TR-278

WHOI

DOCUMENT
COLLECTION

Technical Report

A Surface Recovery Technique
For Deap Moored Vertical Arrays

Marshall A. Paige
Pater J. Hauser
Lee A. Lindquist
Carlos R. Mayoral
John H. Selleck

August 1981

Approved for public release; distribution unlimited

PREPARED BY
COMMANDING OFFICER,

NAVAL OCEANOGRAPHIC OFFICE
NSTL STATION, BAY ST. LOUIS, MS 39522

Ge
/
713 PREPARED FOR

w0-TR-§18 COMMANDER,

NAVAL OCEANOGRAPHY COMMAND
NSTL STATION, BAY ST. LOUIS, MS 39529 oad

oow30

FOREWORD

The requirement to retrieve a malfunctioning
instrumented array anchored in a depth of 4550 meters,
resulted in the development and successful execution
of a surface recovery technique. This engineering
report is intended to provide sufficient design and
procedural details to facilitate utilization of the
technique by others unfortunate enough to find them-
selves in the same predicament.

L are
C.H. BASSETT
Captain, USN
Commanding Cfficer

MBL/WHOI

0 0301 OOb9e0e &

DISTRIBUTION LIST

COMNAVOCEANCOM
CNO (OP 952)
ONR (Code 480)
NORDA

NUSCDET NEW LONDON
NUSCDET NEWPORT
NOSC x
NAVENENVSA

NRL

NAVPGSCOL

DiEEC

WHOL
uUC/SIO
MPL

OSU

UW
UM/RSMAS
TAMU
UNH
JHU/APL

DEEP EP REE PREP

he

PRE EP ERP EER

TR-278

A SURFACE RECOVERY TECHNIQUE
FOR DEEP MOORED VERTICAL ARRAYS

August 1981

UNCLASSIFIED

SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)

REPORT DOCUMENTATION PAGE

i. REPORT NUMBER 2. GOVT ACCESSION NO.| 3. RECIPIENT’S CATALOG NUMBER

4. TITLE (and Subtitle) 5. TYPE OF REPORT & PERIOD COVERED

A Surface Recovery Technique for Deep Moored Final
Vertical Arrays

6. PERFORMING ORG. REPORT NUMBER

7. AUTHOR(s) Marshall A. Paige John H. Selleck 8. CONTRACT OR GRANT NUMBER(s)
Peter J. Heuser
Lee A. Lindquist
Y}

9. PERFORMING ORGANIZATION NAME AND ADDRESS j : ROJECT, TASK
Naval Oceanographic Office sa
NSTL Station
Bay St. Louis, Mississippi 39522

- CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE
August 1981

13. NUMBER OF PAGES

35

. MONITORING AGENCY NAME & ADDRESS(if different from Controlling Office) 15. SECURITY CLASS. (of thia report)
Unclassified

1Se. DECLASSIFICATION/ DOWNGRADING
SCHEDULE

. DISTRIBUTION STATEMENT (of this Report)
Approved for Public Release; Distribution unlimited

. DISTRIBUTION STATEMENT (of the abstract entered in Block 20, if different from Report)
. SUPPLEMENTARY NOTES

. KEY WORDS (Continue on reverse aide if necessary and identify by block number)

Recovery

Salvage

Mine Sweeping

‘Deep Ocean Recovery

. ABSTRACT (Continue on reverse side if necessary and identify by block number)

A surface recovery technique was developed for the retrieval of a
vertically oriented array which had been anchored at a depth of 4550 meters.
The array was severed within 100 meters of the bottom in order to retrieve
two faulty anchor releases as well as the scientific instrumentation.

DD fon", 1473 EDITION OF 1 Nov 65 1s OBSOLETE

S/N 0102-014- 6601 | U
SECURITY CLASSIFICATION OF THIS PAGE (When Data Entered)

iti

(ee ah:
icy ED si

a adalat Y
HO

Contents

ie Introduction
Ie Design Goals and Considerations
III. Moored Sweepline Technique
IV. Design Details - Sweepline Array
\. Determining Array Geometry - Catenary Analysis
Vile Navigation and Positioning
VII. The Recovery Operation
VIII. Acoustic Release Failure Analysis
DX: Conclusions
X. Recommendations

Figures
Figure 1 - MAVA
Figure 2 - Moored Sweepline Technique
Figure 3 - Configuration of Towed Recovery Array
Figure 4 - General Configuration of Sweepline Array
Figure 5 - Anchor/Depressor Weight Rigging Detail
Figure 6 - Sentinel Weight Rigging Detail
Figure 7 - Pinger/Release Rigging Detail
Figure 8 - Grapnel Rigging Detail
Figure 9 - Recovery Array Geometry
Figure 10 - Sweepline Geometry
Figure 11 - Free Body Diagram at Sentinel
Figure 12 - Array Geometry and Imaginary Segment
Figure 13 - Geometry of Imaginary Segment
Figure 14 - Extended Array Geometry
Figure 15 - Catenary Detail - Case 1
Figure 16 - Catenary Detail - Case 2
Figure 17 - Transponder Net
Figure 18 - Transponder Net Geographical Orientation
Figure 19 - Electronic Equipment
Figure 20 - MAVA Deflection vs Current
Figure 21 - Plotting Sheet
Figure 22 - Position Details

Tables
Table 1 - Fixed Constants
Table 2 - Measurable and Calculatable Parameters
Table 3. - Calculated Parameters

way

10
10

16 Introduction

In August 1980 an acoustic array, referred to as MAVA (Moored Acoustic
Vertical Array), was deployed by the Naval Oceanographic Office off the coast
of France and moored in a depth of 4550 meters. Attempts to recover the array,
which was equipped with dual acoustic anchor releases, were unsuccessful. One
release failed to respond in any way to surface interrogation while the other,
although functioning in the transpond mode, would not release on command.

During December of 1980 MAVA was successfully recovered using a tech-
nique which this report will describe in detail.

ies Design Goals and Considerations

The basic goal of the recovery operation was to retrieve the MAVA
instrument package and the data it was expected to contain.

MAVA is a Single, taut line array which consists of a positively
buoyant subsurface instrument package and a vertical hydrophone array. The
array is attached to an anchor by means of % inch diameter kevlar line, a
dual anchor release package and & inch diameter 3X19 construction wire
rope, figure 1. In this particular case, MAVA had been configured to position
the instrument package at a depth of 1900 meters. The water depth was 4550
meters at the mooring location. The anchor releases were located 100 meters
off the bottom.

Bathymetric and geophysical data for the area indicated that the muddy
bottom was nearly flat with no indications of obstructions.

Ocean current data for the area, although sparse, indicated surface
currents up to .5 knot diminishing to .1 knot at the bottom.

Statistical weather information for the area indicated that the condi-
tions would be poor to severe with the probability of the latter increasing
Significantly in January and February. The decision was therefore made to
attempt the recovery early in December.

An important consideration in planning the recovery operation was
navigation. Loran C was not available in the recovery area for geographical
fixes. The only available means of navigation was Satellite Navigation and
dead reckoning was required between satellite fixes. Accurate satellite
fixes were generally available every one to two hours.

During the initial stages of selecting the recovery technique,
several candidate schemes were considered. Submersibles or remote vehicles
were found to be unavailable. Towing a cable with a large grapnel at the end
and trying to snag the array seemed to have little chance for success. A
technique used successfully by Woods Hole Oceanographic Institute at lesser
depths was also considered. Briefly this consisted of circling the array while
paying out cable on the bottom. Grapnels were attached periodically along the
cable length. Having circled thearray, the ship would steam off pulling the
cable and grapnels around the array in an attempt to cut or snag it.

The expectation of poor weather and high sea state, which could greatly
complicate locating the MAVA buoy when it surfaced despite its being equipped
with two radio beacons and two z@non flashers, led to the decision to sever
MAVA below the dual anchor release package. This would permit acoustically
tracking the functioning anchor release hanging from the surfaced array. The
opportunity to retrieve the anchor releases and perform failure analysis was
additional motivation to sever MAVA within 100 meters of the bottom.

Given the conditions of extreme water depth, poor navigation aids and
the expectation of poor weather conditions, it was obvious that the ability
to control and position the recovery apparatus was of paramount importance.

Telples Moored Sweepline Technique

The general concept of the moored sweepline technique is Shown in
figure 2. An acoustically instrumented array, 600 meters long and equipped
with cutting grapnels, was anchored on the bottom at a predetermined location
near MAVA. The ship then steamed away from the anchor point while cable was
being payed out until the sweepline was positioned in a horizontal attitude.
A weight (4500 pounds in water) attached to the upper end of the sweepline
created the desired taut, horizontal configuration. The ship was then maneu-
vered such that the sweepline would pivot about the anchor and intersect
MAVA at some point between the anchor and the dual release package.

Instrumentation of the array with two transponders and a pinger permitted
continuous determination of both horizontal position and vertical configuration
of the array. A transponder net was deployed in the vicinity of MAVA for this
purpose. Details of this instrumentation and its utilization are provided in
Subsequent sections of this report.

The controlling factor in successfully performing the sweepline opera-
tion was the ability to maneuver the array to the intended anchor point and
to control the position of the sweepline in both the horizontal and vertical
planes during the sweeping operation. To maximize control of the recovery
array during the anchoring phase, the approach was made at a speed of less than
one knot, figure 3, and the ship was headed directly into the wind and seas thus
minimizing the tendency for the wind to set the ship off course. During the
sweeping phase of the operation, the wind (20-25 knots) was used to "sail" the
ship around MAVA.

The recovery operation was conducted from the USNS KANE, a 285 foot
long AGS class ship. The KANE is equipped with a trainable bow thruster.
The deck machinery used in the recovery operation consisted of a Western Gear
traction winch capable of a 30,000 pound line pull and a slack tensioner. The
winch was equipped with 12,000 meters of 9/16 inch diameter, 3X19 construction
wire rope.

Ve Design Details - Sweepline Array

The general layout of the recovery array is shown in figure 4, while de-
tails of the construction and hardware are given in figures 5 through 8.
Basically, the array consisted of a sweepline stretched between two weights.
The lower weight, termed the depressor weight, was a steel cylinder weighing

2

2000 pounds in water. Steel was used in lieu of concrete to minimize the
physical size of the weight and to ease shipboard storage and handling. A
110 pound "Bruce" anchor was attached to the depressor weight. This
provided the necessary horizontal holding force to fix the end of the array
on the sea floor. An EG&G Model 322 acoustic release, installed in-line
between the "Bruce" anchor and depressor weight, provided a means to release
the sweepline from the bottom, figure 5.

The upper sweepline weight was designated the sentinel weight. This
4500 pound (in water) steel cylinder was attached to the array 600 meters
above the depressor weight. Its purpose was to create the horizontal attitude
of the sweepline. Attachment of the sentinel weight to the array, where the
cable would make a transition from a horizontal to a vertical orientation,
was made in such a way that a sharp bend in the cable was precluded, figure 6.
A pinger and an acoustic release were attached above the sentinel weight to
provide vertical and horizontal positioning information, figure 7.

Seven grapnels, designed and fabricated at NAVOCEANO, were distributed
along the center portion of the sweepline. These were designed to hook the
4-inch wire rope mooring line and wedge it into a sharp-edged V-notch,
figure 8. It was intended that this cutting action would damage the cable
enough to cause it to fail in tension.

V. Determining Array Geometry - Catenary Analysis

As previously noted, the recovery operation was carried out at very
low speed. This condition was imposed to permit a static analysis of the
cable geometry to be made thus providing the required information concerning
the vertical configuration of the cable during the sweeping operation. Given
the positions of the ship and sentinel as determined from acoustic ranging to
the transponder net and the altitude of the sentinel using a pinger, the exact
altitude of the sweepline at any point along its length was determined. Thus
it was possible to control the sweepline altitude by either varying the ship's
position or by varying the amount of cable payed out.

The geometry assumed by the sweepline and cable with fixed anchor and
Sentinel weight formed two distinct catenaries joined at the sentinel. This
geometry is illustrated in figure 9. The anchor and depressor weight are at
point A with the sentinel weight at point B. The arclength AB is the Ue
Sweepline which forms one catenary while the rest of the cable, arclength BC,
forms the other.

In order to set up the catenary analysis, it was necessary to decide
what parameters were known or could be determined by direct measurements, and
what variables would have to be calculated from the analysis. Tables of fixed
constants, measured parameters and calculated variables were set up. The
nomenclature for these values are given in tables 1, 2, and 3 on page 10.

Table 1 cites the "Fixed Parameters." These values would not change
during the retrieval operation, but were input values that could be changed
at any time up until the actual operation commenced. The linear density of

the cable would not change unless a different cable were used. The depth at
the anchor point, given to be 4515 meters, was a value that could be changed
if required. The depressor and sentinel weights were approximate values
that would be refined when the weights were actually constructed, and the
length of the sweepline would be fixed at the time the sweepline array was
deployed.

Table 2 lists those parameters that could be measured during the
operation. For the catenary calculations these values are redundant and some
could serve as inputs for the calculations while the rest provide a means for
checking the accuracy of the theoretically computed values. Furthermore, any
of these values could be computed from the other inputs using the catenary
equations. If any of the instrumentation failed, the analysis provided a
means for calculating the missing parameter.

Table 3 lists some calculated values used in determining the sweep-
line and cable geometries. This information would be used to ensure that
the sweepline was off the bottom and simultaneously below the MAVA releases.

Usual catenary calculations employ a coordinate system whose origin
is offset some vertical distance from the lowest point of the catenary
(i.e., where the catenary becomes horizontal). This is demonstrated in
figure 10 where the offset is c. H is the horizontal tension at the bottom
and T is the tension at the top of the catenary. The arclength is s and W
is the weight of the cable occurring at the midpoint of the arc.

It could not be assumed that the sweepline would form the geometry
depicted in figure 10. If too much cable were payed out, some of the line
would lay on the bottom. This would have the effect of moving point A farther
down the cable, resulting in a new catenary of shorter arclength. In this
case the standard catenary equations would hold; however, an adjustment in the
position of point A and coordinate system would be required. On the other
hand, if not enough cable were payed out to achieve the geometry shown, then
the assumption that the cable was horizontal at point A would be incorrect.

This led to two cases for consideration:

Case 1 - sweepline catenary is horizontal at point A
Case 2 - sweepline catenary is not horizontal at point A

This first case was solved using standard catenary equations. The second case
required some additional analysis and introduced new parameters for considera-
tion. Both cases will be considered separately.

Catenary Analysis for Case 1

Figure 10 can be considered a free-body diagram of the sweepline
anchored at point A without the sentinel weight. The vertical coordinate
offset c is calculated from:

c= seeendh (1)
2d

where: S = catenary segment arclengtn
d

height of point B off bottom

4

The arclength s would be fixed when the array was assembled. The
height d would be measured during the operation using an acoustic pinyer
attached to the cable. For a given arclength and height, the horizontal
length x of the cable is:

Xie MN) ie (x) 1 (2)

where: y=ctd
The horizontal distance x between the anchor and sentinel would
be determined from positions acquired using acoustic transponder navigation.
A transponder was attached at the sentinel location. This value was, there-
fore, an input to the calculations. However, should the transponder fail
to operate the parameter could be calculated using equation (2). Further-
more, the measured value could be used as a check of the analysis by comparing
it with the calculated value.
The tension at the bottom of the catenary is given by:
H = cw (3)
where: w= linear density of cable
The tension at the top of the catenary is:
T = yw (4)
The weight of the catenary is the linear density times the length:
W = SW (5)

Summing forces in the x and y directions and dividing yields the
angle between the tension vector and the horizontal at point B:

@ = tan! te (6)
A free-body diagram of point B, where the sentinel weight attaches
to the two cable segments, is depicted in figure 11. In this figure, P

represents the sentinel weight, T and 8 represent the tension and its
direction in the lower catenary cable segment, and T' and @ represent the
tension and its direction in the upper catenary cable segment. Summation
of forces in equilibrium yields:

1 Tsin 9 + P

OS een Tcos Q (7)
and
Tcos 9

The cable segment BC forms a portion of a catenary whose loading
conditions are calculated for point B. In order to use standard catenary
analysis, the segment can be extended to include an imaginary portion of
cable which would form a single continuous catenary with the same resultant
loads at point B. This concept is illustrated in figure 12.

The segment A'B is the imaginary portion with arclength s', horizontal
length x' and horizontal tension H'. The vertical displacement for the
catenary coordinate system is c'. Figure 13 depicts a free-body diagram
of this segment. The parameters in this figure are calculated as follows.
The value of y'

is:
pete

y W (9)
The horizontal tension H' is:

All Se GOS (10)

The coordinate displacement c' is:
ie ule

Cras (11)
The height d' is given by:
d' = y' at c! (12)

The horizontal distance is then:
' \2
ieee 8 Miss Meare 3
x 2c! in (x: y(t) ] (13)
The arclength s' is:
Si Be y - (C (14)

The weight of the imaginary segment is:

W' = s'w (15)
For the entire catenary A'C in figure 12, we have:

Dee =) Di tedmvaud (16)
and

VP SD Rae (aim)

The horizontal distance is:

' 1 2
K' = 6! In ie + (*1) t ) (18)

The arclength S' is:
S! = ye a (ce (19)

Equations (9) through (15) provide the parameters for the imaginary
catenary segment A'B, while equations (16) through (19) give the required
parameters for the entire catenary A'C (imaginary and actual segments).

It is now a simple matter of subtracting the imaginary segment from the
entire catenary yielding the required parameters for segment BC, the actual
catenary segment.

The horizontal distance X for the catenary segment BC is:

The total length of cable ABC shown in figure 9 is given by:
S. 254s (23)

The total weight of the cable plus load is:

Me San oh G (24)
The total horizontal distance xX. is:
Ke tx (25)

In summary, the known values are:

linear density of cable
length of sweepline

height of sentinel

weight of sentinel

depth from surface to bottom

[Sp neler, (i) Ex
nou ow wou

The calculated parameters, not including intermediate values, are:

horizontal tension at anchor
tension in sweepline at sentinel
T' = tension in cable at sentinel
= tension in cable at ship
x = horizontal distance of sweepline
horizontal distance from anchor to ship
T total cable payed out from anchor to ship

ae
tou

The calculated values are used to indicate the tensions in the cable
and the geometry of the cable. As previously mentioned, the tension at the
ship, T" , and the distances x, X, and S. could be measured inputs. In fact,
such measurements could serve aS a means of verifying the theory and the
calculations.

Catenary Analysis for Case 2

We will next consider the case when the sweepline is not horizontal
at the anchor. In this case, the sweepline segment forms a portion of a
catenary. Development_of the analysis can be made along the same lines used
for the cable segment BC in Case 1; which is to assume the catenary segment
to be extended so the usual catenary calculations may be applied. This is
illustrated in figure 14.

In this case the loading conditions are unknown. For a direct
analysis of this situation values for at least two parameters must be assumed.
For our purposes the values s, the segment arclength A'B, and Q, the angle
of the sweepline at the sentinel, were chosen. The analysis proceeds as
follows.

Select s and @. Known values are s_, the actual length of the sweep-
line from anchor to sentinel, and w, the linear density of the cable. The
horizontal tension at A', the bottom of the imaginary catenary segment, is
calculated from equation (6) by:

_ W
H = tan 0 (26)

Where W is the weight of the total segment including the imaginary
part and is calculated using equation (5). From equation (3) the coordinate
system vertical displacement is:

c=p (27)
Using equation (1) the vertical distance from A' to B is:
dae 4 Wee + 5 (28)

The horizontal distance x is given by equation (2). The tension
T at the top of the sweepline is calculated from equation (4).

Using equation (28) the vertical height of the imaginary segment

AGAGAIS"
2 We
S

Ql? Se gee Fa (29)

il
nn
1
n

From equation (13) the horizontal length for the imaginary segment

2
x'=ciln [+ (¢) ail (30)

USS
Cc
Where: y' =c+d'

Therefore, the horizontal distance for the actual catenary segment

AB is:
5a pagans ae (31)

The vertical height is:

d

is d -d' (32)

The tension at A, the anchor, is:

t' = y'w (33)
The tension occurs at an angle 9 given by equation (6) as:
g= tan (34)

The tension at point A may be divided into horizontal and vertical
components. The vertical component tends to lift the depressor weight
decreasing the horizontal load that it can hold. The tension components
are given by:

P= t'cos@ (35)

Py t'sing (36)
Equations (26) through (36) may be used to analyze the sweepline
tensions and geometry. Equations (7) through (25) of Case 1 may be used to
calculate the rest of the cable parameters (substituting x, for x and d. for
d where required).
In summary, the following values are assumed:

length of lower catenary segment including imaginary

S =
portion and sweepline
@ = angle of sweepline at sentinel

The known values are:

Sa bin length of sweepline

w = linear density of cable

P = weight of sentinel

D = depth from surface to bottom

The calculated values included:

P. = horizontal tension load at anchor

Py = vertical tension load at anchor
T = tension in sweepline at sentinel
hg horizontal distance of sweepline
d. = height of sentinel

@' = angle of sweepline at anchor

T" = tension of cable at ship

Xt total horizontal distance from anchor to ship
Sy = total line payed out from anchor to ship

Again, many of the calculated values could be measured directly with
instrumentation. Tables were developed usin values of s from 670 meters
to 1200 meters and 9 varying from 5° to 80° (it was estimated that a sweep-
line length of approximately 600 meters would be used). Plots were also
made drawn to scale of the cable and sweepline at the various geometries.
It was felt that from instrument readings visual “interpolation" of cable
geometries between the various plots could be made. In this way a visual
indication of cable behavior could be made in real time, hopefully, with-
out the need of simultaneous calculations. Examples of the catenary plots
which were thus generated are shown in figure 15 and 16, -

TABLES
Table 1. Fixed Constants

linear density of 9/16 in. dia. cable
ocean depth at MAVA

- length of sweepline

- weight of sentinel

unos
!

Table 2. Measureable and Calculatable Parameters

T" - tension of cable at ship

d - height of sentinel off bottom

x -- horizontal distance between anchor and sentinel
X - horizontal distance between sentinel and ship
S - amount of cable payed out

Table 3. Calculated Parameters

- horizontal tension at bottom of catenary

- tension at top of sweepline

- tension at bottom of cable

weight of cable

- angle at top of sweepline

- angle at bottom of cable

- horizontal component of cable tension at depressor

- vertical component of cable tension at depressor

uv us O=f=TtAt
ak
t

10

VI. Navigation and Positioning

The requirement for accurately monitoring the ship and sweepline array
positions during the recovery operation dictated the need to deploy an acoustic
transponder net in the vicinity of the MAVA. The critical assumption had been
made that the one funccioning anchor release on the MAVA would continue to
function long enough to deploy the transponder net. This assumption was made
following an analysis of the life expectancy of the transponder batteries. It
was further assumed that the MAVA transponder would function long enough to
complete the retrieval operation. Thus it was decided to use the MAVA trans-
ponder as one corner of a triangular net therefore reducing the number of
transponder arrays which had to be deployed. A backup transponder array was
available, however, in the event that the MAVA transponder failed during the
operation.

The general configuration of the transponder net is shown in figure 17.
The transponder arrays were rigged such that the transponder was positioned
100 meters from the bottom. Transponders "A" and "B" were deployed 1.5 miles
from MAVA to form a right triangular navigation net. The line from transponder
"A" to MAVA was oriented with respect to the prevailing wind direction for the
area during the month of December.

Determination of the exact geometry of the navigation net was done in
several steps. First, the depth of each transponder ("A", "B", MAVA) was
determined by taking depth soundings above each transponder and also by measuring
the minimum slant range to each transponder. The length of each leg of the
triangular net was determined by measuring the slant ranges to the transponders
‘at each end of the leg while slowly crossing that leg. These slant range data
pairs were then converted to horizontal ranges and added together. When the
sum of the two horizontal ranges reached a minimum, the length of that leg
was defined. By determining the length of each side of the triangle in this
manner, the net geometry was defined.

Orientation of the navigation net relative to magnetic north was done
by running a straight course between two satellite fixes and continuously
plotting horizontal distances to MAVA and transponder "A" from dead reckoning
(DR) positions along the ship's track. The bearing of the line between "A"
and MAVA was thus determined, figure 18.

A precision, hull mounted speed sensor had been installed for the
purpose of determining the DR positions during the operation. The sensor was
designed and built at NAVOCEANO using a 4 cm. diameter rotor. The low thres-
hold (3 cm/sec) and high accuracy (+2 cm/sec) of the sensor was particularly
Suited to the low speed recovery operation.

The sentinel was instrumented with an acoustic transponder and pinger.
The former permitted acoustic ranging to the transponder net while the latter
provided the height off the bottom data required to convert slant ranges to
horizontal range. The altitude of the sentinel was also used in order to
define the catenary shape of the sweepline array.

a

A Benthos Model 2214 bottom finding pinger provided sentinel altitude
information. It was thought that a continuously operating pinger might
interfere with interrogation of the acoustic releases so a modification was
made to the pinger to permit an adjustable on-off duty cycle. A circuit
was designed and installed in the pinger; an operating cycle of 19 seconds
on and 57 seconds off was selected.

AMF/EG&G Model 322 acoustic releases were used for transponders. Eight
were selected for the operation, each with a different receiver channel.
Special attention was given to the two releases to be attached to the sweep-
line. The tilt function was disconnected in these two, and both were retuned
to receive a 9.0 kHz and transmit at 11.0 kHz. Also, the sensitivity was
increased to -92 dB/320 MV P/P (millivolts peak to peak) which later proved
to be a problem. The increase in sensitivity caused random pinging of the
transponders as they were being towed in the water. This was attributed to
flow generated noise. This problem was eliminated when the sensitivity of
the transponders was reduced to -88 dB/320 MV P/P. Command receiver channels
1 and 3 were chosen for the sweepline acoustic releases to minimize the
possibility of adjacent channel interface.

The acoustic releases for the navigational markers were also retuned.
The reply frequency was set at 9.5 kHz for transponder "A" and 10.5 kHz for
transponder "B". Command receiver channels 7 and 5 were selected for trans-
ponders "A" and "B" respectively. A third marker, designated "C", was also
set up but never used. The remaining acoustic releases were prepared as
backups for the navigational markers and sweepline array transponders.

The deck units used to receive each of the different transponder returns
were a AMF/EG&G 206A receiver and a AMF/EG&G 301 receiver. A AMF/EG&G Model
200 coder and amplifier connected to a AMF/EG&G 301 transducer were used to
interrogate the transponders.

The 206A is a 4 channel receiver capable of simultaneous reception and
decoding of four discrete frequencies. The readout is in milliseconds. The
count is started when the interrogation pulse is sent out and each channel
stops when it receives a reply. The 206A was tuned to receive frequencies of
Gro wKHzis) LOOP KHZ. 1 Oman Khizaram cuales On khiz—

The 301 receiver is a single channel receiver and displays bearing
and range in kiloyards. This receiver was modified so that the trigger pulse,
initiated by the 206A also reset the 301 receiver. The receive frequency for
the 301 was tuned to 10.0 kHz. The Model 301 transducer which was used is
permanently hull mounted on the USNS KANE. This arrangement of receiver
equipment was found to be workable. However, early on in the operation it was
discovered that interrogation range was very limited. Two miles from the
transponder net, the ship's engines had to be stopped and forward motion had
to slow before return signals could be received from the transponders.

Upon investigation, it was found that using the hull mounted 12 kHz
wide beam transducer and associated transceiver resultec © = much higher
Signal to noise ratio. This eliminated the need to stop the ship's engines
in order to receive the transponder replies. The AMF/EG&G 206A receiver's

12

input was wired to the audio jack on the PFR recorder being used to monitor
the 12 kHz wide beam system. Range to the transponders was thus increased

to 6 miles. The 12 kHz return from the bottom finding pinger was also
strong. It was discovered that the duty cycle of the pinger could be ignored
because no interference was noted between the pinger and transponders.

Ship and sentinel positions were determined using the equipment setup
illustrated in figure 19. The choice of transmit and receive frequencies
for the various transponders was dictated by the transmit and receive
frequencies of the functioning MAVA release. The ship's position was
determined by transmitting an 11 kHz ping to the transponders in the net
("A","B" ,MAVA). They in turn responded by transmitting pings of 9.5 kHz,
10.5 kHz and 10.0 kHz respectively. These signals were then decoded by
the AMF/EG&G 206A receiver. Conversion of the three round trip travel
times into horizontal ranges to each transponder permitted plotting the
ship's position at the intersection of the three radii. The sentinel posi-
tion was determined by transmitting a 9 kHz ping to the sentinel transponder
which in turn transmitted at 11 kHz. The sentinel transmit ping (11 kHz) was
then received by "A", "B", andMAVAwhich in turn transmitted at their
designated frequencies. Once again, these signals were decoded by the
AMF/EG&G 206A receiver which displayed the following four round trip times:
1) Ship-Sentinel-Ship; 2) Ship-Sentinel-"A"-Ship; 3) Ship-Sentinel-"B"-Ship;
4) Ship-Sentinel-MAVA-Ship. Having determined the travel times from the
ship to "A" (as measured using a direct interrogation at 11 kHz) and from
the ship to the sentinel, these may be subtracted from the overall round trip
time yielding the slant range between the sentinel and "A". Given the
vertical position of the sentinel, as determined by the difference between
the direct and bottom return from the pinger, the horizontal range from the
sentinel to "A" was calculated. This process was then repeated for the "B"
and MAVA transponders. The three radii were then plotted to intersect at
the sentinel position.

VII. The Recovery Operation

The operation commenced on the 8th of December when the still func-
tioning MAVA transponder was located. By the end of the day on the 10th
of December the transponder net had been deployed and its geometry accurately
determined. The major difficulty encountered during this phase was navi-
gation. Without Loran C, we were forced to dead reckon between satellite
fixes and occasionally got "lost".

On December 11th, degrading weather conditions (30 knot winds)
forced a hold on launching the recovery array until midday. By 1830 hours
the sweepline array was in the water, but the pinger had quit working, the
line tension monitoring system was reading zero (line tension was actually
7500 1b) and the sweepline transponders were pinging at random! By 0930
the next morning (12 December) we had desensitized the sentinel transponder,
deactivated the anchor transponder, repaired the bad pinger, and fixed and
recalibrated the line tension monitoring system. At 1000 hours, while
lowering the array to the bottom, the line tension system went out again. It
was decided to proceed without a tension measurement and rely on the catenary
analysis to predict line tension.

13

Having lowered the array to 100 meters off the bottom, the run
toward the targeted drop point for the anchor began. As previously stated,
the run was made directly into the wind and seas at speeds below one knot.
Figure 17 illustrates the targeted drop points, one for passing MAVA
on the right, the other for passing MAVA on the left.

Due to deep ocean currents, a deflection of MAVA from the vertical
was anticipated. Potentially this could cause premature interference between
MAVA and the recovery array if the horizontal sweepline were laid too close
to the MAVA anchor. To avoid this situation, the maximum deflection of MAVA
was calculated using a subsurface moored array design program. Figure 20
shows the results of this computer analysis. The curves represent the
deflected configuration of MAVA in unidirectional, constant velocity current
profiles. Ocean current data for the recovery area indicated a .2 knot current
at the buoy decaying to .1 knot on the bottom. Assuming a .3 knot worst case
Situation, it was determined that the sweepline should be laid no closer than
215 meters from the MAVA anchor, figure 17. The shaded area of figure 17
was chosen as the acceptable area within which we could anchor considering
the length of the sweepline, placement of the grapnels and the desire to per-
form the sweeping operation while heading into the wind and seas.

A plotting sheet was usdd which had a one inch grid divided into
tenths of an inch. Positions were plotted at a scale of 1 inch = 400 meters.
A plotting resolution of 10 meters was easily obtained. Figure 21 isa
reproduction of the plotting sheet.

At 1600 hours on 12 December, the array was lowered to the bottom
and 100 meters of slack cable was payed out. The drop point chosen was the
One on the left in Figure 17 The positioning data indicated that the anchor
was dropped within 20 meters of this point. The ship was maneuvered to guide
the sentinel to the left of MAVA by at least 215 meters. Ship and sentinel
positions were plotted every three to five minutes. For many of the fixes,
the three position radii from the net transponders crossed at a single point
providing welcome reassurance during an anxious period.

Maneuvering of the ship was accomplished by making appropriate changes
in engine RPM, rudder angle, bow thruster power and direction while observing
compass heading, wind speed and direction, and "course-made-good." The major
advantage of deploying and positioning the sweepline while heading directly
into the wind and seas was being able to precisely control the ship's posi-
tion and speed. The 20-25 knot winds were used to advantage in two ways.
First, movement of the ship from the anchor point was limited by throttling
back and using the wind and cable tension to slow or back the ship as
required. Secondly, during the sweeping maneuver, the wind was used to
"sail" the ship around MAVA by maintaining a heading slightly off the wind
thus causing the ship to crab in the desired lateral direction.

As the ship progressed, cable was payed out at approximately 10-15
m/min. The pinger indicated that the sentinel was moving to its desired
altitude. At approximately 1700 hour, one hour after anchoring, the sweep-
line had been extended. The sentinel was 460 meters from the anchor point

14

and 220 meters above the bottom. The minimum distance from the sweepline to
MAVA was 280 meters. The ship was then maneuvered to begin the sweeping
Operation. Contact of the sweepline with MAVA was made at 1850 hours. At
that time the sentinel was 580 meters from the sweepline anchor and 105 meters
above the bottom. Total line out was 5190 meters. Figure 22 illustates the
above sweepline configurations.

Inspection of the catenary plots indicated that the intersection of
the sweepline with MAVA occurred very close to the midpoint of the 100 meter
cable between the MAVA anchor and dual release package. Furthermore, the
catenary analysis indicated a theoretical static line tension of 11,500 pounds.
The distance between the ship and anchor point was limited to 2000 meters
during the sweeping operation to keep line tensions at or below this value.
It was feared that with the addition of dynamic loading caused by fifteen
foot seas, the 23,000 pound elastic limit of the wire rope could be exceeded.
The lack of tension data was aggravated by the fact that the cable slack
tensioner had developed a leak and was inoperative.

It was hoped that MAVA would be cut free by the action of the 9/16
inch diameter sweepline chafing against the 1/4 inch diameter MAVA mooring
line. At 2155 hours, after three hours of sweepline contact with MAVA and a
180 degree sweep, there was no indication that this had occurred. The decision
was made to fire the lower anchor release which held the sweepline to the
"Bruce" anchor. The sentinel was observed to move rapidly following the
release from the anchor. At 2258 hours, one hour and three minutes later,
MAVA surfaced. The surfacing was announced by a shipboard radio receiver
which picked up the signal from the radio transponder on the MAVA buoy.

Subsequent inspection of the MAVA dual release package showed that the
4, inch diameter cable had been severed by one of the grapnels directly below
the Nicopress sleeves at the attachment point. Apparently, the grapnel had
moved up the wire rope and lodged at that point.

VIII. Acoustic Release Failure Analysis

A failure analysis was conducted on the recovered MAVA anchor releases
with the following conclusions. One release failed to operate in either the
transpond mode or the command release mode because the connector between the
transducer and the electronics had come loose. All connectors of this type
have been modified to provide a positive locking feature. The second release,
while functioning in the transpond mode would not release upon command. The
failure was attributed to a faulty squib. Gun 011, used to clean the firing
chamber, had worked its way into the propellant thus causing the failure.

IX. Conclusions
The moored sweepline technique described above worked exceptionally
well considering the depth of water, sea state and wind speed. Control of the
array during anchoring and sweeping phases was not overly difficult.
Throughout the operation, theoretical indications and actual experience

agreed very closely.

15

It is speculated that the primary limitation factor in utilizing this
technique at depths greater than 4500 meters would be the increase in line
tensions.

X. Recommendations

As in any operation of this type, problems were encountered and lessons
were learned. Although the recovery technique described in this report worked
exceptionally well, the following recommendations for improving or simplifying
the operation are provided.

It is recommended that release of the sweepline from the "Bruce" anchor
take place as soon as it is clear that the sweepline is in contact with the
mooring line.

An improvement in the cutting action of the sweepline could be
achieved by using a mine sweeping cable equipped with cable cutters or line
chippers. Grapnels should be attached to the lower end as a back up.

Severe weather conditions at the time would have made retrieval of the
sentinel weight very dangerous. Consequently, after removal of the sentinel
pinger and transponder, the sweepline array was cut away. The result was
the loss of the sentinel, grapnels, depressor weight, 600 meters of cable
and the bottom anchor release. To reduce the possibility of such a loss, it
is recommended that the sentinel transponder (anchor release) be installed
between the sentinel and the array so that the sentinel can be jettisoned
if necessary.

Electronic equipment problems were limited to the bottom finding
pinger. Failures were traced to cold solder joints and faulty batteries
resulting in power loss at low temperature. Modifying the pinger to have an
on-off duty cycle proved to be unnecessary as interference with the acoustic
transponders was not observed.

A major improvement in the conduct of the operation would be the use
of a computer/plotter to ease the burden of manual data reduction and plotting.

16

1900m

MAVA BUOY 4500m

1500m (HYDROPHONE ARRAY)

1000m (KEVLAR)

DUAL RELEASES ee

100m (1/4 DIA.; 3x19 WIRE)

FIGURE 1 MAVA

17

TO SHIP

MAVA

ANCHOR RELEASE——__

PINGER——__,
SENTINEL WEIGHT—__,_

TRANSPONDER v
GRAPNELS (7) A
ve

TRANSPONDER

DEPRESSOR WEIGHT
ANCHOR RELEASE
BRUCE ANCHOR

FIGURE 2 MOORED SWEEPLINE TECHNIQUE

18

DISTANCE ASTERN (METERS)
3500 3000 2500 2000 1500 1000 500 0

500

Cable Scope = 4500Meters 1000

Nay 1500

Vv 2000

2500

3000

.5 Knots

3500

4000

4500

FIGURE 3 CONFIGURATION OF TOWED RECOVERY ARRAY

19

DEPTH (METERS)

Acoustic Release

Pinger

@

<———— 4500 Ib Sentinel Weight

610m

3, Grapnels (7)

152m

2000 |b Depressor Weight

Acoustic Release

hos
15m 3

en ME eas Anchor

FIGURE 4 GENERAL CONFIGURATION OF SWEEPLINE ARRAY

20

———-_ 9/16 inch Wire Rope

I 9/16 inch Wire Rope Clips (5)

ieee 9/16 inch Wire Rope Thimble

( 3/4 inch Shackle

g ~+— 3 Ton Swivel
| — 3/4 inch Shackle
()
ra

a “1 inch Shackle

E>

\f

2000 Ib (Weight in Water) Depressor Weight

<—_ 4.5 Meters of 3/4 inch Proof Coil Chain

5/8 inch Shackle

KK COCO
a
S

EG&G Model 322 Acoustic Release

5/8 inch Pear Link

— ae 5/8 inch Shackle

15 Meters of 3/4 inch Proof Coil Chain

5/8 inch Shackle
3/4 inch Shackle

Ep esstenrssi esssc eG:

——_———. 110 |b Bruce Anchor

FIGURE 5 ANCHOR/DEPRESSOR WEIGHT RIGGING DETAIL

(ail

\
\
\
'
\
'
\

al perenne tants 9/16 inch Wire Rope
iM

|
fl
' PLP Dead End Grip

3/4 inch Shackle

1 Meter 3/4 inch Proof Coil Chain
9/16 inch Wire Rope (Slack)

3/4 inch Shackle

4500 Ib. (Weight in Water) Sentinel Weight

2.5 Meters 3/4 inch Proof Coil Chain

PLP Dead End Grip

FIGURE 6 SENTINEL WEIGHT RIGGING DETAIL

<~— EG&G Model 322 Acoustic Release

5/8 inch Pear Link

1/2 inch Shackle

Pigtail Assy

9/16 inch Wire Rope Clips

9/16 inch Wire Rope

FIGURE 7 PINGER/RELEASE RIGGING DETAIL

23

9/16 inch Wire Rope

9/16 inch Wire Rope Clips

9/16 inch Wire Rope Thimble

1/2 inch Shackle

19"

Grapnel
15lbs

Lt
iat
1/2 inch Shackle

<—— 9/16 inch Wire Rope

FIGURE 8 GRAPNEL RIGGING DETAIL

24

A = Anchor
B = Sentinel
C = Ship

FIGURE 9 RECOVERY ARRAY GEOMETRY

bottom

FIGURE 10 SWEEPLINE GEOMETRY

25

12

FIGURE 11 FREE BODY DIAGRAM AT SENTINEL

FIGURE 12 ARRAY GEOMETRY AND IMAGINARY SEGMENT

26

FIGURE 13 GEOMETRY OF IMAGINARY SEGMENT

bottom

FIGURE 14 EXTENDED ARRAY GEOMETRY

27

T=10340 T=10430 T=10870 T=11110 T=11812

1000
2000
e
ie
a
£
=r
kK
a
uu
a
3000
T=LINE TENSION (Ibs)
d=SENTINEL HEIGHT (m)
S=LINE OUT, SHIP TO ANCHOR (m)
4000 X_-HORIZONTIAL SEPERATION, SHIP TO ANCHOR (m)
Ts = Anchor Release Location
Sentinel Weight Arc
PH =<—

1000 2000 3000
DISTANCE (meters)

FIGURE 15 CATENARY DETAIL — CASE 1

28

T=15474

T=12199 T=12928 T=13945
d=592 d=401 d=240 d=160 d=112
X=280 X.=1122 X7=1953 X1=2776 X1=3704
1000
T=17780
d=80
2000 X=4800
co a
nn
s 8 &
o H +
= on i
— n
Yr
Ge S
- 4
i ops
a bt
(7)
%
a
3000 +- of
oe
f
T=LINE TENSION (Ibs.)
d=SENTINEL HEIGHT (m)
rong ee S=LINE OUT, SHIP TO ANCHOR (m)
. X,-HORIZONTIAL SEPERATION, SHIP TO ANCHOR (m)
oN + =ANCHOR RELEASE LOCATION
\
SENTINEL WEIGHT ARC
== Loma
a
VPy 1000 2000 3000
DISTANCE (meters)

FIGURE 16 CATENARY DETAIL — CASE 2

29

Wind

Transponder

1.5 mi.
(8) :

Targeted Drop Points

Transponder

FIGURE 17 TRANSPONDER NET

30

on FIX

A

DR POSITION

IS

<—__k aoe

BEARING: ATO MAVA

TRANSPONDER A

®

L SATNAV FIX

FIGURE 18 TRANSPONDER NET GEOGRAPHICAL ORIENTATION

31

AMF 301

Tx 11khz
Tx 9khz

301 12khz WIDE BEAM
TRANSDUCER TRANSDUCER
/
/
ae ne
1
TOUMAVAA Becks

NAVIGATION

SENTINEL TRANSPONDER’ TRANSPONDERS

Ty=TRANSMIT FREQUENCY

: Ry=RECEIVE FREQUENCY

ANCHOR TRANSPONDER MAVA TRANSPONDER

FIGURE 19 ELECTRONIC EQUIPMENT

32

DEPTH — (METERS)

2000

2500

3000

3500

4000

4500

> og ~MAVA
x
S
Cc
¥ |e
a [2
om fe
2 be
ni 2
Av
re)
500 1000

DEFLECTION — (METERS)

FIGURE 20 MAVA DEFLECTION VS. CURRENT
33

Q
509
©
me Oe
z
: Ono”
Os?
10
°? > ie A ral
OE | STEN
al
40
@iy 6
>
et!
a?
0
+)
a>
\
ao
40
Cry
ahd
v
a
\o a

FIGURE 21 PLOTTING SHEET
34

1801 HRS 1833 HRS
if WIND

1850 HRS
(@)

1700 HRS
iS)
b
© 1638 HRS
TRANSPONDER B AY : =e
Natt

SWEEPLINE ANCHOR
1600 HRS

2100 HRS

21 57 HRS ee
TRANSPONDER A ‘1320 HRS

= — — SWEEPLINE
O SHIP POSITIONS
@ SENTINEL POSITIONS

SENTINEL TRACK

FIGURE 22 POSITION DETAILS

a ‘
Nua

Pie vil