0.5-1.0 m from the tower.  An
illustration of the site is shown in Figure 6.

Measurements were made in two directions  from the tower base.
"Radial 1" extended approximately north from the tower,, and
"Radial 2M was perpendicular to Radial 1  and in an eastward
direction.  Results are shown in Figures  E-l through E-4 in
Appendix E.

Agreement between measured and predicted values was generally
good for electric field values.  The effect of the metallic fence
on the electric field readings can be seen in Figure E-l, but the
perturbation was not as great as was seen for other stations.
The fence was located just inside the 4-meter point along Radial
1 and just outside the 2-meter point along Radial 2.  Further out
from the tower (Figure E-2) measured values tended to exceed
predictions, but differences were generally not great.

                                20

-------
 t
5,9m
GATE
                      t
                       RADIAL 1
                       3 m
                            2.5m
                   t
                 METAL
                 TUNING
                  BOX
                               RADIAL 2
                               CHAIN LINK
                                FENCE
                               WITH METAL
                                STRIPS
                  6.12m
       6.   DIAGRAM OF STATION E SITE
                 (NOT TO SCALE)
                      21

-------
In the immediate vicinity of the tower base, radial electric
field components were significant, as seen previously.  Along
Radial 1, the radial-field component at 1-meter was almost equal
to the vertical-field component.  There was also a small, but
significant, tangential-field component that was about 10% of
each of the other two components.  The vertical-field component
was predominant further out, but significant radial-field com-
ponents  (>25% of the vertical component) were detected along
Radial 1 at the 2-meter, 4-meter, and 5-meter positions.  Along
Radial 2, a similar situation occurred except that the radial
electric-field was actually greater than the vertical field until
a distance 1 m past the fence (i.e., until the 4-meter reading).
There were also notable tangential-field components at the 1-
jneter and 2-meter locations along Radial 2.

Excellent agreement between predicted and measured magnetic field
values was obtained for Station E (Figures E-3 and E-4) .  As
before, no perturbing effect due to metal fences was observed
with respect to magnetic field strength.

5.6  STATION F

Station F was the only high-powered AM station surveyed during
this study.  This station transmits with an operating power of
50,000 watts at a frequency of 1070 kHz.  The station format is
"all news," making measurement of unmodulated field values easier
due to the greater frequency of pauses in the modulation.

There are two towers on the site, a main antenna  ("Tower l") and
an auxiliary antenna ("Tower 2M).  Both are guyed, uniform cross-
section, steel radiators.  The towers are de-tuned (adjusted to a
non-resonant condition) with respect to one another and are
separated by a distance of about 105 m, with Tower 2 located to
the northeast of Tower 1.  The height of Tower 1  is about 150 m
(0.53 electrical height).  Tower 2 is slightly over 111 m (elec-
trical height of 0.40).  The ground system for Tower 1 consists
of 240, 152-meter radials.  A 15.2-meter radius ground screen is
under Tower 1.  Tower 2 also has a ground system  consisting of
240, 152-meter radials.  A 9.8 X 9.5 meter ground screen is under
Tower 2.  In the area between the two towers the  radials meet at
a common ground strap.

The land surrounding the two towers is flat and mostly covered
with grass.  The area is well maintained, since it is also used
as public park land.  Overhead power transmission lines are in
the area but are at least several hundred meters  from the closest
tower.  The ground was damp on the day measurements were made.

Tower 1 is surrounded by a relatively large cinder-block building
that would have made close-in measurements difficult and not very
useful.  Therefore, it was decided to switch transmitting power

                                22

-------
over to Tower 2 and make all radial measurements relative to it.
Tower 2 is in a more open area and there are fewer obstructions
to interfere with the measurements.

Tower 2 is surrounded by two separate chain-link fences, each 2-
3m high.  There is also a small cinder-block structure, about 3m
on a side, next to the tower.  The tower sits on a concrete
pedestal about 2 m high with an additional 0.65 m provided by the
insulator.  The distance from Tower 2 to the inner chain-link
fence is about 5.5-9.5 m, depending on direction.  The outer
chain-link fence is an additional 1.8 m beyond that.  The closest
wall of the block structure is about 1 meter from the tower, and
much of the tower base is enclosed by a fiberglass shield.
Station F's site is illustrated in Figure 7.

Before power was switched from Tower 1 to Tower 2, several
measurements were made inside the cinder-block building surround-
ing Tower 1, since station personnel spend a significant amount
of time inside this building, and it was desirable to record
typical field-strength values.  Ambient electric field strengths
were generally found to be less than 10 volts/meter.  This is not
surprising in view of the fact that the building has copper mesh
incorporated into the walls and roof.  Higher electric-field
readings could be obtained in certain locations very close to
transmitter cabinets, but significant exposure to personnel is
unlikely at those spots.  Electric-field measurements made inside
the small courtyard occupied by Tower 1, and surrounded by the
building, were much higher (e.g., in excess of several hundred
volts/meter approximately 1 meter from the tower).  This location
is rarely visited by personnel, and, if so, only for a short
time.

After switching from Tower 1 to Tower 2, measurements were made
in two directions.  "Radial 1" was directed to the south away
from Tower 2 and through the small block structure, although no
measurements were made inside the structure itself.  The first
reading along Radial 1 was made between the inner chain-link
fence and the wall of the structure farthest from the tower.
This point was 7.5m from the tower base.  "Radial 2" extended to
the west, perpendicular to Radial 1.  This radial did not pass
through the cinder-block structure, and the first measurement
point was 2 m from the tower base.

Results of measurements made along the radia-ls are shown in
Figures F-l through F-4 in Appendix F.  It is obvious from Figure
F-l that electric-field values measured close-in were signifi-
cantly less than those predicted from the NEC analysis.  This
could be due to perturbing effects of the small block structure
and the metallic fencing or, possibly, to the relatively tall
concrete pedestal upon which the tower rests (measurements were
made 1-meter above ground; the pedestal height was 2 m, see cover
photo), or to a combination of these factors.  It was noted that

                                23

-------
                                     CHAIN LINK
                                      FENCES
        RADIAL 2
         FIBERGLASS
           SHIELD
           IN REAR
                    2.5m
Tower #1
 (Main)
   CINDER BLOCK
     BUILDING
5.5m
Tower
  #2
  / CINDER
    BLOCK
    BUILDING
                             V9.5m
                       1.8m
       RADIAL 1
 Figure 7.   DIAGRAM OF STATION F SITE
               (NOT TO SCALE)
                     24

-------
the radial component of the electric field close-in to the tower
base constituted a less significant fraction of the total field
than was observed at other towers surveyed with lower pedestals.
The block structure also has copper mesh incorporated into its
walls, making electric field perturbations even more likely.

The effect of the fencing on electric-field values can be seen
clearly along Radial 1 in Figure F-2.  The inner fence is located
at about 9.5m and the outer fence at 11.25 m.  After the fences
had been passed there was a steep increase in electric field
strength that continued to about 15 m before leveling off and
finally decreasing.  However, even past the fences the measured
field values continued to be somewhat less than those predicted.
A similar pattern occurred along Radial 2, except the locations
of the inner and outer fences were 5.5 m and 8 m, respectively.
Radial field components were significant, with respect to the
vertical components, close-in to the tower base and near the
metal fences.

Figures F-3 and F-4 show that measured magnetic field values were
also less than predicted.  There was no noticeable effect of the
metal fencing on the magnetic field readings.

Readings were also made of the maximum detectable field strength
in any direction within 1-2 m of the base of Tower 2.  The
highest electric field reading obtained was 830 V/m, and the
highest magnetic field strength reading was 1.6 A/m.  Both of
these readings were obtained at heights above 1 meter and rela-
tively close to the top of the pedestal upon which the tower
rests.

5,7  STATION G

Station G transmits at a frequency of 790 kHz and power of 5000
watts.  The station is non-directional in the daytime, and
directional at night.  Two towers are used for nighttime trans-
mission, but only one tower is used during the day.  All measure-
ments were made during the day when one tower ("Tower 1") was
operational.  The other tower ("Tower 2") is approximately 73.5 m
from Tower 1 to the northwest.

Tower 1 is a uniform cross-section, guyed tower about 146 m high
(electrical height of 0.39).  It sits atop a 0.6-meter high
concrete pedestal and 0.75-meter high insulator.  Tower 2 is a
tapered, self-supporting tower about 85 meter high  (electrical
height of 0.22 wavelengths).  Tower I is located approximately at
the center of a rectangular area enclosed by a chain-link fence
about 1.5 m high.  Dimensions of the fenced enclosure are about
15 m on a side.  A steel-encased tuning box is located 1-2 m from
the base of Tower 1.  Figure 8 is a diagram of the site.
                                25

-------
 Tower #2
 (Night Only)
   CHAIN LINK
    FENCE
                        i,
       RADIAL 2
                                  TUNING
                                  /BOX
7.6 m
                                         --GATE
^Tower
                              7.6m
                              RADIAL 1
Figure s.   DIAGRAM OF STATION G SITE
               (NOT TO SCALE)
                    26

-------
The ground system for Tower 1 consists of 120 equally-spaced,
buried, copper radials extending to the edge of the property.
The ground system for Tower 2 consists of 120 equally-spaced,
buried copper radials extending about 91.5 m or to the edge of
the property.  There is a 15.2-meter square copper ground screen
at the base of Tower 1 and a 10.7-raeter square screen at the base
of Tower 2.  Both ground systems are connected to a common ground
strap.  The surrounding terrain is flat and dry with some scrub-
type vegetation.

Measurements were made along two radials extending out from Tower
1.  "Radial 1" extended to the south, and "Radial 2" extended to
the west.  Results are presented in Figures G-l through G-4 in
Appendix G.

With a few exceptions, there was very good agreement between
measured and predicted values for electric field strength.  From
Figure G-l it can be seen that there was some variation between
predicted and measured values close-in (within 3 m) to the tower
base.  Also, in Figure G-2, lack of agreement was notable in the
10 to 15 m range, with actual values significantly less than
predicted values.  This anomaly could have been due to the
perturbing effect of the chain-link fence that was located at
about 7 to 8 m along each of the two radials.  Although not shown
in the figures, a peak electric-field reading of about 700 to 870
V/m was obtained within 2 m of the tower base near-the antenna
feed line.

With respect to magnetic field strength, Figure G-3 shows fairly
good agreement between measured and predicted values close-in to
the tower base.  The only exception was the 1 m location where
the NEC model over-predicted measured values.  Figure G-4 shows
that past the 10 m location the NEC model under-predicted the
measured values.

5.8  STATION H

Station H was the only station where field strength measurements
were made relative to a self-supporting tower.  Station H is a
non-directional station transmitting with 1000 watts, daytime, on
a frequency of 1450 kHz.  Although only the main tower was
transmitting during the measurements, there is also a (de-tuned)
auxiliary tower on the property about 63 m away from the main
tower.

The main tower is a triangular, self-supporting, steel radiator
approximately 70.1 m in height (electrical height of 0.34 wave-
length) .  It rests atop three insulators  (one per tower leg) 1.0
m above ground-level.  The tower is located near the end of a
paved  (asphalt) parking lot, although the area immediately
surrounding the tower is unpaved and enclosed by a chain-link
fence that is approximately 2 m high.  The overall dimensions of

                                27

-------
the fenced enclosure are about 6 by 7 m.  A paved road surrounds
the station lot, and one- or two-story buildings line most of the
other side of this road.  A tower used for microwave and land-
mobile antennas is nearby, about 50 m to the west.  The ground
system consists of 120 equally-spaced, buried, copper radials
53.3 m in length plus an additional 120 interspaced radials 15.2
m in length.

The triangular base of the tower is approximately 3.5 m on a
side, and the tower is situated eater-cornered inside the fenced
enclosure.  The tower tapers gradually from the base up to about
the mid-point and is then uniform in cross-section to the top.
Figure 9 shows a diagram of the Station H site.

Radial measurements were made in two directions, as shown in the
Figure.  "Radial 1" was directed to the south of the main tower,
in the direction of the auxiliary tower, and "Radial 2" extended
toward the north-northeast.  Both radials extended over asphalt
pavement.  However, at the origin for the radials between the
tower legs the lot was unpaved.  A tuning box is also located
under the tower legs.  Results are presented in Figures H-l
through H-4 in Appendix H.

As can be seen in Figure H-l, there was not good agreement
between measured and predicted electric-field values for dis-
tances within a few meters of the tower base.  However, agreement
was better beyond about 20 m  (Figure H-2).  This lack of agree-
ment close-in was probably due to the difficulty in modeling a
self-supporting tower such as this.  The drop in measured elec-
tric field values between the 3 and 4 m points was likely due, at
least in part, to the chain-link fence that is found at approx-
imately 3-4 m along the radials.  Although not plotted, a peak
reading of about 300 V/m was  found at the tower base under the
tower legs.

Radial components were significant, but not predominant, up to
about 5 m along Radial 2,  However, the radial component along
Radial 1 was predominant at the 3 and 4 m points on either side
of the chain-link fence.  Tangential components were measured at
2 m and 3 m and were also found to be significant components of
the total field.

There was poor agreement close-in to the tower  (less than 3 m)
with respect to vthe magnetic  field, as  Figure H-3 shows.  Better
agreement occurs beyond 10 m  (Figure H-4).  The peak magnetic
field measured  (not plotted)  was a value of about 0.4 A/m,
obtained under the tower legs.  A similar reading was made at 1 m
along Radial 1.
                                28

-------
         RADIAL 2
              3 m
             3.7m
         RADIAL 1
                       CHAIN LINK
                         FENCE
                    Tower ft
                     (Main)
                     63.4m
                            Tower #2
                         y\ (Auxiliary)
Figure 9.   DIAGRAM OF STATION H SITE
               (NOT TO SCALE)
                     29

-------
6.  DISCUSSION AND CONCLUSIONS

The data collected during this study can be analyzed in a number
of different ways.  The following discussion covers various
topics related to the analysis of electromagnetic fields near AM
broadcast towers.  Of particular interest is how the measurement
data and the computer modeling techniques used in this study can
help in more accurately predicting fields near AM towers.

6,1  NEC MODELING AND AGREEMENT WITH MEASUREMENTS

Visual inspection of the figures in this report shows that, in
general, agreement between measurement data and values predicted
by computer modeling was better at larger distances from a tower
than closer in to the tower.  Unfortunately, the region close to
the tower base is of greatest interest because this is where the
highest field exposures can occur.

Figures 10 and 11 summarize data on the difference between mea-
sured and calculated fields versus distance for all the stations'
data combined.  To develop the graphs the average measured field
value was first determined for each station at each distance by
dividing the sum of measurements on different radials by the
number of measurements.  The absolute value of the difference
between calculated NEC value and the average measured field
divided by the average measured field times 100% gives the
percentage error for each distance at each station.  Finally, the
average of these errors over all the stations was used to gen-
erate the plots of error versus distance.  Generally, errors are
largest for electric fields at distances less than 15 meters and
largest for magnetic fields at distances less than 2 meters.

Two problems contribute to errors in calculating fields very
close to the tower base.  One is fundamental and is due to the
approximations in modeling the base region of the tower.  The
second problem is incidental and is due to the presences of field
perturbing objects near the tower base such as fences and tuning
networks.  If a fence is far away from the tower its perturbation
of the fields is probably just as great but is unlikely to be
seen in the data because measurements are not apt to be made near
it.

The antenna tower as a field source may be fundamentally viewed
as a series of radiofreguency current elements.  Fields due to
current elements decrease with the inverse cube of the distance.
Because of this, at points where some elements are much closer
than others, those close elements dominate the field.  This
implies that at locations close to the tower base only the base
region is the field source, whereas at greater distances the
entire tower is the source.  Since the base region is the most
poorly modeled part of the tower, large errors are seen when


                                30

-------
                  Electric Field  Error Analysis (L-lOm)
      a

      •2
      13
      -

      1!
      :i
      a
         150
       
-------
          i
                    Magnetic Field Error Analysis (I-10m)
                                <     J     «

                                 Distance (m.)
                Magnetic Field Error Analysis (I0-100m)
                            Distance (m)
Figure  11.   Magnetic  Field Error Analysis
                               32

-------
approaching the base, especially when the scale of measurements
is small compared to the size of the tower base as for station F.

Detailed modeling of the base region is not practical using NEC.
The NEC rules on wire diameter and length are restrictive for
this application.  The effort to match calculated impedances to
the measured base impedance was intended to better approximate
the known voltage to current ratio at the base at the expense o-f
accurately modeling the height of the tower.  The correct base
impedance was assumed to lead to better modeling of fields near
the base, while small changes in tower height would only affect
fields further away and of less interest.  This approach was only
partially successful.  It has also been suggested that a capaci-
tance be added across the feed point to better model the electri-
cal effects of the detailed base geometry; this has not been
attempted here.  Other possibilities with NEC include modeling
fences as vertical wire segments and modeling guy wires.

6.2  QUASI-STATIC MODELING OF CLOSE-IN FIELD STRENGTH

The immediate base region, nearby conductive fences and other
field perturbing objects close to the tower base are small
compared to the wavelength at AM frequencies.  This implies that
a quasi-static approximation could be used to determine fields
near the tower base.  Qualitatively, at least, the effect of
conductive fences on electric fields can be understood using
electrostatic reasoning.  A metal fence may be considered an
equipotential surface at ground potential.  At large distances
from the tower the fence perturbs the horizontal equipotential
surfaces such that the electric field is normal to the fence
surface, decreases on both sides of the fence, and increases
above the fence.  A fence close to a tower brings ground poten-
tial closer to the high potential tower and increases electric
fields between the tower and fence but shields regions outside
the fence.  This has the practical result that conductive fences
of adequate height around a tower should reduce electric fields
outside the fence, but may increase occupational exposure inside
the fence, as well as disturb the tower impedance by increasing
the capacitance between the tower and ground.  This result is
consistent with measurements in this study.

A quasi-static approach to magnetic fields assumes that fields
are due to electric currents or magnetized material and the speed
of propagation i§ not considered.  For the case of a conductive
fence surrounding a tower the eddy currents induced in the fence
are probably small because the magnetic fields are circum-
ferential around the tower and generally tangential to the fence
surface.  If the fence is also magnetic these effects are assumed
to be localized and secondary to the eddy current effects.  In
either case the effect of fencing on magnetic fields would be
small and this is consistent with the results of the study.
                                33

-------
Simple quasi-static approximations can be used to estimate fields
close to the tower base.  Since for AM stations the complex
impedance (primarily a function of electrical height)  and power
are measured at the feed point, the current and voltage at the
tower base can be easily calculated.  The current is given by the
square root of the power divided by the real part of the impe-
dance, I=yf"/Re(2).  The voltage is given by the current multi-
plied by the magnitude of the impedance.  The voltage can be
divided by the distance between a surface on the tower (above the
insulator) and a grounded surface (below the insulator) to
estimate the average electric field between those two points.
This approach was used to calculate electric fields close to
(mostly at 1 m) the eight stations in this study.  The distance
chosen to divide the voltage is somewhat arbitrary, it was chosen
to be the distance from the bottom of the radiator to the ground
beneath the measurement point.  This distance is calculated by
adding the pedestal height to the insulator length, squaring this
number, adding the square of the horizontal distance from the
center of the tower to the measurement point and taking the
square root of this quantity.  The results of these calculations
are shown in Table 4.  The average electric field measured and
the value calculated by NEC are shown for comparison.  It is not
clear that the NEC calculations are "better11 than this simple
quasi-static estimate at close-in locations.  At least, the
quasi-static numbers are always greater than the average measured
values.

Similarly, the magnetic field can be estimated near the base by
dividing the current by the circumference of a circle around the
tower at the distance of interest.  At the 1 m distance the
circumference is 6.28 m and at the 2 m distance the circumference
is 12.6 m.  The results of this calculation are shown in Table 5.
The average magnetic field measured and the value calculated by
NEC are shown for comparison.  The quasi-static values are
essentially the same as the NEC values except for the self-
supporting tower H, where only the quasi-static number is in good
agreement with measurement.

These results show that the quasi-static approach needs to be
explored further.  It should be emphasized that quasi-static
methods must fail at large distances from the tower.  However,
detailed quasi-static modeling by computer of the base region
including perturbing objects may be able to calculate close-in
fields with high accuracy.  In any case, measurement is probably
the only economical approach to accurately determine fields near
the base of AM towers,

6.3  FACTORS AFFECTING CLOSE-IN FIELD STRENGTH

The question of whether the electric or magnetic fields predomi-
nate near a tower base is important if only one of the fields is
known.  In this case, simple rules based on electrical height

                                34

-------
                       TABLE 4.   QUASI-STATIC ELECTRIC FIELD RESULTS
Station

   A
   B
   C
   D
   E
   F
   G
   H
Base
Voltage
rv>
475
956
964
938
1061
6179
1978
550
Base to Ground
Distance
fin)
1.43
1.46
1.66
1.56
1.22
3.32
2.03
1.41
                                       Quasi-Static
                                      Electric  Field
                                           332
                                           655
                                           581
                                           601
                                           870
                                           1861
                                           974
                                           390
                                                 NEC
                                            Electric Field
                                                 (V/m)

                                                   109
                                                   348
                                                    89
                                                   262
                                                   333
                                                 3289
                                                   739
                                                   125
                             Average Measured
                              Electric Field
                                   fV/m)	

                                   109
                                   271
                                   200
                                   500
                                   311
                                   491
                                   499
                                    95
NOTES:  Calculations and measurements  are  at  1  meter except for Station G at 1.5 m and
Station F at 2 m.  Average measured  field  is  the  average  of values obtained along one or
more radials.
                       TABLE 5.  QUASI-STATIC MAGNETIC FIELD RESULTS
Station

   A
   B
   C
   D
   E
   F
   G
   H
Base Current
    fAl	

     4.50
     1.23
    13.9
     3.58
     7.72
    15.9
     2.73
     1.91
                          Quasi-Static
                         Magnetic  Field
                             0.716
                             0.196
                             2.21
                             0.570
                             1.23
                             1.27
                             0.434
                             0.304
     NEC
Magnetic Field
    (A/mi

    0.68
    0.18
    2.2
    0.55
    1.23
    0.88
    0.43
    0.04
Average Measured
 Magnetic Field
     (h/m)	

     0.22
     0.10
     1.5
     0.53
     1.17J
     0.51
     0.32
     0.31
NOTES:  Calculations and measurements  are  at  1  meter  except  for Station  F at 2  m.
measured field is the average of values  obtained along  one or more  radials.
                                                                                   Average

-------
that determine whether the ratio of electric to magnetic fields
is high or low relative to the free space value of 377 ohms would
be useful.  The magnetic field is expected to be roost significant
near the base of towers with heights close to one-quarter wave-
length since a current maximum exists at or near the base of the
tower.  On the other hand, the electric field should predominate
in the immediate vicinity of the base of towers with heights
close to one-half wavelength since a voltage maximum would exist
at or near the tower base.

This relationship can be examined among the eight stations that
were the subjects of this study.  In Table 6, average values of
electric and magnetic field strength (measured at one meter,
unless otherwise noted) are given for the various stations in
order of increasing electrical height.  Electrical height ranged
from 0.18 to 0.53 wavelength.  As expected, with increasing
electrical heights in this range there is a general qualitative
trend toward a higher electric field strength and lower magnetic
field strength near the tower base.  This can be seen in the last
column in the table where ratios are given for values of E and H
(field impedance).

This observation is supported by the results of a recent study of
electromagnetic fields near quarter-wavelength AM towers done for
the FCC by R. A. Tell Associates [6].  Tell's results indicated
that magnetic fields are more likely to exceed safety limits (in
which electric and magnetic fields are related at the free space
ratio as in the ANSI C95.1 - 1982  RF protection guides) than
electric fields close-in to quarter-wavelength towers.

Another means of analyzing the data is to compare measurements
from different stations according to their operating frequency
and electrical height.  These comparisons are useful for illus-
trating certain patterns that can be detected from the data.
Table 7 shows measurement data from five stations arranged in
four categories.  Field-strength readings  (normalized) are listed
for four representative distances as averaged from the two or
three radials measured with respect to each station.  Note
magnetic field units are mA/m.

Data listed under category (1)  (similar frequency and different
electrical height) show, as expected, relatively greater magnetic
field values near the quarter-wave towers  (Stations A and E) than
near the towers with electrical heights closer to one-half
wavelength (Stations B and D).

Category  (2) in Table 7 compares data from stations with similar
electrical height but different frequencies.  There is some
difference in electric field readings close-in, but at greater
distances relatively little difference is seen.
                                36

-------
              .TABLE 6.   ELECTRIC AND MAGNETIC FIELD STRENGTH
                     AS  A FUNCTION OF ELECTRICAL HEIGHT

          (Measured in the immediate vicinity of each tower base)
STATION' ELEC HT
C 0.
A 0.
E 0.
H ' 0.
G 0.
F 0.
B 0.
D 0.
18
25
27
34
39
40
42
53
POWER
(kW)
4
1
4
1
5
50
1
1
.3
.0
.2
.0
.0
.0
.0
.0
AVG E
(v/m) •.
200
109
311
95
499 •...•-;
(1.5 ni'j
491
(2 m)
271
500
AVG H
(A/m)
1.
0,
1.
0.
: 0.
0.
(2
0.
0.
5
22 '
17
31
32
51
m)
10
53
E*
(V/m )
96
109
152
95 :
223
69
271
500
H*
(A/m)
0,
0.
0.
0.
0.
0.
0.
0.
72
22
57
31
14
07
10
53
E/H
(ohms)
133
495
' 266
306
1559
963
2710
943
NOTES:

     (1)  AVG E and AVG  H values are  averages  obtained along the various
radials at 1 m  (unless otherwise  noted)  from each respective tower.

     (2)  E* and H* are normalized to 1.0 kilowatt of power.
                                    37

-------
Category (3) repeats the data from  (1) but for a different
comparison, similar frequency and similar electrical height.
Quarter-wave towers (Stations A and E) with similar frequencies
should show generally consistent field values along the measured
radials.  However, some differences are noted, particularly for
close-in electric field readings.  With respect to the half-wave
towers (Stations B and D) there is some discrepancy in the close-
in magnetic field data, but otherwise differences were not very
great.

In category (4) of Table 7, stations with different frequency and
different electrical height are compared  (A and E versus G).
Significant differences in relative field readings can be seen,
with the magnetic field predominating close-in for the quarter-
wave towers and the electric field predominating close-in in the
case of the relatively low-frequency, half-wave tower.

     TABLE 7.  COMPARISON OF DATA FROM STATIONS ACCORDING TO
                 FREQUENCY AND ELECTRICAL HEIGHT

    STATION            AVG E-FIELD  fvVm)      AVG H-FIELD  fmA/m)
                      3m    5m   10m   25m    3m   5m   10m   25m

(1) Similar frequency f different electrical height:

E*(1410 kHz, 0.27)   23.9  16.8  11.3  7.0    208  134  68.5  31.3
B  (1410 kHz, 0.42)   59.2  25.0  15.8  7.2    48.4 32.6 21.7  15,6

A  (1350 kHz, 0.25)   10.5   8,1   6.4  5.4    185  130  74.4  27.2
D  (1440 kHz, 0.53)   60.6  32.7  15.6  8.2    176  99.0 34.8  3.48

(2) Pi f f erent  frequency, iimilar electrical jheight;

G* (790 kHz, 0.39)   94.4  49.6  15.4  7.3    72.8 48.9 29.0  15.3
B  (1410 kHz, 0.42)   59.2  25.0  15.8  7.2    48.4 32.6 21.7  15.6

(3) Similar frequency, similar electrical. JieJ-ght;

A  (1350 kHz, 0.25)   10.5   8.1   6.4  5.4    185  130  74.4  27.2
E*(1410 kHz, 0.27)   23.9  16.8  11.3  7.0    208  134  68.5  31.3

B  (1410 kHz, 0.42)   59.2  25.0  15.8  7.2    48.4 32.6 21.7  15.6
D  (1440 kHz, 0.53)   60.6  32.7  15.6  8.2    176  99.0 34.8  3.48

(4) Different  frequency, different  electrical height;

A  (1350 kHz, 0.25)   10.5   8.1   6.4  5.4    185  130  74.4  27.2
E*(1410 kHz, 0.27)   23.9  16.8  11.3  7.0    208  134  68.5 31.3

G* (790 kHz, 0.39)   94.4  49.6  15.4  7.3    72.8 48.9 29.0 15.3

* = data normalized for  1.0 kilowatt  power

                                38

-------
Although most of the stations participating in this study operate
at 1 kilowatt of power, a few operate at higher power  levels.   It
is instructive to compare data from the 1 kilowatt stations with
data from the higher powered stations.  For example, field
strength measurements  from a I kilowatt station may be useful  for
predicting the magnitude of electric and magnetic fields near  a
higher powered station with a similar frequency and electrical
height.  This could be accomplished by multiplying measured
values from the 1 kilowatt station by the square root  of the
power of the high power station.
         TABLE 8.  COMPARISON OF MEASURED FIELD STRENGTH
         BETWEEN  STATIONS WITH DIFFERENT OPERATING POWERS

                                          Histance_ from tower*
                                         2m     5m     10m   50m
STATION B
1.0 kW, 1410 kHz,
0.42 A

STATION D
1.0 kW, 1440 kHz,
0.53 A
     E-FIELD(V/m):   105     25     16      4
     H-FIELD(A/m):  .059   .033   .024   .012
     E-FIELD(V/m):  124     33     16      3
     H-FIELD(A/m): .296   .099*   .035    .010
AVERAGE FOR B AND D:
     E-FIELD(V/m):  115     29     16      4
     H-FIELD(A/m): i 178   .066   .030   .011
STATION G
5.0 kW
790 kHz
0.39 X
STATION F
50.0 kW
1070 kHz
0.40 A
     E-FIELD(V/ra);  361    111     34      9
 AVG(B&D)*/5"(V/m) :  257     65     36      9


     H-FIELD(A/m): .291   .114   .069   .025
 AVG(BScD) **/!*(A/m) : .398   .148   .067   .025
     E-FIELD(vym) :  436    169     44     26
AVG(B&D)*y50"(V/m) :  813    205    113     28

     H-FIELD(A/m); .505    .293    .207    .075
AVG(B&D)*/5Q"(A/m) : 1.259    .467    .212    .078
*NOTE:  Actual field strength values  were  taken from the measured
data from Stations B,  D,  F,  and  G  at  the indicated distances
along 1, 2, or 3 radials,  depending on  the station.  An average
reading was used if 2  or  3 radials were measured.
                                39

-------
 To test this proposition, Table 8 was constructed.  This table
 shows a comparison of data from Stations B and D  (both 1 kilowatt
 of power with similar frequencies and electrical heights) with
 actual and "expected" field strength values from Station G (5
 kilowatts) and Station F  (50 kilowatts).  Stations G and F both
 have elctrical heights that are not very different from Stations
 B and D, but the frequencies are not similar.  The "expected"
 field strength values for Stations G and F were obtained by
 multiplying average values from Stations B and D by the square
 root of 5 or 50, as appropriate.  Field strength comparisons were
 made at 2, 5, 10, and 50 meters along the radials measured from
 each tower base.

 An interesting observation from Table 8 is the generally good
-• agreement between measured data and the extrapolated, 1-kilowatt
 values beyond 10 m from a tower.  However, agreement was not so
 good within 10 m, illustrating the variability of measurements
 very close-in to a tower base.  In the case of Station G, ex-
 pected E-field values were exceeded by actual values within 10 m
 while H-field values were lower than might be expected.  On the
 other hand, actual E and H values from Station F within 10 m were
 lower than those that might be expected from the data for the l-
 kilowatt stations.

 6.4  PERTURBATION OF FIELDS DUE TO OBJECTS NEAR A TOWER BASE

 Our results show that conductive objects close to the base of an
 AM tower can have a significant effect on the electric field
 strength in the immediate vicinity.  Chain-link fences, in
 particular, had a noticeable effect on electric field strength.
 Measurements made on either side of chain-link fences near towers
 showed an attenuation in the total electric field.  The vertical
 component of the electric field was particularly reduced near
 these fences, while radial field components tended to be enhanced
 relative to values farther away from the fence.  This is consis-
 tent with an electrostatic model of the fence as an equipotential
 surface held to ground potential.  We did not obtain data on
 electric field strength directly above the fences.  However, a
 significant enhancement would be expected.

..As discussed by Tell [6], this effect of chain-link fencing is
vdue to the tendency of electric field lines to terminate on
 metallic fences which are at a lower (ground) potential.  This
 results in a lower value for the total electric field near the
 fence than would be found if the fence were not present.  Tell
 also found that in cases where wooden fences surround tower bases
 there was virtually no perturbation in the electric field.

 A good example of the effect of metallic fences was observed at
 Station F.  station F is a relatively powerful AM station,
 transmitting at 50 kilowatts, and we had expected to find sig-
 nificantly higher close-in field strengths.  However, the two

                                 40

-------
chain-link fences that surround the tower base apparently were
the primary cause of our results showing values considerably
lower than predicted.  Only when we measured very close to the
tower base (about 1 meter) and substantially within the fenced
area did we obtain higher readings that were closer to those that
would be expected.

Chain-link fences are not the only objects that can affect field
strength.  At other station locations we found that any large
metallic object can alter electric field readings, particularly
if such objects are relatively close to the tower base.  As for
magnetic field readings, they can be affected by the presence of
nearby tuning coils or inductive loops.  However, as also noted
by Tell [6],  magnetic field readings are generally•• more stable
and less susceptible to the local field perturbations seen for
electric fields.

6.5  VARIATION OF FIELDS ALONG DIFFERENT RADIALS

A comparison of measured values obtained at the same distance
from a tower base, but in different radial directions, also
suggests the perturbing effect of the environment in the im-
mediate vicinity of the tower.  In general, there appeared to be
better agreement from radial to radial farther out from a tower's
base than closer in, at least for electric field measurements.
In an attempt to illustrate this, Table 9 was constructed using
data from four single-tower stations.  The table gives field
strength values obtained at the same distances along radials from
each tower.  At distances beyond 10 m the differences tend to be
less, in general, than the differences closer in.  This table
shows that the direction in which measurements are made can make
a difference in readings close-in to a tower.  Differences tend
to be more noticeable for electric field readings than for
magnetic field readings.  This may be related to the electric
field readings being more susceptible to perturbation by conduc-
tive objects near the tower base that may be encountered along
one radial but not along others.

Tell made similar observations in his study [6],  He obtained
variations in field strength readings at fixed distances around
AM towers as a function of direction from the tower.  As Tell
pointed out,  this implies that field strength measurements along
a single radial may hot be representative of maximum exposure
levels.  Along with field perturbation by conductive objects and
tuning structures, Tell suggested that disturbances in a tower's
ground system may be a factor in explaining the differences in
readings.  Tell proposed that measurements to show compliance
with safety limits be made along the perimeter of controlled
areas near AM towers in order to be certain of locating maximum
field levels.
                                41

-------
       TABLE 9.  VARIATION OF FIELD STRENGTH VALUES AS A FUNCTION OF
                  RADIAL  DIRECTION  FROM SINGLE  AM  TOWERS*
                                   DISTANCE ALONG RADIAL  (m)
No.
Station Radials
E FIELD:
(V/m) A 3
B 2
D 3
E 2
H FIELD:
(A/m) A 3
B 2
D 3
E 2
2
15.9
107.1
102.9
124.7
133.1
114.4
124.6
164.4
0.23
0.06
0.06
0,28
0.31
0.27
0.56
0.68
3
13.6
8.3
9.6
62.3
56.2
73.5
34.6
73.8
49.0
0.18
0.23
0.14
0.04
0.05
0. 16
0.18
0.42
5
10.5
5.6
8.2
27.0
23.1
28.6
36.9
31.4
37.5
0.13
0.14
0, 12
0.03
0.03
0.10
0.09
0.27
0.28
10
7.4
6.0
5.8
13.9
17.7
8.7
22.6
22.6
23.6
0.07
0.07
0.07
0.02
0.02
0.03
0.04
0.14
0.14
Hf-m
25
5.5
6.0
4.7
7.2
7.2
7.5
8.9
13.8
14.8
0.03
0.03
0.02
0.02
0.02
o.oi
0.01
0.06
0.07
50
4.9
5.3
4.5
3.9
4. 1
3.4
3.0
11.9
9.9
0.02
0.02
0,02
0.01
0.01
0.01
0.01
0.03
0.03
 At .the  indicated distances from a tower's base,.measured values are listed
for each of the respective  radials.  Values have  been  rounded off.
                                   42

-------
6.6  AGREEMENT WITH VALUES GIVEN IN FCC BULLETIN 65

One of the reasons for this study was an attempt to determine the
accuracy of some of the predicted values for AM electromagnetic
fields given in the FCC's OST Bulletin No. 65 [1].  Table 1 on
page 49 of this bulletin gives distances at which fields from AM
stations are predicted to fall below various field strengths.
That table is reproduced here as Table 10,  It should be kept in
mind that the values given in this table were based on NEC
computer models and represent "worst-case" situations.  They were
intended to apply to any station, regardless of the tower height
or frequency.  Therefore, it was expected that in most cases the
values in the table would be conservative.  The results of this
study generally confirmed that assumption.

Table 10 may be used to obtain distances necessary for compliance
with the exposure guidelines of the American National Standards
Institute (ANSI) that are used by the FCC for purposes of evalua-
ting RF radiation in the environment (7).  According to Table 10,
all of the stations in this survey should be candidates for
exceeding the ANSI guidelines at various distances from their
respective tower bases.  In fact, however, we found that only at
Stations D and F did we obtain readings that actually exceeded
the ANSI limits.

In Table 11, values from Table 10 are compared with actual
distances where the indicated electric and magnetic field streng-
ths were measured or were interpolated from our data.  The
measured distances given in the table represent the maximum
distance observed from among the two or three radials surveyed
for each station.  For simplicity, distances obtained from our
data were rounded to the next highest whole number.

From Table 11 it can be seen that the recommended distances from
Table 10 are generally conservative with regard to the eight
stations studied.  An exception occurred with regard to Station C
where the magnetic field distance at 0.06 A/m exceeded the recom-
mended distance.  However, this reading was obtained on a radial
between two active towers, and there may have been a significant
contribution from the second tower to the magnetic field reading
at the measurement location.

Tell also observed that the values in Bulletin 65 tend to be
conservative.  His report included a table similar to Table 11
for the four stations he surveyed.  In no case was a recommended
distance exceeded by the measured distance for a given field
strength level.

Even though our results and those of Tell illustrate the conser-
vative nature of the values given in Bulletin 65, it should be
emphasized again that those values were meant to represent
"worst-case" approximations and were intended to apply to any

                                43

-------
                                TABLE 10.  (From OST Bulletin No. 65)
                       DISTANCES  (IN METERS) AT WHICH  FIELDS  FROM AM STATIONS
                         ARE PREDICTED TO FALL BELOW VARIOUS FIELD STRENGTHS
                                          (*See notes below)
Electric Magnetic
Field Field
Strength Strength 50.00
fV/m) (A/nO
25
50
75
100
: 150
200
300
400
500
632 (ANSI)
750
1000
*Notes: 11)
0.06
0.13
0.19
0.25
0.38
0.50
0.75
1.00
1.25
1.58 (ANSI)
1.88
2.50
This table
109
65
49
40
30
25
20
16
14
12
11
9
can be
25.00 10.00
83
51
38
31
24
20
16
13
11
9
8
7
used
60
37
28
23
18
15
11
9
8
7
6
5
for any AM
Transmitter Power (kW)

5.00
47
29
23
19
15
12
9
7
6
5
5
4
frequency

2.50
37
23
18
15
11
9
7
6
5
4
4
3

1.00
27
18
13
11
8
7
5
4
3
3
3
<2
or electrical

0.50
22
14
11
9
6
5
4
3
3
<2
<2
<2
height.

0.25
18
11
8
7
5
4
3
<2
<2
<2
<2
<2


0.10
13
8
6
5
4
3
<2
<2
<2
<2
<2
<2

         (2)  The entries in this table apply to both electric  field  strength  and  the  corresponding
magnetic field strength (assuming impedance of free-space  equals  400  ohms).

-------
          TABLE  11.   COMPARISON OF RECOMMENDED AND MEASURED DISTANCES FOR FIELD STRENGTHS
                  USING  VALUES  FROM OST BULLETIN 65 AND DATA COLLECTED IN THIS STUDY

              (Order  of  entries is:   OST 65 distance/E-field distance/H-field distance)

                                                   STATION
E
(V/m)
25
50
75
100
150
200
300
400
500
632
750
1000
H
(A/m)
0.06
0.13
0.19
0.25
0.38
0.50
0.75
1.00
1.25
1.58
1.88
2.5
A B
27/3/12 27/7/3
18/3/6 18/4/-
13/2/4 13/3/-
11/2/3 H/3/-
8/-/- 8/2/-
"!/-•/- 7/2/-
5/-/- 5/-/-
4/-/- 4/-/-
3/-/- 3/-/-
3/-/- 3/-/-
3/-/~ 3/-/-
f
C
44/9/65
27/2/23
22/2/16
18/2/12
14/2/10
11/1/8
a/-/ 6
7/-/4
6/-/3
5/-/-
5/-/-
4/-/-
D
27/7/8
18/4/5
13/4/4
11/3/3
8/2/2
7/2/2
5/2/-
4/2/-
3/2/-
3/2/-
3/2/-
<2/-/-
E
44/10/29
27/4/12
21/4/9
18/3/6
14/3/4
11/2/3
8/2/2
7/-/2
6/--X2
5/-/-
5/-/-
4/-/-
F
109/55/100
65/25/22
49/15/12
40/9/10
30/5/4
25/5/2
20/4/-
16/3/-
14/2/-
12/-/-
ll/-/-
9/-/-
G
47/17/12
29/9/5
23/7/3
19/6/2
15/5/-
12/4/-
9/3/-
7/3/-
6/2/-
5/-/- •
- 5/-/-
4/-/-
H
27/6/7
18/4/4
13/3/3
11/-/2
S/-/2
7/-/-
- 5/_/_
4/-/-
3/V-
3/-/-
3/-/~
2/-/-
NOTES:  (1)  Entries from survey data have been rounded  to  next  highest  whole  number.
        (2)  Dash indicates that a field strength  of that magnitude  was  not  measured at
that station.

-------
station, regardless of electrical height or frequency.  It should
also be noted that the recommendations in Bulletin 65 did not
take into account the perturbing effect of conductive objects
such as metal fencing on localized field values.  For example,
our results and Tail's results show that chain-link fencing may
significantly reduce close-in field readings from those that
would be expected if a fence were not present.  Another point
that should be made is that we only measured along a few selected
radials.  It is apparent that there can be significant variation
from radial ;o radial, particularly if conductive objects are
encountered along the radial.

6.7  RELATED STUDIES

The results of Tail's study should be considered along with our
results to obtain a better understanding of the eleetirontagnetic
field environment in the immediate vicinity of AM broadcast
towers.  In that connection, a few other observations made by
Tell should be mentioned.

Tail's study investigated directional tower arrays as well as a
non-directional tower.  He found that for both the non-direction-
al tower and directional arrays RF fields near quarter-wavelength
towers were directly related to the base current in the tower
under investigation.

For towers in arrays, the influence of other towers in the array
was found to be minimal with respect to compliance with the ANSI
guidelines.  Tell also observed that "non-driven" towers in a
directional array exhibited strong electric fields but essential-
ly no magnetic fields.  He suggested that this was due to the
process of "floating" the non-driven tower by disconnecting it at
the base.

Tell also investigated "contact currents" measured at guy wires
and at a chain-link fence near AM towers.  He found that such
currents can exceed the 100 milliampere level on guy wires and
can result in localized RF burns if these objects are touched.

In another field study conducted by the FCC and the U.S. Environ-
mental Protection Agency (8) we measured body currents induced in
an individual climbing a transmitting AM tower.  In that study,
•induced body current of up to 110 milliamperes was measured, and
the magnitude of the current appeared to be correlated with the
radial component of the local electric field.  Such currents may
be more relevant to the question of exposure of tower personnel,
since body current may be more closely related to absorption of
RF energy than field strength [9].
                                46

-------
7.   SUMMARY

(1)   A survey was made of electric and magnetic field strength at
     distances within 100 m of the transmitting towers of eight
     AM broadcast stations.  These measurements were made along
     various radial directions from the towers.  The purpose of
     the survey was to acquire data that can be used to more
     accurately assess the potential for human exposure to high
     radiofrequency fields near these towers,

(2)   Using the Numerical Electromagnetic Code (NEC), models have
     been developed for AM towers of varying electrical height
     and operating frequencies.  The models predict electric and
     magnetic field strength values at locations relatively close
     to a tower's base.  However, it is difficult to predict the
     effects of conductive objects such as metallic fencing on
     electric field strength.

(3)   Measurements generally yielded results in good agreement
     with the NEC model for distances greater than several meters
     from a tower's base.  The agreement was generally .not as
     good closer in to the tower base.

(4)   The NEC modeling technique developed for calculating near
     fields from AM towers involves representing a given tower as
     "three wires" and adjusting the tower height to match impe-
     dances.  Further work on numerical modeling may be useful. •

(5)   Electrical height of an AM tower is important in determining
     whether the electric field or the magnetic field will
     predominate close-in to the base of the tower.

(6)   Measurement data near AM towers showed that significant
     effects on electric field strength can result from conduc-
     tive objects such as chain-link fencing.  Metallic fences
     tend to reduce electric field strength on either side of the
     fence.

(7)   Field strength values near single, non-directional AM towers
     may differ when measured at the same distance but in dif-
     ferent directions from the tower.

(8)   Results of this study tend to confirm that the values given
     in the FCC's Bulletin No. 65 are generally conservative with
     regard to recommended minimum distances for compliance with
     various field-strength limits.
                                47

-------
                         •  •REFERENCES

1,   "Evaluating Compliance with FCC-specified Guidelines for
     Human Exposure to Radiofrequency Radiation."  Federal
     Communications Commission, OST Bulletin No. 65, October
     1985.  Copies may be purchased from the National Technical
     Information Service (NTIS),  (800)  336-4700.  Order number
     is:  PB 86-127081.  A limited number of copies are also
     available from the FCC at (202) 653-8169.

2.   Cleveland, R.F., Jr.,  and E.D. Mantiply, "Estimating Human
     Exposure Near AM Broadcast Towers:  Comparison of Theoreti-
     cal and Measured Electromagnetic Fields."  In Eleventh
     Annual Meeting Abstracts. Bioelectromagnetics Society,
     Tucson, AZ, June 18-22, 1989,  p. 77.

3.   Mantiply, Edwin D., "An Automated TEM Cell Calibration Sys-
     tem."  EPA 520/1-84-024, U.S.  Environmental Protection
     Agency, Las Vegas, NV, October 1984.

4.   Mantiply, E.D., and N.N. Hankin, "Radiofrequency Radiation
     Survey in the McFarland California Area."  EPA 520/6-
     89/022, U.S. Environmental Protection Agency, Las Vegas, NV,
     September 1989.

5.   Burke, G.J. and A.J. Poggio, "The Numerical Electromagnetic
     Engineering Design System (NEEDS)."  Version 1.0,, February
     1988.  Distributed by The Applied Computational EM Society.

6.   Tell, R.A., "Electric and Magnetic Fields and Contact
     Currents Near AM Standard Broadcast Radio Stations," August
     1989.  Prepared for the Federal Communications Commission by
     Richard Tell Associates, Inc., Las Vegas, NV.  Copies may be
     purchased from the National Technical Information Service
     (NTIS), (800) 336-4700.  Order number is:  PB 89-234850.

7.   "American National Standard Safety Levels with Respect to
     Human Exposure to Radio Frequency Electromagnetic Fields,
     300 kHz to 100 GHz."  ANSI C95.1 - 1982.  Copyright 1982 by
     the Institute of Electrical and Electronics Engineers, Inc.
     Copies may be obtained from the American National Standards
     Institute  (ANSI), 1430 Broadway, New York, NY  10018.  These
     guidelines are also discussed in Reference 1.

8.   "Radiofrequency Electromagnetic Fields and Induced Currents
     in the Spokane, Washington Area," April 1988, U.S. Environ-
     mental Protection Agency, Report No. EPA/520/6-88/008,
     Copies may be purchased from the National Technical Informa-
     tion Service  (NTIS),  (800) 336-4700.  Order number is:  PB
     88-244819/AS.
                                48

-------
Tell, R.A., "An Investigation of RF Induced Hot Spots and
their Significance Relative to Determining Compliance with
the ANSI Radiofrequeney Protection Guide," July 1989.
Prepared for the National Association of Broadcasters (NAB),
Washington, DC  20036,  Copies available from NAB;   (202)
429-5346.
                           49

-------
   APPENDIX A
STATION A GRAPHS

-------
Figure A— 1:
Station A Electric Fields (0-10m)
      110
o

-------
Figure A-2:   Station A Electric Fields  (10-100m)
   OT
   
   I— t
   w
            Radial 1
40     50     60



 Distance (m)



 Radial 2
                                              70
      HO
    JOO
Radial 3
„ NEC

-------
Figure A—3:
Station A Magnetic Fields (0-10m)


-------
Figure A-4: Station A  Magnetic Fields  (iO-iOOm)
   tn
   a
   K
  i


  •a
  PH
   

   
-------
   APPENDIX B
STATION B GRAPHS

-------

-------
        Figure B—1:
Station B Electric Fields  (0-10m)
OQ

t3
Q>

-------
   Station B  Electric  Fields (10-100m)
Figure  B—\
                                             100
              Distance (m)
Radial 1
                              Radial 2
NEC

-------
Figure B —3:  Station B Magnetic Fields  (O^lOm)

fc
o
•l-<
*-»

§

-------
Figure B-4:  Station B  Magnetic  Fields  (10-100m)
         10
20
   100
           Radial 1
           Distance (m)



                Radial 2
A NEC

-------
   APPENDIX C
STATION C GRAPHS

-------

-------
     Figure C-T:Station C  ElecLric~FieTds~(0-iOm)
           200
o
O
CD
OTQ

T3
Of
OTQ
CD

• 21
a,
      S
      K
      ,"-"s
      a
      o
      O
      4)
      ^^
      U
                 Radial 1
 10
                                    Distance (m)
Radial 2
* Radial 3
NEC

-------
o
      Figure  C-2:  Station  C Electric Fields (10-100m)
        s
        X
         O
         •r-l
         ^
         •M
         O
         (D

         W
                                                               too
                               Distance (m)
                 Radial t
Radial 2
Radial 3
NEC

-------
Figure C —3:  Station C  Magnetic  Fields (0—10m)
       3,0
OT
a
 QJ
 O
 •^
 V
 H)
 d
 tUJ
 (fl
           Radial 1
   Distance (m)


Radial 2       » Radial 3
A NEC

-------
Figure C-4: Station C  Magnetic Fields  (10-100m)

   (U
   •rH
   o
   •l~t
   4->
   
-------
   APPENDIX D
STATION D GRAPHS

-------

-------
       Figure D— 1:
                Station D Electric Fields  (0-10m)
CD
O
CO
CX

3"
flR
•O
o»

•s

or
      DQ

      2
      CC
       o
              aoo
             700 __
600 ___
              500 __
              400 ._
              300 __
              200 _ _
              100 _^
       Radial 1
     Distance (m)



Radial 2          »  Radial 3
                                                                                    10
NEC

-------
u
         td
                       Station D Electric Fields (10-100m)
                                                                    iuu
                  Radial 1
Distance (m)



      Radial 3
NEC

-------
Figure D-3:      Station D Magnetic  Fields (0-10m)
      0.70
-d
fummt


-------
Figure  D—4:
         0.040
         0.000
             10
Station  D Magnetic Fields (10-IOOm)
                Radial 1
         40     50      60




           Distance (m)




                 Radial 3
                                                      70
80
                                            90
                                                  100
            NEC

-------
   APPENDIX E
STATION E GRAPHS

-------

-------
Figure  E- 1:
Station E  Electric Fields (0-1 Om)
      350
GO
S
o
,rm4
I_
4->
o

-------
           Station  E Electric Fields (10-100m)
Figure  E-2:
w
      10
                                              BO
                                                 90
   100
                      Distance (m)
        Radial  1
                              Radial 2
A NEC

-------
CO
     Figure  E-3:  Station E Magnetic Fields (0-1 Om)
     C/l
      o
      •r-l
      4->
      0)

      fl
                                  Distance (m)
                Radial 1
Radial 2
A NEC

-------
Figure E-4: Station E Magnetic Fields (10-iOOm)
   2
   a;
   •a
   §— i
   
-------
   APPENDIX F
STATION F GRAPHS

-------

-------
      Figure  F— 1:     Station F Electric Fields  (0-lOm)
            3500
TJ
-*
CD
O
CD
O_
CPQ

•o
Q)
QTQ
CD
      u-

      O
      O
      «
      s
3250 ^


3000 __


2750 __


2500 ._


2250 __


2000 __


1750 ._


1500 _^


1250 T_


1000 __


 750 __


 500 ._


 250 __


  0

    0
       Radial 1
Distance (m)


    Radial 2
A NEC

-------
       Figure F-2:   Station F Electric  Fields (10-100m)
            220
ro
                10
20
                  Radial 1
30
40     50     60




Distance (m)




      Radial 2
80
90
100
                                           A  NEC

-------
Figure F-3:   Station F Magnetic  Fields (0-10m)
w

S

 d
 bO
 a
           Radial 1
Distance (m)



    Radial 2
                                                            10
NEC

-------
Figure F-4: Station F Magnetic  Fields (10-100m)
   
-------
   APPENDIX G
STATION G GRAPHS

-------
Q
     Figure G-l:   Station G  Electric Fields (0-10m)
      45
      o
      a

      4)
      f*"*

      m
                Radial 1
                                   Distance (m)
Radial 2
                                                                10
A NEC

-------
Figure G-2:    Station G Electric Fields  (10-75m)
     CO

     33

     A3
      o

     •c
     4-1
      o
      U
              10
30
                  Radial 1
35   40  45   SO




Distance (m)




  Radial 2
                                              55
65
70   75
                              NEC

-------
      Figure G-3:     Station G  Magnetic Fields (0-10m)
           0.45
Q
 I
CO
           0.00
                                                                  10
                 Radial 1
* NEC

-------
O
 I
      Figure G-4:     Station G Magnetic  Fields  (10-75m)
             a
s
O)
b£
13
C!
                 o.oa
                 0.07
                 o.oa
                 0.05
0.04
                 0.03
                 0.02
                 0.01
                 0.00
                     10   15
                25   30
                          Radial 1
                             35   40   45   SO

                             Distance (m)

                            •  Radial 2
                                                         55   60   65  70  75
                                                    NEC

-------
   APPENDIX H
STATION H GRAPHS

-------
Figure  H-l:   Station  H Electric Fields  (0-lQm)
Figure H— 1:



a
^
?
t
•d
i—i
•i-i
o
+j
o
1— 1
W




140
130 _
120 _
110 _
100 .
go _
ao .
70 .
60 .
50 _
40 .

30 -

20 _
10 -
o
















0
• Ra
                             Distance (m)
                                 Radial 2
NEC

-------
Figure H-2:
Station H Electric Fields (i0^50m)
  OQ
   U
   u
   0>

  s
                             Distance (m)
              Radial  1
           .  Radial  2
NEC

-------
 Figure H-3;
                 Station II  Magnetic Fields (0—10m)
        0.40
C/l

04
•o
"3
 4J
 o
 tin
 at
33
        0-35 _ _
        0,30
        0,25
0,20
        O.|5 _ _
O.IU
        0.05 _ -
        0.00
                Radial 1
                                4       S       6


                                  Distance (m)


                                   «   Radial 2
                                                                                  10
NEC

-------
Figure H-4;
Station H Magnetic Fields  (10-50m)
      0.050.
      0.045J
      0.040.
VI
S3
OS
If
•o
~s
0
W»
03
S


J
0.035_
0.030_
0.025^
0.020.
0.015.
0.010-
0,005_
0.000
\v
• \i\
• ^

^.

1 1
	 —•-""" 	 — :: : : • -r-l---^...-^..................™™.....- f
10 15 20
                              25
                  30
35
                               Distance (m)
H—
 40
—j—
 45
50
              Radial  1
               Radial 2
                  NEC

-------