EPA-600/3-77-035
April 1977
Ecological Research Series
     INSTRUMENTATION  TO  MONITOR  LOCATION  OF
             FISH  CONTINUOUSLY IN  EXPERIMENTAL
                                             CHANNELS
                                      Environmental Research Laboratory
                                     Office of Research and Development
                                    U.S. Environmental Protection Agency
                                           Duluth, Minnesota  55804

-------
                 RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of  traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.   Environmental Health Effects Research
      2.   Environmental Protection Technology
      3.   Ecological Research
      4.   Environmental Monitoring
      5.   Socioeconomic Environmental Studies
      6.   Scientific and Technical Assessment Reports (STAR)
      7.   Interagency Energy-Environment Research and Development
      8.   "Special" Reports
      9.   Miscellaneous Reports

This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal spe-
cies, and  materials. Problems are assessed for their long- and short-term influ-
ences. Investigations include formation, transport, and pathway studies to deter-
mine the fate of pollutants and their effects. This work provides the technical basis
for setting  standards to minimize undesirable changes in living organisms in the
aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                              EPA-600/3-77-035
                                              April 1977
     INSTRUMENTATION TO MONITOR LOCATION

OF FISH CONTINUOUSLY IN EXPERIMENTAL CHANNELS
                      by

               Joseph R. Jahoda
        Bayshore Systems Corporation
         Springfield, Virginia  22151
           Contract No. 68-01-0752
                Project Officer


            Kenneth E.F. Hokanson
    Monti cello Ecological Research Station
   Environmental Research Laboratory-Duluth
         Monti cello, Minnesota 55362
   ENVIRONMENTAL RESEARCH LABORATORY-DULUTH
      OFFICE OF RESEARCH AND DEVELOPMENT
     U.S. ENVIRONMENTAL PROTECTION AGENCY
           DULUTH, MINNESOTA 55804

-------
                                 DISCLAIMER
      This report has been reviewed by the Environmental Research Laboratory-
Duluth, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation
for use.

-------
                                  FOREWORD
     Our nation's freshwaters are vital  for all  animals and plants,  yet our
diverse uses of watei	for recreation,  food, energy,  transportation,  and in-
dustry—physically and chemically alter lakes,  rivers, and streams.  Such
alterations threaten terrestrial  organisms, as well  as those living  in water.
The Environmental Research Laboratory in Duluth, Minnesota develops  methods,
conducts laboratory and field studies, and extrapolates research findings

    --to determine how physical and chemical pollution affects
      aquatic life

    --to assess the effects of ecosystems on pollutants

    --to predict effects of pollutants on large  lakes  through
      the use of models

    --to measure bioaccumulation of pollutants in aquatic organ-
      isms that are consumed by other animals, including man.

    This report describes development of an ultrasonic tracking system used
to monitor position and temperature history of a mobile fish population in
an experimental stream channel, and discusses design problems and solutions
associated with operations near a nuclear power electrical generating facil-
ity.

                                            Donald I.  Mount, Ph.D.
                                            Director
                                            Environmental Research Laboratory
                                            Duluth, Minnesota
                                     iii

-------
                                    ABSTRACT

      This study resulted in the development and construction of equipment
to monitor continuously the position and the temperature of up to 20 fish
in a water channel 486 meters long, 3 meters wide, and 1 meter deep.  The
system utilized miniature sonic transmitters (tags) operating in the 51 kHz
to 366 kHz frequency range which were implanted in 500 gram or heavier
fish.  The battery operated tags were pulse modulated and designed for over
1 year operational life.  A temperature sensitive thermistor controlled
the repetition rate of the tag providing the temperature of the fish to an
accuracy of 1 degree C.  The nominal range of the polyurethane encapsulated
tage was several hundred feet.  Nominal tag size was 16 mm OD x 32 mm long
(4.6 - 5.4 g in water).  Sixteen hydrophones were located at 30.5 meter
intervals in the water channel.1  A control console contained a manually-
operated, frequency-stepped receiver which could select any i. dividual
hydrophone, thus locating the fish to within +_ 15.25 meters.  Up to 20
individual fish could be monitored.  Automatic operation and recording of
the data was considered in the design of the system for future equipment.
      Severe radio frequency interference problems were encountered, requiring
extensive precautions and modification of the channel equipment and wiring.
      Also investigated were passive fish monitoring and tracking of small
fish fry.  An experimental system was completed for limited monitoring
applications.
      This report was submitted in fulfillment of Contract Number
68-01-0752, by Bayshore Systems Corporation under the sponsorship of the
Environmental  Protection Agency.   Work was completed as of October 1974.
                                      IV

-------
                                  CONTENTS
Foreward                                                      iil
Abstract                                                       iv
List of Figures                                               .Yl
List of Tables                                               v1^
Acknowledgements                                               1X
Sections
    I  Introduction                                             1
          Large Fish Monitor System                             3
          Small Fish Monitor System                             6
   II  Conclusions                                              9
  III  Recommendations                                         10
   IV  Description of the Large Fish Tracking System           12
    V  Large Fish Tracking Sub-Systems                         17
          Ultrasonic Temperature Transmitters (Tags)           17
          Hydrophone/Preamplifier                              33
          Receiver                                             42
          Temperature Decoder                                  45
          Receiver-Temperature Decoder Console                 49
          Installation                                         51
   VI  Problems and Solutions                                  55
          Radio Frequency Interference                         55
          Electromagnetic Interference                         56
          Stability of Equipment            •                   57
          Tag Leakage and Aging                                57
          Acoustic Attenuation Problems                        57
References                                                     58
Appendices                                                     62
    A  Electromagnetic Field Measurements                      62
    B  Investigation of Inductively Coupled Circuits for
       Use in Fish Tags                                        87
    C  Fry Detection System                                   105
    D  Supplementary Materials                                116
Glossary of Abbreviations                                     117

-------
                                     FIGURES
 No.

 1
 2
 3
 4
 5
 6
 7
 8
 9
 10
 11
 12
 13
 14
 15
 16
 17
 18
 19
 20
 21
 22
 23
 24

 25
 26
 27
 28
 29
 30
 31
 32
 33
 34
 35
 36
 37
 38
 39
40
                                                 Mounting
                                 kHz, Vertical Mounting
                                 (Tag) Assembly
Test Channel at Monticello Ecological Research Station
Monticello Ecological Research Station Test Site
Large Fish Tracking System, Hydrophone and Channel Conduit Layout
Ultrasonic Temperature Transmitters
Blocking Oscillator
Modified Colpitts Oscillator
Hartley Oscillator
Acoustic Temperature Transmitter Schematic
Beam Pattern - Tag Frequency: 82 kHz, Horizontal
Beam Pattern - Tag Frequency: 82
Acoustic Temperature Transmitter
Dual Sided Receiving Hydrophone
Hydrophone Preamplifier Schematic
Hydrophone Assembly
Hydrophone Assembly
Receiver Schematic
Temperature Decoder
Signal Descriptions of Temperature Decoder
Receiver Decoder Console
Steel Conduit Runs from Blockhouse
Junction Box
Interconnecting Box
Interconnecting Box Wiring Diagram
Electromagnetic Test Sites - Monticello Ecological Research
Station Channels, Monticello, Minnesota
Electromagnetic Field Measurements - Composite Tests
                      Measurements
                      Measurements
                      Measurements
                      Measurements
                      Measurements
                      Measurements
                      Measurements
                      Measurements
                      Measurements
                      Measurements
                      Measurements
                      Measurements
Ground Conditions at the Monticello Ecological Research Station
Ground Loop Circuits
Induction Coupling into Pool Conductor
Electromagnetic Field
Electromagnetic Field
Electromagnetic Field
Electromagnetic Field
Electromagnetic Field
Electromagnetic Field
Electromagnetic
Electromagnetic
Electromagnetic
Electromagnetic
Electromagnetic
Electromagnetic
Field
Field
Field
Field
Field
Field
- Test
- Test
- Test
- Test
- Test
- Test
- Test
       Number
       Number
       Number
       Number
       Number
       Number
       Number
- Composite Tests
- Test Number 1
- Test Number 2
- Test Number 3
- Test Number 1
1
2
3
4
5
6
7
 2
 2
13
14
18
18
18
21
27
28
31
37
38
40
41
43
46
48
50
52
52
53
54

64
65
66
67
68
69
70
71
72
73
74
75
76
77
80
81
83
                                      VI

-------
                                  FIGURES

No.                                                                     Page

41   Electric and Magnetic Fields Around a Half Nave Vertical
     Antenna                                                             88
42   Variations of Q Against Frequency for Various Litz Wound
     Solenoid Inductors                                                  90
43   Variations of Q Against Frequency for Various Litz Wound
     Solenoid Inductors                                                  91
44   Variations of Q Against Frequency for Various Litz >   r
     Solenoid Inductors                                                  92
45   Curves Showing the Relative Received Energies for a vuryin^
     Source Energy for Q-l and H Type Core Materials                     94
46   Curve Showing the Relative Received Energy for Constant
     Source Energy                                                       95
47   Schematic Showing the Configuration of a Power Pickup and Con-
     verter for Use as a Power Source for an Ultrasonic Tag              96
48   Externally Triggered, Battery Operated Ultrasonic Tag               97
49   Test Equipment Configuration for the Evaluation of High Current
     Inductive Field Sources and Receivers                               99
50   High Power Test Data Showing Received Signal as a Function of
     Drive Current at a Distance of 6 Feet                              101
51   High Power Test Data Showing Received Signal as a Function of
     Distance from the Transmitter                                      102
52   Schematic of Externally Powered Ultrasonic Tag                     103
53   Photograph of Ultrasonic Receiver for Reception of 35 kHz
     Signals                                                            104
54   Instrumentation Sites for Fry Detection System                     106
55   Sample Test Data, Equilibrium Potential System                     111
56   Acoustic Fry Detection System Block Diagram                        113
                                     vn

-------
                                   TABLES

No.                                                                     Page

 1    Transmitter (Tag) Frequencies (kHz)                                 24
 2   Corrmercially Available Mercury Cells                                29
 3   Overall  Dimensions, Weight and Volume of Each Tag                   34
 4   Tag Repetition Rate and Pulse Width Characteristics                 35
 5   Tag Cylinder and Coil  Specifications                                36
                                   vm

-------
                              ACKNOWLEDGEMENTS

      This program was completed based on a team relationship between the
EPA, Monticello Ecological  Research Station as directed by Dr.  Kenneth Hokan-
son, and his assistant, Mr.  Charles Kleiner, and Bayshore Systems  Corporation
(BSC).  The many contributors at BSC were Don Anderson, Fred Gagnon,  Homer
Guerra, Francis Sherwood, Robert Greenwell, Richard Rockwell and Joseph R.
Jahoda.  Their contributions include advances in the state-of-the-art in low
current and high efficiency ultrasonic transmitters, RF interference  elimina-
tion, tag potting techniques, step tuned sensitive rf receiver,  hydrophone
preamplifier development, temperature A/D decoders, and light emitting diode
displays.  Note must be made of the days of working in cold/hot weather side-
by-side with EPA personnel  installing the equipment.
                                       IX

-------
                              SECTION I
                              INTRODUCTION

      To solve some of the more complex ecological problems concerning
the discharge of waste heat into the nation's surface waters, the Environmental
Research Laboratory-Duluth constructed a field station at Monti cello,
Minnesota.  Eight  (8) identical experimental stream channels, each about
486 meters  (1,600  feet) long were constructed at the Monticello test site
(Figure 1).  Each  channel is composed of alternating riffle and pool
areas.  Each channel contains a total of eight (8) riffle areas of
2.4 meters  (8 feet) wide and .3 meters (1 foot) deep and eight (8) pools
of 3.7 meters (12  feet) wide and .92 meter (3 feet) deep; each about
30.5 meters (100 feet) long.  To date, one channel was instrumented,
channel number 1,  and placed into operation.  A small laboratory provides
office space, sample processing areas, and houses the control and
monitoring  equipment (Figure 2).
      Raw Mississippi River water is used for the water supply.  This
water is heated to the desired temperature in a heat exchanger using the
heated condenser water of Northern State Power (NSP) nuclear power plant.
The condenser water is returned to the NSP discharge where it is monitor-
ed against  state water-quality standards.  The rate of flow of water in
channel number 1 was .028 cubic meters per second (1 cubic foot per
second) with mean  velocity less than .03 meters per second (0.1 foot
per second).
      Four different annual temperature regimes in duplicate will be
programmed and controlled at the upstream end of the channels.  Thermal
gradients will develop in these channels except for one pair which will
follow ambient Mississippi River temperatures.  Temperatures will vary
            *
diurnally and seasonally along thermal gradients.  It is intended to
monitor water temperature and other water quality and meteorological
parameters in some detail to accurately describe conditions in each
channel.

-------
     Figure 1.   Test channel  at  Monticello
                Ecological  Research  Station,
Figure 2.   Monticello Ecological  Research Station
           test site.  Building to the right is the
           laboratory, the building in the center
           is a garage and storage building.

-------
      Natural river plankton and insect fauna are expected to colonize
all of the experimental channels.  Once an abundance of food organisms
is present, one fish species will be stocked at natural densities.  The
distribution of fish is known to be strongly influenced by the environ-
mental temperature and varies with the different age groups.  Young fish
can thrive at higher temperatures than adult fish.  Larval fish will
most likely be found near the surface in pools and in areas of low
current velocity.  Juvenile and adult  fish may have a wider distribution
and may be found on the riffles near the substrate.  Recognizing the
problem of monitoring temperature of mobile organisms, this contract was
awarded on 30 June 1972 to develop the instrumentation to monitor the
location of large fish on a continuous basis in the experimental channels.
This instrumentation is necessary to permit determination of the thermal
experience of each fish in the channel so that the effects of temperature
can be properly assessed.  In addition, a small fish monitor system,
required to monitor continually the location of small fish fry, was
demonstrated.  The large fish monitor system was completed and delivered
by October 1974.
      As part of the overall program other items which were delivered
included Instruction and Maintenance Manual and Engineering Drawings
of System Prototype.
LARGE FISH MONITOR SYSTEM
                                           r,
      A system was required by the EPA to monitor instantaneous positions
and temperature of fish in the water channels.  The system requirements
included a temperature monitoring transmitter (tag) which would be affixed
to fish and measure the temperature of the fish or the surrounding water
to the nearest 1 C.  Further, the tag must have a minimum life of 1 year
and be small enough to be placed on, or in, a 25.4 cm (10 inch) fish
without significantly interfering with growth or survival.  Each fish
position and temperature should be recorded once an hour.  The tags will
require a minimum range of 30 to 150 meters (100 to 500 feet).  Either
different frequencies or pulse rates should be used for each fish to permit

-------
distinguishing 10 to 20 individual fish in each channel.   The final
system design must be capable of interfacing with a real-time, digital
recording system at a later date.
      An additional factor which must be considered in the final  design
of the system is the presence of two high voltage overhead transmission .
lines (345 and 230 kilovolts) which cross the channels and will unques-
tionably cause radio frequency interference (RFI) problems (see Appendix A
for details).
      During the early phases of the program several system approaches
were considered.  They all utilized the basic concept of attaching a small
transmitter to the fish and thereby transmitting the location and internal
temperature to pick-up units on shore.  Several variations were possible
and accordingly considered.
Radio Frequency System
      A significant amount of research has been done in related military
                                            1 St Y\f\ 9
and wildlife programs for developing rf tags         for use in fresh
water.  Ranges of over 500 meters have been reported.  The use of rf tags
would have required rf receivers, multiple frequencies (to identify
individual fish), elaborate direction finding antenna systems  (to locate
each fish), and means to eliminate the effects of rf signals present
from local and distant transmitters and interference from the overhead
power line.  These problems would prove insurmountable and the rf system
was dropped from consideration early in the program.
Sonic System
      The use of sonic tags has been used for well over 25 years for fish/
manual location,  to   , temperature monitoring,   and 26, and other
                                       O"7 i-. *5 O
fish/mammal/water parameter monitoring,        .  All of the efforts to
date had concentrated on manual tracking over relatively short periods
of time, usually 2 to 4 weeks.  The ranges usually involved 100 to
500 meters (110 to 550 yards).

-------
      Early in the program it became evident that to meet the one year
operating life requirement of the tag in a compact unit, a highly efficient
tag would have to be developed.  Since long range was not required in the
experimental channels, a system was designed which would decrease the
range requirement of the tags to 40 meters (132 feet) and thereby
considerably reduce size and increase battery life  .
      The minimum fish size of 25.4 cm (10 inches) dictated the use of
internally mounted tags with lengths of less than 3.2 cm (1.26 inches)
and outside diameters of less than 1.6 cm (.63 inches).   This size
                                                        9  35
limitation was arrived at by researching the literature,     , discussions
with fish biologists, and experimentation at the Monticello EPA laboratory.
      All of these requirements imposed severe restrictions on the
final tag design.  Factors considered were:
            (a)  Long term (over one year) waterproof sealing.
            (b)  Ultra-miniature packaging.
            (c)  Circuitry stabilization to minimize frequency drift
                 as a function of short and long terms and temperature.
            (d)  Non-toxic outer coating.
            (e)  Long life batteries.
            (f)  Minimum weight (in water).
Passive System
      The use of passive systems has found very little application in fish
                2
tracking systems .  The approach basically utilizes sonic or magnetic
energy beamed to a remote location (fish tag) from a shore station.  This
energy is converted to dc   and used to power a transmitter which then
responds with an appropriate signal.  The limiting factor is the range
between the tag and shore location.  After considerable effort, a maximum
range of only 3 meters (9.9 feet) was accomplished.  By utilizing larger
power levels at the shore based transmitters this range could be increased.
However, after considering all factors, this system was deemed economically
unfeasible (see Appendix B for details).

-------
SMALL FISH MONITOR SYSTEM
      Initially, the EPA required the counting and/or location  within
the water channels of freeswimming offspring from natural  reproduction
in relation to temprature recording stations.  A remote sensing device
was required to locate and/or count (i.e.  + 10%) larval and juvenile
fish (i.e. 4 to 40 mm total length) at fixed stations.   Fish count or
position  recordings were to be made once  an hour at each  station.
Relative abundance indices at each station were required,  rather than
a total fish count in each channel.  Such  a means of detection  would not
require handling or tagging individuals.   A design goal was to  discriminate
different size fish so that large and small fish could  be  counted as
individuals regardless of their direction  of movement.   The system was
to have the design capability of interfacing with a real  time,  digital
recording system at a later date.
      There is very little precedent for a system of this  type  which
concerns itself with a female fish spawning 15,000 to 40,000 eggs in
the channel bottom.  Initially, the survival rate is 60% to 70% and one
should be prepared to monitor several thousand 4 to 12  mm  (.16  to .47  inches)
fish fry.  In the early stages the fry may school together or assume a
uniform distribution.  As they approach 10 mm (0.4 inches) they start  to
school, covering an area of approximately  one cubic meter  (1.31 cubic
yards).
      The following are typical of the methods proposed and/or  actually
utilized by researchers, fisheries, universities, commercial fisherman
and industrial, NASA, military and non-profit agencies  for tracking
schools of fish:
      1.  Passive - acoustic
            a.  Swimming noise.
            b.  Vocal sounds.
            c.  Strumming, beating or scratching noises.
            d.  Noise produced by contact  with external objects.

-------
      2.   Passive - general
            a.  Water-color spectrometer - color/luminescence.
            b.  Luminescing fish detection.
            c.  Fish oil slick detection.
            d.  Infra--red detection of fish swimming near the water
                surfaces.
            e.  Visual-by air and shipborne observers.
            f.  Photographic - IR and color.
            g.  RF signal emmissions from fish.
      3.  Active
            a.  Sonar - transmission and reception of acoustic signals.
            b.  Laser - transmission and reception of coherent light.
            c.  Microwave/millimeter radiometers.
      After considering all of these techniques, those which offered
promise for locating fish fry in the narrow channels were sonar, laser,
and infra-red.
Sonar
      This is probably the most often taken approach.  Sonar systems for
fish location have been built over the frequency range of 10 kHz to
600 kHz.  They have been used to:  count salmon going upstream, measur-
ing biomass (kilograms) of fish in rivers, lakes, and portions of the
ocean, etc. with some success.  However, the transmission of an acoustic
signal in a small narrow channel offers additional problems.  These are
due to such factors as multiple reverberations, pickup: from the main
and side lobes of the transducers, unpredictable variation in reflection
from the fish as a function of its position, reflection from the water
surface and irregular contours, the unpredictable variation in signal
strength returned from fish of different sizes, and doppler shifts due
to rapidly moving fish.  Many techniques have been proposed or attempted
to obviate these disadvantages.  Such factors as swim bladder, vertebral
column and body tissue resonances can be used to isolate fish

-------
acoustically,    °   , especially small  fish, from others of different
sizes.  To properly utilize resonance effects would require transducers
operating at multi-frequencies.  The problem of reflections, echos,
false responses, etc., can be solved by time gating, FM techniques
and multiple transducers.  One could use specially built sonar arrays,
funnels, baffles, etc., to assist the sonar system, however, the small
size of the fish and the desire to avoid unusual mechanical intrusions
into the water further complicate the use of sonar techniques.  The
final system chosen, after careful consideration of all factors and
extensive experimentation, utilized a modified multi-frequency sonar
technique, see Appendix C for system details.
Laser
      This would prove expensive due to the complexity of the ancillary
equipment required.  In addition, the data processing would prove too
complex and expensive and it would suffer from shading effects from
fishes nearest the laser.  The same comments would apply to the use  of
similar techniques based on light source sonar.
Infra-Red
      Limited to fish at surfaces of the water.  Shading effects and
difficulty of implementation prevented serious consideration.
                                    8

-------
                                  SECTION  II
                                CONCLUSIONS
      The final system, which accomplished this objective was composed of
16 hydrophone/preamplifiers, sonic fish tags, shielding cabling and conduits,
and a receiver/decoder console.  The 16 hydrophone/preamplifiers were
located at 30.5 meter (100-foot) intervals along the channel.  The operator
can select the operating frequency of the desired tag (unique frequency in
the 51 kHz to 366 kHz range for each fish) and the appropriate hydrophone
which has the strongest response.  Knowing which hydrophone received the
strongest signal would locate the fish to within an accuracy of +_ 15.25 meters
(50 feet).  The decoder converted the repetition rate to a numeric display
of the fish temperature.
      The extrapolated life of the tag meets the one (1) year objective.
With the advent of the new lithium battery, the life can be further
increased by almost a factor of two.
      The seriousness of the radio frequency interference (RFI) from the
overhead high voltage power lines cannot be overstated.  Signals were
present at unpredictable times on frequencies very close to those being
used.  At times, the frequencies in use had to be changed.  Based on the
RFI difficulties experienced with one instrumented channel, additional
shielding, filtering and isolation will be required for the instrumentation
to be installed in the future for the remaining 7 channels.  This is mandatory
because the RFI problem can be expected to be cumulative in its impact,
      Monitoring of the small fish fry was a most difficult task.  Their
small size and their physical properties, being very similar to that of
water, precluded the use of conventional sonar, laser, and infra-red
techniques to monitor their presence.  Drifting algae also presented
background interference problems during the initial system development.
After considerable experimentation a modified sonar technique (utilizing
critical frequencies and the transmission of sonic energy through a mass
of material having characteristics different from that of the water) was
developed which was capable of monitoring small fish fry under conditions
of limited floating algae and in areas of minimum turbulence.

-------
                                SECTION III
                              RECOMMENDATIONS

       Based on the data and results of this  program,  recommendations
for future effort and system expansion are;
             (a)  The availability of advanced  shielding material,  improved
 coupling and balancing circuits,  and a more  selective phase-lock receiver
 will  greatly improve system operation.  The  RF interference  problem,  which
 inherently is unpredictable, will necessitate  these modifications.   In
 particular, the use of a phase-lock receiver should be stressed, since  it
 will  permit automatic frequency lock-on to each tag.   Thus,  should the  tag
 frequency drift or the fish be successful  in hiding behind growth  for
 a short time, the equipment would not require  manual  retuning  and  loss
 of data would not result.  Specifically, the following should  be implemented
 during the next program phase:
                  (1)  Hydrophone Preamplifiers - Modify output circuitry
                       to provide improved sensitivity and decrease ground
                       loop and RF interference from power station.
                  (2)  Improved shielding of  console and receiver.
                  (3)  Automatic phase lock receiver modification.
                  (4)  Improved tags, through the use  of temperature
                       compensated components,  to minimize frequency drift.
                  (5)  As additional instrumentation is added for the
                       remaining 7 channels,  extreme care  must  be exercised
                       in proper utilization  of interference  filters, isolation
                       and shielding to avoid mutual interference problems.
             (b)  The present receiver-decoder  system  control panel is
 manually operated.  This greatly limits the  data available  to  the  normal
 working hours of available personnel and increases the system  operating
 costs.  The present design was predicated upon ease of conversion  to
 automatic operation with a computer and recorder.  This should be  imple-
                                       10

-------
mented after the initial phases of system check-out are completed and
the biological experiment is satisfactorily structured.
            (c)  Lithium batteries should be used when they become
available.  They will permit as much as a 10% decrease in length and
weight for the same life expectancy or a longer life expectancy for the
same size.
            (d)  It is recommended that a small fish fry detector be
implemented in biologically critical areas of the channel.  An efficient
and simplified system can be completed and calibrated especially with
the availability of fish fry on site.  They were not available during
much of the original experimentation.  As a minimum such a system
would provide position location and approximate volume of the fish fry
school.
            (e)  Initiate a small program to study the impact of the
projected automated system on the computer software requirements.  A
computer with sufficient capacity will be required and a study made of
the specific details of modifying the fish location system.   Specifically,
it is important that the computer have sufficient capacity to handle the
data from eight channels.  Typical of the critical factors which need to
be considered is the switching rate of the receiver and hydrophones.  This
will be limited by reverberation rate of the channels, time required to
locate each fish, etc..  This effort will interface with the biological
work in progress at the Monticello test site and paragraph (a) above.
                                    11

-------
                               SECTION IV
              DESCRIPTION OF THE LARGE FISH TRACKING SYSTEM

      The  large fish tracking system installed at the Monticello Field
Station consists of a number of component equipments arranged as shown in
Figure 3.  The system is comprised of three main subsystems/components;
a  number of ultrasonic temperature transmitters (tags, inserted within
the fish), an array of hydrophones, and a receiver-decoder control console.
      Ultrasonic temperature transmitters (or fish tags), transmit a
pulsed, acoustical signal through the water (Figure 4).  The repetition
rate at which the tags transmit is determined by a temperature sensor
within the tag.  At cold temperatures the repetition rate is slow and
for warm temperatures the repetition rate is faster.  This form of trans-
mission is called pulse interval modulation (PIM).  This type of modulation
permits location and temperature monitoring with minimum hardware.  Complex
telemetry  modulation techniques would have required more electronics in
the tags as well as in the receiver.  As an additional advantage, the system
is much less susceptible to reverberation effects than multi-carrier,
phase or constant wave modulation techniques.  The tags are designed to
be pulsed  on for a relatively short interval of time in order to obtain
a  low duty cycle which minimizes the drain on the internal batteries and
increases  the effective life of the tag.
      The  hydrophones are used to pick up pulsed acoustical-signals
transmitted by the tags and convert them to electrical signals that can
be detected by a receiver.  A total of 16 hydrophones, located at 30.5 m
(100 ft) intervals are used.  By observing which of the hydrophones receives
the strongest signal, one can determine the location of the fish within
^ 15.75 m  (50 ft).  Due to the wide range of frequencies covered by the
tags, 51  to 366  kHz, a special wide-band hydrophone was designed with
special filters and preamplifiers.  Due to the extremely high radio
frequency  interference level, rf shielding was used in the hydrophones/
preamplifiers and at all junctions and conduits.  Each preamplifier was
                                     12

-------
    4"  CONDUIT
      " CONDUIT
     N3
BLOCKHOUSE
O  HYDROPHONE
D  JUNCTION  BOX

    RECEIVER/ DECODER CONSOLE
               Figure  3.  Large fish  tracking system,
                          hydrophone  and channel  conduit layout.

-------
             Figure 4.  Ultrasonic temperature transmitters.
protected against high rf levels and high induced voltage  from
electro-fishing instruments and possible lighting discharges by the
use of diodes at their inputs.
      The receiver selects the one desired frequency from among the
20 present frequencies at the output of the hydrophone/preamplifier and
converts the pulsed electrical ultrasonic signals to a low frequency
pulsed signal to the decoder.  The decoder then measures the time between
pulses and gives a corresponding readout of temperature on a digital dis-
play meter.
      Technical specifications of each of the major subsystems are as
follows:  The value of the parameters are average or nominal values.
All tags delivered were within 10% to 15% of these values.  The dimensions
varied more widely since at the higher frequencies smaller ceramic elements
were used.  Specific specifications are discussed in greater detail in
Section V.
                                    14

-------
 (1)  Ultrasonic Transmitters
      Frequency:
      Pulse Width:
      Power Output:
      Repetition Rate:

      Duty Cycle:

      Expected Life:

      Size:

      Weight:

      Frequency Stability:
            SI to  366 kHz (18 Tags)
            2.5 milliseconds
            +40 dB re 1 JLIbar/meter
            0.8 pps at 0 C
            1.6 pps at 30 C
            .2% at 0 C
            .4% at 30 C
            450 days at 0 C
            275 days at 30 C
            (OD) 15 to 17 mm
            (L) 30 to 39 mm
            (Air) 10.7 to 12.6 grams
            (Water) 4.6 to 5.4 grams
            0. 15% (0 C to 30 C)
(2)  Hydrophone/Preampl
     Frequency:
     Gain:
     Output Impedance:
     Input  Power:
     Shielding:
       (RF)
       (Electrostatic)
     Beamwidth:
       (Each Section)
     Sensitivity:
(3)  Receiver (Located
     Frequency:
     Sensitivity:
     IF Bandwidth  (3dB)
     Input  Impedance:
ifier (16 per channel)
    51  to 366 kHz
    20 dB (Minimum)
    10 ohms (Maximum)
    12 V DC

    Special Housing/Cable
    Coating for Sonic Window
    180° Horizontal
     40° Vertical
    -88 dB re 1 nbar/meter
in Receiver/Decoder  Console)
    51  to 366 kHz  (20  steps)
    0.1 microvolts (S/S+N  = 1)
:    1.2 kHz or 4.0 kHz
    400 ohms
                           15

-------
(4)  Receiver/Decoder Control  Console
     Temperature Readout:   0  to  30 C
     VCO Readout:
     Controls:
     Jacks:
     Meter:
Voltage to VCO (Frequency
analog voltage)
Audio Gain
RF Gain
Tag Selector (20 Frequencies)
Freq. Adj. Fine Tuning
  (20 Frequencies)
Temp. Calibration (20 tags)
Hydrophone Selector (16 stations)
Oscilloscope Output
Audio Output
AGC Level
Signal Strength
                           16

-------
                                SECTION V
                     LARGE FISH TRACKING SUB-SYSTEMS

ULTRASONIC TEMPERATURE TRANSMITTERS  (TAGS)
Oscillator Circuits
      During the design phase of the contract three oscillator circuits
were selected to be evaluated:  squegging, modified Colpitts and Hartley.
These oscillators were designed to serve dual purposes:  frequency generator
and transmitter, in order to reduce the number of components used in the
transmitter circuit.  The size of the tag was'of primary concern, there-
fore emphasis was placed on the use of minimum parts along with the
selection of miniature sized components.  Since the electronics would
only occupy less than 25% of the final tag (the battery and ceramic
element are the largest size contributor and cannot readily be miniaturized),
the possible size and weight saving on the final tag of using micro-
circuitry do not justify their high cost and accordingly were not considered.
      The squegging or blocking oscillator was found to be the simplest
circuit as far as the number of components used (Figure 5).  However,
the circuit was very unstable with changes in battery voltage and resulted
in a 25% change in repetition rate over the life of the battery.  A small
tag was assembled, potted and evaluated.  The size of the tag was 10 mm
in diameter and 20 mm in length.  Test results showed an interaction
between the repetition rate and pulse width with load changes in the
oscillator circuit.  As the tag was lowered into the water the repetition
rate would increase and the pulse width would decrease.  This circuit,
although simple and small, could not be used for accurately determining
temperature of the water due to large changes in the repetition rate with
battery voltage decay and water loading.
      The modified Colpitts oscillator (Figure 6) was breadboarded and
evaluated.  The circuit had good frequency stability with changes in
battery voltage (1.4 to 1.0 volts), less than 0.15% frequency drift
                                     17

-------
                            -A/W
               Figure 5.    Blocking oscillator.
                                                       V
V
                         T
               Figure 6.  Modified colpitts oscillator.
         Figure 7.  Hartley oscillator.



                             18

-------
resulted.  The tuning of the circuit was critical because the ratio of the
two feedback capacitors had to be changed with different frequencies.
The circuit required four capacitors to tune to the operating frequency
and two of the four were large value capacitors which required larger
sized components.  The power output of the Colpitts oscillator was
limited due to the required emitter resistor in the oscillator circuit.
      The Hartley oscillator (figure 7) was breadboarded and evaluated.
The frequency stability of the circuit was less than 0.12% with the
change in battery voltage.  The frequency of the oscillator could be
changed by changing the value of a single capacitor.  The voltage required
across the cylinder could be varied by changing the turns ratio of the
transformer.  The power output of the oscillator could be adjusted by
inserting a low value resistor in series with the emitter of the oscillator
transistor.  Because of the simple circuit with few components the
Hartley oscillator was chosen to be used as the transmitter in the tags.
Timer Circuits
      Due to the limited battery power the tag should operate at the
minimum duty cycle consistent with satisfactory reception at the receiver.
The duty cycle is merely the ratio of transmitter "ON" (drawing battery
power) to transmitter "OFF" (drawing no battery power)..  To obtain low
duty cycles the transmitter must be turned ON and OFF at certain periods
of time.  To do this a timer circuit is used that required little
current drain from the battery.  Two timer circuits were evaluated;
astable multivibrator and a complementary astable multivibrator.
      The astable multivibrator is a two transistor oscillator that
periodically produces a square or rectangular pulse at its output.  The
frequency at which this pulse is produced can be controlled by changing
the value of a resistor in one of the transistor stages.  By using
a thermistor, which is a device that has a negative temperature coefficient,
in place of the resistor, large changes in resistance occur wtth changes
in temperature.  This provided a means of turning the transmitter ON
                                    19

-------
and OFF at different time intervals that corresponded with the temperature
of the thermistor.  The astable circuit proved to be stable with voltage
changes with less than a 5% change in the repetition rate over the life
of a single battery.  The use of two batteries reduced the change in
repetition rate to 2.5%.  The one disadvantage of the astable multivibrator
was that at all times one of the two stages was in an ON condition,
thereby continually drawing current from the battery.  The average current
drawn was reduced to less than 20 micro-amps but this was considerable
as the total average current for both the transmitter and timer had to be
kept below 25 micro-amps to obtain a one year life.  This is based on an
available battery capacity of .18 amperes hours.  A year has almost
9000 hours, thus at 20 microamperes (or 20 x 10   amperes) consumption
per hour, the battery would require a capacity of 9000 x 20 x 10"  or
0.18 ampere hours.
      The complementary astable multivibrator uses a complementary pair
of transistors (Ql and Q2, Figure 8) so that both transistors are either
ON or OFF at the same time.  The only time current is drawn from the
battery is during the pulse on time which is in the order of a few
milliseconds.  With the use of high beta transistors and high value
resistors the average current can be kept below two micro-amps for the
timer circuit.  This circuit was selected for the timer because of its
good stability (less than 5% variation in repetition rate over the
voltage variation which results over the life of the battery), low
power consumption and the simplicity of the circuit.  It should be noted
that an integrated circuit (1C) commercially available version of the
complementary astable was also evaluated and found to have excellent
stability, less than 1% change in repetition rate with 20% changes in
battery voltage.   The 1C required a minimum of 3.0 volts for its
operation and its overall size was larger than that of the discrete
components.  Because of the additional batteries required and the larger
size,  the 1C was  not selected as the timer in the final design.
                                    20

-------
                                                                  Battery
ro
                 R - Resistors
                 Q - Transistors
                 Y - Ceramic PZT Cylinder
                 C - Capacitor
                 T - Transformer
                RT - Thermistor
- Battery
                                Figure 8.   Acoustic temperature transmitter schematic.

-------
Transmitting and Receiving Elements
      The transmission and reception of underwater sonic signals is made
possible by piezoelectric devices.  They have the property of vibrating
mechanically when a voltage is applied to them.   Piezoelectric cylinders
and discs were evaluated to be used as acoustic  transmitters and receivers
for the tags and the land-mounted hydrophones.  The requirements of the
receiving hydrophone were that it be directional to reduce noise pickup
and be highly sensitive so as to reduce the power output requirements
of the acoustic transmitters.  As an example, since the water is shallow,
there is no  need to have the hydrophone receive signals from the bottom
and top of the channel.  Rather it should be most sensitive up and down
stream.  Thus extraneous signals and noises coming from the top of the
channel would not be received strongly and would not interfere with the
desired (tag) signals.  Lead zirconate titanate  (PZT) discs were
selected as the receiving elements because of their directional capabilities
and their high sensitivities.
      Three PZT cylinders of different dimensions were selected as the
acoustic transmitters.  Initially, one size of cylinder was used.  However,
the sensitivity of the unit was poor at the low and high frequencies.
This is due to a quality of resonance which all  cylinders have.  The
cylinder vibrates best at its resonance and this is primarily due to its
size and shape.  To cover the large frequency range several units with
different resonant frequencies were required.  The size of the cylinders
and their resonant frequencies are as stated below.  Note that the high-
est frequency is 185 kHz.  Since this unit operated satisfactorily up to
365 kHz, higher resonant frequency cylinders were not necessary.  This
is fortunate since higher frequency units are not readily available
and are fragile.
            (1)  12.9 mm OD x 6.4 ram L and 11.4 ram ID
                 Frequency-Circumference Mode 80 kHz.
            (2)   9.6 mm 00 x 6.4 mm L and 8.0 ram ID
                 Frequency-Circumference Mode 125 kHz.
                                    22

-------
            (3)  6.4 mm OD x 6.4 mm L and 4.8 mm ID
                 Frequency-Circumference Mode 185 kHz.
Thermistor Element
      In order to monitor the temperature of the tag environment a
small temperature sensitive element (thermistor) was used in the tag.
Its resistance varies with the temperature.  Thus, by using the thermistor
element in a circuit whose repetition rate varies with the variation in the
thermistors,   an    , its temperature can be monitored at a remote
location by measuring the repetition rate of the received signal.
      The thermistor has a resistance of one megohm with a tolerance of
+_ 20%.  With a +_ 20% tolerance the resistance variation is 800K ohms
to 1.2 megohms.  Such a wide range will result in wide variation from
tag to tag.  To compensate for the different repetition rates to be
expected, a gain and slope adjustment was required in the decoder unit which
converts repetition rate to temperature.
Final Design Circuit
      The schematic of the final tag circuit is shown in Figure 8.  The
timer circuit consists of transistors Ql and Q2 and their associated
components.  Transistors Q3, Q4 and Q5 act as switches and buffer amplifiers.
The oscillator and transmitter circuit is transistor Q6 and its components.
The PZT cylinder is represented by the symbol Yl and the thermistor by the
symbol RT1.  The frequency determining components are the PZT cylinder Yl,
capacitor C3 and the secondary inductance of Tl.   Thus for each tag, these
3 parameters must be carefully evaluated to assure that the correct
final operating frequency results.  The thermistor RT1 along with
capacitor Cl determine the repetition rate of the transmitter, with the
thermistor being the variable with temperature changes.  The operating
frequencies of the transmitters were selected so as to maintain a minimum
of 6.5 kHz i 2% of the frequency between each transmitter (Table 1).
                                     23

-------
Table 1.  TRANSMITTER (TAG) FREQUENCIES
                    (kHz)

Tag
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
20
Fundamental
frequency
50.0
59.35
69.2
77.2
88.2
102.0
113.6
123.9
133.0
143.0
160.25
170.2
184.1
194.4
218.4
239.7
365.0
345.0
2nd
Harmon.ic
100.0
118.7
138.4
154.4
176.4'
204.0
227.2
247.8
266.0
286.0
320.5
340.4
368.2





3rd
Harmonic
150.0
178.05
207.6
231.6
264.6
306.0
340.8











4th
Harmonic
200.0
237.4
276.8
308.8
352.8













                    24

-------
      The pulsing ON and OFF of the transmitter creates harmonically
related spurious signals that are of sufficient power levels to interfere
with certain selected operating frequencies.  Therefore the second,
third and forth harmonics of each transmitting frequency must be considered
when selecting the operating frequencies, in order that they are different
from the fundamental frequency of all of the tags.  This is particularly
important in selecting the lower frequencies.  These will have harmonics
which could fall in the range of higher frequency tags.  See Table 1,
and note the final selection of frequencies and how the harmonics are
different from all of the fundamental frequencies.
      Another factor in determining the operating frequency is the
cylinder.  Each cylinder has three modes of operation:  Circumference,
length and thickness;  each mode resonates at a different frequency.
At these mode frequencies large changes in cylinder impedance occurs,
resulting in critical tuning at, or near these frequencies.  These
frequencies must also be omitted in the final section of operating
frequencies.
      Another consideration for the selecton of the tag frequencies,  and
one which was not considered until the problem surfaced during
final systems testing at Monticello, was interference from signals being
transmitted through the overhead high voltage wires.  These signals
are unusually strong and are picked up through the water and into the
hydrophones, at the junction boxes and cables, and at the blockhouse
in the receiver.
      The frequency stability of the oscillator at a constant temperature
is determined by the aging process of the ceramic material used in the
manufacture of the cylinder.  After the manufacture of the cylinder a
continuing change in the dielectric constant occurs for a period of
several weeks.  This dielectric change results in a change in the
capacitance of the cylinder which results in a change in the oscillator
frequency.  The aging rate decreases with time and can be accelerated by
placing the ceramic in a heated oven for a period of time.
                                    25

-------
      The frequency drift of the oscillator with changes in temperature
 is caused by the  temperature coefficients of the cylinder, the transformer
 and  the tuning capacitor.  This change in frequency over temperature
 was  compensated in the final tags to a marked degree with temperature
 compensating capacitors.  They are larger in size than the miniature
 tuning capacitors used in the present tags but do not affect the final
 size significantly.
      The acoustic power output of the tags average 1 milliwatt with
 an average of 7 milliwatts input power for an efficiency of approximately
 14%.  The range of most tags exceeded 46 meters (150 feet) in clear
 stable water, with the design goal set for a minimum of 15.25 meters
 (50  feet).
      The temperature sensing thermistor was placed inside the tag to
 reduce external leads.  External leads were kept to a minimum to prevent
 any  weak points in the tag that might be susceptible to water leakage,
      Directional patterns of an,82 kHz tag are shown in Fiaures 9
 and  10.  The horizontal pattern, Figure 9, clearly indicates that
 maximum power is  radiated from the sides of the cylinder, as expected.
 The  vertical pattern, Figure 10, is symmetrical and virtually
 omnidirectional.  All of the tags will exhibit similar patterns.

 Batteries
      The mercury and the silver oxide batteries are at this time the
 only available miniature high capacity batteries on the market, with
 a relatively constant voltage over the life of the battery (Table 2).
 The  silver oxide battery MS76 has an initial voltage of 1.5 v and a maximum
 capacity of 160 milliampere hour for the required size used in the present
 tags, 11.4 mm OD, 5.2 mm long, and a weight of 2.8 grams.  The mercury
 battery MP675 was selected for the tags because of its higher capacity
of 220 milliampere hour.  Its size is 11.4 mm OD, 5.1 mm long, and  it
weighs  2.8 grams.
                                    26

-------
                                                    210°
                                                    ISO0
Figure 9.  Beam pattern-tag frequency:  82  kHz,
           horizontal mounting.
                         27

-------
Figure 10.  Beam pattern-tag frequency: 83 kHz,
            vertical mounting.
                          28

-------
                                Table 2.   COMMERCIALLY AVAILABLE MERCURY CELLS
ro
vo
Type3
RM 212
RM 312
E 312
Hg 312
W-l
Mai lory
RM136H
MS13H
RM 400
E 400
Hg 400
Chemistry
Mercury
Mercury
Mercury
Mercury
Silver Oxide
Mercury
Voltage,
volts
1.4
1.4
1.4
1.4
1.5
1.4
Service
capacity,
ma -hour
16
36
36
60
60
80
Weight,
grams
0.5
0.64
0.64
0.98
0.98
1.15
Diameter/
height, cm
0.33/0.55
0.36/0.79
0.36/0.79
0.535/0.79
0.535/0.79
0.345/1.16
                    ^Letter prefixes in "type" show mfgr. RM, Ms Mai lory;

                    e, Everready;  Hg Burgess

-------
       The  battery  is the major limiting factor in the life of the tags.
 The  capacity  of  the batteries vary depending on when they were manufact-
 ured and where they have been stored since their manufacturing date.
       Batteries  of the same type but made by different manufacturers
 vary in their capacity.  The batteries used in the present tags were
 made by Mallory  because of their higher capacity rating 220 ma/hr as
 compared to Everready with 210 ma/hr and Burgess with a rating of
 180  ma/hr.  Batteries purchased from Mallory were ordered directly from
 the  manufacturer and were placed in plastic bags and put into a refrigera-
 tor  when received.  The temperature of the refrigerator was maintained
 at approximately 5 C.
       A higher capacity lithium battery has been developed and is being
 manufactured  in  the larger battery sizes.  The lithium battery has
 approximately twice the capacity of an equivalent volume mercury battery.
 There are  several  manufacturers of the lithium battery, however, at
 this time  there  are no miniature batteries of this type being sold.
 They should be considered for future tags, however.
 Tag  Assembly
       The  tag as shown in Figure 11 is assembled in four steps.  First,
 the  printed circuit board containing the timer is assembled and tested
 for  operation.   The oscillator circuit is next assembled on the back side
 of the printed circuit board.
       The  second step of assembly is the glueing (Eastman 910 Glue) of
 the  transformer  pot core with the cylinder.  The pot core is an iron core
 upon  which the transformer is wound, providing an efficient and smaller
 transformer than if an air core were used,  witn the use of the 125 m
 (0.5  inch) OD cylinders the pot core is inserted into the cylinder to
 reduce the length.  With the smaller high frequency cylinders the
 pot core is attached to the end of the cylinders.
      The third  step glues the cylinder and pot core to the printed
circuit board.  At this point the tag can be tested for operation and
                                    30

-------
          BATTERY
Wire connected to
activate power
                                                                         3
                                                                                .XL.
                                                                                      'A-  StAIML£:SS  STEEL

                                                                                       SCREW
              Figure 11.   Acoustic temperature transmitter (tag) assembly.

-------
can be tuned to its operating frequency.  This involves changing the value
of capacitor C3 until the desired frequency is achieved.  The last step
consists of attaching the battery, applying epoxy to all components except
the cylinder and the potting of the completed tag.
Potting Compounds and Molds
      Potting compounds from several manufacturers (Dupont and Products
Research and Chemical Corporation) were evaluated.  The compounds evaluated
were epoxy resins, silicone rubber, and urethane rubber.  Parameters used
in the evaluation were:  Workability, presence of air bubbles, odor,
hardness, fish toxicity, and moisture absorption.
      The two-part epoxy resins tested had a durometer hardness of 35 to
80 on the Shore D scale.  The moisture absorption rate of the resins was
dependant to an extent on the hardness of the cured epoxy.  The harder
the epoxy the lower the absorption rate.  The moisture absorption rate
varied from .1% to 1.2% after ten days with a relative humidity of 96%
as quoted by the manufacturers.
      The silicone rubber tested had a durometer hardness of 20 to 35 on
the Shore A scale and was very susceptible to water absorption.  This
single part room temperature curing, Silicone Rubber Sealant (RTV) com-
pound was available in tubes and when squeezed from the tube was in a
thin paste form.  Two or more coatings of RTV were required to prevent
water leakage into the test circuits.
      The urethane rubber, a two-part compound, required an evacuation
process to remove trapped air bubbles from the compound.  With a duro-
meter hardness of between 50 to 80 on the Shore A scale, the urethane
rubber proved to be satisfactory for the potting of the tags.  The
semi-flexible urethane adhered well to the cylinders  and presented a
good acoustical  match between the cylinder and the water.
      The manufacturer of the urethane selected was Products Research
and Chemical Corporation (PRC).  The PRC number of the first urethane
used was PR 1524.  This compound was used for over one year before  it
                                    32

-------
was known that it contained a cancer suspect agent, 4, 4'-methylene-bis-
(2-chloroaniline) or "Moca" as named by Dupont.  At that time PRC came
out with a "Moca" free compound PR 1564.  This new compound was used
in the potting of the tags and the hydrophones delivered to EPA.
      Plastic molds were used initially for the potting of the tags.
With the use of the higher viscosity compound PR 1524, trapped air
bubbles remained at the bottom of the mold aften evacuation.  This
resulted in weak spots in the cured compound which made the tags
susceptible to water leakage.  Attempts were made to dip the tags in
the potting compounds, but resulted in an uneven coating around the tags
which after a short period of time in water resulted in water leakage
into the tags.
      With the use of the lower viscosity compound PR 1564 and the use
of the teflon molds the majority of the trapped air could be removed by
evacuation for about 30 minutes.  This resulted in a uniform coating
around the tag with less chance of water leakage.
      Several types of molds were evaluated.  The final mold used was a
"split" mold made from 2 sheets of Teflon (each 2.54 cm x 12 cm x 8 cm)
held together in a vise to form a 5.08 cm x 12 cm x 8 cm sandwich.
A 1.5 cm diameter hole was drilled at the joint between the 2 sheets.
The depth of the holes was approximately 6 cm.  After the potting was
completed the tags were easily removed by separating the teflon sheets.
      Tables 3, 4, and 5 summarize the individual tag overall dimensions,
weight, volume, repetition rate, pulse width, cylinder dimensions, and
coil winding information. Section IV also provides technical specifications
for the tags.
HYDROPHONE/PREAMPLIFIER
Piezoelectric Elements
      Three different sized PZT disc were selected for use in the receiving
hydrophone to construct a wide band hydrophone with good sensitivity.  The
radial  and thickness resonances are used to obtain overlapping frequency
                                    33

-------
              Table 3.   OVERALL DIMENSIONS,  WEIGHT AND
                         VOLUME OF EACH TAG

Tag
number
2
4
5
6
6*
7
8
10
12
13
14
15
16
Length,
mm
32.4
36.4
34.6
33.3
32.5
33.7
30.4
30.8
32.4
31.8
38.9
37.0
38.8
Diameter}
mm
15.5
17.4
15.7
15.5
16.1
15.5
16.1
16.2
15.4
15.3
16.2
16.1
15.8
Weight
in water,
g
4.77
4.72
4.64
5.12
5.17
5.47
4.98
4.69
4.84
4.91
4.77
4.60
4.79
Weight
in air9
g
11.08
12.58
11.50
11.71
11.68
11.98
11.05
10.67
11.18
11.17
12.44
12.01
12.62
Volume
mg/1
6.3
7.8
6.8
6.7
6.4
6.4
6.1
5.9
6.3
6.2
7.7
7.4
7.9
*No longer in use.
                               34

-------
Table 4.  TAG REPETITION RATE AND PULSE WIDTH CHARACTERISTICS

Tag number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
20
Repetition rate,
PPS
.75
.74
.74
.87
.84
.75
.77
.78
.80
.80
.78 •
.76
.84
.78
.84
.84
.82
.74
Pulse width,
msec
2.5
2.6
2.8
2.6
2.6
2.6
2.8
2.8
3.0
2.8
3.0
2.6
3.5
2.4
3.0
2.6
3.4
2.0
Frequency
air
49
58.5
68.4
78.5
89.2
103
108
123
133
145
161
171
184
195
219
242
365
340
water"
51
62
69
80
91
96
no
122
134
145
162
173
183
198
221
236
366
341
                              35

-------
Table 5.  TAG CYLINDER AND COIL SPECIFICATIONS

Cylinder
Tag diameter x length Coil (number of turns)
number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
20

1.3 x
1.3 x
.97 x
.97 x
.97 x
1.3 x
1.3 x
1.3 x
1.3 x
1.3 x
1.3 x
1.3 x
1.3 x
.97 x
.97 x
.64 x
.64 x
.64 x
cm
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
.64
primary
15
15
15
15*
15
13
13
13
13
11
11
11
10
8
8
8
8
8
secondary
90
81
80
64
90
80
71
64
62
60
50
45
40
45
40
50
45
30
tap
20
20
20
18
18
18
18
18
18
15
15
15
15
15
15
15
15
15
Nominal
frequency in water,
kHz
51
62
69
80
91
96
110
122
134
145
162
173
183
198
221
236
366
341
                       36

-------
ranges to cover the frequency range of 51 to 366 kHz.
      The single low frequency disc 31.75 mm OD x 15.4 mm thick has
a radial resonance of 70 kHz and a thickness resonance of 150 kHz.
The single medium frequency disc 19.1 mm OD x 6.35 mm  thick has a
radial resonance of 125 kHz and a thickness resonance  of 300 kHz.
The three high frequency discs (10.7 mm OD x 4.7 mm thick) have a
radial resonance of 220 kHz and a thickness resonance  of 400 kHz.
Preamplifier
      The dual sided receiving hydrophone (Figure 12), consists of  three
sets of PZT discs on each side with each set of discs  having its own
bandpass filter and buffer amplifier to isolate the different sets  of
discs from each other to prevent interaction between the discs.
The acoustic signals received by the discs are filtered by the bandpass
filters (LI L2/C1C7 and L3L4/C8C14	etc.) and are sent to the buffer
amplifiers  (Ql, Q2, Q3...Q6) (Figure 13)    The  sinnals  from each of
the buffer amplifiers are added together and amplified by the low output
impedance power amplifier.  The output of the power amplifier (Q12  and
               Figure 12.   Dual  sided receiving  hydrophone.


-------
CO
CO
        HYDROPHONE
        FROKT SIDE
                                                                                                                LI4
                         LI
        CERAMIC
        1.25" QD.     __
        xO,50"THK   T]'      C7-T
 ^-^r^-j-^m^
•   ••      "-^      cisr:
        CERAMIC
        0.75"OD.
                    1 1
                    1 1
       L3
        CAPs800pf
        CERAMIC
        OA Z" 0,0.
                                    C"
                     r
                     i
                        L5
                                  L6
                                 nnrrir^
        CERAMIC
        1.25" QD.
                           i
                                     C"
                     r
                         L7
.r^
                   cie
          r
          t
        CERAMIC
        0.75"O.D.  ..
        x0.25"THK
        CAp-.800pf
        CERAMIC
        0.4 2" 0.0.
                    11
                    ,,  _ U9_
                 LIO
               Tnnrr
          C"
         r
I	^T
                   C"
                         Lit
                                  LI2
         3

         f
                                            CIS
        HrCROPHONE
       I REAR SIDE
                                                                                                  HYDROPHONE PHEAMP HOUSING
                   J L
                                               Figure  13,   Hydrophone  preamplifier  schematic.

-------
Q13) is coupled into a low impedance balanced transformer (Tl).  The
output of the balanced transformer is connected to the signal cable
through contacts of a relay located inside the junction box at the
selected station.
      Power to the preamplifier and the relay is obtained from the
hydrophone selector switch on the receiver-decoder console.  The necessary
dc power supply is included in the console.
Hydrophone Assembly
      The hydrophone consists of two aluminum housings bolted together
as shown in Figure 14 and Figure 15.
      The rear housing contains three sets of discs mounted on rubber
spacers to insure their vertical centering within the housing.  Cork
rubber is glued to the bottom and the side of the discs.
      The front  housing contains the preamplifier printed circuit
 board and three  sets of discs that are also mounted on rubber spacers
 to provide centering.
      The two  housings are filled, one at a time, with a urethane potting
 compound to within 3 mm of the top edge of each housing.  The housing
 is placed in an  evacuation chamber for 60 minutes to remove air bubbles
                                                             \,
 from the urethane compound.  The urethane is allowed to cure for
 approximately  24 hours.
      Special  care was taken to protect against rf interference (RFI).
 Any rf signal  present in the vicinity of the channel will be picked up
 by the hydrophone assembly.  Some signals will of course be picked up
more readily than others;  such factors as wavelength, power level,
 depth of the hydrophone under water, direction from which the signal is
 generated, etc., will determine the final intensity of the signal which
 is received.   Thus, to eliminate false readings or even mask the
 signals from the fish tags, the RFI must be greatly diminished.
                                   39

-------
41  ELECTROSTATIC-
                                                 Cr Cg, etc.
                                                 PZT Cylinders
           Figure 14.  Hydrophone assembly.
                            40

-------
      Since the hydrophone ceramic  elements must be exposed to the
sonic signals, a solid metal  shield cannot be used to eliminate the
RFI.  What is required is a material which will  inhibit RF signals
and allow sonic signals to pass unattenuated.  Screen materials
(20 x 20 mm to 100 x 100 mm meshes) have been used with varying success
to solve similar problems associated with military sonar systems.
Unfortunately, when tried on the present hydrophones, the attenuation
to power line frequencies and the RF signals present on them, was  not
sufficient to prevent masking of several tag signals.  The final
solution to most of the RFI was the use of liquid silver paint
(Acheson-Electrodag # 5041) which was brushed on the outer surface
of the hydrophone  (Figure  14).
                    Figure  15.  Hydrophone assembly,


-------
RECEIVER

General

      The sonic receiver is designed to monitor various sonic sources
within a frequency range of 50-365 kHz.   It is a sensitive and selective
receiver using super heterodyne circuitry built around modern integrated
circuits and discrete components.  Tuning is accomplished by varying the
voltage controlled local oscillator thus giving the receiver the feature
of optional remote control tuning.  The modification for EPA application
provides a 20 step frequency selection feature with vernier adjustment
for each step, rather than continuous tuning and a mechanical mounting
arrangement which is compatible with the other units of the entire EPA
fish tracking system.  The final receiver is mounted in the console,
Description
      The selected signal source (described elsewhere in this document)
is connected to the 600 ohm winding of the input transformer (Figure 16).
      A band pass filter consisting of capacitors Cl, C2, C3, and C4
and inductors LI, 12 and L3 is included to reject signals outside the
tuning range of the receiver.  The signal input level to the first mixer,
Yl, is adjustable by front panel control Rl - RF Gain,  The output of Yl
is the difference frequency between the input signal and the local
oscillator, Y6.
      The local oscillator frequency range is determined by the
components, C13, C15, L4 and the voltage across CRT, the tuning diode.
The diode tuning controls are connected to pin 9 on the PC board.
Rotating TAG SELECTOR'switch connects resistor networks to the  tunina
diode which sets the receiver channel frequencies.  Each channel frequency
may be adjusted over a limited range by the TAG TUNI FIG CONTROLS.
      The dual outputs of the mixer are connected to two band pass  filters
at the IF frequency, 455 kHz.  These filters establish  the selectivity
                                   42

-------
                                                                                                              «{-
CO
                             C2  LI
                                           Figure  16.   Receiver schematic.

-------
characteristics of the receiver.  The choice of bandwidths is made by
selector switch BANDWIDTH.  The 455 kHz signal is amplified by two
transformer coupled integrated circuit amplifiers, Y2 and Y3.
      Signal detection occurs by the product mixer action of Ql, a dual
gate transistor.  Transistor Q2 and its associated circuitry form a
beat frequency oscillator (BFO).  The output siqnal  of the product
detector is the difference frequency between the IF frequency and the
BFO frequency.  The BFO is adjusted for approximtely 100 Hz.
      The audio signal is amplified to earphone level and speaker level
by 1C U4.  The audio level is adjustable by the AUDIO GAIN control  (AGO).
      The output of the product detector is also fed into the AGC
amplifier consisting of transistors Q3, Q4, and Q5.   The AGC output, pin 5,
is connected to the IF amplifiers, integrated components (1C) U2 and LI3,
pin 4, through the SIGNAL STRENGTH meter.
      Jacks are provided on the front panel for earphone, loudspeaker,
and oscilloscope monitoring of the audio signal and a jack is provided
to monitor the AGC level.
Operation
      The receiver is made operative by applying power to the system.
To monitor a specific tag, rotate TAG SELECTOR switch to  the  desired
channel.  Controls AUDIO GAIN,  RF GAIN and the appropriate control
of the TAG TUNING CONTROLS group may be adjusted to  produce  a satisfact-
ory signal.  The relative signal strength of the received tag signal
may be observed by the deflection on the SIGNAL STRENGTH  meter.   This
is a relative reading only, any deflection above the steady state level
can be compared with other signals.
Adjustments
      There are only two adjustments to be made on the receiver,  the local
oscillator and the beat frequency oscillator (BFO) frequencies.   These
adjustments are to be found on the back of the enclosure holding  the
                                    44

-------
receiver electronics.  There are two access holes in the compartment cover,
marked |_0 level oscillators and BFO beat  frequency oscillator.  To
adjust the local oxcillator, rotate the TAG SELECTOR switch to Tag 1, and
set the Tag Tuning Control to mid scale,  and the BANDWIDTH switch to 1.5
kHz position.  Connect a signal generator to the input of the receiver and
adjust its frequency to 50 kHz.  Using an insulated hexigon alignment tool,
adjust the case of L4, maximum signal strength, as indicated in the SIGNAL
STRENGTH meter.  The output level of the  signal generator should be about
11 microvolt for this adjustment.  The beat oscillator may be adjusted to
produce a beat of about 1000 Hz.

  TEMPERATURE DECODER
        The  fish tags used in  this  system  produce a continuous series of
  energy bursts whose period varies as  a function of temperature.  The
  temperature information is contained  in  the  period of the tag, and the
  temperature decoder extracts this information by measuring the time
  between the signals from the tag  and  applying to that quantity a
  multiplication constant.
        The  measurement of time is  accomplished by counting internally
  generated  clock pulses during the interval between tag  signals.  From
  this point the decoding process  involves a scaling operation which is
  performed  by first converting the contents of the counter to an analog
  voltage and then modifying that  voltage  by means of two operational
  amplifiers whose function it is  to compensate for the variations in the
  period-versus temperature characteristics of the tags.  The output of
  the operational  amplifiers is then displayed as temperature on a digital
  (voltmeter) display on the front panel.
  Operation
        The  temperature decoder operates as follows:  A timing module  (U3),
  shown in Figure 17, is used  as a  source  of clock pulses (approximately
  900 Hertz) which are accumulated  in a counter (U4).  Each time a  signal
                                     45

-------
                                                                            ONE SHOT
                                                                                            ASTABLE
CT»
        TEMPERATURE  DECODER
         AUDIO
         INPUT
         FROM
        RECEIVER
                                C!
R44
2.2K
                                                                                                      D4IN4I48
                                                                                                        KJOTE 3
                                                                                                        R36220K
                                                                                                 &AIW/OFFSEJ
                                                                                                   OUTPUT
                                                Figure  17.  Temperature decoder.

-------
is received by the decoder, the contests of the counter are loaded into
a shift register, (U5, U6 and U7), and the counter is reset.  This is shown
in Figure 18.  Immediately after  the counter is reset, it begins a new
counting cycle which continues until the subsequent audio pulse is
received.  If another pulse is not received, the counter will accumulate
to maximum capacity and load "all ones" into the registers, corresponding
to "mininjum"temperature.
      Monostable 111, which is triggered by the initial signal from the
receiver, has a pulse width long  enough to prevent the monostable from
being re-triggered by echo pulses of the tag^generated acoustic signal.
The signal which loads the register and resets the counter is generated
by the monostable U2 which was fired by the leading edge of a pulse
generated by the monostable Ul (Figure 18).
      A digital to analog converter, 1)8, provides an output voltage
which is proportional to the contents of the register.  The resistance
versus temperature curve of the fish tags are to a large extent linear.
Non-linearities occur, however, at the higher and lower temperature
extremes.  These nonlinearities are corrected by the Op Amp U9 and the
feedback circuitry composed of R39, R40, R41, and D4 at one extreme and
by R36, R37, R38 and D5 at the other extreme.  Offset and gain adjust-
ments of the Op Amp translate into calibration of 0 C and at 25£C.
These are performed by potentiometers on the temperature decoder unit
which are accessed through holes  in the front panel.  Each tag has a
corresponding set of potentiometers for high and  low temperature
calibration.
Calibration
      Calibration is performed as follows:  The TAG SELECTOR is set to
select the desired tag.  The tag is immersed in a water bath at 0  C
for about-10 minutes,  allowing: it to~ stabilize.  & portable
hydrophone is also inserted in the water and connected to the hydrophone
input on the rear of the console.  The potentiometer, identified under
the zero degree column by the appropriate tag number, is adjusted until
                                    47

-------
LOCATION
       Audio
       Input
                                                                 FUNCTION
               Signal From Tag
       Filter
       Output
       Output  	
       Ul  (Pin 3)
t^SOmsA	 inhibit  Re-Trigger
Ar-
       Output  —
       U2 (Pin 3)
                            150/JS
              Load Register
              (Leading  Edge)
       Base
       of Q4
    U4 Pin 11
                        t * 50 ns
      A/—
              Reset Counter
NOTE:  Not To Scale
          Figure  18.  Signal  descriptions of temperature decoder,
                                  48

-------
the temperature display meter  indicates zero.  The tag is then immersed
in water at 25 C for 10 minutes and the corresponding potentiometer
in the 25 C column is adjusted until the display reads 25 C.
      Ordinarily there is very little  inter-relationship between the
zero and 25 degree potentiometers;  however.it is recommended that
the cycle be repeated at least once.
RECEIVER-TEMPERATURE DECODER CONSOLE
      The receiver-temperature decoder console provides the controls
and displays by means of which fish are located in a channel and their
temperature is determined.  The. console contains three elemental assemblies.
They are:  The receiver, the temperature decoder and the power supply.
Operation
      Up to a maximum of twenty tags may be addressed using the TAG
SELECTOR switch on the panel of the console .(Figure-19).  Each tag is
individually selected by one of 20 TAG TUNING CONTROLS.  A separate
selector, VCO VOLTAGE switch, is available to display the VCO voltage
associated with each tag.  The VCO voltage is directly related to frequency.
This is an excellent check on the tag selected for temperature monitoring.
This feature was readily available due to the circuitry chosen to tune
the receiver.  By using voltage tuned diodes to accomplish frequency
tuning, one need only read the voltage on the tuning diode to have a
frequency analog.  By using the digital display on the panel to indicate
VCO voltage, a precise analog measurement of frequency is available.
This measurement is representative of the frequency to which the receiver
is tuned and can be recorded for reference purposes.
      The frequency bandwidth of the receiver may be set to either
1.5 kHz or to 4 kHz by a selector switch located on the panel.  The 4 kHz
bandwidth permits a broader spectrum of signals to be monitored by the
receiver, while the 1.5 kHz bandwidth restricts the reception to a smaller
spectrum of frequency and at the same time reducing the amount of noise
picked up by the receiver.  This has the effect of increasing the signal-
                                     49

-------
                   Figure  19.   Receiver  decoder  console.

to-noise ratio of the reception by approximately the ratio of the band-
widths.  Thus, by switching from 4 kHz to 1.5 kHz the signal  reception
is improved by approximately 2.6.  This is an approximately 50% increase
in reception range.
       The  panel contains  controls for adjusting  both the RF and audio
gains  of the  receiver.  It also  provides a meter to monitor signal
strength,  and phone jacks for  additional monitoring of the receiver
output and AGC voltage and for connecting external phones or an external
speaker.   Since the tag temperature and  tag  tuning parameters can be
represented as analog voltages,  it has  been  convenient to display both
on the same display.  In  order to do  this the appropriate scaling circuit-
ry has been designed into both the temperature  decoder and the receiver.
The display is, in fact,  a digital voltmeter.   The desired parameter may
be displayed  by selectively  throwing  a  toggle switch.
Calibration
       Calibration  of the  temperature  tags is performed at 0 C and at
25 C.  Twenty sets of calibration potentiometers are provided on the
console panel for  this purpose.  These  are related to the 'TAG  TUNING


-------
CONTROLS-so that a tag is selected by the TAG SELECTOR, the appropriate
set of calibration  pots  is  applied  to  the temperature decoder.
      The power  assembly is located in the console  and  is attached  to
the bottom  frame.   The assembly  houses two commercial power  supplies,
and a terminal strip.  A New Jersey Electronics Model RL 1-5 provides
3 amps at 5 volts.   The  plus and minus 15 volt source may be either
of the following units:   Zeltex  Model  Z 15AT100DP,  Analog Devices
Model 902,  or Semiconductor Circuits Model  P741-1015-

INSTALLATION
      System installation was completed during the  period of 25 to
28 June 1974 at  the Monticello Ecological Research  Station  (Figure  1).
The receiver-decoder console was located inside the blockhouse
adjacent to the  conduit  box provided for the  cable  runs to  the channels.
A ground strap was  connected from the  console to  the conduit.
      The power and signal cables from the console to the junction
boxes at the channels were run through a  10.2 cm (4 inch)  steel  conduit
(Figure 20) to provide additional shielding from the power lines.   The
conduit is strapped to each junction box along the channel  to insure that
a low resistance path is provided for better shielding of the cables
from the power lines.  Both ends of the conduit were connected to  4 meter
rods driven completely into the ground.  A cable inter-connecting  box
was installed within the junction boxes at each station to provide
additional  shielding and to protect the exposed cable wires from moisture
(Figure 21).  These grounding and shielding precautions are all  essential
to minimize ground loop currents within the overall  hydrophone,  preamplifier,
cables and  console system.
      The inter-connecting  boxes  have  romex  connectors mounted on  them
for the entry of the signal  and  power  cables.  The  boxes contain a
terminal strip,  provided  for  the  inter-connection of the cable wires,
and a relay.  This  relay  is  used  to activate  the  appropriately selected
preamplifier.
                                      51

-------
Figure 20.  Steel conduit runs from blockhouse.
            Figure 21.   Juction box.
                       !

-------
      The shield of each power and signal cable entering the box is
connected to the box by an adjustable clamp on the romex  connector
(Figure 22).  The shields are connected together again on the terminal
strip.
      The output cable from each hydrophone is run through a 19 mm  (3/4 in.)
conduit to  the junction boxes.  The conduit is connected to the hydrophone
housing by a compression type connector that is mounted through one
wall of the hydrophone housing (Figure 23).
                   Figure 22.  Interconnecting box.


-------
     r
COiOSOLf
                                     leiug
                                    VFLLOVJ
                                   S. Hi £1.0
                                   SHISL&
                                            0  0
                                            0^0
                                           -0*0
                                  WHlTf
=D-tO
2)-^0
   3_

0^<
                                            0   0
                                                      WHITE
                                                      SREY
           BLOC
                                                      NEU.OW
                                                       QLUE
           WHITE
                                                                 tt

                                                                    -^
          Figiire  23.   Interconnecting  box wiring diagram.


 NOTE:   Each cable feeds through 1.27  on (1/2 inch) romex electrical
        connector mounted through side of box.
                                54

-------
                              SECTION VI

                         PROBLEMS AND SOLUTIONS
      Tests conducted at the Monti cello Ecological Research Station
revealed two  interference  problems  that resulted  in loss of reception
of signals transmitted  by  ultrasonic transmitters.
      The two problems  were electromagnetic interference (EMI) and
radio frequency interference (RFI), both of which were being generated
by the overhead power lines.
RADIO FREQUENCY INTERFERENCE
      During  the early  phases of the program an RFI evaluation of the
area was made, see Appendix A, over the frequency range of DC to 50 kHz.
A quick search at higher frequency was also made;  however, no signals
were evident  up to, and beyond, 200 kHz.  The signal  frequencies below
50 kHz appeared to be of a level which would not cause serious RFI
problems. Subsequently, the power companies started to transmit signals
which are now causing problems.  Telemetry signals used for relay switch-
ing are transmitted along  the power lines.  The frequency of the signals
105 kHz, 146  kHz, and 192  kHz fall within the receiving bandwidth of
the ultrasonic receiver, and are close to the frequencies of tags
numbers 6, 10, and 14.  These signals were being picked up at low levels
by the signal and power cables and at higher levels by the hydrophones.
The signals and, to some extent, their harmonics, could also be detected
at low levels within the blockhouse.  The level of these signals were
two or three times higher with the hydrophone located in channel 1 as
compared to the hydrophone being located in channel 2.
      Testing of electrostatic shielding was conducted to determine to
what degree the interference could be reduced.  Shielding materials
consisting of mu-metal, aluminum screen and steel screen, were evaluated
for their attenuation properties.  A frequency generator was used to
produce the RFI signals.  The use of aluminum screen placed over the
                                   55-

-------
hydrophone and connected to the housing resulted in attenuation of
RFI signals by 40 dB where as the steel screen resulted in an attenuation
of only 27 dB.
      A solid mu-metal shield placed over the hydrophone and connected
to the hydrophone housing resulted in an attenuation of 58 dB of the
RFI signals.  Additional testing using silver paint resulted in RFI
attenuation of 67 dB.  Two hydrophones were built and shielded with the
silver paint.   These hydrophones were tested at the test site and
resulted in the attenuation of the RFI signals to a near acceptable
level, that is one can reasonably isolate and recognize tag and RFI signals.
      Additional attenuation of the undesired RFI signals was gained
when the hydrophone cable was inserted in the 19 mm  (3/4  in.)  conduit and
when the signal and power cables were connected inside the inter-connecting box
within the junction boxes.  Further attenuation was noticed when the
front panel of the receiver was cleaned on the inside with steel wool
allowing better contact of the front panel to the receiver housing.
      The RFI signals at this point had been reduced to an acceptable
level and resulted in little, if any, interference with the reception
of the tag signals.
ELECTROMAGNETIC INTERFERENCE
      The EMI signals were picked up by the signal and power cables
and were induced into the receiver at levels high enough to block
the reception of the lower level acoustical signals.  A common mode inter-
ference problem resulted due to these high level signals that required
changes to be made in the hydrophone preamplifiers and the receiver.
To reduce the common mode interference, a balanced transformer was
placed at the receiver input and at the hydrophone preamplifier output.
With the addition of the balanced transformers and additional  filtering
within the preamplifier circuits the EMI problem was reduced to an
acceptable level.  See Appendix A on electromagnetic testing for additional
detailed information.
                                      56

-------
STABILITY OF EQUIPMENT
      The reciever-decoder components were designed to be operated
within a controlled temperature environment.  Temperature variations
of more than a few degrees centigrade will cause variations in the
frequency of the local oscillator within the receiver.  This oscillator
has to be capable of being continuously tuned over a wide frequency
range.  Crystals could not be used in the receiver unless the tags
also contained crystals to maintain a constant frequency at each of
the tag frequencies.
      The hydrophone preamplifiers were designed to operate over a
wide temperature range due to their location within the channels.  The
operation of the preamplifiers is constant over the temperature range
of 0 C to 40 C and over a voltage range of 10 to 18 volts.
TAG LEAKAGE AND AGING
      Some problems were experienced with water leaking into the
 electronics  of the  tags. The problem was solved by covering the external
2 wires which are used to activate the tag with GE Silicone Rubber Sealant,
RTV-102.  To minimize aging effects of tags, all units were stored under
refrigeration.
ACOUSTIC ATTENUATION PROBLEMS
      Serious tag signal attenuation was observed which severely limited
the system range.  Under some conditions, the range was limited to 3 meters,
The problem was caused by vegetation, drifting algae, and, in some cases,
bubbles in the vicinity (within a 3 meter area) of the channel inlet.
To solve the bubble problem, the area was screened off and the fish
were excluded from the inlet area.
                                     57

-------
                                 REFERENCES
 1.   Lonsdale,  E.M.  and G.T.  Baxter.   Design  and  Field Tests of a Radio-Wave
     Transmitter for Fish  Tagging.   Prog.  Fish-Cult.   30_:47-52, 1968.

 2.   Mackay,  R.S.   Bio-Medical  Telemetry:  Sensing and Transmitting Biological
     Information From Animals and  Man.   New York, John Wiley and Sons,  Inc.,
     1968.   388 p.

 3.   Adams,  L.   Progress in Ecological  Biotelemetry.  Bioscience.  15:83-86,
     1965.

 4.   Albers,  V.M.   Underwater Acoustics Handbook.   University  Park,  Pennsyl-
     vania State University Press,  1960.   290 p.

 5.   N.W.  Fisheries Centre.  Electronic Tags  Used to Study  Adult Salmonid
     Movements  in the Columbia River Basin.   National Marine Fisheries  Ser-
     vice, Seattle, Wash.  1974.  p.  1-7.

 6.   Baldwin, H.  Marine Biotelemetry.   Bioscience.  l_5_:95-97, 1965.

 7.   Chapman, C.J., A.D.F. Johnstone, and  G.6.  Urquhart.  Preliminary Acous-
     tic Tracking Studies  of Nephrops norvegicus.  Department  of Agriculture
     and Fisheries for Scotland, Marine Laboratory Internal  Report  IR 74-2
     (Unpublished typescript), 1974.  14 p.

 8.   Hart, L.G.  and R.C.  Summerfelt.  Homing Behavior of Flathead  Catfish
     Tagged with Ultrasonic Transmitters.   Proc.  27th Ann.  Conf. S.E. Assoc.
     Game & Fish Comm., Oct. 14-17, 1973,  pp. 520-531.

 9.   	and H.F. Henderson.  Instrumentation  Problems in the Study  of
     Homing in Fish.  In:   Biotelemetry, Slater,  I.E.  (Ed.).   New  York, Per-
     magon Press, 1963.

10.   	 and I. Mohus.   1973 Equipment.  Fish  Telemetry Report  2.  SINTEF
     and Reguleringsteknikk, 7034 NTH - Trondheim, Norway.   Report STF48
     A73051,  1973.  29 p.

11.   Horrall, R.M., H.F. Henderson, and A.D.  Hasler.   Ultrasonic Tracking of
     Migratory Fishes with an Internal Tag.    Presented  at Amer.  Fish.  Soc.
     Meeting.  .Portland, Oregon. 1965.   2 p.

12.   Ichihara,  T., M. Soma, K. Yoshida, and K. Susuki.   An Ultrasonic Device
     in Biotelemetry and Its Application to Tracking a Yellowtail.   Bull.
     Far Seas Fish. Res. Lab.  7:27-48, 1972.
                                     58

-------
13.   Johnson, J.H.  Sonic Tracking of Adult Salmon at Bonneville Dam, 1957.
     U.S.D.I., Fish and Wildl. Serv., Bur. Comm. Fish. Fishery Bull.   60:471-
     485, 1960.                                                       ~"~

14.   Johnson, R.K.  Acoustic Tracking of Individual Fish.  Oregon State Uni-
     versity, School of Oceanography,Corvallis, Oregon, 97331.

15.   Koo, T.S.Y and J.S. Wilson.  Sonic Tracking Striped Bass in the  Chesa-
     peake and Delaware Canal.  Trans. Amer. Fish. Soc.  1_0]_:453-462, 1972.

16.   Li scon, K.L.  Sonic Tags in Sockeye Salmon, Oncorhynchus nerka,  Give
     Travel Time Through Metropolitan Waters.  Marine Fisheries Review (U.S.).
     35:38-41, 1973.

17.   Malinin, L.K.  Use of Ultrasonic Transmitters for Tagging Bream  and
     Pike.  Report I.  Reaction of Fish to Net Webbing.  Biologiya Vnutren-
     nikh Vod (USSR) Informatsionnyi Byulleten (Biology of Inland Waters.  In-
     formation Bulletin).  7;64-69, 1970.

18.   Monan, 6.E. and D.L. Thome.  Sonic Tags Attached to Alaska King Crab.
     Marine Fish. Rev.  35:18-21, 1973.

19.   Novotny, J. and G.F. Esterberg.  A 132-Kilocycle Sonic Fish Tag.  Prog.
     Fish-Cult.  24:139-141,  162.

20.   Poddubnyi, A.G., L.K. Malinin and V.V. Gaiduk.  Experiment in Telemetric
     Observations Under Ice of the Behaviour of Fish.  Biologiya Vnutrennikh
     Vod  (USSR),  (Biology of  Inland Waters).  6:65-70.  Fish. Res. Board Can.
     Transl. Ser. 1817, 1970.                 ~"

21.   Steadman, J.W.  A Short  Range Underwater Biotelemetry System.  Trans.
     Int. Telemetering Conf.  3;316-324, 1967.

22.   Tesch, F.W.  Experiments on Telemetric Tracking of Spawning of Migra-
     tions of Eels (Anguilla  anguilla) in the North Sea.  Fish. Res.  Board
     Can. Transl. Ser. 2724:  29 p.  1972.

23.   Trefethen, P.S.  Sonic Equipment for Tracking Individual Fish.  U.S.D.I.
     Fish and Wildl. Serv., Spec. Sci. Report (Fish) No. 179, 1956.  11 p.

24.   Ziebell, D.C.  Ultrasonic Transmitters For Tracking Channel Catfish.
     Prog. Fish-Cult.  35_:28-32, 1973.

25.   Lawson, K.D. and F.G. Carey.  An Acoustic Telemetry System for Transmit-
     ting Body and Water Temperature From Free-Swimming Fish.  Woods Hole
     Oceanographic Institution, Woods Hole, Massachusetts, May 1972.

26.   Rochelle, J.M. and C.C.  Coutant.  Temperature Sensitive  Ultrasonic Fish
     Tag, Q-5099.  Oak Ridge  National Laboratory, Oak  Ridge,  Tenn.,  Report
     ORNL-TM-4438.  1973.  26 p.
                                      59

-------
27.  Ferrel, D.W., D.R.  Nelson, T.C.  Sciarrotta,  E.A.  Standora  and  H.C.  Car-
     ter.  A Multichannel  Ultrasonic Marine Bio-Telemetry System for Monitor-
     ing Marine Animal Behaviour at Sea.   In:   Instrumentation  in the Aero-
     Space Industry, Washburn, B.  (Ed.).   Pittsburg,  Instrument Soc.  Amer.,
     J9_:71-84, 1973.

28.  Gaiduk, V.V. and L.K. Malinin.  An Informational  Ultra-Sonic Transmit-
     ter for Biotelemetric Investigations.   Akademiya Nauk SSSR, Institut
     Biologiya Vnutrennikh Vod, Informatsionnyi  Byulleten (The  Academy of
     Sciences, USSR, Institute of Biology of Inland Waters, Information Bul-
     letin).  ]_2:74-78, 1971.

29.  Roland, B.  1973 Experiments.  Fish Telemetry, Report 3.   SINTEF and
     Reguleringsteknikk, 7034 NTH - Trondheim, Norway.  Report  STF48 A73052.
     1973.  22 p.

30.  Luke, D.McG., D.G. Pincock and A.B. Stasko.   Pressure-Sensing  Ultrasonic
     Transmitter for Tracking Aquatic Animals.  J. Fish. Res.  Board Can.  30:
     1402-1404, 1973.

31.  Malinin, L.K. and A.M. Svirskii.  Application of Biotelemetry  to Ich-
     thyology.  TsNIITEI of Fisheries Economic Utilization of the World Ocean
     Fish Resources.  J_:17-39 (in Russian).  English Translation:  Fish Res.
     Board Can. Transl. Ser. 2707.  1972.  29 p.

32.  Slater, A., S. Bellet and D.G. Kilpatrick.   Instrumentation for Teleme-
     tering the Electro-Cardiogram From Scuba Divers.  IEEE Trans.  Bio-Med.
     Engng.  BBE-16:148-151, 1969.

33.  Summerfelt, R.C. and L.G. Hart.  Performance Evaluation of a 74 Kilocy-
    cle/Second Transmitter for Behavioral Studies on Reservoir Fishes.  Proc.
     25th Ann. Conf. S.E. Assoc Game & Fish Comm., Oct. 1971, pp. 607-622.

34.  Lin, W.C. and W.H. Ko.  A Study of Microwatt-Power Pulsed Carrier Trans-
     mitter Circuits.  Med. Biol. Engng.  6;309-317, 1968.

35.  Shirahata, S.  Effect of Externally Attached Sonic Tags on  Fish Behav-
     ior.  (Freshwater Fisheries  Research Laboratory, Nikko) Bulletin  of
     Marine Biotelemetry Research Group.  No. 4, p. 3-12,  Dec. 1971.

36.  Andreeva, I.B.  Scattering of Sound by Air Bladders  of Fish in  Deep
     Sound-Scattering Ocean Layers.  Soviet Phys.-Acoust.  1_0:17-20,  1964.

37.  Batzler, W.E.  Acoustic Target Strengths of Some Marine Animals.    Pa-
     per UW3, 73rd Meeting of the Acoustical  Society of America, 19  April
     1967.

38.  Batzler, W.E., R.M. Regan and G.V. Pickwell.  Resonant Acoustic Scatter-
     ing from Air-Bladder Fishes.  Paper D7,  75th Meeting of Acoustical  So-
     ciety of America, 21 May 1968.  Abstract J.  Acoust.  Soc.  Amer.   44:356,
     July 1968.                                                       ~
                                      60

-------
39.  Hasler, A.D.  Underwater Guldeposts.  Madison, University of Wisconsin
     Press.  1966.  155 p.

40.  	 and W.J. Wisby.  The Return  of Displaced Largemouth Bass and
     Green Sunfish to a 'Home1 Area.  Ecology.  34_:289-293, 1963.

41.  Mersey, J.B. and R.H. Backus.  Sound Scattering by Marine Organisms.
     In:  The Sea, Vol. I.  New York, John Wiley & Sons, Inc., 1962.  p. 499-
     507.

42.  Love, R.H.  Maximum Side-Aspect Target Strength of an Individual Fish.
     J. Acoust. Soc. Amer.  46(3-Part 2):746-752, September 1969.

43.  Nickles, J.C. and R.K. Johnson.  A  Digital System for Volume Reverbera-
     tion Studies.  J. Acoust. Soc. Amer.  J50_:314-320, 1971.

44.  Noerager, J.A., E.J.  Rice and  C.E.  Feiler.  A Method For Reducing
     Groud Reflection Effects From  Aoustic Measurements.  NASA Technical TN
     D-6666.  March 1972.  p. 1-67.
       \
45.  Raitt, R.W.  Sound Scatterers  in the Sea.  J. Mar. Res. 7^:393-409, 1948.

46.  Rochelle, J.M.  Design of Gateable  Transmitter for Acoustic Telemeter-
     ing Tags.   IEEE Trans. Bio-med. Engng.   BME-21(L):63-66, 1974.

47.  Weston, D.E.  Sound  Propagation in  the Presence of Bladder  Fish.   In:
     Underwater  Acoustics, Vol. 2,  Albers, V.A. (Ed.).  New York, Plenum
     Press, 1967.  p. 56-88.
                                      61

-------
                              APPENDIX A
                   ELECTROMAGNETIC FIELD MEASUREMENTS

 TEST OBJECTIVES
      The objectives of these tests are to determine the design  criteria
 for the instrumentation system with respect to the possibility  of
 interference from the power lines that are located overhead  or  from
 the power generators themselves.
 TEST SPECIFICATIONS
      The electromagnetic fields present at the experimental  channels
 will be measured at various points along the channels  in order  to  establish
 a profile of the field at the site.
      In order to relate the on-site test to a known set of conditions
 a set of measurements will  be made in Virginia on  power lines of
 similar construction and power capacity.  By comparing the two  sets of
 data the on-site radiation  contributed by the generators may be
 determined,
      A third set of measurements  will be made in the  Bayshore Systems
 laboratory area where the fish tanks are located and  where the  in-house
 experimental work is undertaken.   The level of interference  in  the lab
i
 can then be compared to that at the actual site and establish a bench
 marker to work from in the  design of the instrumentation.
      The above mentioned tests are to be conducted in the frequency
 range of from DC to 50 kHz, with  equipment having  a  sensitivity of
 1  microvolt, a sensitivity  equal  to, or better than,  that of the propos-
 ed test instrumentation. A search from 50 kHz to 200 kHz will  also be
 made to observe possible spurious signals.
                                      62

-------
TEST RESULTS
     The electromagnetic radiation tests were run &t the EPA test
channel at the Monticello Nuclear Generating Plant at Monticello,
Minnesota on October 5, 1972.
     The Monti cello test site is 40 miles northwest of Minneapolis on
the Mississippi River.  The soil condition at the site is that of sandy
loam to the channel bottom level.  The overall site is flat and 8 to 10
feet above the river level.
     Conditions of the test were with heavy fog during the first half
of the day and heavy overcast during the afternoon.  The temperature was
in the high 40's and low 50's.  Some water was evident on the ground
with surface mud indicating a rainfall earlier in the week.
     The power plant was in operation during the test.  Two or more
pieces of construction equipment were in operation during most of the
test.  During the noon break of the construction equipment a test was
run to establish the effect of the equipment.  There was no apparent
effect, with the radiation from the power line masking out any of the
interference from the construction equipment.
     The map in Figure 24 illustrates the positions of each of the
tests that were run.
CONCLUSIONS
      From the test fcesults, Figures 25 to 37, it is evident that the
signal levels present on the site, resulting from power line radiation,
is of a magnitude that will cause interference to the fish tracking
systems.  The tests have indicated that the design of any system
instrumentation to be used on this site must be designed for operation
in a high nedsc  level environment.
     From the various tests that were run, both on site and local, the
radiation originates from the power lines rather than from the power
generating equipment or the nearby distribution station.  This point
                                   63

-------
CT>
                       SCALE
                       1" = 200'
             Figure 24.   Electromagnetic test sites - Monti cello  Ecological  Research Station channels,
                         Monti cello, Minnesota.

-------
    200
S-

O)
S-

O
S_
o
•=c
H-
—I
O
Q
LoJ
 UJ
 Qi
      20
                                                                        TEST LOCATION:  MONTICELLO

                                                                        TEST DATE:  10/5/72
          10                            100                            1000

                                             FREQUENCY,  hertz

                 Figure  25.   Electromagnetic field measurements  -  composite tests.

-------
en
       Ol
       O)
       Q.
           200
           180
           160
       •5   140
       OJ
       >
       O
       JD
       "3
       
-------
200




180




160




140




120



100




  80




  60



  40




  20
                                                                       TEST LOCATION:  MONTICELLO


                                                                       TEST DATE:  10/5/72
cu
4->
o>


s_
a)
o.

1/1
 o
 s_
 O>
 >
 o
 CD

 H-

 O


 O
 LU
 tJ
          10
                                                                                 T"
                                    100                            1000

                                         FREQUENCY,  hertz


               Figure  27.    Electromagnetic field measurements - test number 2.

-------
CT>
OC
       O)
       O
       s_
       O

       'i

       O
o>
>
O
JO
(O
       CD
       
-------
XQ
       s_
       o>
       -M
       0)
       o>
       Q.

       I/)
       O
       S_
       (J
01
>
O
.Q
iTJ
cs

I—

O


Q
UJ


UJ

UJ
o:
           200
    180
            160
            140
            120
            100
             80
             60
             40
             20
                 10
                                                                              TEST LOCATION:   MONTICELLO

                                                                              TEST DATE:   10/5/72
                                        100
1000
                                                                                                           HI
                                                                                                   IT
                                                                                                          II
                                                    FREQUENCY, hertz


                              Figure  29.   Electromagnetic field measurements - test number 4.

-------
'—1
o
           200
      Ol
      -»->
      CD
      E

      s_
      OJ
      o.

      1/1
       o
       i-
       c_)

      'i

      o
       O
       JD
       fO
       LiJ
       CS
       O
       UJ
       o
       UJ
       DC.
20
                10
                                                                               TEST  LOCATION:  MONTICELLO


                                                                               TEST  DATE:   10/5/72
                                   100                            1000

                                        FREQUENCY,  hertz

                   Figure 30.  Electromagnetic field measurements  - test,
                                                                                                5,

-------
200



180



160



140



120



100



 80




 60



 40



 20
                                                                   TEST LOCATION:  MONTICELLO


                                                                   TEST DATE:  10/5/72
s_

o>
s_
O)
o.
o




LU


LU

LU
o:
     10
                                        100
1000
                                        FREQUENCY,  hertz
                    Figure 31.  Electromagnetic field measurements  -  test number 6.

-------
ro
            200
        £   180
        O)
        E
        O
        >
        O
            160
            140
        u
        e   120
        O
       JD
        (O
    100


     80
       <    60
o
>
o
       UJ
       Qi
             40
             20
                10
                                                                             TEST LOCATION:  MONTICELLO

                                                                             TEST DATE:  10/5/72
                                       100
1000
                                                   FREQUENCY, hertz


                           Figure 32.  Electromagnetic field measurements - test number 7.

-------
CO
         i-
         CD
         O>
         O)
         Q_
         1/1
         O
         >
         O
         s_
         o
O)
>
o
-Q
fC

.a
T3
         C3
         LU
         LU
              40
              20
                  10
                                                                                TEST LOCATION:   VA. TEST


                                                                                TEST DATE:   10/5/72
                                        TOO                            1000

                                             FREQUENCY,  hertz


                       Figure 33.  Electromagnetic  field measurements - comnosite tests.

-------
Ol
•t-1
O)
Ol
CL
     200



     180



     160
o    140
>
O
S-
u
0)
>
O
_Q
(O
CD

I—

O
>

O
UJ
•>
I—I
UJ

UJ
     120



     100



     80




     60



     40



     20
                                                           TEST  LOCATION:   FAUQUIER COUNTY, VIRGINIA


                                                           TEST  DATE:   9/15/72
                                                                      1000
                                             FREQUENCY,  hertz

                             Figure  34.   Electromagnetic field
                                                                              -  test, numhpr 1.

-------
CJ1
       s_
       OJ
       +->
       o>
       s_

       Q.

       l/l
       O
       S-
       u
       O
       -Q
       (T3
       UJ
       UJ
       a:
             20
                10
                                                                  TEST  LOCATION:   FAUQUIER COUNTY,  VIRGINIA


                                                                  TEST  DATE:   9/15/72
                100                            1000

                     FREQUENCY, hertz

Fiaure 35.  Electrnmaanetir. fipld measurements  - test mimhpr ?,

-------
     200
O)

CD
E

S-

Q-
O
S-
o
QJ
>
O
J3
(O
CD

t—«
LU

LU
o:
      60
     40
     20
         10
                                                                  TEST LOCATION:  FAIRFAX  CO,  VA.

                                                                  TEST DATE:  9/20/72
                                        100                            1000

                                             FREQUENCY,  hertz

                              Fiqure 36.  Electromaqnetic field measurements - test number 3,

-------
                                                                 TEST LOCATION:  BAYSHORE LAB AREA

                                                                 TEST DATE:  10/7/72
i-
OJ
4->
QJ
S-
OJ
Q.
O
>
O
s_
O
01
>
O
-Q
fC
T3
LU
CD
Q
LU
CJ
LU
a:
      40
20
                                                                       1000
                                             FREQUENCY,  hertz
                        Figure 37.   Electromagnetic field measurements  -  test numhpr 1.

-------
is made obvious when matching the curves from the different test
locations.
      There are two distinct areas of interference:
            1.  Those of power lines related frequencies up to about
                10 kilohertz generating levels that will interfere
                with the sensor signals.
            2.  Above 10 kilohertz there are those carriers that are
                imposed on the power lines.  The carriers are used for
                the relaying of switching and distribution information
                by the Northern States Power Company.  These signals
                usually fall in the range from 14 to  50  kHz.
                It should be noted that none of these type of signals
                were detected.  It is anticipated that when the power
                plant is fully operational their presence will be
                apparent.  Normally, the measured level  of the signals
                are not as high as the power line frequencies but are
                well above the ambient noise level on site.  Since the
                signals are in the range of the sensor information
                they are not easily filtered out of the system.
      As expected the signal levels vary over the site and are directly
related to the distance from the power lines.  The area  of high test
levels are those that are directly under and run parallel to the power
lines.  Unfortunately the laboratory building itself is  set in a high
level area.
      The most severe problem to face the system designer is the poor
ground that is present on the site.
      The soil condition at the pond is a fairly high grade of sand,
with the top  15 - 25 cm mixed with an organic base.  The  sandy  soil,
which showed good drainage characteristics during the test, will not
hold moisture to any great extent.  This point is also illustrated
by the need of an additional sealer on the bottom of the channel to
prevent leakage and errosion.
                                  78

-------
      The problem presented by the sand is similar to many instrumentation
and communication systems installed on islands or in soil that does
not exhibit good ground characteristics.  Although ground conductivity
tests were not made, a good earth ground should be present at the
water table, which in this soil should be at the river level, about
3.0 - 3.7 meters below the surface.' This would place the reference ground
at about 2.4 meters below the bottom of the pond.  In order to maintain
a reliable earth/ground the ground plane should be somewhat below
this in order to compensate for a shift in the water table.  It is
conceivable that the ground level may be somewhat higher if there
are deposits of minerals.  In observing the soil in the channels this
does not seem the case.
      As stated the sandy soil will present a variable ground condition
to the system designer.  During or after a rainfall the ground conditions
will be good with a low ground resistance.  When the soil dries out, the
ground resistance becomes higher, the sand becomes an insulator and the
so-called ground is now above the earth/ground reference and a resistance
is presented between the instrumentation and the ground reference.
      The sketch in Figure 38 illustrates the channel in the sand base
and the ground resistance (Rg). -;As shown, the channel when full of water
is a rather large, long electrical conductor.
      As shown in the site plan the four channels are located parallel to
a power line for a distance of about 183 meters.  The four channels are
spiced out from the power lines to a distance of about 45.7 met'ers (150 feet),
      As shown in Figure 39, the channels act as four isolated conductors
spaced progressively away from the 345KV power line.  The channels, like
any other conductor under the same conditions, will act as the secondary
of a transformer with a voltage generated in the channels due to the
CQM&ling to the power lines.  Where the instrumentation lines are run
above electrical ground, a voltage will also be induced into the
conductors.  A greater problem will exist where a sensor is "grounded"
                                      79

-------
                                      POWER LINES
                    SENSOR
        immiimiiii [minimi
GROUND
RESISTANCE >
 WATER TABLE OR ELECTRICAL GROUND
       38,  Ground con#iti«fls at the
            KeltttteCTe' Ecological  Re-
            search Station.
                80

-------
—55—


 345  KY
00
                •e
        r
               OJ
                   I,
        r
        r
                              x \\  \
                              INDUCTIVE  COUPLING
                                     Channel  No. 1
                                     Channel  No. 2
                                     Channel No. 3
                                     Channel No. 4
KL
                                                                              Re i
                                                                            WV—|
                               Figure 39. Ground loop circuits,

-------
at the channel but will actually conduct the channel  voltage (V  )  back to
the instrumentation at the laboratory.
      Figure 40, illustrates the methods in which the various interfering
voltages are introduced into the system.  There are sources  of interfering
voltage on both the sensor end and the  load end of the system.  On the
sensor end a voltage is generated when  the induced current from  the shield
flows through the ground resistance (Rg) with a voltage drop I Rg.  An
addition to the ground loss is the voltage developed  from the signal
current in the sensor.  Since the two generators are  in series both of
the voltages will be present on the center conductor  of the  line.   In the
case of the channels nearest the power  lines, the voltage induced  may be
greater than the voltage generated by the sensor.  The induced voltage
(or noise) from the power line may be of such a magnitude that makes  the
detection of the sensor signal impossible.  Some relief from the ground
loop may be had by using twin conductor cables with a floating ground
(common) to the sensor.  Using the two  conductor cables will keep  the
line induced signal off the signal line but will not protect against
the noise from being induced on the shield, and conducted back to  the
load end.
      On the load end there are two.-possible sQueees ofggwo^d loop
voltages.  One of the sources is the same as in the sensor end where the
loop voltage is developed across the ground resistance which is  in series
with the load.  Again, like the sensor  this condition may be relieved
if two conductor cable are used.  If the two conductor cable or balanced
system is used then some type of transformer or other balanced type of
transducer input circuitry must be used such as differential amplifiers.
In the case of the balanced input the ground loop components will  be
dependent upon the balance of the system, at best the ability to hold
balance is not better than about 0.1%.   In the differential  amplifiers
or op amps this is referred to as common mode rejection.
      Another problem associated with a pper ground system is cross
talk that is generated by the common sensor signal currents  flowing
                                    82

-------
       (fa  345 KV
03
co
                           PI
                           JL
                               Dt
                                                                              K,
                                              Channel  No. 1
                                                             ~~AAACL
              T1TR
                                              Channel  No. 2
                                                             LVW-L
                                              Channel No. 3
                                                             —WV-L
j-jAAAr—
                                              Channel No. 4
—AA/V-L
                             Figure 40.   Induction coupling into pool conductor.

-------
through a common ground resistance.   In this case a portion of each
of the sensors will show up at the input of every other sensor load.   This
means that in every channel there will  be the desired signal  and three
undesired signals at various levels.   In some cases this can drastically
reduce the value of the received information causing a low signal/noise
ratio.
      A serious problem can also arise  from a poor ground with lightning
or similar phenomena.  With the channels setting above ground it is
conceivable that the channel could build up a high potential  thhough  the
various discharge phenomena.   For  this reason alone the channels should
be well grounded, in the same way that  a building or other structure
is grounded.
RECOMMENDATIONS
      As a result of the testlthe following recommendations are made:
            a.  Receiver Sensitivity:
                The sensor receiver  sensitivity should be maintained  as
                high as possible (insensitivity) while retaining the
                necessary system performance.  A receiver with high
                sensitivity will be  prone to over load (blocking) and
                generation of intermodulation products and spurious
                outputs.
            b.  Frequency Selection:
                The operating frequencies of the system should be main-
                tained as high as possible while retaining the other
                performance factors.   By all means the operating frequency
                should be above 50 kHz.
            c.  Signal Levels:
                The signal level present in all areas of the system must
                be maintained at a relatively high level with high
                signal to noise ratios.  The use of local preamplifiers
                                    84

-------
c.  Signal Levels:  (Continued)
    at the output of the sensors and intermediate amplifiers
    along the line is recommended.
d.  Shielding:
    Shielded two conductor cables should be used for all
    signal lines.  In addition to the braided shield on the
    cable, the cables should be routed through conduit.
    The conduit should be of a type affording-magnetic shield-
    ing (cold rolled or drawn steel).  At all joints the
    conduit should be bonded with straps or welded.  Each
    of the sensor inputs should be terminated in a shielded
    box.  The termination boxes should be located along the
    line and like the conduit should be well bonded to the
    conduit itsalf.  In addition to the shielding capability
    the boxes should be weatherproof.
e.  Filtering:
    High pass filters will have to be iRCOroc?rated into the
    design of the sensor and the receiver.   The cutoff of the
    filters should be as close to the operating frequency as
    possible, preferably near 50 kHz.  The filters at the
    sensor end of the system may have to reject the line
    frequencies as much as 40 to 50 dB which will require
    5 or 6 section filter.
f.  Earth/Ground:
    The grounding problem that is present on the site must
    tje surmounted.  The twin conductor will help.  For
    the purpose of instrumentation the channel should be
    paired, the lines running up the dike between the
    channels or ponds.  By pairing, a common ground can be
    established for the ponds.  A series of ground rods
    must be driven into the dikes to a level that will provide
                             85

-------
f.  Earth/Ground: (Continued)
    a good ground system.  At this time it is not known to
    what depth or at what intervals the rod will  be set.
    An estimate would be that the rods should be  driven not
    less than 3.7 meters (12-feet) deep and should be spaced
    every 15.2 meters (50-feet) along the line or at each of
    the junction boxes.   At best this will require a great
    number of ground rods to be installed.  The rods should
    be one inch in diameter and made of copper clad steel.
    The junction boxes should be bonded to the rod or better
    yet mounted on the rods themselves.
                           86

-------
                                APPENDIX B
              INVESTIGATION OF INDUCTIVELY COUPLED CIRCUITS
                            FOR USE IN FISH TAGS

      Magnetically coupled transmitters and receivers are not new .
Contemporary portable broadcast receivers use a Ferrite antenna rod
which uniquely responds to the magnetic field of electromagnetic
radiation.  As shown in Figure 41, a typical vertical antenna radiates
electric flux lines vertically with closed horizontal magnetic flux  lines.
Hence, the electrical line is vertical and at right angles to the
magnetic lines.  Since magnetic lines generate an electromagnetic fiild
when they cut electrical conductors, a multi-turned inductor resonant at
the traasmitted signal generates voltage at points a and b.  The unique
sensitivity of the inductor to the magnetic field alone, causes the
broadcast portable receiver to be directional.  Maximum signal is
received when the Ferrite rod is orthogonal to the radios:l1ne?f#Qffl"-tbee
transmitter and minimum when the end of the rod is pointed at the station.
      The expression for E., the induced electromagnetic field (emf)  is:

                       E1  =  wn HA /Mr  =  n d0/dt

                         (The generator equation)
where  H  = received magnetic field intensity,
       A  = cross sectional area of the rod,
       M  = effective rod permeability,
The actual output voltage is a function of the electrical loss parameters
of the tuned circuit.   "Q" is used to express these losses.  Generally,
"Q" is expressed as a bandwidth function, that is f/Af.  ButAf is
actually constrained due to the losses.  In an air core coil these
losses are due to fringe effects, parasitic capacitances and leakage
                                    87

-------
VOLTAGE
STANDING
WAVE
MAGNETIC
FLUX LINES
   Figure  41,   Electric  and magnetic  fields  around
               a  half  wave vertical antenna.
                      88

-------
flux.  For a cored inductor the fringe effects, parasitic capacitances,
eddy losses and frequency constraints (loss tangents) of the core material
are the significant loss factors.  Leakage flux is significant dti the
reduction of "Q" in air core inductors and certain ferrites reduce this
leakage and thereby, significantly increase the "Q".
       Therefore, the significant requirements for a miniaturized induction
link system may be generalized.  First, it should be noted that the general
electromagnetic radiation equation show the electromagnetic field radiation
                                                      p
is reduced as the inverse of the distance squared (1/r ), and the induction,
or near field, falls off as the inverse of the cube^of the distance (1/r3)
for distances greater than 10 coil diameters.  The range of the induction
field communication is limited then to very short distances.  The exact
distances are ultimately dependent of the size of the transmitter and
receiver inductance loops, although specific coil parameters seriously
affect the received signal strength.  The coil parameters must be developed
empirically, and the range must be experimentally determined.  For these
reasons, it is extremely difficult to calculate many of the performance
chanacteristies.  Hence, when the receiving loop size is constrained,
various inductors must be made within these constraints and measured
electrically.  From these data trends may be shown, trends which lead to
the final design figures.  Such data are described in the next section.
EXPERIMENTAL DATA
       Several experiments conducted to evaluate the qualities of solonoid
inductors are discussed in the following section.  For these inductors,
selected core materials representing optimal choices from off-the-shelf
components were used.  Various sizes of litz wire were used.  These results
are shown in Figures 42, 43, and 44.  Once having tabulated the basic
properties of these inductors with regard to core material, core size,
wire and inductance, further tests evaluting thefr performance as receiver
and transmitters were conducted.  The data shown in these curves evaluate
the "Q" against frequency for Indiana General's Q-l and H material.  The
                                     89

-------
500  r
400
300
200
100
-   HOT 30/46  Q
         HOT 7/41  Q
    HOT 30/46  H

    HOT 7/41 H
                            CORE  SIZE
                                                    _i
                10          100        1000

                       FREQUENCY,  kHz
                                              10000
Figure 42.   Variations of Q against frequency for various
             litz wound solenoid inductors.  Core size
             as shown.
                         90

-------
500  -I
400  -
300  -
200  -
 100-
           54T 18/42 Q HAT
            54T 7/41 Q MAT
            54T  18/46 M MAT
54T 7/41  M MAT
                 10         100          1000
                        FREQUENCY,  kHz
                                       10000
     Figure 43.  Variations of Q against frequency for
                 various litz wound solenoid inductors.
                 Core size as shown.
                          91

-------
                               CORE SIZE
200 _
150 -
TOO -
 50 -
        140 T 7/46 Q MAT
         150 T 7/41  Q MAT
       HOT 30/46 H MAT
        140 T T/41  H MAT
~1	1	
 10         100
    FREQUENCY, kHz
                                       1000
10000
    Figure 44.  Variations of Q against frequency for various
                litz wound solonoid inductors.   Core shape and
                size as shown.
                           92

-------
litz wire size is shown.  Since the highest output voltage is a function
of figure of merit "Q", then the higher "Q" values are desirable.  Note
that "Q" is variable with material, number of windings, and wire type.
"Q" peaking is at various frequencies and is possible through judicious
selection of core material and wire type.  "Q" was measured with a Boonton
Model 260 A Q-meter.  It should be noted that the Boonton "Q" meter measures
"Q" with 8 ma. rms through the inductor at a scale multiplication of 1.
It does not measure the actual "Q" at low signal levels such as would
be experienced when the inductors are used in receiving systems as antennas.
In these cases, performance can only be measured and, in general, curves
are a good rational fit with the "Q" data.
       For the actual use data, the test setup in Figure 45 was used for
the accompanying data in both Figure 46 and Figure 47.  Power into the
transmitting inductor is provided by a current source with a maximum out-
put voltage of 25 volts rms.  Since litz wire is not easily available in
larger sizes, some data is shown which obviously conceals some of the
 2
I R losses in the transmitter inductance.  Additional data for higher
powered sources was obtained by using a larger solid copper wire.  The
"Q" of the transmitter coil is not especially  important.  Lower "Q" coils
merely require more voltage to drive the same amount of current as would
be obtained by a higher "Q" coil.  This does not affect transmitter
efficiency, an efficiency which is not important for these tests.  As
stated before, a high "Q" for the receiver inductor is essential as the
signal varies directly as the "Q".
EXTERNALLY POWERED AND EXTERNALLY TRIGGERED TAGS
       To externally power or trigger a tag requires adequate energy
to be transmitted to the tag receiver to either be rectified and hence
provide power to the tag, Figure 48; or to be applied to a sensitive
                                    93

-------
10
                                  35 khz
                                  78 inches
                          2.3 MILLIHENRY
                          Q-l INDIANA GENERAL
                              CORE MATERIAL
                          LENGTH  1 INCH
                          220 TURNS 25/44 LITZ
                             20
40           60          80
 TRANSMITTER CURRENT, mi Hi amperes
                     Figure 45.    Curves showing the relative received energies for a varying source
                                  energy for Q-l and H type core materials.

-------
               1.0,—
VO
en
               0.8
               0.6
CIS
i—i
co
o
           LU
           CJ
0.4
               0.2
  RECEIVER COIL

Ql CORE L=l" D= .125
400T  7/41 LITZ
f= 35khz
                                                   (XMTR COIL = 10.2 mh)
                                                   L = 7"  D= .5"
                                                   INPUT I = 57ma.nms
                                                                              10
                                                     DISTANCE, feet
                                                                               12
                                                                                       14
                  Figure 46.   Curve  showing  the relative received energy for constant
                               source energy.

-------
          1/8 inch diameter  core  Q-l material
          150 turns 25/44  litz
          f 35 khz
                                            BRIDGE
                                            RECTIFIERS
                                   X
D. C. OUT TO
POWER TAG
Figure 47.  Schematic showing the configuration of a
            power pickup and converter for use as a
            power source for an ultrasonic tag.
                       96

-------
                                              LIB
                                       MATERIAL,  "T5KOID
uc
AT
                          -f
Figure 48.   Externally triggered,  battery operated
            ultrasonic tag.
                      97

-------
switch which would turn-on a battery powered tag, see Figure 48,  and
thereby save battery life.  Each of these choices required an evaluation
of the useable power one could transmit to the tag and over what  kind
of distances this was practical.  To externally trigger a battery operated
on, 100 millivolts rms had to be available at the output of the tag receiver
antenna and power (standby) for the switch activation is all that is required
during off times.  Since bridge rectifiers are used, the forward  drop of the
diodes is lost and this contributes to the inefficiency of an externally
powered tag.  In addition, the signal being pulled off the receiving inductor
must be developed across a resistance and this spoils the circuit "Q";
therefore, the current that can be drawn from the antenna is limited.  The
load of the rectifiers and filter can be approximated by a 1000 ohm load.
Any configuration can be evaluated for sufficient voltage by this technique.
Germamium diodes, if available in miniature form, have a forward  drop of
0.3 volts.  For  the two used  in the bridge circuit, 0.6 V is lost.   Earlier
tests at less than 100 ma drive on the transmitting antenna indicated that
higher drive currents would be required to reach the necessary voltage.
Initial estimates were approximately 500 ma.
       Preliminary experiments with higher power were discouraging.  With
the early transmitting inductors which were with litz wire, losses became
excessive above 100 ma. rms.  Increasing the drive power provided little
increase in inductor current.  The effect seemed to be due to the limited
current handling ability of the small litz wire.  When another inductor
was wound with #21 enamel  wire, the'current limit effect was not observed
and the inductor coil could be driven to the limit of the power amplifier.
       The test set-up is illustrated in Figure 49.  The operating frequency
is determined by the signal genertor, an Hp 3310B.  This signal is fed
to a power amplifier, an Hp 467A,which in turn drives the inductor.  The

-------
vo
VO
HP 331 OE>
35 Kite.
A A
y u
^ X
HP %7A
A fi (.-P
u 1 1 y\ \

\
               SIGNAL.
             GENERATOR.
  FbWER.
AMPLIFIED
             Figure  49.   Test equipment configuration for the evaluation of high current
                           inductive field sources and receivers.

-------
inductor is series resonated at the generator frequency with a variable
capacitor.  A current probe, Hp 456A, was connected around one of the
leads from the power amplifier to the inductor.  The inductor consisted
of 80 turns of #21 Formwar insulated wire on a 1.27 cm (0.5 in.) diameter
by 17.7 cm (7 in.) long Indiana General Q-l ferrfte rod.  The inductor
and capacitor were mounted on a non-metallic stand which allowed a variety
of positions and angular adjustments.
       The receiver inductor was wound with a 25/44 litz wire on a
3.2 mm (1/8 in.) diameter ferrite rod of Indiana General Q-l ferrite rod
2.54 cm (1 in.) long.  It was parallel resonant with a capacitor at the
transmitter frequency.  The voltage across the resonant circuit was carried
by coax to a switch which allowed the drive current and received signal
to alternately be measured with a Hp 3400A rms voltmeter.
       The received signal versus drive current and distances are shown
in Figures 50 and 51, respectively.  Since the ultra-sonic tag requires
                                                    ^
1.2 volts for minimum operation, over 2 volts rms must be generated by
the receiver inductor.  From the earlier data showing the received signal
strength, Figure 50, over 10 amps must be driven through the transmitter
inductor to effectively be operational over a distance of 2.125 M (five ft).
The ultrasonic tag used for this test was optimized for low voltage low
current operation (Figure 52).  Some frequency and pulse duration instabil-
ities exist because of the variable voltages generated at the receiver
due to distance and angle variations.  When the antenna is nearly orthogonal,
the signals are so low that the tag will not be activated.  These are
statistically small occurrences and should not seriously ramify the applica-
tion.
ULTRASONIC RECIEVER
       To receive the ultrasonic signal in a laboratory environment a
versatile receiver is required.  Such a receiver was fabricated for this
                                    100

-------
    250  _
                                                              0
    200  -
 in
 E
   150  _
2 TOO
oo
Q
UJ
O
UJ
a:
    50  -
                    T
T
                   500         1000        1500        2000

                    SOURCE DRIVE CURRENT, mi Hi amperes
                      2500
        Figure 50.   High power test data showing received signal  as a
                     function of drive current at a distance of 1.82
                     meters (6 feet).
                                    101

-------
   500-^
   400 _
                                   2000  ma  P-P drive
   300 -
CL.

D.

-------
Figure ^.  Schematic of externally powered ultrasonic  tag.
                        103

-------
program, and it is shown in Figure 53.   It is similar to the receiver
of Figure 16.  The input is a PZT microphone.  The received signal  is
amplified and hetrodyned down to video for audio detection.  No attempt
was made to filter out frequencies beyond the tuning range and should
they be present, they will  be heard as clicks in the loudspeaker.
Basically, it is a tuned frequency receiver and uses one integrated
circuit amplifier which serves also as the mixer for audio conversion
               Figure 53.   Photograph  of ultrasonic  receiver
                           for reception of 35  kHz signals.
                                  104

-------
                               APPENDIX  C
                           FRY DETECTION SYSTEM
GENERAL
       The investigation of  thermal  pollution effects at the Monticello
Ecological Research  Station  requires a  system to  detect and count  free-
swimming  fish  fry  entering and  leaving  a  test channel.  This condition
must be monitored  if meaningful  results offish production are  to  be
measured.  Consultation with fish  hatcheries has  provided a better
understanding  of the habits  of  fish  fry,  their size, quantity,  physical
characteristics  and  general  behavior during the early stages of their
life.
       Basically,  one is concerned with a fish spawning several thousand
eggs in the river  bottom.  Initially, the survival rate is 60% to 70%
and size varies  from 4 to  12 mm  long and  1 to 2 mm in diameter.  In the
early stages the young fish  will be  clumped together on the river bottom.
As they approach a 100 mm  size they  start to school, covering an area of
approximately  one  cubic meter.   The  schools range further afield as they
grow and generally follow  the flow of the current.  Detection of single
small fry would  be exceedingly difficult  but in schools the fry are detect-
able.
       The physical  layout of the Monticello test station provides a
restricted area  at both ends of  each channel.  River water is introduced
at the head of each  channel  through  a well 30.5 cm (1 ft) x 46  cm  (18 in.)
and at the lower end  through a spillway with a V-notch weir gate (Figure 54)
This means that  a school of  fry  entering  or leaving the test channel would
be forced through a  confined area providing an ideal point to establish
a fry monitoring station.
                                     105

-------

            Test channel  entrance.
                                              liiiii


III

111



                   Exit  gate.
Figure 54.   Instrumentation sites for
            fry  detection system.

                      106

-------
 BACKGROUND
       Much  effort  has  been  expended  over  the  last 30 years in attempts
 to  locate, track, classify and  count  fish.  A  literature search was
 conducted to determine  what  methods had  been tried and the results of
 research  in  the  fish  counting area.   The following areas were found to
 be  worthy of further  development for  our specific problem:
       a.  Magnetic field.
       b.  Impedance  technique.
       c.  Electrostatic method.
       d.  Equilibrium  potential  electrodes.
       e.  Sonar pulsing.
       The scope of investigation was  limited  to a detection method
 without consideration of target  classification or data acquisition and
 processing.
 INVESTIGATION OF FISH FRY DETECTION SYSTEMS
 Magnetic Field Techniques
       The detection  of fry  as a  result  of perturbation of an establish
 magnetic field was  investigated  using  a  series of flat wound coils
 bottom-mounted in the channel.   The electrical  Q of the coils could
 not be maintained in  water and the system was  insensitive to fry activity.
 Impedance Techniques
       The impedance  of a section of the channel can be measured as a
 function of  frequency.  This impedance,  R + JX, as a function of frequency
 would take into  consideration the type of fish, its size, and quantity.
After sufficient empirical data  was obtained,  extraneous objects,
 pollutants,  etc., could be isolated due  to their different composition.
To increase  the  sensitivity  of the method, the impedance would have to
                                    107

-------
be normalized for varying temperatures.  This will offer no problem since
the temperature will be readily available for each channel section.
Because this technique is parallel to the electrostatic method most
effort was devoted to the capacitive approaches.
Electrostatic
       The capacitance of a section of the channel can also be measured
as a function of frequency.  This method  is similar to the impedance
technique except it can be implemented more easily.  An oscillator, using
the capacity of the channel section as one element of the frequency deter-
mining LC network was investigated.  A typical impedance or capacitance
plate configuration to be used for either the capacitance or impedance
techniques would in its simplest form be a 2.5 cm (1 in.) pipe laid on
both sides of the channel and one on the channel bottom.  The pipe would
be composed of 3.05 m (10 ft) sections connected by a flexible insulated
coupling.  Inside the pipes each section would have an isolating electronic
circuit to permit making the measurement at each section and allowing the
data to be connected by relatively long wires to the on shore electronics
and control center without these wires introducing stray impedances.
       The use of pipe on the side and bottom of the channel allow more
accurate measurements to be made on a school of small fishes located in
the small depth areas near the shore.
       Obviously, other impedance plate configurations are possible.
Other plate configurations could be fiber glass mesh with silver painted
surface, conventional wire mesh fence material, etc..  In all cases the
metal  surfaces must be covered to insulate the plates from the water.
       Two types of plates were fabricated for this series of experiments.
Aluminum plates 30.48 x 30.48 cm and type G10 copper-clad glass filled
                                    108

-------
epoxy circuit board  15.24 cm x 30.48 cm were used.  It was necessary to
completely insulate the plates from  the water and  the G10 material with
the copper clad sandwiched between glass  insulators proved easiest to
seal.  The G10 boards were mounted on four wooden  dowels separating the
plates  by 15.24 cm.
       Direct current (dc) insulation resistance was checked with the plates
in position in the filled fish tank  and resistance was greater than 1 megohm.
A 1.5 m (5 ft )  long  pair of test leads,  separated by  at least 15.24  measured
7 picofarads capacity.  The complete system capacity with plates in air
was 10 picofarads and the same system when placed  in water exhibited  a
capacity of 1200 picofarads.
       Because of the small variation in  capacity expected as a result of
fish located in the dielectric field, a second set of capacitor plates
were constructed and placed in the water.  The use of two sets of plates
allows us to use a bridge circuit.  The second set of plates are isolated
from the fish and form the reference leg  of the bridge.   The capacity
change caused by the fish is the differential capacity (AC) of the bridge
circuit.
       The passage of a 3.8 cm (1.5  in.)  goldfish  through the test area
produces changes in capacity of 0.25 to 3.2 picofarad depending on the
aspect of the fish during transit.  Maximum capacity change occurs when
the fishes length is perpendicular to the capacitor plates.  Testing
neo-tetras in place of goldfish under the same conditions produced
variations of 2.5 to 3.0 picofarads  capacity.
       This system proved to be insensitive and adjustment very critical.
Stray capacity due to lead dressing  and shifting caused considerable
problems in reproducing test results.
                                    109

-------
Equilibrium Potential Method
       The equilibrium potential approach to fry detection was investigated
following the outline of the paper presented by Spoor and Drummond of
ERL-Duluth.  Only one set of electrodes were used and one channel was
recorded.
       After the potentials in the test tank had stabilized and in the
absence of electrical interference, which would affect the normal chemical
equilibrium between electrodes, significant readings of repeatable data
were recorded, using goldfish, as shown in Figure 55.  Swimming movements
could be  identified, but no indication of breathing activity was observed.
We seemed to be limited in ambient level due to AC power line pickup.
       It was necessary for the fish to pass very close to the electrode
for detection (2cm or less).  Any activity in the tank other than fish
movements produced responses similar to fish movement.  It appears that
a fairly elaborate electrode grid would be required to effectively monitor
a small segment of the channel and would represent a definite impediment
to the flow of large fish and debris.
Sonar Method
       Since the simpler systems originally proposed did appear feasible
from preliminary testing a sonar system concept was set up and tested at
the Monti cello facility.  This is probably the most often taken approach.
Sonar systems for fish location have been built over the frequency range
of 10 kHz to 600 kHz.  They have been used to:  Count salmon going upstream,
measuring the biomass (kilograms) of fish in rivers, lakes, and portions of
the ocean, etc.  with some success.   However, the transmission of an acous-
tic signal in a small area (such as the narrow channel of the Monti cello E-
cological  Research Station)  presents many problems.  This is due to such
factors  as multiple reverberations, side lobes of the transducers,
                                    110

-------
                                             END OF TEST
                                                RUN
                              SMALL GOLDFISH  PASSED
                              CLOSE TO ELECTRODE
                                     ATTEMPTED TO MOVE
                                     FISH INTO TEST ZONE
                             SMALL GOLDFISH PASSED CLOSE
                             TO POTENTIAL PROBE WIRE
                            1 GOLDFISH  PLACED
                             IN TEST AREA
                                          1  MARCH 1973
                                          START OF
                                          TEST RUN

                                          RECORDER:
                                          BECKMAN RP
Figure 55.  Sample test data, equilibrium
              potential  system.
                             Ill

-------
unpredictable variation in reflection from the fish as a function of
its position (can vary as much as 30 dB), reflection from the water
surface and irregular contours, the unpredictable variation in signal
strength returned from fish of different sizes, and doppler shifts due to
rapidly moving fish.  Air in the water and plant life also add to the
problem.  Many techniques have been proposed or attempted to obviate
these disadvantages, such factors as swim bladder, vertebral column
and body tissue resonances can be used to isolate fish acoustically,
especially small fish, from others of different sizes.
       The sonar system tested at Monti cello is shown in block diagram
in Figure 56.  A tone burst pulse generator provides the signal to the
transmitting transducer at one side of the channel.  The signal travels
the acoustic path to two receiving transducers separated  15.24 cm on the
receiving side of the channel, where the two receiver outputs are compared.
A variation in either path is an indication of an abnormality present.
The direction of movement can be determined from phase information.
Target classification is possible by examination of the received waveform
and period of perturbation.
DEMONSTRATION OF FRY SYSTEM AT MONTICELLO.
       The fry detector system test program was conducted to demonstrate
the feasibility of detecting small fry as they enter or leave the test
channels at the Monticello Ecological Research Station, using an ultrasonic
detector system.
       The system will detect fry, and in addition, grass clippings,
clumps of algae, and air bubbles.  The system will not count the number
of fry, but will indicate the time period fry are in the  sonic beam
allowing an estimate of the number of fry in the school.  (There is
                                   112

-------
co
         V  &ATE   AT
          LOWER   E.ND
           TEST   CHAMKIEL
                     Figure 56.   Acoi
                                                                 VERT.
                                                                    OSCILLATORS
                                                                  FULSE  GENERATOR.
                                                                    INTERSTATE
                                                                    HOD.
                                                                          90
stic fry detection system block diagram.

-------
 error  here  unless  it  is determined that the system is indeed looking
 at  fry).
        The  detector transducers were placed in the outlet at the end of
 Channel No.  1 which is 92 cm  (3 ft) wide with a water depth of  76  cm
 (30 in.).   The  sound  field was established 46 cm (18 in.) upstream from
 the "V" slot weir  and 38  cm  (15  in.) below  the water surface.
 There  is  a  source  transducer  on  one side of the channel  and two
 receiving transducers on  the  opposite side.  The signal  received is
 applied to  the  plates of  an oscilloscope and then differences in the
 scope  pattern indicate one interruption of the sound field.  Two receivers
 in  a balanced configuration are the most sensitive to changes in the
 field.
        A group  of  approximately 25 fry (identified as red horse, Moxostoma spj
 20  to 40 mm  in  length were swimming in the current before the  "V" notch
 weir in the  end of the channel and would drift back and forth  through the
 sound field.  These transients were easily detectable in the oscilloscope
 display.  There was considerable floating algae in the area and flowing
 through the  sound field.   Large clumps of algae extending a foot or more
 below the surface,  disturbed the scope display in the same manner as the
 fry.  Algae  at the surface did not interfere with the system.   Agitation
 of  the surface of the water did not interfere with data displays.
       The system was operated at frequencies from 40 kHz to 400 kHz and
 it was found that most satisfactory results were obtained at resonance of
 the transducers  - 90 kHz.   There was no indication of air bladder resonance
or  improved  target acquisition as a function of frequency.
       No attempt was made during this demonstration to optimize test zone
acoustic path length, receiver spacing or acoustic window placement in
relation to the  bottom or distance from the "V" notch.
                                    114

-------
       The tests demonstrated a viable sonic fry detection method,
although further work would be required to properly implement a final
system for enumerating fish fry without interference.
                                    115

-------
                          APPENDIX D

                   SUPPLEMENTARY MATERIALS


     Oversize blueprints and other specific information of limited

interest  are available only from the project officer.   Those materials

are:

     (1)  Engineering blueprints
          —System block diagram
          —Receiver/temperature decoder
          —Acoustic temperature transmitter assembly
          — Interconnecting box wiring diagram
          —Hydrophone assembly

     (2)  Parts lists for some of the five preceding components

     (3)  Instruction and Maintenance Manual  for Large Fish Tracking
          System, 27 p.

Readers who want to review these materials may borrow them by writing

to Ken Hokanson, Ph.D., Chief, Monticello Ecological Research Station,

P.O. Box 500, Monti cello MN  55362.
                                  116

-------
                         GLOSSARY OF ABBREVIATIONS
AGC       Automatic Gain Control
cm        Centimeters
cw        Constant Wave
dB        Decibels
dc        Direct Current
emf       Electromagnetic Field
1C        Integrated Circuits
ID        Inside Diameter
kHz       Kilohertz
mm        Millimeter
OD        Outside Diameter
pm        Pulse Modulation
pps       Pulses Per Second
PZT       Lead Zirconate Titanate
Mbar      Microbar
rf        Radio Frequency
Shore A  Scale of Relative  Hardness
Shore D  Scale of Relative  Hardness
VCO       Voltage Controlled Oscillator
                                  117

-------
                                    TECHNICAL REPORT DATA
                             (Please read Instructions on the reverse I if ore completing)
  1. REPORT NO.

    EPA-600/3-77-035
                                                          3. RECIPIENT'S ACCESSION NO.
  4. TITLE AND SUBTITLE
    Instrumentation to Monitor  Location  of
    Continuously in Experimental  Channels
                                          Fish
                       S. REPORT DATE

                         April  1977
issuing date
                                                          6. PERFORMING ORGANIZATION CODE
  7. AUTHOR(S)
                                                          8. PERFORMING ORGANIZATION REPORT NO.
    Joseph R. Jahoda
  9. PERFORMING ORGANIZA1 ION NAME AND ADDRESS
    Bayshore Systems Corporation
    5404 B Port Royal Road
    Springfield VA  22151
                                                          10. PROGRAM ELEMENT NO.

                                                              1BA608
                                                          11. CONTRACT/GRANT NO.

                                                             Contract 68-01-0752
  12. SPONSORING AGENCY NAME AND ADDRESS
    Environmental Research Laboratory-Duluth,
    Office of Research and Development
    U.S.  Environmental Protection Agency
    Duluth, Minnesota  55804
                                                          13. TYPE OF REPORT AND PERIOD COVERED
                                                           Final.  Completed 10/74	
                                                          14. SPONSORING AGENCY CODE

                                                              EPA/600/03
  15. SUPPLEMENTARY NOTES Q
  isfartorily under fie
  cost1*'.   Supplementary materials,
                       .,  the large fish tracking unit was  delivered ana performed sat-
                       d conditions.  However, the technology  was  not continued as too
                                                   ints  and Maintenanct
                                                     Tf  are aval Table
including blueprints and~Maintenance and Instruction
             Manual, are  aval Table from the Pro.iect
  This  study resulted in the development  and  construction of equipment to        Officer.
  continuously monitor the position  and temperature of up to 20 fish in a water  channel
  486 meters long, 3 meters wide, and  1 meter deep.   The system utilized miniature  sonic
  transmitters (tags) operating in the 51  kHz to 366 kHz frequency range which were
  implanted in 500 gram or heavier fish.   The battery operated tags were pulse modulated
  and designed for over 1 year operational  life.   A temperature sensitive thermistor con-
  trolled "the repetition rate of the  tag  providing tne temperature of the fish  to  an ac-
  curacy of 1  degree C.  The nominal range of the polyurethane encapsulated  tag  was sev-
  eral  hundred feet.  Nominal tag size was 16 mm OD x 32 mm long (4.6 - 5.4  g in water).
  Sixteen hydrophones were located at  30.5 meter intervals in the water channel,  A con-
  trol  console contained a manually-operated, frequency-stepped receiver which could
  select any individual hydrophone,  thus  locating the fish to within +_ 15.25 meters.  Up
  to  20 individual fish could be monitored.   Automatic operation and recording of the
  data  was considered in the design  of the system for future equipment.  Severe  radio
  frequency interference problems were encountered, requiring extensive precautions and
  modification of the channel equipment and wiring.   Also investigated were  passive fish
  monitoring and tracking of small fish fry.   An experimental system was completed  for
  limited monitoring applications.
 17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                             b.lDENTIFIERS/OPEN ENDED TERMS
                                                                         c. COSATI Field/Group
*Biotelemetry,  *Temperature measuring  in-
struments,  *Acoustic receivers, *Monitors,
*Electromagnetic noise, *Ultrasonic fre-
quencies, *Pulse modulation, *Miniaturiza-
tion, ^Hydrophones,  *Preamp1ifiers", *$onar
*Inductive  reactance,  *Impedance        :
          fish location,  control
          console, polyurethane
          encapsulated, mercury
          batteries, .temperature
          decoder, Hartley oscilla-
          tor.,...complementary
         •astable multivibrator
                                                                              06B
                                                                              09F
 8. DISTRIBUTION STATEMENT
     RELEASE TO PUBLIC
                                              19. SECURITY CLASS (ThisReport)'
                                                 UNCLASSIFIED
                                                                        21. NO. OF PAGES

                                                                           128
                                             20. SECURITY CLASS (Thispage)

                                                UNCLASSIFIED
                                                                         22. PRICE
EPA Form 2220-1 (9-73)
                                            118
                                                                *
-------