EPA-670/2-73-069
September 1973
Environmental Protection Technology Series
Fabric Boom Concept For Containment
and Collection of Floating Oil
5
S2SZ
r4
UJ
(3
Office of Research and Development
U.S. Environmental Protection Agency
Washington, D.C. 20460
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and
Monitoring, Environmental Protection Agency, have
been grouped into five series. These five broad
categories were established to facilitate furtber
development and application of environmental
technology. Elimination of traditional grouping
was consciously planned to foster technology
transfer and a maximum interface in related
fields. The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
t4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL
PROTECTION TECHNOLOGY series. This series
describes research performed to develop and
demonstrate instrumentation, equipment and
methodology to repair or prevent environmental.
degradation from point and non—point sources of
pollution. This work provides the new or improved
technology required for the control and treatment
of pollution sources to meet environmental quality
standards.
-------
EPA-670/2-73-069
September 1973
FABRIC BOOM CONCEPT FOR
CONTAINMENT AND COLLECTION OF
FLOATING OIL
By
Philip E. Bonz
Contract No. 68-01-0139
Project 15080 FWM
Program Element B12041
Project Officer
John S. Far low
Edison Water Quality Research Laboratory
National Environmental Research Center
.Edison, New Jersey 08817
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
For sale by the Superintendent of Documents, U.S. Government Printing Office, Washington, D.C. 20402 - Price $1
-------
EPA Review Notice
This report has been reviewed by the Office of Re-
search and Development, EPA, and approved for
publication. Approval does not signify that the
contents necessarily reflect the views and poli-
cies of the Environmental Protection Agency, nor
does mention of trade names or commercial pro-
ducts constitute endorsement or recommendation
for use.
1-i
-------
ABSTRACT
The feasibility of applying the concept of oil-water separation by means
of woven hydrophilic fabric to a floating oil containment boom was in-
vestigated through a series of model tests A preliminary model boom
configuration was developed and towed at speeds to 0.686 meters/sec
(2.25 ft/sec) in both calm water and waves. Oil retention performance
of this model was clearly superior to that of a conventional flat plate
boom of comparable draft in the environment investigated. A larger
model of similar configuration demonstrated no oil leakage when towed
at 0 .77 meters/sec (1 .5 kt) in calm water.
While further detailed analysis, engineering, and testing is required to
fully examine this concept, it appears that a properly designed flexible
boom which uses a hydrophilic skirt material offers significant potential
both as a containment device for floating oil in high current situations
and as a high-speed collecting device.
This report was submitted in fulfillment of Project Number 15080 FWM,
Contract Number 68-01-0139 by CONSULTEC, Inc., Rockville,
Maryland, under sponsorship of the U.S. Environmental Protection
Agency. Work was completed as of August 1972.
ill
-------
C ONTENTS
Page
Abstract iii
List of Figures VI
List of Tables VII
Sections
I Conclusions 1
II Recommendations 3
III Introduction 5
W General Concept 7
V Mathematical Analysis 9
‘JI Fabric Selection 13
VII Laboratory Testing 17
VIII Demonstration Tests 21
IX References 39
X Appendices 41
V
-------
FIGURE S
No. Page
1 CONFIGURATION DETAILS OF TEST MODELS, SECOND 22
TEST SERIES
2 OIL SLICK PROFILE, SECOND TEST SERIES
(SAE 30 WEIGHT OIL) 25
3 OIL SLICK I SECOND TEST SERIES
(SAE 30 WEIGHT OIL) 26
4 OIL SLICK PROFILE, SECOND TEST SERIES
(NO. 2 DIESEL FUEL) 28
5 CONFIGURATION DETAILS OP TEST MODELS, THIRD
TEST SERIES 30
vi
-------
TABLES
No. Page
1 Summary of Fabric Characteristics 15
2 Summary of Third Test Series 32
vi i
-------
SECTION I
CONCLUSIONS
The following conclusions are based on preliminary efforts aimed at
establishing the feasibility of utilizing the oil/water separation qualities
of selected woven fabrics in a floating oil containment device and
specifically determining whether such a concept could offer better per-
formance than conventional flat plate booms. It should be recognized
that this effort was primarily exploratory in nature and that a final or
prototype design was neither developed nor tested.
The results of this effort, although limited, were encouraging. The con-
cept of employing a woven fabric in a flow-through containment boom
was demonstrated to be feasible. Towing tests of a preliminary model
showed that the oil retention capability of such a device in calm water
was superior to that of a conventional flat plate boom of comparable
draft at the speeds and with the oil tested. Towing tests conducted in
the presence of waves indicated that the preliminary model conformed
well with the wave profile and that no oil losses were caused by wave
action.
Although it is not possible as a result of these efforts to accurately
predict full-scale performance, they do indicate the potential of such a
device for containment of floating oil in currents where the conventional
flat plate boom is of limited effectiveness and the adaptability of such
a device to high-speed oil collecting.
While the preliminary design showed promise, it was far from optimum.
Design refinements both in fabric selection and cross-section configu-
ration are expected to result in improved performance.
In terms of oil containment capability, the boom concept tested during
this effort possesses the following advantages over a conventional flat
plate boom under conditions of current and waves:
• Entrainment losses are reduced (or eliminated) since
water flow under the oil slick must pass through the
separation fabric which rejects penetration by entrained
oil droplets.
• Losses by drainage are eliminated (or reduced) through
proper selection of the throat opening such that the bag
draft is at least as great as the oil slick depth at the boom.
1
-------
• The inherent buoyancy of the oil-filled bag contributes
to the wave conformance capability of the total boom,
reduces the requirements for floatation, and minimizes
oil loss due to wave action.
Because oil encountered by the boom is swept by the current into the
bag where it is contained, this fabric boom concept appears to offer
potential as an oil-collecting device. While details of this application
were not addressed, only a minor modification to the configuration
tested would be required for the bag to act as the sump for suction hoses
of an oil removal system. In this situation, throat opening can be sig—
nificantly less than that required for a retention boom since a large
buildup in slick depth is precluded by continual (or periodic) oil removal.
2
-------
SECTION II
RECOMMENDATIONS
Additional detailed fabric analysis and testing should be conducted.
Critical to any further consideration of a hydrophilic fabric for the
separation of oil and water is a complete understanding of the flow and
rejection process through these fabrics. Detailed analysis - theoretical
supplemented by laboratory testing - should be undertaken to identify
and quantify those factors which influence fabric behavior as an 011/
water separator. Factors which should be-addressed include fabric
characteristics, fluid properties, fluid interfacial properties, and the
degradation of fabric performance due to presence of probable solutes
and suspended solids. The results of this effort would provide a sound
basis for fabric selection and for predicting the optimum and limiting
operating conditions for oil separation for various type oil in various
environments.
Detailed engineering analysis to support optimization of the boom con-
figuration should be conducted concurrently with fabric analysis and
testing. The initial purpose of this effort should be to develop the capa-
bility to predict dynamic pressures, internal and differential across the
fabric, for various values of current, slick profile, and internal oil
volumes as a function of bag configuration. Upon the completion of the
fabric analysis effort wherein the optimum operating conditions for oil
separation are defined, an optimum bag configuration would be developed
and its performance capability analytically assessed.
If it appears at this time that indeed the fabric boom concept has suffi-
cient performance potential to warrant further investigation, then the
following major efforts should be undertaken sequentially:
• Full—scale two-dimensional testing in calm water of
the predicted optimum configuration for a variety of oils.
• Engineering design of a complete boom assembly including
optimization of the floatation and mooring/towing systems.
• Full—scale three-dimensional testing in both calm water
and waves for a variety of oils.
• Design, development, and testing of an oil removal system.
3
-------
SECTION III
INTRODUCTION
Under Federal Water Quality Administration Contract Number 14-12-878,
CONSULTEC, Inc. demonstrated the basic feasibility of separating oil
from water using selected woven fabrics, reference (1). Among the
potential applications of this phenomenon is the concept of utilizing oil-
separating fabrics in oil containment boom systems. Under Environmental
Protection Agency Contract 68-01-0139, CONSULTEC developed and
tested a preliminary design of a fabric oil containment boom system
which makes use of this concept. Due to its improved hydrodynamic
behavior and the ability of the fabric to pass water while retaining
collected oil, the fabric boom system has the potential to perform better
than currently available flat plate boom systems. It has successfully
done so at speeds up to 0.77 meters/sec (1.5 kt).
Efforts under the current contract were conducted as follows:
(1) Initial testing in both calm water and waves of very
preliminary boom concepts which employed fabrics
identified in reference (1) as having the best oil-
water separation characteristics, arid the analysis
of those test results.
(2) Identification of boom configuration criteria, limited
additional fabric analysis, and development of a
refined preliminary boom design.
(3) Preliminary testing.
(4) Demonstration testing.
Initial tests, and analysis of the results of those tests, are discussed
in Appendix A. All other efforts are described in Sections V through
VIII. A discussion of the characteristics of various oils is provided in
Appendix B.
5
-------
SECTION IV
GENERAL CONCEF
OF FABRIC OIL CONTAINMENT BOOM
A literature search revealed that existing (1971) oil containment booms
all behave essentially as flat plate systems (references 2, 3, 4, 5, 6,
7, 8, 9, and 10). Some configurations are rigid plates, and some con-
sist of flexible flat skirts. Some rigid and semi-rigid systems, although
not being actually configured as flat plates, behave much as a flat plate
boom would behave in currents. Flat plate booms have been shown to be
effective only in low currents, i.e., currents of 0 .515 meters/sec (1 kt
or less). At moderate and high current velocities stripping occurs,
wherein collected oil is entrained in the water and carried under the boom
by the water flow. In currents, non-porous flat plates cause flow streams
to accelerate as water passes under the boom. The higher velocity flow
causes lower pressures at the submerged base of the boom and collected
oil is sucked into this low pressure region and allowed to escape under
the boom. In addition to this “drainage” there is a second mode of
failure wherein, above a certain critical current speed (for given oil and
water properties), droplets of oil are torn off the headwave and “en-
trained” in the water. If gravity and buoyant effects are not sufficient to
cause the oil droplets to rise and regain the slick before reaching the
boom, the flow will carry the oil droplets under the boom.
The fabric boom concept improves upon the performance of flat plate
systems through the use of a porous fabric skirt which, by permitting
water flow through skirt, reduces the tendency for oil to be sucked under
the boom. This also redirects entrained droplets into, and not under,
the boom. The flexible fabric skirt is in the shape of a bag and is held
open at the mouth. The fabric is a “hydrophilic” fabric, which exhibits
the property to “wet” with water and not with oil. With an appropriate
configuration of fibers, the surface tension properties of the oil are
sufficient to preclude its penetration of the water-wetted fabric. Thus,
the fabric boom system redirects flow lines and carries entrained oil into
the fabric skirt, where it will be separated from the water and allowed to
return to the surface slick due to gravity and buoyant effects.
Except for those droplets of oil entrained in the entering water flow
which have not yet risen to join the surface oil slick within the bag, and
those droplets which may be temporarily torn off from that slick by local
internal turbulence, the oil contained within the bag takes the form of a
surface slick, the buoyancy of which causes the bag to float with its top
7
-------
or upper portion above the water surface. Because water-wetting of a
hydrophilic fabric is an essential factor in its rejection of passage by
oil, this upper portion of the flexible skirt is constructed of a non-porous
fabric. The collected oil, floating under this non—porous fabric, provides
a significant degree of floatation for the boom system.
Water flowing into the bag through the mouth flows out through the
bottom of the bag 0 In currents, the bag assumes a bulged shape toward
the back. The increase in bag depth causes accelerated flow under the
bag and results in a low pressure region on the underside of the bag.
This increases the water flow through the fabric skirt and thus enhances
flow into the bag.
Being flexible and lightweight, the fabric boom conforms readily to wave
shapes, thereby minimizing oil losses over the top of the boom, as well
as under the boom. In more advanced applications, oil collected in the
bag-shaped fabric skirt could be pumped out, thus providing a high rate
oil collection and separation device.
8
-------
SECTION V
MATHEMATICAL ANALYSIS
Bag design equations were developed on theoretical grounds as an assist
in the overall boom concept development. From the standpoint of cloth
selection, three conditions apply:
(1) The effective hole diameter between threads must be
sufficiently small that entrained bubbles of some
designated size will not pass through the cloth.
(2) The stagnation conditions experienced by oil bubbles
of a larger size will not produce extrusion of oil
through the cloth opening.
(3) The head loss for water flow through the cloth must
be equal to that occurring in adjacent streamlines
so that there is no diversion of flow around the cloth.
For purposes of illustration, consider a cloth with thread diameters and
spacing when wet as shown in the sketch below:
I HEdt
11
The free face area for a single stream of flow is
A 1 =w.l
The contracted area through which the flow stream must pass is
A 2 = (w - d ) (i - d )
For convenience sake, the contracted area may be redefined as
A =d .d
2 w 1
9
-------
where d and d 1 are the smaller and larger (respectively) dimensions of
the individual fabric hole.
In consideration of the first condition listed above, the maximum di-
ameter, d 0 , of an oil droplet that will pass through in the flow stream,
disregarding hydrophilic action, will be
d
-------
If the entering flow is suddenly blocked by a glob of oil, the contracted
flow pressure remains the same (at least instantaneously), but the enter-
ing water flow stagnates 1 resulting in a pressure increase equal to the
kinetic energy of the flow. The resulting pressure difference acting to
force oil through the contraction increases to
____ /A 1 \2
2 A 2 )
The cloth selection criterion, then, is that this pressure difference
remain less than that requirc i to create an oil bubble, or expressed
mathematically:
PV
P= 21 -
This inequality may also be expressed in terms of V 2 , the water flow
velocity in the contracted region, which is equal to V 1 (A 1 /A 2 ):
d
and rewritten in terms of the Weber, i .e.,:
pV
W=
In consideration of the third condition, the head loss through the cloth
can be expressed as
)2 2
or, in terms of pressure drop, as
v 1 2 /A 1 \2
2
These expressions are quite simplistic and assume that A 9 = d 2 and there
is full flow through A 2 . In point of fact, the effective varue of A,, in the
head loss equation is a function of the cloth’s geometry and the 1 eynolds
11
-------
number of the flow. TO further complicate the is sue, the effective values
of d and A 2 in the extrusion equation depend as well upon the hydrophilic
characteristics of the cloth used. The actual behavior of a specific cloth
must be determined by experiment, but the equations given above were
used for initial design; and the second and third design conditions can
be achieved by meeting the following basic design equation:
Pv 1 2 ( A 1 \ 2
2 \A 2 / d
Given a design headwave thickness and a bag draft, b, sufficiently large
to collect all oil bubbles generated by the headwave in a current of
velocity V. the bag design must meet the constraint given above. As-
suming the head loss ratio is sufficiently small, the flow entering the
bag is q = ‘Tb (cubic feet per second per unit width). From continuity
considerations, this flow is maintained at the cloth face. Therefore
b
=Vb=V 1 Lor; - j--
where L is the bag length and V 1 satisfies the inequality given above.
The adequacy of this analysis for initial design efforts was reasonably
confirmed by the tests reviewed in subsequent Sections.
12
-------
SECTION \TI
FABRIC SELECTION
Under EPA direction, minimum effort was to be expended in fabric analy-
sis and testing during the conduct of this contract. Although it was
recognized that the performance of the flow-through boom is directly
dependent on the characteristics of the fabric employed, it was felt that
if the concept could be demonstrated to be attractive without optimiza-
tion of the fabric, then later efforts could be undertaken in this area.
Following unsuccessful initial tests (see Appendix A), a modest effort
was undertaken to identify additional basic fabric selection criteria and
to select a fabric(s) for further demonstration testing. The fabric selec-
tion process was restricted by EPA to the quick appraisal of a limited
number of readily available fabrics and the selection of the most prom-
ising from that group.
Based on prior laboratory testing, and the limited analysis presented in
the preceding section, the following criteria were developed for the
initial selection of fabrics for evaluation:
• The thread diameter should be as small as possible.
• The fabric should be a plain weave.
• A cloth count (numerical ratio of warps to wefts per square
inch) should be close to 1 .0 (square count).
Following discussions with various fabric manufacturers, it was decided
to concentrate on various cotton fabrics, and five sample fabrics were
obtained for evaluation.
Weave type, cloth count, and fabric dimensions were obtained by view-
ing water-soaked specimens of each fabric with a compound microscope.
A simple test was then conducted on each fabric to determine its capa-
biity to resist oil extrusion. The setup for this test is shown in the
sketch on the following page.
13
-------
LARGE DIAMETER TUBE
A tube of sufficient diameter to minimize capillary effects was attached
to a funnel on which a cloth specimen was mounted. The funnel was
placed in water so that the cloth face was a defined distance, L, below
the water surface. The cloth specimen was allowed to become water-
soaked and water was allowed to seek its own level in the funnel. SAE
30 weight additive-free test oil was then slowly added in the tube from
the top, displacing water until the funnel was filled with oil. Additional
oil was carefully added until minute oil bubbles was extruded from the
cloth. At that point the total oil height, Z, was measured. Under the
test setup the bubbles formed were stable and of minimum size. The
differential pressure at which oil will extrude was calculated as
= (p oil) (Z) - (p water) (L)
A summary of the properties of the five cotton fabrics investigated is
provided in Table 1. Geometric characteristics of the fabrics used in
the initial (unsuccessful) tests discussed in Appendix A are also shown
for comparison.
From a review of that data, fabric No. 5 was selected as having the best
overall performance potential in terms of concurrentjy providing the
greatest resistance to oil extension (maximum AP) and the smallest head
loss due to water flow through the fabric (maximum A 2 ).
WATER
14
-------
TABLE 1 — SUMMARY OF FABRIC CHARACTERISTICS
Fabric used for Initial model testing of preliminary boom configurations (fixed fabric models) described in Appendix A.
Fabric used for initial model testing of preliminary boom configurations (free—to—float fabric model) described In Appendix A.
cri
Calculated
Calculated
AP at
Sample
No.
Fiber
Manufacturer &
Pattern
Nominal
Thread
Count
Measured
Thread
Count
Measured
Number of
Holes/In 2
Measured
Thread
Diameter
(10 3 in)
Measured
Minor Hole
Dimension
(l0 3 in)
Measured
Major Hole
Dimension
(lO 3 in)
Area of
Individual
Hole
(104in2)
Percent of
Fabric Area
Open to
FloW
Which
Extrusion
Occurs
(lO 2 psi)
1
100% Cotton
Burlington Industries
60 x 64
56 x 63
3528
7.15
8.60
l0 80
0 93
32 8
3.81
(Unbleached)
Leslie, Catlin
*13766
2
100% Cotton
Lowenstein
64 x 56
70 x 56
3290
8.60
571
9 30
0.53
20 8
5 89
(Bleached)
*88815—59
3
100% Cotton
Lowenstein
80 x 80
93 ‘t 93
8649
7.86
2.86
2.86
0.08
7.1
11.82
(Bleached)
*N—3982
4
100% Cotton
Logantex
72 x 80
64 X 70
4480
4.30
10.00
11.40
1.14
51.0
294
(Organdy with
*12 35
Heberlin
Finish)
5
65% Polyester
West Point Pepperell
56 x 54
56 ,c 61
3294
5.72
10,70
12.85
1.38
45.0
3 34
30% Cotton
*24140
A/ L
Linen (plain)
(unknown)
———
47 x 34
1598
10—25
0.53
8.5
-——
8 LL
Linen (Irish)
Couturier Fabric
———
20 x 28
560
30—45
———
———
1.43
8,1
———
WPL 9893
—________
-------
SECTION VII
LABORATORY TESTING
In addition to the testing involved with fabric selection, a series of
laboratory tests was conducted to determine a reasonable cross-sectional
configuration for the flow—through boom. Based on the analysis of initial
configuration testing (see Appendix A), it was decided to employ an im-
pervious (to both oil and water) fabric as the material for the upper portion
of the bag, and to control the throat opening of the boom by rigid separa-
tion of the lower leading fabric edge from the upper leading edge (boom
floatation). The objectives, then, of this test series were to identify
water flow lines into and through the bag and to determine the influence
of various parameters such as bag length, throat opening, location of
rear seam, and shape of the lower spacer bar on those flow lines.
To support this testing, CONSULTEC funded the construction of a simple
flow tank shown in the sketch below. An insert in the tank provides a
test section 30 .48 cm (12 in) wide, 137.16 cm (54 in) long, and up to
25.40 cm (10 in) deep. Cross-sectional viewing in this test section is
provided by vertical plexiglass windows. A variable speed electric out-
board motor was used to create controlled flow in the tank and, with this
configuration, current speeds of up to 0.23 meters/sec (0 .75 ft/sec)
were obtained. Flow speeds were monitored with an accurate flow
veloci meter.
1.22 METERS
(4 FEET)
In the first flow test, a piece of test fabric was positioned in the flow
channel as shown in the sketch on the following page. One end of the
fabric was held under water by means of a dowel inserted through a sewn
3.66 METERS (12 FEET)
-j
TEST SECTION
17
-------
seam, while the other end was positioned above the water by hand. Dye
was injected into the stream to study the behavior of the flow through the
fabric 0 The length of the bag was varied by moving the loose end. A bag
with an impervious back was simulated through the use of a flat plate or
sluice which could be positioned to any combination of immersion depth
and distance from the fixed leading edge of the fabric. With this tech-
nique, various bag configurations could be investigated; for example, a
bag of cross—sectional symmetry with an impervious upper section and a
fabric lower section could be represented by immersing the sluice to a
depth half as deep as the leading edge fixed dowel.
In the absence of the sluice, flow through the bag was seen to be very
good and essentially straight. When the sluice was inserted, dye traces
revealed that flow streams were directed downward and through the
bottom of the bag. The sluice was held at various depths to observe the
effects of the depth of the impervious portion of the back of the bag.
This test provided basic insight into the influences of bag length and an
impervious upper and rear section of the bag.
Concurrently, a limited examination was conducted into the effect on
flow patterns of the shape of the lower support rod. Support rods of
various cross-sectional shapes were inserted into the leading edge
fabric seam and their effect on flow lines was observed through the use
of dye trace. Shapes tested included triangular, oval, streamline foil
and circular, all of various sizes. Because no advantage was observed
for shapes other than circular, and since it was felt that lower support
in the prototype design could most easily be provided by a cable, a
circular cross—section was utilized in all further tests.
Based on the results of these tests, a tentative bag configuration was
developed and tested. In this configuration, bag length was 76.20 cm
SLUICE
DYE
LOOSE END
o
FABRIC
18
-------
(30 in), throat opening was maintained at 10.16 cm (4 in) by two 2.54 cm
(1 in) diameter dowels, and the top of the bag was fabricated of an im—
pervious polyethylene fabric which extended halfway down the back of the
bag, while the remainder of the bag was fabricated of the selected cloth
fabric. Dye was injected at various depths upstream of the bag. The
test arrangement is shown in the sketch below.
30”
I — — I
The dye traces showed that all flow streams to a depth of 10 16 cm (4 in)
were directed into the bag; and that flow inside the bag was uniform and
went well into the back of the bag before exiting through the cloth fabric.
However it was noted that at the upper velocity range, the maximum draft
of the bag exceeded one-half test channel depth. Since, as discussed in
Appendix A, this is a limiting test condition, it was felt that additional
testing should be conducted with shorter bags and smaller throat openings.
Additionally, it was concluded that the effect of the fabric seams on inter-
nal flow should be determined more precisely.
A third series of tests was conducted using a bag length of 25.40 cm
(10 in) and a throat opening of 5.08 cm (2 in). Throat opening was main-
tained by using two 1 .27 cm ( - in) dowels. For this series, four models
were constructed--identical except for the fabric seam, which was
located at 1/4, 1/2, 3/4, and full bag draft, respectively--as shown in
the cross-sectional sketches on the following page.
These tests conducted in a 0.23 meter/sec (0.75 ft/sec) current, indi-
cated that the best flow lines as shown by dye injection occurred when
the fabric seam was located midway between the top and bottom (i.e.,
at bag’s longitudinal centerline). With that configuration, all water flow
lines above the lower support passed into the bag and were smoothly
directed out of the bottom of the bag through the entire length of the
woven fabric. (As noted previously, in tests of the longer bag, water/
— 4”
DYE —
— -‘-0 -
9%’,
19
-------
flow lines passed well into the bag before exiting through the woven
fabric.) Maximum bag draft was observed to be 10.16 cm (4.0 in).
‘A DRAFT SEAM V 2 DRAFT SEAM
% DRAFT SEAM FULL DRAFT SEAM
As a result of these laboratory tests, the influence of various design
parameters on water flow lines into and through the bag were identified
and basic criteria for subsequent bag configuration were established.
20
-------
SECTION VIII
DEMONSTRATION TESTS
The primary purpose of efforts conducted under this contract was to
demonstrate the feasibility of applying the concept of oil/water separation
by means of a hydrophilic fabric to a floating oil containment system. At
different stages of the contract effort, various model tests were conducted
to examine and/or demonstrate the performance of an evolving boom con-
cept in oil. These tests are summarized in this section.
With one exception, testing was limited to SAE 30 weight motor oil, since
that oil was used in prior efforts, reference (1). Additionally, no effort
was under taken to Weber-scale the oil by addition of surface active
agent, since the effect of such a procedure on the oil rejection character-
istics of the fabric was unknown.
First Test Series
Initial model tests of preliminary boom designs were conducted in the
Hydronautics 24.4 meter (80 ft) towing tank. Models for these tests
were fabricated using fabrics identified in previous efforts, reference (1),
as displaying desired oil/water separation characteristics. However,
these fabrics, while resisting penetration by oil, offered excessive
resistance to water flow at moderate current velocities. Because of this,
preliminary booms behaved essentially as flat plates. In addition, in-
stabilities due to tank configuration were observed. The results of this
test series are discussed in detail in Appendix A.
Second Test Series
Following the initial tests, analysis of those initial tests and further
laboratory testing led to the development of an improved boom configura-
tion. Oil containment tests of this configuration were conducted in the
Consultec flow channel to observe performance in oil; to demonstrate
superiority of that design over the configuration originally tested; and
to provide a basis for further design refinement and testing at higher speed.
Two models were constructed for this test series, the dimensions of
which are provided in Figure 1. Both models were identical except for
the woven fabric: in the first model, one of the two fabrics employed in
the initial tests described in Appendix A was used (Fabric B, Table 1);
the second model was constructed using a fabric (Fabric No. 4, Table 1)
21
-------
10”
- 1
II
L i
I
FIGURE 1 — CONFIGURATION DETAiLS OF TEST MODELS, SECOND TEST SERIES
W’ DIA. UPPER SUPPORT DOWEL
2
W DIA. LOWER SUPPORT DOWEL
I
/
22
-------
which by the selection process described in Section VI was determined to
have suitable characteristics. (It should be pointed out that the fabric
selection efforts had not been completed, and that based on those efforts
a different fabric was selected for subsequent demonstration testing.)
Although no attempt was made in the development of these demonstration
models to carefully scale the properties of the boom, a rough estimate of
the scale of the boom can be obtained by comparison with typical proto-
type boom systems. If the 5 . 08 cm (2 in) mouth opening of these demon-
stration models is compared to the draft of existing flat plate booms
(using an average draft of 76.20 cm (30 in)), the scale of the demonstra-
tion model would be approximately 1/15.
The models were held in a plexiglass fixture which provided a flow
channel 30.48 cm (12 in) wide and approximately 24.77 cm (9.75 in) deep
upstream of the model. Flow velocities were monitored upstream and
downstream of the model. Dye was injected into the flow upstream of the
model to study the behavior of the flow around and through the model
booms. For the oil containment tests, oil was poured into the water just
upstream of the boom. Entrained oil droplets were created by dropping the
oil from a height above the water, in front of the boom.
The following demonstration tests were conducted:
Test No. 1 Flow test with the first model. No oil was used in
this test, but dye was used to observe the flow
conditions.
Test No. 2 A similar test with the second model.
Test No. 3 An oil containment demonstration test using the second
model and SAE 30 weight oil (Sears additive-free).
Test No. 4 An oil containment demonstration test using the second
model and No. 2 Diesel fuel oil.
Test No. 1, conducted at a current velocity of 0.17 meters/sec (0.55
ft/sec), demonstrated that with a skirt made of Fabric B, very little flow
occurs through the fabric, and flow is primarily directed under the lower
dowel of the model. Some dye, however, did flow through the mouth of
the boom, and the behavior of the dye showed that the conditions in the
bag were almost stagnant. The conclusion is that an oil retention boom
which uses Fabric B for the skirt, results in essentially stagnation in
front of the boom and does not behave much differently than a flat plate
boom of the same draft.
23
-------
Test No. 2, conducted at the same current velocity, demonstrated that
with Fabric No. 4 for the skirt, flow was directed through the mouth of
the bag and through the fabric. Dye traces indicat d that flow went
almost all the way to the back of the bag before passing through the fabric.
Test No. 3 demonstrated the oil retention capability of the fabric boom.
SAE weight motor oil was added until a total of 5400 ml was held by the
boom. While it appeared that the model could contain a greater volume
of oil without failure, a limiting bag draft of half channel depth was
reached with 5400 ml of oil. Based on the analysis presented in Appendix
A, it was felt that test results with a greater amount of oil would provide
false performance indications because of test channel blockage. The
5400 ml test volume 3 corresponds to a specific volume of oil of 17.65
liters/meter (0.19 ft ,‘ft) of boom. If the scale factor of the demonstration
model is assumed to be- 1/15 and if. the specific volume of the oil is
assumed to scale up geometrically, then the full-scale specific volume
of oil would be 3 .96 meters 3 /meter (42 .6 ft 3 /ft). If the estimated current
speed of 0.17 meter/sec (0.55 ft/sec) is assumed to Froude scale up,
then the simulated prototype speed would be 0 .65 meter/sec (2 . 13 ft/sec)
or 1.26 knots.
Figure 2 shows the configuration of the boom and that of the oil slick at
0.17 meters/sec (0.55 ft/sec) and 5400 ml of oil. In this test, a head—
wave about 6.35 cm (2.50 in) deep existed inside of the bag, and the
slick extended all the way to the back of the bag. Figure 2 indicates
that the bag assumed a bulged shape under these flow conditions. This
shape is a desirable feature, since, by creating higher velocities (and
resultant low pressures) under the bag, flow through the fabric and, thus,
into the mouth of the boom is enhanced. Due to edge effects in the flow
channel, and possibly also to surface tension effects, the slick configu-
ration was not uniform across the width of the channel. The current speed
was lowered from 0,17 meter/sec (0.55 ft/sec) to approximately 0.14
meter/sec (0.45 ft/sec) to allow the oil slick to form more in front of the
boom. Under these conditions, a headwave formed just forward of the
boom, and the depth of oil inside of the bag decreased. A sketch of the
slick configuration of 5400 ml of oil and 0 .14 meter/sec (0.45 ft/sec)
current velocity is shown in Figure 3.
Test No. 4 was conducted to look at the effects of the viscosity of the oil
on the oil retention ability of the boom. No. 2 Diesel fuel has a viscosity
which is of the order of 1/100 that of the SAE 30 oil used in Test No. 3.
The oil was added until the oil retention limit of the boom was reached.
The No. 2 Diesel fuel oil slick extended beyond the mount of the boom
and the primary headwave was outside of the bag. After 1400 ml of oil
24
-------
10”
SPECIFIC OIL VOLUME — 17.65 LITERS/METER (0.19 FT. 3 /FT.)
CURRENT VELOCITY 0.17 METERS/SEC. (0.55 FT./SEC.)
N 8”
FLOW
2%
(OIL)
5”
CHANNEL SIDE
01
FIGURE 2
- OIL SLICK PROFILE, SECOND TEST SERIES (SAE 30 WEIGHT OIL)
-------
8”
cr -- - - - - -
,
I
/
,
SPECIFIC OIL VOLUME - 17.65 LITERS/METER (0.19 FT. 3 /FT.)
CURRENT VELOCITY = 0.14 METERS/SEC. (0.45 FT./SEC.)
FIGURE 3 — OIL SLICK PROFILE, SECOND TEST SERIES (SAE 30 WEIGHT OIL)
o )
6”
-------
was poured into the channel, the headwave was so deep—-approximately
5 .08 cm (2 .0 in) deep-—that oil droplets which were torn off the trailing
edge of the headwave did not rise and rejoin the slick, but were swept
under the boom. Because of the test facility configuration, once any oil
got past the boom, it was carried into the propeller of the outboard motor
and dispersed into the current as tiny droplets. Unless these droplets
rose to the surface before reaching the boom, they were swept under the
boom in succeeding passes. With Diesel oil this effect was particularly
limiting, and further testing was prevented by oil droplets entrained in
the water flow itself. Figure 4 shows the slick configuration for 1400 ml
of No. 2 Diesel fuel at-an estimated 0.17 meter/sec (0.55 ft/sec) current
velocity. If the demonstration model results are scaled up to prototype
size in the same manner as described for Test No. 3, it would correspond
to a specific volume of 2.07 meters 3 /meter (22 .3 ft 3 /ft) at a current
velocity of 0.65 meters/sec (2.13 ft/sec) or 1.26 knots.
In both Tests 3 and 4, oil droplets carried into the bag by the water flow
were observed, either to rise directly to, and join, the oil slick within
the bag or, in some cases, to 11 ro1l 1 ’ along the woven fabric toward the
back of the bag where they rose to join the slick. No oil droplets en-
trained in the water flow were observed to pass through the fabric. Addi-
tionally, in neither test was oil observed to extrude through the fabric.
In addition to demonstrating feasibility of the basic concept and indica-
ting performance potential, these tests provided the opportunity to eval-
uate various model fabrication and mounting techniques and provided the
basis for developing model configuration (in terms of throat opening,
throat-to-length ratio, and plastic/fabric seam location) for further testing.
Third Test Series
While the previous oil containment test series conducted in the flow
channel demonstrated model performance capability within the limitations
of that test facility, additional testing was necessary at higher speeds
and in a deeper tank which would permit greater boom drafts and greater
oil volumes.
A series of final demonstration runs was conducted in the 24.4 meter
(80 foot) Hydronautics oil test tank. These tests consisted of essentially
two events—-a comparison of the fabric boom and flat plate in calm water
using SAE 30 weight motor oil, and a performance demonstration of the
fabric boom in waves also using SAF 30 weight motor oil. The rationale
for the selection of these tests was as follows:
27
-------
. —Y- ________ _______________________ ___________
16W ’
/
/
SPECIFIC OIL VOLUME = 9.20 LITERS/METER (0.099 FT. 3 /FT.)
CURRENT VELOCITY 0.17 METERS/SEC. (0.55 FTJSEC.)
FIGURE 4 — OIL SLICK PROFILE, SECOND TEST SERIES (NO. 2 DIESEL FUEL)
1 2”
-------
• By comparing performance in calm water, the significant
engineering effort necessary to accurately model the flat
plate for wave response was eliminated.
• If the fabric boom performed better than the flat plate in
calm water and the fabric boom could be shown to perform
successfully in waves, then it may be assumed that the
fabric boom will perform better than the flat plate under
all conditions.
The Hydronautics towing tank is 0.61 meters (24 in) wide and was filled
to a depth of 0.48 meters (19 in) for this tpst series. One side is made
of clear acrylic for viewing. A rubber -tired carriage rides on steel rails
along the top of the tank walls and is pulled by an electric motor-driven
endless cable. Waves are generated by a wave maker installed at one
end. Photographic coverage of all runs was provided by a 16mm camera
mounted on an outrigger attached to the tow carriage.
Three models were fabricated for this test series: a flat plate boom, a
calm water fabric boom, and a free-to-float fabric boom. Sketches of
these models are provided in Figure 5. Except for the floatation and the
support and mounting systems, both fabric boom models were identical
and employed a cotton polyester fabric, Fabric No. 5 of Table 1, as the
lower bag material. As pointed out previously, fabric testing efforts had
not been completed when the second demonstration tests were conducted 0
Although in that demonstration series, the boom configuration which used
Fabric No. 4 (considered at that time to be the best available) performed
satisfactorily to the limits of the test facility and no oil penetration of
the fabric was observed, it was recognized that the higher speeds of the
third demonstration test series could result in greater differential
pressures across the fabric and thus an increased probability for oil
penetration of the fabric. Therefore, based on the final results of the
fabric selection effort described in Section VI, Fabric No. 5 was chosen
for use as the lower bag material in this demonstration series since,
while its weave was slightly less open than Fabric No. 4 (calculated
percentage of open fabric area of 45% versus 51%), the differential
pressure at which oil extrusion occurred was greater (3.3 psi versus
2.9 psi).
It should be noted that the floatation and support systems of the free-to-
float model were derived empirically and thus do not necessarily represent
an optimum configuration. The basic purpose of the free-to-float test was
to demonstrate concept performance in other than calm water and, specif-
ically, to show the inherent response of the oil-filled boom to wave
action.
29
-------
ROD (FIXED)
1” DIA. UPPER SUPPORT
3.. /
__ —--I
- -
1/8”
ROD (FIXED)
I 15”
/
,
1/8” DIA. LOWER SUPPORT ROD
15”
FLAT PLATE
FIXED FABRIC BOOM
FREE-TO-FLOAT FABRIC BOOM
FLOATATION DEVICE & UPPER SUPPORT
THROAT SPACER (3)
0)
0
NOTE: 23 INCH BOOM WIDTH FOR ALL THREE TEST MODELS
FIGURE 5 — CONFIGURATLON DETAILS OF TEST MODELS, THIRD TEST SERIES
-------
All models were mounted in a clear plexiglass support assembly which,
in turn, was suspended from the wheeled tow carriage. This carriage
was capable of speeds to 0.69 meters/sec (2.25 ft/sec) with the model
boom installed. The flat plate and calm water fabric boom models were
rigidly mounted directly to the sides of the plexiglass support assembly.
The free-to-float model was towed by four nylon threads which, at one
end, were attached to either end of the floatation device and to either
end of the lower spacer bar. At the other end these lines were secured
to the inside of plexiglass support device at a height such that the angle
of pull was horizontal. Sufficient length was provided so that the ability
of the model to follow the wave profile was not inhibited by the tow system.
The fabric boom models were constructed with clear polyethylene sides
which were equipped with frontal flaps. These flaps were taped to the
inside of the plexiglass support structure to prevent oil from passing
between the boom and that structure, and thus invalidating test results.
Passage of oil between the plexiglass support structure and the tank was
prevented by installing flexible wipers on exterior leading edges of the
plexiglass.
All tests were conducted using the constant volume technique. In this
technique all of the oil to be encountered by the model is collected and
contained just ahead of the model by a cofferdam which is removed just
prior to boom acceleration. Steady state is achieved after the desired
carriage velocity is reached and the oil slick has adjusted to a constant
length. When oil losses are large, no steady state condition is achieved.
In the two-day period available for testing, a total of 33 test runs were
conducted with oil volumes of up to 30,000 ml and at speeds of up to
0.69 meters/sec (2.25 ft/sec). The superior oil retention capability of
the fabric boom over the flat plate in calm water was demonstrated; and
the free—to-float boom was shown to be capable of oil retention while
exhibiting stable performance and satisfactory wave response at all
speeds tested. Since the primary purpose of these tests was demonstra-
tion rather than engineering testing, oil losses (when they occurred)
were measured in only a few instances and estimated in others, and all
runs were filmed. The results of these tests are summarized in Table 2.
In addition to clearly demonstrating the feasibility of applying the con-
cept of oil/water separation by means of a hydrophilic fabric to a float-
ing oil containment system, these tests provided valuable data for
developing an improved boom configuration and an indication of the
adaptability of this concept to a floating oil collection system.
31
-------
TABLE 2 - SUMMARY OF THIRD TEST SERIES
Model
Oil Volume
(ml)
Carriage S
need
Wave Charac—
teristics
011 Loss
Run No.
(meters/sec)
(ft/sec)
Flat Plate
I
I
I
I
t
5,000
5,000
5,000
10,000
10,000
10,000
0.23
0.30
0.38
0.23
0.30
0.38
0.75
1.00
1.25
.75
1.00
1.25
Calm
I
I
I
I
$
None
85 L
30 La
64%
95% -
21—1
21—2
21—3
20—1
20—3/21—15
20—2/21—4
Fixed Fabric
10,000
0.23
0.75
Calm
None
20—4/21—5
10,000
0.30
1.00
None
20—5/21—6
10,000
10,000
10,000
10,000
10,000
0.38
0.46
0.53
0.61
0.69
1.25
1.50
1.75
2.00
2 25
None
Possible Minute
2 - - Minute
2%
/
Extrusion
Extrusion
20—6/21—7
21—8
21 _ 9 / 21 _ 10 t_
21—li
21-12
20,000
0.23
0.75
None
20—8
20,000
20,000
20,000
30,000
0.30
0.38
0.46
0.38
1.00
1.25
1.50
1.25
None
None
Pos I Minute
5%
Extrusion
20—9
20_10Li/21_ 13
20—1i /20—12 /2i—l4
20—13
Free—to—float
10,000
0.23
0.75
None
21 19 L
Fabric
20,000
0.23
0.75
None
21—20
20,000
0.30
1.00
L .
None
21—20
Estimated
/3 Measured
Oil loss continues throughout duration of run.
Although no oil droplets could be seen extruding, 6—10 small droplets were visible on water surface aft of the boom at conclusion of run.
10—12 tiny droplets were observed to extrude through fabric during run.
Loss appeared to have been caused by extrusion.
Some oil passage occurred due to vortex action at Junction of model and plexiglass support plates.
Loss occurred near end of run when, as bag became full, internal flow pattern caused spillage out of mouth of bag. No loss through fabric.
)‘— Length — I • 83 meters (8 ft): Height — 3.18 cm (1 .25 in).
Run No. 20—7 experienced large oil loss due to vortex action. Corrected in 21-series runs by repositioning model within plexiglass support plates.
Runs 21—16, 21-17, and 21-18 were conducted without oil to establish towing and mounting techniques.
-------
As shown in Table 2, oil extrusion through the fixed fabric model was
noted at speeds above 0.46 meters/sec (1.50 ft/sec). If the 9.21cm
(3 .63 in) throat draft is compared to an average flat plate boom draft of
76.20 cm (30 in), the scale of the fabric model would be approximately
1/8.3. Under conventional Froude scaling the model velocity of 0.46
meters/sec would represent a full—scale velocity of 1 .32 meters/sec
(4.32 ft/sec) or 2.6 knots. However, the applicability of Froude scaling
to extrusion is not clear since the fabric characteristics would not be
changed in a full—scale system. Furthermore, the draft of a full—scale
fabric boom will not necessarily be 76.20 cm (30 in)——further analysis
and testing must be conducted before a full-scale configuration can be
developed. One of the factors causing oil extrusion is the differential
pressure across the fabric. For a given fabric, differential pressure can
be reduced by increasing the bag length/throat opening ratio, thus pro-
viding a greater fabric area for water discharge.
Table 2 also shows that the maximum volume of oil which could be con-
tained by the fabric model configuration before loss by spillage or drain-
age occurred was between 20,000 and 30,000 ml. An oil volume of
20 ,000 ml equates to a specific volume for the fabric model of 34.24
liters/meter (0.37 ft 3 /ft), which is far greater than the oil retention capa-
bility of the flat plate model, and represents a full-scale specific volume
of 2 .36 meters 3 /meter (25 .49 ft 3 /ft)——again, assuming a scale factor of
1/8.3. While the full-scale specific volume predictions appear to exceed
a prototype design goal of 0.93 meters 3 /meter (10 ft 3 /ft), the design of
a full-scale configuration has not been developed and thus the scaling
laws are not completely valid. However, it is clear that the specific
volume contained by the boom before spillage or drainage occurs can be
increased by increasing the bag length, thus providing a greater holding
volume.
It was concluded from these tests that,with minor configuration changes,
the oil retention capability of the fabric boom could be improved in terms
of increasing the specific volume of oil contained and increasing the
speed at which loss by extrusion commences.
Fourth Test Series
Following the successful demonstration of the basic fabric boom concept
at Hydronautics (third test series), CONSULTEC was asked to investigate
the applicability of that concept to high—speed oil collecting. The initial
phase of that effort was to be the demonstration of the capability of the
fabric boom to retain a “pumpable” volume of oil at a towing speed of four
33
-------
knots (2.06 meters/sec, 6.76 ft/sec): the basic considerations being that
the boom must retain some specified, but undefined, quantity of oil; and
that as this entrapped volume of oil is increased by sweeping, the excess
oil would be removed at the same rate it is collected.
The approach envisioned for this phase was the development of an ac-
ceptable or preferred bag configuration through iterative testing and modi-
fications of both design and procedures based on the review of test
results. Upon the determination of an acceptable bag configuration, a
compatible oil pump-out system would be developed and incorporated
for demonstration purposes. (However, as will be pointed out, due to
funding constraints, only one series of bag configuration test runs was
conducted.)
The tow tank at EPA, Edison Water Quality Research Laboratory, was
selected as the test facility for this phase. This tank is 30.48 meters
(100 ft) long, 3.66 meters (12 ft) wide, and can be filled to a water depth
of 0 .91 meters (3 ft). A wheeled carriage which spans the tank width
and runs on rails mounted on the tank sidewalls can be towed at speeds
to 4 knots by means of a variable speed electric motor and can carry both
test equipment and an observer/operator.
In the conduct of this phase it was desired to utilize the results of pre-
vious testing to the maximum extent possible. Since the majority of
prior test efforts were in calm water with no appreciable effort expended
on the analysis and design of boom floatation and towing systems, pre-
liminary testing was planned for calm water using the same type of
carriage/boom support employed in previous testing. Additionally, in
order to compare performance with previous efforts, the first boom se-
lected for testing was fabricated from the same material, Fabric No. 5,
Table 1, arid was similar in configuration to the boom used successfully
at Hydronautics. The frontal width of the model was increased from
58.42 to 116.84 cm (23 to 46 in) to take advantage of the wider tank,
while the throat opening was decreased from 7.61 to 5.08 cm (3 to 2 in)
and the length was increased from 38.10 to 142 .24 cm (15 to 56 in)
because of the increased towing speed: however, the rear plastic—to—
fabric seam was maintained at the same (mid-draft) relative position.
Upon completion of necessary modifications to the test facility, the in-
stallation of required underwater lighting and viewing systems, and the
preliminary debugging of both equipment and procedures during tests with-
out oil, one series of calm water tests using SAE 30 weight oil was con-
ducted on August 11, 1972. In these tests, the boom configuration
described above was held rigidly between two plexiglass plates,
34
-------
suspended from the tow carriage. Underwater performance was observed
and recorded via periscope by the portable SONY BRover TV camera
mounted on the carriage. A total of seven runs was conducted at
carriage speeds ranging from 1 to 2 knots (0 .52 to 1 .03 meters/sec;
1 . 69 to 3 .38 ft/sec).
Because these tests were considered at the time to be the first in a
series of iterative tests, emphasis was placed on evaluation of pro-
cedures and qualitative performance rather than on the gathering of quan-
titative data. However, since no additional testing of the fabric boom
was performed under this contract, and since this test series represents
the most severe testing to which the fabric boom concept has been
exposed, it is considered appropriate to discuss these tests in some
detail.
As in the case of all prior testing, the oil to be encountered by the boom
during the test run was contained in a small area just ahead of the boom.
For these tests, the oil was contained by the vertical plexiglass plates
(to which the boom was mounted) and by a removable athwart-tank
cofferdam which was lifted clear as the carriage began to accelerate.
(At the termination of the run, the cofferdam was dropped back into place
to retain the oil contained by the boom which otherwise would have been
lost during deceleration.) No physical measurement of the oil which got
by the boom was undertaken although, at the end of each run, the oil on
the water surface behind the boom was swept and collected at the ex-
treme (starting) end of the tank. It should be pointed out that while
estimates of oil loss are given in the following discussions, these ex-
tremely crude estimates are based on visual impressions of the relative
amount of oil floating on the water surface following the passage of the
boom. In addition to oil which penetrated the fabric or otherwise legit-
imately escaped from the boom, these estimates include oil losses
which may have been artificially induced by the test set up such as oil
loss over the boom or oil loss between the boom and the plexiglass side
plates due to vortex action.
Run No. 1 . A speed of 2 knots (1.03 meters/sec or 3.38
ft/sec) was selected for the first run since it represented
a compromise between maximum previous test speeds and
the desired goal of this phase, and thus the first step of
a bracketed approach. Initial oil volume was 9.46 liters
(2.5 gal). Although the periscope, lights, and TV camera
had been adjusted to provide full coverage of the bag
during a preliminary two-knot run without oil, the intro-
duction of oil caused the bag to submerge such that
35
-------
during this run only the plastic top of the bag was in
the camera’s field of view. Consequently, although
oil droplets were noted by tank-side observers to rise
in the wake of the boom, the source and cause of this
oil loss could not be determined. Total oil loss was
estimated at 15 to 20 percent.
Run No. 2 . After adjusting TV zoom to increase field of
vision, 28.39 liters (7.5 gal) of oil were added to the
residual oil from Run No. 1 and another two-knot run
was conducted to determine the path of oil loss.
Again, although oil droplets were noted rising to the
surface in the boom’s wake and an oil loss of 15 to
20 percent was estimated, the TV camera position was
unsuitable to determine the source of oil loss.
Run No. 3 . The positions of the periscope and under-
water lights were lowered to provide underwater TV
viewing of the entire rear portion of the bag and another
two-knot run was conducted using only the residual oil
from Run No. 2. During this run the pattern of oil
droplet movement within the bag indicated that the
water flow was essentially from front to rear. Addi-
tionally, extreme turbulence in the rear portion of the
bag was noted. Oil droplets caught both in the straight
flow lines and in the turbulence were observed to pene—
trate the fabric in the after section of the bag and
escape. Total oil loss was again estimated at 15 to 20
percent.
Run No. 4 . In order to confirm results of prior test
efforts, the carriage speed was reduced to 1 .0 knot
(0 .52 meters/sec or 1 .69 ft/sec) for this run. An
additional 18.93 liters (5 gal) of oil was added to the
residual oil remaining from Run No. 3. During this run,
internal turbulence was significantly less than that
noted at 2 knots and oil droplets could be seen drifting
along inside the bag (sometimes 1 ’rolling” along the
fabric) and rising to join the oil slick within the bag.
This would indicate that the water exit flow lines were
more evenly distributed along the length of the fabric.
No oil droplets were seen to pass through the fabric
and no oil loss was observed.
36
-------
Run No. 5 . Carriage speed for this run was increased to
1.5 knots (0.77 meters/sec or 2.53 ft/sec). No oil was
added to the residual oil of Run No. 4. Although internal
turbulence was considerably greater than that observed
during the previous run, no oil was observed to pass
through the fabric and no oil loss was observed.
Run No. 6 . In an attempt to bracket the speed at which
oil loss is initiated, the carriage speed for this run was
increased to 1.75 knots (0.90 meters/sec or 2.96 ft/sec).
No new oil was added. Again, an increase in internal
turbulence was noted and some- passage of oil through the
fabric was observed. Total oil loss was estimated at
5-10 percent.
Run No. 7 . With an estimated 35 liters (9 .25 gal) of oil
remaining from the previous run, the final run of this series
was conducted at a carriage speed of 2 knots, with the
desire to more accurately determine, if possible, the
mechanics of failure. As expected, internal turbulence
was greater than that observed at 1 .75 knots. During
this run, oil droplets were observed to “hang 1 ’ in
position on the inside of the fabric toward the rear of the
bag and then suddenly be forced or extended through.
Total oil loss was estimated at 15-20 percent.
Analysis of the results of this test series can be summarized as follows:
• The boom tested appeared to retain oil at speeds
to 1 .5 knots (0 .77 meters/sec, 2 .53 ft/sec).
• Oil leakage occurred at 1.75 knots (0.90 meters/sec.
2.96 ft/sec) and at two knots (1.03 meters/sec,
3.38 ft/sec)-.
• Oil leakage rate of 2 .0 knots appeared greater than
at 1.75 knots.
• The boom support system employed for those tests
was not satisfactory for high-speed operations
• Underwater viewing is a necessity for performance
analysis.
The failure of the boom to retain oil at speeds in excess of 1 .5 knots
can be attributed to the following factors:
37
-------
• Because the upper boom support was fixed, the boom
itself was not floating on the surface of the water. In
fact, the water surge wave generated at higher veloc-
ities increased both the pressure and the water flow
into the bag itself due to the rigidity and shape of the
upper support. This surge wave was estimated to be
as high as 6.35 cm (2 .50 in) at a carriage speed of
2 knots.
• Flow inside the bag was far from optimum. Successful
boom performance is dependent on flow distribution
across the entire bag length. From review of the tele-
vision tapes of these tests, it appears that water flow
within the bag at speeds in excess of 1 .5 knots was
essentially from front to back with little flow exiting
along the bottom of the bag. Additionally, because of
this flow pattern, extreme turbulence was generated
within the after-end of the bag itself.
• The location of the plastic/fabric seam at the rear of
the bag directly influences the water flow pattern.
While positioning this seam at the mid-point produced
best results at low speed in terms of retention capacity,
this location may not be optimum for high-speed
collection operation.
Other factors which may have influenced performance were the fact that
the boom tested was installed for numerous checkout runs during the
preceding month, and thus was repeatedly exposed to water of high
chlorine content. Although the deterioration aspect of the fabric has
never been examined, all prior tests were conducted with “new ’ t booms.
Although this test series did not result in the immediate demonstration of
a boom configuration which could be directly applied to high speed oil
collecting and additional testing could not be conducted due to funding
constraints, it should be recognized that these tests were only the first
step in an iterative test/design effort. Further analysis and evaluation
of the results of these and preceding tests will provide valuable input
for future efforts. CONSULTEC is confident of the adaptability of the
fabric boom concept to high-speed oil collecting.
38
-------
SECTION IX
REFERENCES
1. “Concept for the Recovery of Floating 011,11 Final Report to Federal
Water Quality Administration — Contract No. 14—12-878,
CONSULTEC, Inc., March 1971.
2. Miller, E. R., W. T. Linde nmuth, Hydronautics, Inc., Lehr, CDR
W. E., and Abrahams, CDR R. N., United States Coast Guard,
“Experimental Procedures Used in the Development of Oil Retention
Boom Designs,” Chesapeake Section, The Society of Naval
Architects and Marine Engineers, March 17, 1971.
3. Lindenmuth, Miller, and Hsu, “Studies of Oil Retention Boom
Hydrodynamics,” HYDRONAUTICS, Incorporated Technical Report
70 13-2 (USCG Office of Research and Development Report No.
714l02/A/008), December 1970.
4. Lehr, CDR W. E., United States Coast Guard, and Scherer, J. 0.,
Hydronautics, Inc., “Design Requirements for Booms,” Joint
Conference on Prevention and Control of Oil Spills, (API-FWPCA),
New York, December 1969.
5. Frank, R. L., Cornell Aeronautical Laboratory, Inc., Buffalo, New
, “Oil Pollution Control on the Buffalo River,” Joint Conference
or’ Prevention and Control of Oil Spills, (API-FWPCA), New York,
December 1969.
6. Hoult, David P., Editor, Department of Mechanical Engineering,
Massachusetts Institute of Technology, Cambridge, Mass.,
Oil on the Sea , Plenum Press, New York-London, 1969.
7. March, Frank, Ocean Systems, Inc., “Dynamic Keel Oil Contain-
ment System,” Joint Conference on Prevention and Control of Oil
Spills, (API), Washington, I D. C., June 1971.
8. “Oil Pollution Menace Grows . . . Ga mien Booms Attract Greater
Universal Interest,” The Dock and Harbor Authority , October 1969.
9. Mhz, E. A., Manager, R&D Laboratory, Shell Pipeline Corp.,
“Evaluating Oil Spill Control Equipment and Techniques,” Ocean
Industry , July 1970.
39
-------
10. “Swedish Pilot Invents Disposable Oil Boom,” Institute of Marine
Engineers, London Transactions, November 14, 1970.
11. Streeter, Victor L., Fluid Mechanics , Second Edition, McGraw-
Hill Book Company, Inc., p. 389, 1958.
120 Hoult, D. P., et al, “Concept Development of a Prototype Light-
Weight Oil Containment System for Use on the High Seas,” Johns-
Manville Research and Engineering Center, Manville, New Jersey,
June 1970.
13. Marks, W.., et al, Poseidon Scientific Corp., ‘ Theoretical and
Experimental Evaluation of Oil Spill Control Devices,” Joint
Conference on Prevention and Control of Oil Spills, (API),
Washington, D. C., June 1971.
14. Schwartzberg, Henry C., Chemical Engineering Dept., New York
University, “Spreading and Movement of Oil Spills,” Federal
Water Pollution Control Administration, Department of the Interior,
Program No. 15080, Contract No. WP 01342—O1A, March 1970.
15. Wicks, Dr. Moye, III, Supervisor of Fluid Mechanics, Shell Pipe
Line Corporation Research and Development Laboratory, Houston,
Texas, “Fluid Dynamics of Floating Oil Containment by Mechanical
Barriers in the Presence of Water Currents,” Joint Conference on
Prevention and Control of Oil Spills, (API-FWPCA), New York,
December 1969.
40
-------
SECTION X
APPENDICES
Page
A. Initial Model Testing of Preliminary Boom Configurations . 43
Figure A-i: Configuration Details of Initial Test Models . 44
Table A-i: Initial Program of Two-Dimensional Model
Tests of Preliminary Boom Configurations
Conducted in the Hydronautics Towing Tank . 46
Figure A2: Oil Loss Rate as a Function of Boom Velocity
in a irn \ATater . . . . . . . . . 4 7
Figure A—3: Oil Slick Characteristics for Flat Plate Boom. 48
Figure A—4: Oil Slick Characteristics for Flat Plate Boom . 49
Figure A—5: Oil Slick Characteristics for Fixed Short
Fabric Boom 50
Figure A-6: Oil Slick Characteristics for Fixed Short
Fabric Boom 51
Figure A-7: Oil Slick Characteristics for Fixed Long
Fabric Boom 52
Figure A—8: Oil Slick Characteristics for Fixed Long
Fabric Boom . 53
Figure A-9: Comparison of Oil Slick Characteristics for
Fabric Booms 54
B. Oil Characteristics 61
Table B-i: Properties of Various Test Oils 62
41
-------
APPENDIX A
INITIAL MODEL TESTING OF
PRELIMINARY BOOM CONFIGURATIONS
Test Program
Model tests of the initial boom designs were conducted in the 24.4 meter
(80-foot) towing tank at Hydronautics, Inc., (Laurel, Maryland). This
program consisted of the following tests:
a. Exploratory two-dimensional tests of two boom configu-
rations in calm water to measure their oil containment
ability at various speeds for different quantities of oil.
b. Tests of a rigid flat plate boom of the same draft as the
models tested in (a), above, to obtain oil containment
properties for comparison with the fabric boom designs.
c. Tests in waves of a free-to-move fabric boom model to
determine the effect of waves on the oil containment
properties of the system.
Four models were employed during this test series: a fixed flat plate
boom, a fixed short fabric boom, a fixed long fabric boom, and a free-
to-f bat fabric boom. Configuration details of the fabric booms are pro-
vided in Figure A-i. The flat plate model was fabricated of plywood and
had a draft equivalent to that of the fixed fabric models (5.75 in). The
fixed fabric booms were fabricated from a plain linen fabric previously
determined as having the best oil—water separation properties of the
fabrices examined-—reference (1), fabric sample no. 0701. In an attempt
to improve on the performance of the fixed booms, a coarser linen fabric,
Couturier Fabric WPL 9893, was used for the free—to—float boom model.
Geometric characteristics of these fabrics, designated A and B, respec-
tively, are listed in Table 1 of the basic report.
The Hydronautics towing tank is 24 .4 meters (80 ft) long, 0.61 meters
(24 in) wide and was filled to a depth of 0 .46 meter-s (1 8 in) for the tests.
Models were towed at steady speeds ranging from 0.15 to 0.69 meters/sec
(0.5 to 2.25 ft/sec). A flat-plate wavemaker at one end of the tank was
used to generate regular waves which varied in length from 0.61 to 2.44
meters (2 to 8 ft) and were 3 . 1 8 or 6 . 35 cm (1 . 25 or 2 . 50 in) high. Using
the constant oil-volume method of testing, the specified quantity of oil
was placed just ahead of the boom. After each run, the amount of oil
43
-------
FIXED BOOMS
3” WA. STYROFOAM
CYLINDER
FREE TO FLOAT BOOM
WIPER
FABRIC
LOWER STIFFENING DOWEL
DIRECTION OF TRAVEL
l o f
WIDTH 1W) 23” 23”
THROAT IT) 4.25” 4.25”
LENGTh CL) 5” 10”
TOW CARRIAGE
MOUNTING SUPPORT
LOWER SUPPORT
ROD
UPPER TOW
LINES
PLEXIGLASS SIDE PLATE (2)
LOWER TOW
LINES
FIGURE A-i — CONFIGURATION DETAILS OF INITIAL TEST MODELS
-------
which passed the boom was measured. Sixteen mm film coverage was
obtained for each of the tests.
This initial test program is summarized in Table A-l. While the per-
formance of the free-to-float fabric model was such as to preclude the
reporting of any meaningful data, oil loss rates for the flat plate model
and the fixed fabric models (in calm water) are shown in Figure A-2.
The flat plate boom tests provided engineering data on oil slick geometry
and oil loss characteristics for flat plates in currents. SAE 30 weight
motor oil was used for all the tests. Fabric boom tests of both the fixed
and free-to-float models showed that due to excessive resistance to
water flow through the fabric, the boom system did not behave as in-
tended. Flow in the bags was almost stagnant for the fixed-bag con-
figurations, and the overall behavior of the boom was much like that of
a flat plate.
The fixed—bag tests provided engineering insight into the effect of the
bag and influence of bag length on oil slick characteristics. These tests
demonstrated that having a hydrophilic material on the top side of the
bag is not an advantage. It is, in fact, a disadvantage because the top
of the bag is pushed out of the water by the floating oil. Since this pre-
vents water-wetting of the hydrophilic fabric and reduces its effective-
ness in preventing oil passage, collected oil passed through the top.
This situation can be prevented by using a non-porous fabric on the upper
portion of the bag.
The free-to-move model did not assume the anticipated shape when towed
in calm water; and in waves, the opening of the mouth of the bag became
extremely unstable. These observations have led to the conclusion that
the mouth of the bag should be held mechanically to the appropriate
dimension.
Film coverage of the tests was analyzed using a photo-optical data ana-
lyzer. Oil slick characteristics, oil dynamics and boom behavior were
studied. Figures A-3 and A-4 show the characteristic shapes of the oil
slicks formed by a flat plate boom at various speeds for different speci-
fic volumes of oil. Figures A-S and A-6 show slick geometry for the
fixed short cloth boom and Figures A-7 and A-B show the same data for
the fixed long cloth boom. Comparison of the cloth boom data with the
flat plate data shows that the slick geometry is essentially the same for
equal speeds and specific volume of oil, and supports the assertion that
these cloth booms behaved essentially as flat plates. Figure A-9 com-
pares the long cloth boom to the short cloth boom for a given speed and
45
-------
TABLE A-i — INITIAL PROGRAM OF TWO .DIMENSIONAL MODEL TESTS OF PRELIMINARY
BOOM CONFIGURATIONS CONDUCTED IN THE HYDRONAUTICS TOWING TANK
1 A to E
2 A to E
3 A to E
4 A to E
5 A to E
6 A to G
7 A to E
8 A to E
9 A to D
10 A to D
11 A to D
12 A to D
13 A to D
14 A to D
15 A to D
16 A to D
17 A to D
18 A to D
19 A to D
20 A to D
21 Ato D
22 A to D
23 A to D
24 A to D
25 A to D
26 A to D
27 A to D
28 A to D
28 E
28 F
Fixed,Short, Fabric
Boom
4,
Fixed,Flat Plate
Boom
Fixed, Long , Fabric
Boom
Free, Long, Fabric
Boom with Wooden
Dowel
w/Aluminum Dowel
w/Steel Dowel
2 x 1.25
3 x 1.25
4 x 1.25
6 x 1.25
8 x 1.25
2 x 2.50
3 x 2.50
4 x 2.50
6 x 2.50
8 x 2.50
2 x 1.25
3 x 1.25
4 x 1.25
6 x 1.25
8 x 1.25
2 x 2.50
3 x 2.50
4 x 2.50
6 x 2.50
8 x 2.50
8 x 2.5
8 x 2.5
9.29
46.46
18.58
18.58
9.29
9.29
18.58
9.29
0.15—0.61
0.15—0.61
0.23—0.53
0.23—0.53
0.23—0.53
0.23—0.69
0.30—0.61
0.23—0.53
0.15—0.46
0.23—0.46
0.23—0.53
0.23—0.53
0.23—0 .46
0.23—0.38
0.23—0.46
0.30
0.30
0.50-2.00
0.50—2.00
0. 75—1. 75
0.75—1.75
0. 75—1. 75
0.75—2. 25
1.00—2.00
0. 75—1. 75
0.50—1.50
0.75—1.50
0.75—1 .75
0. 75—1. 75
0.75—1 .50
0. 75—1. 25
0.75—1.50
1 .00
1.00
Calm
Test
Number
Model
Wave Condition (Length x Height)
(meters x cm) (ft x in)
Specific Oil Volume
(liters/meter) (ft 3 /ft)
Velocity Range
(meters/secj (ft/sec)
Calm
0.1
0.5
0.2
0.2
0.1
0.1
0.2
0.1
0.2
0.61
0.91
1.22
1.83
2.44
0.61
0.91
1.22
1 .83
2.44
0.61
0.91
1.22
1.83
2.44
0.61
0.91
1.22
1.83
2.44
2.44
2.44
x 3.18
x 3.18
x 3.18
x 3.18
x 3.18
x 6.35
x 6.35
x 6.35
x 6.35
x 6.35
x 3.18
x 3.18
x 3.18
x 3.18
x 3.18
x 6.35
x 6.35
x 6.35
x 6.35
x 6.35
x 6.35
x 6.35
18.58
-------
0.05
1.0 1.2 1.4 1.6 1.8
MODEL VELOCITY — FT/SEC
I
10 20 30 40 50
MODEL VELOCITY CM/SEC
2.0
60 70 80
FIGURE A-2 — OIL LOSS RATE AS A FUNCTION
OF BOOM VELOCITY IN CALM WATER
(1132.8)—
(849.6)—
(566.4)—
Zero Loss
Rate Velocity
Limit = 0.50 fps
(15.24 cm/sec)
Li
(283.2L...
(141.6)—
(113.3)_
(85.0) —
C.,
Lu
-J
C.,
Lu
I-
U.
U.’
I.-
0
-I
-j
0
Boom
0.04
0.03
0.02
0.01
0.005
0.004
0.003
0.002
0.001
0.0005
0.0004
0.0003
0.0002
0.0001
(56.6)_
Zero Loss Rate Velocity Limit
(28.3)_
Ft/Sec
Flat Plate 0.75
Short Cloth Boom 0.75
Long Cloth Boom 1.00
22.9
22.9
30.5
(14.2)_
(11.3)......
(8.5)_
(5.7)—
Li
Slick Specific Volume
0.5 Ft 3 /Ft (46.5 I/meter)
LEGEND
0.4
Slick Specific Volume
0.2 Ft 3 /Ft (18.6 I/meter)
0 02
0.6
0.8
2.2
2.4
2.6
47
-------
V = 0.23 meters/sec. (0.75 ft./sec.)
—16 —6
— 14
—12
—10 —
C ,)
—8
—6
—2
—4
—2
— 0
SAE 30 WEIGHT OIL
SPECIFIC VOLUME = 9.29 LITERS/METER (0.1 FT 3 /FT)
16 14 12 10 8
I I I I I I I
40 30 INCHES
I I I i I i i i i I
10 0
11 LL
CENTIMETERS
FIGURE A-3 —
OIL SLICK CHARACTERISTICS FOR FLAT PLATE BOOM
C ’)
LU
I-
LU
I-
z
LU
C.,
V = 0.89 meters/sec. (1.5 ft./sec.)
0
V = 1.04 meters/sec. (1.75 ft/sec.)
6 4
2
I I
48
-------
r 6 O 7 Sft/sec E
V = 0.23 meters!
W i
! [ 2
24222018 1614 1210 8 64 2 0
1 I I I I I I I I I I I I
V = 0.89 meters/sec. (1.5 It/sec.)
SAE 30 WEIGHT OIL
SPECIFIC VOLUME = 18.58 LITERS/METER (0.2 FT 3 /FT)
V = 1.04 meters/sec. (1.75 ft./sec.)
6 4
i I i I
16 14 12 10 8
I i I i 1 I
30 INCHES 20
I I I I I I I I I i
CENTIMETEkS
10
I I I I
0
I I I I
OIL SLICK CHARACTERISTICS FOR FLAT PLATE BOOM
49
INCHES
60 50 40 30
I I
CENTII
10
—6
V = 0.74 meters/sec. (1.25 ft./sec.)
0
0
W
I-
W
I-
2
LU
U
—16
—14
—12
—10
—8
—6
—4
—2
—0
16
14
w 12
; 10
8
- 6
2
LU
02
0
FIGURE A-4 —
U,
Lu
I
0
2
-.4
—2
—0
40
2
-------
I L I I I I I I I
50 40 30
CENTIMETERS
I I I I I
20
INCHES
‘ I • I I I
10 0
—0
-r .--
I. — — - \ r— --------
• 1• . s _
..
• -S.
— —- - I •
.. -‘
5’ -—
S_ •I j — ——S.-
5 ’ , ‘. ,
_______ \ ..‘•
\ f
.1
0.30 meters/soc. (1.0 ft.fsec.)
0.74 meterslsec. (1.25 ft./sec.)
— — — — — 0.89 meters/sec. (1.60 ft./sec.)
— —. — 1.04 meters/sec. (1.75 ft/sec.)
FIGURE A-5 — OIL SLICK CHARACTERISTICS FOR FIXED SHORT FABRIC BOOM
I I ‘I I ‘ I ‘ I ‘ I I 1
20 18 16 14 12 10 8 6 4 2 0
- U i
I
0
- z
— 10
—0
—2
—4
—6
Ui
Ui
I -.
U I
C.,
—20 SAE 30 WEIGHT OIL
SPECIFIC VOLUME 18.58 LITERS/METER (0.2 FT 3 IFT)
-------
CENTIMETERS
I I
I I I ‘
‘
I I I
1
I ‘ ‘
I
50
40
30
20
10
0
INCHES
I
I
I
I I
SAE 30 WEIGHT OIL
SPECIFIC VOLUME = 46.46 LITERS/METER (0.5 FT 3 /FT
FIGURE A-6 — OIL SLICK CHARACTERISTICS FOR FIXED SHORT FABRIC BOOM
I I
20 18 16
—0
14
—0
—2
( rI
I I t I
12 10 8 6 4 2 0
U,
- LU
I
-
—10
U,
LU
I-
LU
LU
0
—4
—6
— 20
0.30 meters/sec. (1.0 ft/sec.)
_______________ 0.74 meters/sec. (1.25 ft./sec.)
0.89 meters/sec. (1.50 ft/sec.)
— • — • — 0.61 meters/sec. (2.0 ft/sec.)
N
.-=--— —,‘— —
-------
I
50
I I
40
I I I
30
CENTIMETERS
I I
I I I I 1 1 I I I
20 10 0
I I I I I
INCH ES
I I I I I
20 18 16 14 12 10
I I I I
I I I I I
$ 6 4 2 0
SAE 30 WEIGHT OIL
SPECIFIC VOLUME = 9.29 LITERS/METER (0.1 FT 3 /FT)
FIGURE A-7 — OIL SLICK CHARACTERISTICS FOR FIXED LONG FABRIC BOOM
C.1
—0 r—0
C,,
w
I
C-,
2
C,)
uJ
IL l
I .-
2
U I
0
—2
—4
—6
—10
—20
0.23 meters/sec. (0.75 ft.Isec.)
0.30 meters/sec. (10 ft./sec.)
—. —. 0.74 meters/sec. (1.25 tt./sec.)
0.89 meters/sec. (1.50 ftjsec.)
-------
CENTIMETERS
I I I I I I u 1 ii
50 40 30 20 10
‘ I ‘
20 18 16
‘ I
14
‘ I
12
INCH ES
10
I ‘ I I ‘ I
8 6 4 2 0
SAE 30 WEIGHT OIL
SPECIFIC VOLUME = 18.58 LITERS/METER (0.2 FT 3 /FT)
FIGURE A-8 — OIL SLICK CHARACTERISTICS FOR FIXED LONG FABRIC BOOM
—0
(‘I
—2
—0
UI
—10
U,
U i
I-
Ui
I-
U i
C.)
—4
—6
—20
— 0.30 meters/sec. (1.0 ft./sec.)
—. —. — . 0.74 meters/sec. (1.25 ft./sec.)
0.89 meters/sec. (1.50 ft.Isec.)
0.61 meters/sec. (2.0 ft./sec.)
—. —. —- —.
-
•‘.— • \ ,— • . .
\. ‘\j
..
S-” r’
-------
C ,’
CENTIMETERS _______________
i r i t t i I I 1T I I !
50 40 30 20 0
p
2’
_________ INC FIES ___________
T I I 1
20 18 16 14 12 10 8 6 4 2 0
SAE 30 WEIGHT OIL
SPECIFIC VOLUME = 18.58 LITERS/METER (0.2 FT 3 FT)
V = 1.04 meters/sec. (1.75 ft./sec.)
FIGURE A-9 — COMPARISON OF OIL SLICK CHARACTERISTICS FOR FABRIC BOOMS
-------
quantity of oil. This figure shows that the configuration of the slick
outside of the bag is essentially the same for both bag lengths, and con—
sequently the depth of oil inside the bag increases as the bag length
decreases. Even with poor water flow through the bag, the oil slick
extends well into the bag. It appears that water flow conditions at the
back of the bag prevented the slick from going all to the back of the bag
for the long cloth configuration. The analysis of the film data provided
insight into behavior of flexible fabric booms, and useful engineering
information for the design of an improved system.
Films of the free-to-move boom in waves were studied, but due to the
angle of the camera it was not possible to accurately measure the slick
geometry.
The flat plate test data is in general agreement with the experimental and
analytical results of references (2) and (3). The headwave is generally
not smooth. Interfacial instabilities arising from viscous shear forces
cause iriterfacial waves to grow along the headwave, reference (3). This
causes inaccuracies in the measurement of the headwave geometry. Com-
parison of the headwave geometry for the flat plate tests using SAE 30
weight motor oil with the results presented in reference (2) for No. 2
Diesel Fuel shows the effect of viscosity. Increased viscosity appar-
ently causes the headwave to be shorter and thicker. The viscosity of
the SAE 30 weight motor oil is of the order of one hundred times that of
No. 2 Diesel Fuel
Understanding Some Observed Flow Phenomena
In some of the free—to—float model tests, the boom was subject to wild
oscillatory motions. Subsequent analysis led to a satisfactory explana-
tion of this behavior in both qualitative and mathematical terms. This
explanation lies rather in the tank geometry than in a fundamental in-
stability in the design of the boom itself.
Qualitative Discussion
The combination of bag material and the testing geometry produced a
sluice-like flow condition. The high velocity and consequently low
pressure region under the bag caused the lower edge of the bag to drop
closer to the bottom of the tank. This constriction in flow areas, with no
compensating head increase (due to the fixed upper edge), produced a net
decrease in total flow. The normally stable hydraulic jump downstream
of the bag then became an advancing surge wave, filling the low pressure
area behind the bag with an onrush of water sufficient to cause back
pressure and the restoration of the lower bag lip to its previous position.
55
-------
This cyclical behavior could have been prevented by either limiting the
lower free edge to less than one-half the tank depth (limiting the sluice
effect) or fixing the relative distance between the bottom and top edges
and freeing the whole assembly. The latter would have been self—com-
pensating and consequently stable, i.e., an initial decrease in flow
would have increased the upstream depth, lilting the entire boom and
reopening the flow area.
The extension of the instability analysis to further testing was based on
the sluice/hydraulic jump phenomenon. The Froude number for flow be-
neath the bag characterizes the form of the hydraulic jump behind the bag.
Under general flow conditions, the Froude number is a direct function of
the ratio of water depth below the bag to the total water depth and is
independent of velocity. In very deep water the Froude number approaches
1 .0 for a free-floating boom, inducing a very smooth standing wave behind
the boom. (With a Froude number of between 3 and 6, the standing wave
is called a “smooth .prejump ’, converting to wild oscillation between 6
and 20, as observed in the Hydronautics tests.) To maintain this same
characteristic condition in a shallow tank, the Froude number must be
restricted to less than 3. This can be readily done by keeping the depth
below the bag somewhat greater than one-half the total water depth.
Analysis of Instability
Assuming that the booms tested in the Hydronautics tank were relatively
impenetrable to water flow, their behavior moving in the tank is analogous
to that of water flow past a fixed sluice 0
The assumed condition is illustrated below:
V 0 . _____
4 vwo
h
0
Cb
C
T
‘V
v i
V 2
h 2
56
-------
Under steady flow conditions
Vh =VCb=Vh (1)
00 ic 22
where Cb is the sluice contraction coefficient and
C f (b )
From continuity considerations
2gh / 1 + C b/h 0 (2) for the sluice
+ gC b ‘ (i + CCb/h \ (3) for the hydraulic
2/ jump
where V = 0 for a standing wave
w
V < 0 for a receding surge wave
Since the bag was not in fact a flat plate, the venturi effect produced by
the high velocity, V 1 , underneath the bag and the more or less stagnant
region in the bag combine to push down the lower edge, reducing the dimen-
sion b.
The resulting constriction in flow creates the unsteady condition shown
below:
57
-------
Although the increase in head due to water storage increases the flow
velocity (V 1 > V ), the total flow decreases due to the reduction in flow
area. Since the boom top is fixed and lower edge unrestricted, there is
no compensating lift of the entire boom to restore steady flow conditions.
The surge, V advances into the boom, drowning the sluice and
restoring the static pressure balance for a repeat of the cycle. At high
velocities, the advancing surge wave is not a smooth jump, but a wild
oscillating wave. Under these circumstances the surge is violent enough
to propagate forward of the boom, resulting in the phenomenon referred to
as “galloping” when seen in the test films.
Numerical spot checks, based on observed velocities and boom configura-
tions and using the equations given, provide the basis for the narrative
description given above.
Extension of Analysis for Test Design
Under deep water conditions
Cb
1
h
0
and the tail wave will be a smooth standing wave.
The tail wave, or hydraulic jump, is known to be a function of the Froude
number of the flow, V 1 . By definition, a standing wave occurs when
V = 0. Therefore, using equation (3) and defining the Froude number,
F 1 , as
v2 /
F = 2gh 1 gCb(l+C - --
1 gCb o/ci ch
C \ 0
(4)
For a deep water condition the tail wave is a standing wave, correspond-
ing to a Froude number of near 1. This same characteristic tail wave is
retained up to a Froude number 3, changes to a “smooth prejump 1 ’ until
F 1 6, becomes wi1d1y oscillatory for F 1 between 6 and 20, and becomes
2
58
-------
a well—formed hydraulic jump for F 1 greater than 20, reference (11). In
order to simulate the deep water conditions in a shallow tank, the Froude
number should be kept below 3. Solving (4) iteratively we find
C b/h 0 0 .457. Since the analysis is based on a sharp-edged sluice
gate, we can assume that in our case C = 1 and require that b/h 0 .5
as a design constraint consistent with o served tests. o
59
-------
APPENDIX B
OIL CHARACTERISTICS
The quantity of oil to be retained by a boom may be specified in terms of
the specific volume, or the volume of oil contained per unit length of
boom. The full-scale tests of refer nce (12) involved a calculated aver-
age specific volume of .316 meters /meter (3.4 ft 3 /ft); and based on the
shape of the boom and the oil slick under towi -ig conditions , the peak
value could have been as high as .929 meters /meter (10 ft /ft)
The constant volume method is better suited for tank testing than the con-
tinuous slick method, and a scaled volume of oil corresponding to a full-
scale specific volume of .929 meters 3 /meter (10 ft 3 /ft) is considered to
be a reasonable design goal.
A literature search has revealed that the fluid properties of the oil can have
a significant influence on the dynamic behavior of the oil slick and the
ability of a boom to retain oil. Specific gravity, viscosity and inter-
facial tension (between oil and water) vary for different oils. A list of
some of the oils that have been used in previous testing is given below:
Test Oil Reference
No. 2 Diesel Fuel (2, 3, 7, 9, 13)
No. 4 Fuel Oil (13)
No. 5 Diesel Fuel (9)
No. 6 Fuel Oil (9)
SAE 30 (3)
Industrial Lubricating Oils (3, 13)
Soybean Oil (12)
Crude Oils (9, 14)
Properties of various test oils are provided in Table B-i. Specific grav-
ities and interfacial tensions do not appear to vary over a wide range,
whereas viscosities do. Studies reported in reference (3) indicate that
viscosity has an effect on the headwave. Comparisons between No. 2
Diesel Fuel and SAE 30 motor oil show that increased viscosity apparently
causes the headwave to have greater thickness at a given speed.
The velocity of the current at which oil droplets are torn off the head-
wave is dependent on the interfacial tension between the oil phase and
the water phase, and is characterized by a critical Weber Number,
61
-------
(1) Synthetic Sea Water
(2) Reference (9)
(3) SEARS ALLSTATE regular motor oil SAE 30 (additive-free)
TABLE B-i — PROPERTIES OF VARIOUS TEST OILS
0 )
Type of Oil
Kinematic
Viscosity
(centi—stoke)
Viscosity
(Saybolt
Universal
Seconds •—
at 40°F)
Absolute
Viscosity
(centi—poise)
Specific
Gravity
Surface
Tension
of Oil
(dynes/cm)
Interfacial
Tens ion
Oil/distilled
water
(dynes/cm)
Interfacial
Tension
Oil/sea water
(dynes/cm)
Source
No. 2 Diesel Fuel
43
0.860
28(2)
27(2)
Ref.(3)
No. 2 Fuel Oil
6,45
47
5.6
.868
26,8
20,6
Ref.(2)
(ESSO) Faxom 35
19
.883
Ref (3)
(ESSO) Faxom 50
100
.900
Ref.(3)
No. 4 Fuel Oil
113.2
525
99.6
.880
28.1
27.3
Ref.(2)
(Humble) Nuso 38
118
.954
Ref . (3)
SoybeanOil
120 ,8
560
111,5
.923
31.4
23.6
Ref.(2)
SAE 30
380
0.906
33,5(3)
Ref.(3)
Kuwait Crude
28
22
Ref.(6)
Intermediate Sweet
Crude
4.6
.821
24
31
18(1)
-
Ref.(9)
Mercy Crude
30
.812
27
29
10 U)
Ref.(9)
Shallow Yates Crude
61
.910
28
31
— 14(1)
Ref.(9)
Fullerton Crude
45.8
.81
24.9
18.7
Ref , (14)
Tia Juana
297
.89
20.9
34.3
Ref.(14)
Urania
2591
.92
31.5
17.5
Ref.(14)
Safaniya
270
.87
27.3
30.1
Ref.(14)
-------
reference (15). In order to simulate oil entrainment in the small-scale
demonstration model, the interfacial tension coefficient would have to be
Weber-scaled if the model velocities are to be Froude—scaled. Simul-
taneous Weber and Froude scaling is achieved if the ratio (Velocity) 4 /
(interfacial tension) is maintained constant. Surface active agents are
available for reducing the interfacial tension of the oil. The effect, if
any, that such additives to the oil would have on the hydrophylic proper-
ties of the boom fabric, must be explored.
No. 2 Diesel Fuel and SAE 30 weight motor oil were used in preliminary
developmental tests and in tests to demonstrate the performance of the
fabric boom system. Since the viscosity of these two oils varies by an
order of magnitude of 100, and since they encompass the range of den-
sities most likely to be encountered in actual cases, they represent
realistic conditions for preliminary evaluation. In more advanced ex-
ploration, other oils should also be considered.
63
-------
SELECTED WATER Report No. L 3. Acessioz No.
RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
4. Th1 5. J port Date
FABRIC BOOM CONCEPT FOR CONTAINMENT AND COLLECTION
OF FLOATING OIL 8. rf m . Orge ati n
Repor# No.
7. Aurhor( )
Bonz , Philip E. 10. ProjeetNo.
15080 FWM
9. Organization
11. ContraetJGrantNo.
Consultec, Inc. - -
Rockville, Maryland — —
12. Typc. f Repo and
Period Covered
12 Sr nsorfrv Organi ation 0
iS. Supplt mentary
U.S. Environmental Protection Agency Report No. EPA-670/2-73-069,
September 1973.
16. .4bstract
The feasibility of applying the concept of oil-water separation by means of woven
hydrophilic fabric to a floating oil containment boom was investigated through a
series of model tests. A preliminary model boom configuration was developed and
towed at speeds to 0.686 meters/sec (2.25 ft/sec) in both calm water and waves.
Oil retention performance of this model was clearly superior to that of a con-
ventional flat plate boom of comparable draft in the environment investigated.
A larger model of similar configuration demonstrated no oil leakage when towed
at 0.77 meter/sec (1.5 kt) in calm water.
While further detailed analysis, engineering, and testing is required to fully
examine this concept, it appears that a properly designed flexible boom which
uses a hydrophilic skirt material offers significant potential both as a con-
tainment device for floating oil in high current situations and as a high-speed
collecting device.
i7a. Descriptors -
*Water Pollution Control, *Ojl Spills, *Flow Separation, *Running Waters, *Fabric,
Feasibility Study, Oil, Oil-Water Interfaces, Oil Pollution, Separation Techniques,
High Flow
!7b. Identifiera
*Floating 011, *011 Containment, *Ojl Collection, *Ojl Boom, Hydrophilic Fabric
I . COWRR E id & G :; 05F
18, A iia 0 ili o’ 19, Security Class 0 ‘2!. No,of Send To:
Rep ) -, Pages
WATER RESOURCES SCIENTIFIC INFORMATION CENTER
9. ,es ‘ It3’ 4. i S. . I-Z : U S. DEPARTMENT OF THE INTERIOR
(Page). WASHINGTON. D. C. 20240
Ahstractor Philip E. Bonz I ‘ Consultec. Inc.
w ’
------- |