Andrew Kankey, Ph. D. Candidate

1
Focusing Sound in Harbor Environments
Experiment and Theory

• Andrew Kankey, Ph. D. Candidate
• Dr. Gary Koopmann, Dr. Chris Rahn,
• Dr. David Bradley, Dr. Kyle Becker

2
Background

• ONR was charged with a research initiative to
  devise nonlethal methods of deterring divers with
  destructive intent in order to defend U.S.
  harbors.
• A method has been devised that uses low frequency
  acoustic energy to deter any unwanted swimmers in
  coastal areas.
• Multiple sources are phased in such a way to
  focus acoustic energy at a designated swimmer
  location.

3
Multiple Phased Sources

• Using multiple phased sources in an array will
  allow for beam forming of the sound field.
• Focus the main lobe in the direction of the
  swimmer, and try to reduce the intensity of the
  side lobes.
• Array equation exists for a linear array in open
  space, how well does that work in a reverberant
  area?
• Other, more accurate methods for phasing?
• Developed a phase search algorithm

4
Phasor Representation of Complex Pressure and the
Optimal Phase
Three sources, one receiver, no boundaries.
Phase search algorithm change the phase, so
that the angle of the phasors line up, leading to
fully constructive interference
One source, one receiver, upper/lower boundaries.
5
July 2007 Coddington Cove Trip Summary

• Proved the phasing algorithm worked, gained
  pressures similar to ideal addition of sources.
• No overwhelming pattern for phasing the sources
  was apparent better test set up required.
• Changing the height of hydrophones did not affect
  the phasing.
• Frequency was too low for the given array spacing
  to get any prominent directivity need ½
  wavelength spacing
• Planned final trip May/June 2008 Testing
  100/200 Hz with appropriate array spacing

6
Coddington Cove June 2008 Hydrophone
Locations
USS Saratoga
USS Forrestal
7
(No Transcript)
8
Four J-15(3)sThree HLF-1Ds
Acoustic Sources
7.4 m (24.3 feet) spacing
J-15(3)
HLF-1D
9
Experimental Procedure

• Interrogate the harbor to determine optimal
  phases for either the HLF or J-15 array.
• Hold reference source at zero phase, and cycle
  through 360 degrees of phase on a second source,
  repeat for third, fourth, etc.
• Run demo with optimal phases and record SPL at
  each hydrophone location when energy is focused
  there.
• Run demo with classic array theory phases and
  record SPL at each hydrophone location when
  energy is focused there.

10
(No Transcript)
11
Setup 1 had sources 1 and 3 reversed from how we
had planned. We realized this the last day of
testing, so the rest of the data uses Setup 2.
12
The phases agree within an average of 10 /- 5
for source 2 and 12 /- 7 for source 3 between
the two methods.
13
The sources are still Spaced ideally for 100 Hz.
14
Pressure at each hydrophone when acoustic beam is
focused towards its location. Over the course of
one day.
15
Pressure at each hydrophone when acoustic beam is
focused towards its location. Over the course of
one day. Array theory and optimal phases were
used. HLF source array was used. 100 Hz
excitation.
16
Pressure at each hydrophone when acoustic beam is
focused towards its location. Over the course of
one day. Array theory and optimal phases were
used. HLF source array was used. 200 Hz
excitation.
17
Unwrapped Phase (degrees)
Hydrophone
18
Conclusions from DataCoddington Cove - 2008

• Optimal Phase Search (OPS) Method can be used
  successfully to calculate phases for an array of
  underwater sources in a harbor environment
• Results are similar to classic array theory
  results
• OPS method will take into account any relevant
  reflections and phase discrepancies between
  sources which are not addressed with classic
  array theory
• Requires the a priori interrogation of the
  harbor, but with the data collected it could be
  inferred that although the pressure magnitude may
  change, the phases required to create a maximum
  pressure remain fairly consistent
• More studies need to be performed, but it may be
  necessary to re-interrogate the harbor only when
  there are major bathymetry changes such as a new
  ship in the harbor or a new pier

19
Theory - Acoustic FEM

• Coddington Cove, Narragansett Bay, Newport, RI

20
Acoustic FEM

• Developed an acoustic FEM code in Matlab to model
  the harbor using cubic elements with one DOF per
  node and linear shape functions
• Bathymetry is voxelized to create FEM model
• Voxel smallest distinguishable box-shaped part
  of a three-dimensional space
• Code verified with simple theoretical models
• Duct, Box, Impedance Tube, Underwater Wedge

21
1k m
200 m
High Press.
Low Press.
Underwater Wedge -Pressure Release top and
slope -?-c boundary on left -Source at 100m
depth -Receiver at 30m depth -Transmission Loss
compared to analytical equation
22
Coddington Cove - Modeling
23
In Phase
Rigid Bottom
Complex Bottom
Out of Phase
24
Future FEM Work

• Preliminary verification has been completed, but
  comparisons between experimental harbor data and
  FEM model data need to be completed.
• The model is large and requires extensive
  computer memory and time for computations.
• Use the code to predict required phases to
  eliminate or reduce need for a priori
  interrogation.