•A ATEH I-OL'.VTIGN COVTROi. RKSEARTH STRIFE
           * ^
        CONCEPT DEVELOPMENi ut-
        HYDRAULIC SKIMMER SYSTEM
      FOR  RECOVERY OF FLOATING OIL
EN V I RON MEN TU PR<» !£(("[(> \
NV \T K R (i T V 1.1TY OFFICE

-------
           WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes
the results and progress in the control and abatement
of pollution in our Nation's waters.  They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Water Quality
Office, Environmental Protection Agency, through inhouse
research and grants and contracts with Federal, State,
and local agencies, research institutions, and industrial
organizations.

Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Office of Research and Development, Water Quality
Office, Environmental Protection Agency, Room 1108,
Washington, D.C.  20242.

-------
CONCEPT DEVELOPMENT OF A HYDRAULIC SKIMMER
    SYSTEM FOR RECOVERY OF FLOATING OIL
        Battelle Memorial Institute
        Pacific Northwest Laboratory
        Richland, Washington  99352
                  for the
      ENVIRONMENTAL PROTECTION AGENCY
           WATER QUALITY OFFICE
         Project # 15080FWP 04/71
           Contract # H-12-884
                April 1971

-------
             EPA Review Notice
This  report has been reviewed by the Water
Quality Office, EPA,  and approved for publication.
Approval does not  signify that the contents
necessarily reflect the views and policies of
the Environmental  Protection Agency,  nor does
mention of trade names or commercial products
constitute endorsement or recommendation for
use.
 For sale by the Superintendent of Documents, U.S. Government Printing Office
            Washington, D.C. 20402 - Price $1.25
                Stock Number 5501-0068

-------
                            ABSTRACT
          Efforts are being directed to develop effective counter-
measures against floating oil slicks.   Mechanical recovery methods,
which do not cause additional environmental insult,  are most attractive.
Such a concept, a hydraulic skimmer, was investigated.

          Floating headers, providing a linear water spray pattern on
the water surface, are attached to an open sea workboat.  Sea water is
pumped through spray nozzles mounted on the headers  to  move an oil slick
toward the boat.  Side mounted chambers are positioned  to collect the
concentrated floating oil.  Recovered fluid is pumped to an onboard
separation system from which the oil is transferred  to  floating tanks or
barges and the water is recycled to the spray system.

          Experimental work was directed toward component development
and evaluation of a large system model in a simulated environment.  A
35 foot support vessel, a 40 foot spray header, and  a 9 foot collection
chamber provided the 23 1/2 foot model sweep width.   Speeds of advance to
five knots, random waves to 30 inches significant height and three oil
types were used in evaluating this system.  Other equipment such as an
ultrasonic oil thickness gauge, process pumps and tankage were also used.

          Model experiment results showed, for light oils, 80 to 100
percent effectiveness and oil recovery rates of 6600 to 8700 'gph.
Results with Bunker fuel were not as good,  being on the order of 1300 to
1800 gph and 12 to 30 percent effective in recovering oil from the water
surface.  However, program time constraints did not  permit experimental
verification of modifications expected to increase performance on heavy
oils.

          Further development of the concept, including additional model
testing with heavy oils plus open sea evaluation of  a prototype system
was recommended.

          This report was submitted in fulfillment of Contract No. 14-12-
884 between the Federal Water Quality Administration and Battelle-Northwest,
a division of Battelle Memorial Institute.

          KEY WORDS:  Concept, equipment development, oil skimmer,
open sea, evaluation, oily water, separation techniques, technical fea-
sibility, efficiencies, hydraulic studies, hydrodynamics, jets, math-
ematical studies, testing, water pollution treatment.
                               Ill

-------
                            CONTENTS
Section






I




II




III




IV




V




VI




VII




VIII




IX
Conclusions




Recommendations




Introduction and Summary




Spray Boom Development




Collection Chamber Development




Concept Evaluation Tests




Acknowledgments




References




Appendix
                                                Page
 1




 3




 5




15




47




65




83




85




87

-------
                             FIGURES


Number


  1       Oil Recovery Concept                             8

  2       Process Flow Chart                               9

  3       Concept Evaluation Arrangement and Model        12
            System Component Details

  4       Illustration and Force Balance                  16
            Diagram of the Spray Induced Skimming
            Mechanism

  5       Illustration and Vector Velocity                20
            Diagram of the Side-Skimming Arrangement

  6       Speed of Advance as a Function of Boom          22
            Configuration and Relative Oil to Water
            Velocity

  7       Speed of Advance Required for 50,000 gph        23
            Oil Recovery

  8       Flow and Pressure Requirements as a Function of 25
            Speed of Advance

  9       Maximum Skimming Speed as a Function of the     28
            Relative Oil Density for Given Spray
            Impingement Angles of 5, 10 and 20°

  10      Spray Boom and Pivot Connections                29

  11      Spray Boom Moments                              31

  12      Skimming Configuration During a Turning         32
            Maneuver

  13      Spray Simulator for Spray Nozzle Evaluation     36

  14      Nomenclature of Spray Simulator Variables        39

  15      Structural Connections Between Boom Sections;    44
            Between the Boom and the Support Vessel;
            Including Maximum Expected Forces and Reactions

  16      Final Design of Model Spray Boom                 46
                              Vl x

-------
Number                                                   Page

  17      Side Entry Collection Chamber Concept            48

  18      Collection Chamber Internal  Circulation         59
            Pattern Illustration

  19      Collection Chamber Displacements  Illustration    51

  20      Flow Channel  Test  Section                        53

  21      Final Collection Chamber Arrangement             55

  22      Internal Collection Chamber  Effectiveness as a   56
            Function of Free Stream Velocity

  23      Spray and Chamber  Effectiveness Tests           57

  24      Spray and Chamber  Effectiveness as a Function   59
            of Free Stream Velocity

  25      Schematic Showing  One  of Three Duplicate Oil    61
            Pickup  Arrangements  and Its Relationship
            to  Other  Parts of  the Collection Chamber

  26     Large  Scale Model  Collection Chamber            63

  27     Concept Evaluation Model Operating in 30 to     67
            36  Inch Waves

  28     Oil Recovery Rate as a Function of Slick        73
           Thickness

  29      Oil Recovery Rate as a Function of Speed of     74
           Advance

 30      Empirical Factor as a Function of Oil           75
           Recovery Rate

 31      Model System Effectiveness  as a Function        76
           of Average Oil Slick Thickness

 32      Model System Effectiveness  as a Function of     78
           Speed of Advance

 33      Empirical Factor as a Function of Model         77
           System Effectiveness

 34      Oscilloscope Trace  from Slick Thickness         88
           Measuring Device
                             Vlil

-------
                             TABLES


Number                                                    Page


  I       Nozzle Characteristics                           38

  II      Simulator Test Data                              41-42

  III     Small Model Interior Chamber Effectiveness       53
            Data

  IV      Small Model System Effectiveness Data            58

  V       Concept Skimming Performance Test Results        71-72

-------
                               SECTION I

                              CONCLUSIONS
1.  The hydraulic skimmer system concept is a feasible approach to the
    mechanical removal of floating oil from the open sea water surface.
    The test program has proven the concept to have sound technical and
    practical bases.  Advancement of the concept to a full scale
    operating oil pollution abatement tool has a high probability
    of success.  A modular system involving linear floating spray
    headers, floating chambers to concentrate and collect the oil,
    plus deck mounted equipment—pumps, tanks, winches, etc.—is con-
    ceived to be easily and quickly transported to an oil spill cleanup
    deployment site and then fitted to an existing workboat, barge or
    tug.

2.  System performance appears to be superior to known methods of oil
    slick recovery in open sea environments.   Conclusions are based on
    experimental results from a half system model approximately 40 per-
    cent of full scale (23 1/2 foot sweep width) operating on several
    types of oil slicks on fresh water.

    (a)  Model performance varied with speed of advance with all
         types of oil used in the experimental program.  The follow-
         ing values represent maximum performance achieved:  2.0 knots—
         97 percent effectiveness and 6800 gph recovery rate; 3.0 knots—
         55 percent effectiveness and 8700 gph recovery rate, and; 4.0
         knots—24 percent effectiveness and 3000 gph recovery rate.

    (b)  Model performance was affected by the characteristics of the
         oil being recovered.  Performance on light oils ranged to
         80 percent effectiveness and 6800 gph recovery rate for diesel
         fuel (No. 2 fuel oil) and 100 percent and 8700 gph for crude
         oil (40-42° API).

         Operation was less efficient for heavy oil recovery—30 per-
         cent effectiveness and 1870 gph recovery rate for Bunker
         fuel (No. 5 fuel oil).  Additional evaluation with heavy
         oils and with system and component modifications would be
         expected to show significant performance improvement.  Time
         constraints precluded such efforts.

    (c)  No model system performance degradation was detected with
         operation in waves up to 30 inches significant wave height
         (short "chop" partially suppressed).  The ability of system
         components to follow wave action indicated that longer and higher
         waves would cause no detrimental effect.

3.  An empirical factor involving system variables (spray flow rate,
    spray pressure and speed of advance) and oil characteristics  (oil
    thickness and oil type—No. 2 fuel, No, 5 fuel, 40-42° API  (crude)
    became evident in data reduction.  This factor^-sflow times the  square
    root of pressure divided by the product of oil thickness and

-------
     and  square of  the rate of system advance—correlated with
     effectiveness  and oil recovery rate for the oil types used.
     Extrapolation  indicated system requirements for various levels
     of performance for oil thicknesses for 1 to 3 mm.  For No. 5 fuel,
     the  extrapolation was necessary over a range in system variables
     not  evaluated  by model test, i.e., for an effectiveness greater
     than 30 percent.

4.   Measurement of system effectiveness for oil slick recovery techniques
     is difficult especially in the presence of wind and waves.  Con-
     tinuous measurement of oil thickness was mandatory for such per-
     formance evaluations and was used in this model test effort.
     The  ultrasonic method employed, although not optimized in this
     program, represents a feasible method for instantaneous measure-
     ment of oil layers down to 0.025 mm (.001 inch) thickness.

5.   Recovered materials (oil, water and emulsion) are of adequate
     quality to eliminate the need for sophisticated treatment equip-
     ment, such as  centrifuges, in the onboard process flow system.  No. 2
     and  crude oil  recovered contained from 11 to 46 percent free water
    by volume.  The oil-water mixture remaining  when allowed to separate,
     contained 3 to 70 percent water, 8 to 94 percent stable emulsion, and
     3 to 25 percent free oil.  No.  5 fuel oil (Bunker fuel) formed no
     emulsion with water, although it did contain 8 to 43 percent air
    by volume.  An initial coalescence, in order to remove the free
    water,  plus pumping direct to tankage for eventual landside pro-
    cessing and disposal  is  concluded to be an effective means of
    handling recovered oil.

6.  Behavior of the model system under dynamic conditions presented no
    direct  difficulties.   Tests  at  speeds of advance to seven knots
    without waves and to nearly  three knots with waves produced no
    instabilities or detrimental effects on performance of the system.
    Evaluation of the high speed mode (boom angled to 10° of the vessel
    centerline)  at up to 3.75 knots proved not to be structurally
    detrimental to the system.   The effects of high intensity winds
    were  not evaluated by actual test.   However, their effect cannot
    be of the order of magnitude of the water currents and waves sus-
    tained  during testing on  the low profile linear floating spray boom
    and the low draft collection chamber.   Maneuverability analyses
    of a  prototype oil recovery  system mounted on a support vessel show
    that  low speed turns  should  be  approached with caution.  The
    liklihood of  overrunning  one of the side mounted boom assemblies
    can be  decreased by  operation of the spray system.

-------
                           SECTION II

                         EECOMMENDATIONS
1.  It is strongly recommended that the concept be carried to the
    next step of development—prototype design and development.   The
    following specific activities are recommended:

    (a)  Secure additional system requirement design data to ascertain
         system performance improvements expected with minor system
         changes and with heavy oils.  Perform additional model  testing.

    (b)  Develop a feedback device for automatic control of the  suction
         head level, with respect to the oil-water surface, within the
         collection chamber.  Model testing is recommended.

    (c)  Review vessel characteristics—length, deck space and dimensions,
         horsepower rating, bulwark height above the deck and overall
         freeboard—as bases for the design of components which  must be
         compatible with workboats, barges and tugs available in major
         West Coast, Gulf Coast or East Coast ports as possible  support
         vessels.  A vessel inventory is recommended.

    (d)  Design a full scale system such that it can be palletized and
         air transported and then easily fit to existing available
         vessels.

2.  Advance the concept to a state of readiness for future oil pollution
    incidents by performing the following recommended activities:

    (a)  Fabricate all required components of a total system prototype
         for use in open sea test evaluation and for eventual use on
         offshore spills.

    (b)  Plan and perform open sea tests to confirm the prototype system
         performance, operability and seaworthiness.

    (c)  After minor system modifications, anticipated as a result of
         open sea testing, plan and perform a demonstration for  all
         interested parties—federal representatives, oil company
         industry conservation coordinators and major port authority
         representatives.  The presentation should include first-hand
         observation of the system in operation, plus a plan of  action
         necessary for further use of the system in actual spill
         incidents.  The plan should be designed by the coordinated
         efforts of government and industry.

    (d)  Prepare a complete set of installation, maintenance and
         operating procedures of such simplicity and completeness

-------
     that competence to operate the system is possible within a
     very few hours by relatively inexperienced personnel.

Further develop the oil slick measurement device which was used
in this program.  Potential applications include industrial process
control, enforcement of regulations regarding floating pollutants,
monitoring of oil spill cleanup progress, as well as evaluation of
oil spill abatement equipment.

-------
                           SECTION III

                    INTRODUCTION AND SUMMARY
          The Federal Water Pollution Control Administration desires to
develop new and efficient devices and techniques for removing spilled
oil from the water surface in both protected and unprotected waters.
This study was directed toward the development and evaluation of a con-
cept for the recovery of oil from unprotected waters as defined in
Request for Proposal (RFP) No. WA 70-23, "Recovery of Floating Oil".

CONCEPT DESCRIPTION

          The concept evaluated was based on the employment of generally
available vessels (workboats, barges or tugboats) of a class compatible
with open sea waters and capable of sustaining the speeds and environmen-
tal conditions to be encountered in oil recovery in such unprotected
areas.

          The concept features a general "Vee-shape" arrangement formed by
linear spray headers on each side of a support vessel.  The concentra-
ting effect of the advancing "Vee" is obtained by water spray which
induces surface currents toward a collection point on each side of the
vessel.

          The so called "spray boom" is an array of spray nozzles pro-
viding a continuous spray impingement pattern on the water surface.
When the nozzles are arranged in a linear pattern and the spray is
directed generally perpendicular to the direction of travel, sweeping
of floating oil toward a collection point is affected.  A continuous
floating manifold provides pressurized water to the nozzles.  The con-
ceptual boom is a low profile floating assembly with relatively short
flat members pivoted at each connection.  Attachment is to the support
vessel by mechanical pivot and flexible hose connections; the spray boom
orientation is maintained by cable rigging to the bow.  A possible pro-
totype arrangement derived from this concept for three classes of off-
shore .support vessel is shown on Figure 1.

          The "collection chamber" forms another major component of the
concept system.  It consists of a deep, narrow and long hull shape with
an entry for floating oil at the side near the rear of the chamber.  The
spray system is required to direct the oil into this side entry.  Vertical
baffles and an open bottom, plus the use of internal spray nozzles pro-
vides for moving oil from the side entry location to the front stagna-
tion area.  A floating suction head provides for recovery of the thickened
oil.  A collection chamber is attached to the vessel at each side, toward
the rear of the vessel.

          Finally, the concept must provide equipment which processes
the material recovered from each collection chamber into a high quality

-------
 oil for storage and subsequent disposal.  The oil recovery  process  employs
 oil recovery pumping units, oily water separation units  and tankage
 filling units.  The process flow chart in Figure 2 indicates the  inter-
 relationship of these units.  One particular characteristic is  that all
 water from the separation unit may be recycled through the  spray  boom
 and ahead of the skimmer.  This approach makes it possible  to reprocess
 any oil which may be lost in the water discharge of the  separator.

           Oily water pumped from the oil-water separator may be dumped
 to waste (land fill or incineration) or processed for bunkering or
 refining.   The use of landside polishing process units may  be required.
 Space and weight limitations on board typical support vessels will
 likely preclude carrying such final process  equipment.

           Storage and transporting of recovered oil to a landside base
 is assumed to be by moderately sized pillow  tanks.

 EVALUATION PROGRAM DESCRIPTION

           Major concept components,  the spray boom and the  collection
 chamber were  developed separately by experimental model  work.   All
 components and support facilities were then  fabricated for  evaluation
 of the total  system.

           Development of the spray boom entailed analysis,  experimental
 component  design,  and evaluation.  Analysis  revealed  the basic  mechanisms
 involved in the induced motion of oil slicks  by water spray.  System
 parameters were determined  from basic assumptions and were  used as
 initial  design criteria.

           A simulated spray boom,  consisting  of a mobile spray  nozzle
 manifold supplied  with high pressure water, was  cantilevered over the
 side of  a  large water basin.   It was used  in  determining nozzle char-
 acteristics which  most efficiently provide oil slick  movement.  This
 simulator, mounted on a truck which  could be  driven parallel to the
 basin  wall, also provided reliable information on the most  promising
 spray  flows,  pressures,  nozzle heights,  and orientations.   Additional
 information was drawn from  observation of a similar device  undergoing
 open sea evaluation at Santa Barbara.   This device, as part of  the
 American Petroleum Institute-Federal Water Quality  Administration sup-
 ported Project  Sea Dragon employed a spray boom.   It  was supported  ,. „ „.
 above  the water surface  by  discretely spaced  catamaran hull floats.  ' '

          The  final design  was  accomplished and  the 40 foot spray boom
was fabricated  as  shown  in  Figure  3  which follows:

          Development  of  the  collection  chamber  was an iterative  pro-
cedure involving small models  testing  and analysis of the mechanisms
involved.  Major deficiencies were overcome and  a concept evaluation
model was designed by  scale-up  of  the  models  tested.  Figure 3  shows
the final design used  to  fabricate the nine foot  long chamber.

-------
                                                                                                                                                Figure 1




                                                                                                                                          Oil Recovery Concept
-- x
                                                                                                                                                      7-8

-------
Water
Water
                        Pressure
                         Pump
                                               Separator
                                                                         Pillow Tank
                                Sludge or
                                Trash Pump
                                   Support Vessel
                                 Figure 2
                            Process Flow Chart

-------
/"- — 1
-_-Lj
— ^
t



1
. i

r ^ >•
v-.:-^
^J>'i<<£ ;:"•.: :••
-------
        The support vessel,  spray pressure pumps,  process  pumps,
tankage, towing arrangements and interfaces were also  designed for
use in the evaluation testing phase as  shown on Figure 3.

CONCEPT EVALUATION

        Concept evaluation was performed in a large water  test basin
209 x 432 x 16 1/2 feet deep.  A flap-type wave generator  capable of
producing random three foot  choppy wave patterns on the entire sur-
face was used.  Support equipment included: electric power supplies, towing
harnesses, booms for restriction of oil spreading and  devices  to  meas-
ure waves, currents and oil  slick thickness.

        Initial experiments  were performed to identify minor modifi-
cations required for operability. Speeds of advance up to  seven knots
were achieved on calm waters and speeds to approximately five  knots
in waves to 36 inches significant wave  height—average height  of  the one-
third highest waves, crest to trough.

        Subsequent skimming  evaluation  experiments showed system
effectiveness—oil recovered versus oil encountered—to be up  to
100 percent for crude oil and diesel fuel at speeds of advance in the
2.5 to 3.5 knot range.  Recovery of Bunker fuel was less effective
with a maximum of only 30 percent.  Oil recovery rates for the 23 1/2 foot
sweep width of the model system were 7000 to 8000 gph  for light oils
and 1000 to 1500 gph for Bunker fuel.   Improved skimming performance
on heavy fuels is likely based on spray boom performance analysis.  At
the low temperatures during the heavy oil tests, 32 to 45°F, the high
viscosity Bunker fuel tended to be in the form of a cohesive mass which
sluggishly rolled beneath the water surface when being moved by the
spray system.  Its high density also caused it to be more readily sub-
merged by impingement of the water spray.  The sluggish motion combined
with the minimal buoyancy of the heavy oil contributed to this poor per-
formance.  It is probable that empirical adjustment of parameters
(speed of advance, water pressure, water flow) would significantly  improve
performance.  Also, locating the spray nozzles at or near the water
surface so that the maximum horizontal spray force can be transmitted
to the slick would improve performance.  However, time limitations  pre-
vented experimental verification.  It was concluded that there is no
inherent reason to prevent  the efficiency of heavy oil recovery from
approaching that demonstrated for light oils.

        The oil-to-water ratios  in recovered materials varied consid-
erably.  Some high oil-to-water  ratios were obtained,  however.  This
suggests  that adjustment of  the  system parameters and operation on  a
continuous basis will result  in  good recovered  oil quality.   Formation
of emulsions  did take place, but only on  the liyht oil products.  Proper
design  of pumping systems  and  continuous,  once through, recovery does
not appear to present problems  of a significant magnitude, at least for
those oils used in this development project.
                                 13

-------
                           SECTION IV

                     SPRAY BOOM DEVELOPMENT
          Development activities entailed an initial analysis,  followed
by model evaluation.  A large scale device was then designed for overall
concept demonstration.

SPRAY BOOM ANALYSIS

          In the concept under investigation, the floating spray boom has
two functions.  First, it produces a water spray impingement on the water
surface.  Second, it is the buoyant support of a spray header.   The first
function produces the effect of moving an oil slick in a controlled direc-
tion and rate; the latter is an assembly arranged in a configuration to
support the header and to produce this effect.  The two functions are
separately treated as; (a) spray effectiveness and, (b) boom stability
and integrity.

          SPRAY EFFECTIVENESS

          The movement of surface oil slicks has been characterized for
many wind and current situations and for several spreading regimes from
instantaneous spillages.(4,5,6)  However, these empirical formulations
are not easily adapted to describe induced movement of oil slicks by high
pressure water sprays.

          Consider an oil slick retained from following the underlying
current by the use of high pressure spray.  The spray reaction force
arrests the surface flow of oil and thus contains the oil slick.  This
is essentially the mechanism which produces the effect of oil skimming—
a spray manifold is angled into the direction of travel thus imparting a
side motion to the surface in such a way that oil travels down the boom
length and into a collection device.

          The oil slick is assumed to be a flat plate slipping across the
underlying water column.  The underlying current may be equated to advan-
cing the water spray at a rate equal to the hypothetical current.  An
effective mass equal to the mass of oil is assumed.  Vertical movements
of oil relative to the adjacent water would be effected by viscosity
(drag).  However for this analysis such an effect is considered as
second order and is neglected.

          Figure 4 illustrates the water spray skimming mechanism.

          The force balance which causes the skimming effects can be
obtained by considering that, in the horizontal direction, the drag
force on a stationary oil lens due to an underlying water current  (or
advancement of the spray) plus a force which resists the acceleration of
an effective mass of material must be balanced by a horizontal component
of the spray reaction force.  This is also expressed by equation  (1),
following.  In the vertical direction, as a first approximation, assume

                                 15

-------
              F, (buoyancy)
                             F  (spray)
          ^
(drag + inertia)
                                               ty
                                  (spray to     I
                                  surface angle) L-    ^
                                               v^^       3
                    I
F  + F.
      ly
                   w
                             j
                      (weight + inertia)

          Force Balance Diagram
                                          >• Coordinate
                                            system  fixed
                                            to spray
                                            nozzle
                       direction of spray advance
                                        water spray from
                                        nozzle
               water current.
        (relative  to  spray  nozzle)
                Figure 4

Illustration and Force Balance Diagram of
   the Spray Induced Skimming Mechanism
                   16

-------
that the vertical component of /the spray  reaction  force must be balanced

by the relative oil buoyancy in  water  and by  a  vertical acceleration of

the effective mass of material.   This  statement is expressed as equation

(2).



     Fd + Fix - FsCos(V
     Fb - Fw - Fiy = FsSin(ei}                                     (2)


     where,


     F, = drag force between oil and water considering  the  slick

          to be a rigid flat plate moving across  the water  surface.



        = C. Area D (V  2)/2
           f       w  ow


      F.  = Force to accelerate an effective mass in the  horizontal

            direction



          = D T Area a /g
             o        x


      F   = Water spray reaction force generated  through  the  concept

            spray boom
      F,   = Buoyancy due to water displaced by the oil  slick

             i

              D T Area,  with no spray


              D T Area,  when oil is submerged
             • w


      F   = Weight of oil slick
       w


          = D T Area
             o


      F.  = Force to accelerate an effective mass in vertical direction



          = D T(Area)a /g
             o        y


      6..  = Angle of water spray impingement on the water surface
      C   = Drag coefficient


      Area= Spray impingement  area per foot of boom length



      T   = Oil slick thickness


      V   = Relative oil to water velocity
       ow

      0   = Spray flow rate per foot of boom length


      H   = Spray pressure  at the nozzles




                                 17

-------
     g   = Acceleration of gravity
     D   = Density of water
     D   = Density of oil
     a   = Acceleration of oil in horizontal direction
      x
     a   = Acceleration of oil in vertical direction
      y
         The solution to force balance equation (1), the summation of
forces in the horizontal direction (x-direction) will imply the water
spray flows required to horizontally move an oil slick.  Inserting the
force relations into equation (1) and algebraically manipulating the
result produces the following relationship:
     a   - ID /(D T)
      x    [ w x o '
                                                                  (3)
          Also now, assume that in order to retain the oil ahead of the
spray boom, the average acceleration  a  Avg. must be such that the
relative oil to water velocity is reduced to zero by the time it tra-
verses the first half of the spray impingement area.  Average acceleration
over a given distance equals the relative velocity change squared divi-
ded by twice the distance, or ex (Avg.) = V  ^/2x.  In this case, x would
be half the spray impingement length or L.  With this assumption and
after rearranging, note the following equation.

       „ 0.5   [Area V  2    1  f D T    C.'
       H     =|       ow      II.  o   + __f
                                 D L    2
                                  w

          Equation (4) identifies the important variables and to what
extent they affect performance as follows:

     •  Increases in drag coefficient, possibly caused by eddy forma-
        tion at the oil/water interface produces an added flow and
        pressure requirement.  Poor nozzle types could have this effect.

     •  Increases in oil slick thickness requires that a greater mass
        be accelerated to retain the oil lens,  thus increased flow
        and pressure.

     •  Decreases in the effective spray impingement length (a function
        of the spray pattern and height above the water surface) requires
        greater acceleration of the oil as the spray advances.

     •  Increases in the relative oil-to-water velocity, V  , have
        important effects on the spray flow and pressure requirements.
        High rates of  advance are required to obtain a high oil recovery
        rate.   A trade off here is necessary—minimum speed for adequate
        recovery rates,  maximum speed for good performance with a given
        spray flow and pressure.
                                18

-------
     •  Changes in spray angle (9,)  near zero degrees  for  this
        application have practically no effect on spray requirements.

     •  The relative oil density (D /D ) varies over a range  of  about
        20 percent—suggesting that oil type has minimal effect  on
        spray flow and pressure requirements.

     •  Required hydraulic force for effective oil slick movement is
        proportional to the spray flow rate and to the square root
        of spray pressure.  Therefore, arrangements which  operate at
        low pressure and high flow will require the least  pumping
        power for a given level of performance.

          The previous analysis may be expanded to the case of a spray
boom angled from the direction of the current (or spray advance)  by an
angle (9~) and with spray nozzles angled from this boom by (9_).  Figure
5 illustrates this side-skimming configuration.

          The vector summation of the velocities shown in  Figure 5
results in the following:

     V     = V   + V                                              (5)
      ov      ow    wv

     where,
     V     =  oil current vector
      ov
     V     =  water current vector
      wv
     V     =  relative oil to water current vector (same direction
              as the spray forces)

          The critical value of V   is that velocity which results in
a vector of oil current in essentially the direction of the spray boom,
or 9- relative to the vessel.  This is the situation depicted by the
velocity vector diagram of Figure 5.  If the water current (speed of
advance) were increased and the spray force were reduced,  the oil current
vector would have a reduced angle, resulting in oil escapement along
the length of the boom.  Conversely, the reverse situation would result
in moving the oil well ahead of the spray, thus wasting pumping power.

          Equation (5) when broken into sealer equations and when the
critical value of V   is assumed, results in the following equation:
     V     = V   Sin(90)/Sin(9_)                                   (6)
      ow      wv      2.       3
This expresses the relationship between the vessel speed (V  ), the
oil to water relative velocity (V  ) and the spray boom configuration
(9«, 9_).  From equation (6) it can be determined that to be the most
effective, the spray nozzles should be perpendicular to the boom
(9^ = 90e).  This case enables the greatest value of vessel speed  for
a given spray flow and pressure and level of effectiveness as in equation
(7).

                                 19

-------
              ow
                                                 Coordinate system fixed
                                                 to spray boom (or the
                                                 support vessel)
                     Vector Velocity Diagram
Support Vess 2
                                                       Spray Boom
                            Figure 5

            Illustration and Vector Velocity Diagram
                of  the Side-Skimming Arrangement
                                20

-------
     V     = V   Sin(e_)                                           (7)
      ow      wv      2

          By substituting equation (7)  into equation (4) ,  derived for
the simplified case of a current perpendicular to a spray  boom,  a
relationship useful to the side-skimming technique is obtained:
         0.5   rAreaveT   Cl                      (8)
        W        0 .0.5.  ,- ,      ID L
                (2g)    Cos (6.^        w

          Implications derived from equation (8)  are:

     •  Reduced angles of boom to vessel require lower spray pressure
        and flow.

     •  Reduced vessel speeds significantly reduces required spray pres-
        sure and flow.

          The effect of angling the skimming mechanism deserves additional
attention.  Equation (7) is graphically expressed in Figure 6 to show
the effect of this variable on the vessel speed for given oil to water
relative velocities.   (Also implied by a given oil to water relative
velocity, is a given spray flow and pressure requirement — see equation
(4)).

          The sweep width is also directly related to the boom to vessel
angle for a given boom length.   This would control the system capacity,
all other things being equal.

          A simple derivation reveals the relationship of sweep width
to system variables as shown in equation (9).
                wv
 where :
     W      = sweep width

     Q      = oil recovery rate

     V      = vessel to water velocity
      wv
     T      = oil thickness

          This equation is shown graphically in Figure 7, using the
design goals of 50,000 gph oil recovery rate at an oil thickness of
1.5 mm and less.

          Decreasing the spray boom angulation (99) , to achieve improved
skimming performance is constrained by the sweep widths required to
achieve the recovery rate goal and by the maximum boom lengths which
are practical for available support vessels.  These vessels can be
expected to be normally about 100 feet in length and 30 feet in beam.
                                21

-------
           10
Vessel Speed

   V
    wv

  (knots)
            0
                                               V   =10 ft/sec
                                                ow
Vow = 5 ft/sec
                         .Sin(e2)
               0                            45



                          Boom to Vessel Angle (9 )- (degrees)


                                  Figure 6


        Speed  of Advance  as a Function  of Boom Configuration and Relative

                          Oil to  Water  Velocity


                                     22
                                      I

                                     90

-------
            250
 CO
*tJ

 (D
 (D
 O.
o.


i
O
n>
(D
.0
o
*•
o
o
o

00
tt>
o
o

to

3
       200



Beam of



Sweep  -  W



 (ft)  150
    H
           100 __
        50
                                                                                      t  = 0.3
                                                                                              nun
                                                 Vessel Speed -
                                                             V    (knots)
                                                              wv

-------
 If the maximum practical boom length  is  taken  to  equal  the vessel
 length (to enable proper guy cable support), then the spray boom
 angulation for oil collection from slicks  1.5  mm  and less in  thickness,
 would be limited to about 45°.

           Equation (4)  and (8)  can now be  solved  to obtain approximate
 spray flow and pressure requirements  as  a  function of oil to  water
 relative velocity or vessel  speed.  Equation (10)  below expresses the
 result of these equations using the following  assumptions—(see
 Figure 8 for a graphical expression of equation (10):

      C.   =  Turbulent  friction factor

           -  0.74Re-°-2

      Re   =  Reynolds Number

           =  D V  L/y
               w ow

      D    =  1.985 slugs/ft3

      V    =5.6 ft/sec for  Reynold Number Determinations
       ow                      J
      L    =  10 ft

      y    =  sea water  viscosity

           =  2.5 x 10~5 Ib sec/ft2
                   2
    Area  =  10 ft /ft  of boom

      DQ    =  1.786  slugs/ft3

      T    =  1.5  mm (0.06 inches)

      g    =  32.2 ft/sec2

      QI    =  10°

      02    =  45°


      Qw(gpm/ft)Hw°-5(Psi) =  67.6 V^Cft/sec)

      °r           0  5                         2
      Q  (gpm/ft)H     (psi) =  78.5 Vessel  speed  (kts)

          The  solution  to force  balance  equation  (2), the summation
of vertical forces, suggest  problems with  oil  loss resulting  from
oil slick  submergence and subsequent overruning.   Since this represents
a possible  failure mode,  it  was  examined using the same general nom-
enclature given at  the  introduction of this section.  By inserting the
force relations previously developed into  equation (2)  and algebrically
manipulating  this result, the following  equation  results:
                                24

-------
         500
    Nozzle
Flow per foot
Floi
of boom
         400
(gpm/ft)
         300
         200
         100 _
                                        ow = 5.8 ft/sec
                                        Vessel Speed = 5 knots
                               4.6 ft/sec
                                4 knots
                           3,5 ft/sec
                               3 knots
                             2.3 ft/sec
                                  2 knots
                            1.16 ft/sec
                                       not
                         0.58 ft/sec - 0.5 knot
                          25
                                                                        125
                                      H  , Nozzle Pressure (psi)
                                  Figure 8

           Nozzle Flow and Pressure Requirements as a Function of Speed
                                  of Advance
                                      25

-------
= (D /D ) lo (2gH)°'5Sin(e )/(gT Area)  - (D -D )/D 1     (11)
    W  O  I  W    W         -L                W  O   W I
      a
      y
where the oil  is submerged and being accelerated downward  (a  ^ 0).

          Overruning an oil slick will occur when the net  oil buoyancy
 (total buoyancy - weight) is overcome by spray forces and  when sufficient
excess force is present to accelerate the slick to a depth where nozzle
sprays are not able to accelerate it horizontally.

          As the density of the oil approaches that of water, several
limitations on system variables are indicated.  A statement of this
is given in the following equation where the vertical ccceleration is
assumed to be  zero, but where the oil layer is just stable—any
increased downward force would produce oil submergence.

     (D -D )/D  = Q (2gH )°'5Sin(e )/(g T Area)                     (12)
      W  O   W    W    W         X

          Using this failure criteria, conditions which may enhance oil
submergence are indicated as possible system limits.

      •  As the relative oil to water density approaches zero, system
        variables become quite limited where either the flow and pressure
        must be reduced (this implies slower vessel speeds and thus
        lower recovery rates—assuming equal slick thicknesses) or
        the spray impingement angle must be reduced to approach zero.

      •  The effect of variations in impingement angle, for snail impinge-
        ment angles, are quite significant—approximately  proportional
        to the absolute angle change—a change in angle of from 5  to 10°
        would roughly double the downward spray reaction force component.

      •  Decreases in oil slick thickness decrease the buoyancy force
        and thus cause a greater susceptibility to submergence of an
        oil slick by a given spray condition.

     •  Decreases in the spray impingement area,  such as would be  the
        case if shallow angle nozzles were used or when nozzle height
        and impingement angles produce reduced areas, cause an increasing
        tendency to submerge the oil.  Wide pattern nozzles may be the
        preferred type.

          The implication of a speed-of-advance limitation as the
relative oil-to-waCer density is reduced is critical and must be further
treated.   By combining the submergence criteria from equation (12) and
the skimming criteria from either equation (4) or (8), eliminating the
common spray term,  0 (2gH )'     , and using the same assumed variables
(but letting 9.. vary),  the following equation results:


     D /D  = (1-0.009 V  2tan 9.. )/(1+0.0035 V  2tan 9,) , or        (13)
      O   W             OW      -L             OW      1
                        26

-------
     D /D  = (1-0.012 Vessel speed2 tan 6 )/(1+0.047  Vessel  speed2
          tan 91)                                                  (13)

          This equation is shown in Figure 9  for given  spray impinge-
ment angles, thus  relating the maximum vessel speeds  to a given  oil
type (density).

          BOOM STABILITY AND INTEGRITY

          Structural support of a spray header which  produces the correct
spray effectiveness is required.  Continuous  rigid floating  sections
with discretely flexible pivots is the support concept.  Floats  are
suspended as outriggers for additional roll stability.   The  total boom
length is supported at an appropriate angle to the vessel by cables  from
each boom section  to the support vessel bow.   At the  aft extreme of  the
boom, attachment to the vessel is made with a free pivot and a lever
arrangement to allow for relative vessel to boom surges.  The flexible
pivots between boom sections are free to rotate only  in one  plane and  not
in the other two.   Twist from one section to the next is constrinaed.
Boom sections are  also prevented from forming other than the desired
straight length.  Figure 10 shows the configuration derived  from these
constraints.

          The several possible hydrodynamic instabilities are considered in
the following paragraphs:

          Instability No. 1 (roll in the plane of the end view of the
boom).  Two stable conditions (2,4) and two meta-stable conditions (1,3)
are possible in considering this roll possibility as  noted below in
generalized cross  section views.
      Condition 1        Condition 2       Condition 3       Condition 4

Condition 4 is the desired condition, applicable during skimming.  Con-
dition 3 is possible if the boom front becomes submerged during forward
speeds and all counteracting moments (cable force moments, spray force
moments and roll stabilizer weight moment) are insufficient to prevent
the motion.

          Figure 11 and the following discussion indicate how this
condition is prevented.  The movement due to the spray reaction force is
a function of the nozzle flow and pressure characteristic and the moment
arm, height from boom center to nozzle.  The moment arm is quite short,
thus producing a relatively small moment to counteract roll.  The roll
stabilizer moment is a constant once the floating portions becomes
raised out of the water.  For this condition it would be the stabilizer
weight times the moment arm from center of roll of the boom to center of
                                27

-------
     LOOT
•H
CO
C
Q)
»    0.3 _L

•H
o
8>    0.2 4-
•H
      0.0
           in0 APT
     0.9Q.25.70 API
     0.804-45.3° API


     0.7 170.6° API


     0.6


     0.5


     0.4
                           9  =  10
                           9n =  20C
                                 Skimming must  be performed  a
                                    system  variables  consiste
                                    with values below those  s
                                    by  the  curves.
                                                               5°.9729
                                                              10°.9457
                                                              20°.8898
                                                  it
                                                  lown
0.0

0.0
2.0

1.72
                                4.0 V
                                     ow
                                3.44
                        Vessel Speed (kts)
                           Maximum Speed
6.0(£t/sec)
5.16
8.0
6.88
10.0
 8.6
                              Figure 9

           Maximum Skimming Speed as a Function of the
         Relative Oil Density for Given Spray Impingement
                     Angles of 5, 10 and 20°
                                 28

-------
        Forward
        Direction
on/
 Support
 Vesse
                                                 \      \
                                          No  Angulation along
                                               the Length
           Bcom-to-
           Vessel Pivot
Support
Vessel
                              PLAN VIEW
                            ELEVATION VIEW
                                         Free Rotation  to
                                              Follow Waves
                                                            Water Surface
                   Water Surface
                                              Twist Between
                                           Sections
                               END VIEW
                              Figure 10

                  Spray Boom and Pivot Connections

                                  29

-------
 gravity  of  the  stabilizer  (quite long).  This moment is relatively large.
 A moment due  to the  cable  forces would be the tension force of the cable
 times  a  moment  arm which extends from the center of roll of the boom
 to the cable  attachment point and also multiplied by the Sin of the angle
 between  the direction of the cable force and the boom roll position (as
 seen in  the end view).  Thus, the cable moment increases with cable
 forces and  with roll angle.  The only positive moment which can produce
 roll is  due to  drag and wave forces on the boom.  The moment due to drag
 will be  a drag  force resultant through the center of drag on the boom,
 the moment  arm  being the distance from the center of drag to the center
 of roll,  multiplied by the Sin of an appropriate angle (boom roll position
 to the equivalent direction of the drag force).

          By  a  summation of forces, it is found that the cable forces
 must be  equal to the spray reaction force plus the drag force*, therefore
 cable  forces  must necessarily exceed drag force.  Also, the cable
 moment arm  will be longer  than a moment arm due to drag forces.  Logically,
 therefore,  a  moment due to drag cannot exceed the counteracting cable
 moment and  no exaggerated  rolling can be produced when the cables are
 intact and  properly adjusted.  The effect of waves produces a similar
 result with the exception  of a rolling inertia which will allow small
 momentary roll  deviations  from the equilibrium.  Figure 11 graphically
 expresses this  inherent roll stability.

          Condition 2 or the upside-down position is only possible if
 condition 1 or  3 has been  encountered first.  Although condition 2 is
 quite  stable, considering  that condition 3 is unlikely and condition 1
 is  quite  doubtful (no significant moments would be present to cause this
 condition), the inverted position will not likely occur in even quite
 harsh  wave  conditions.

          Catastrophic boom roll is considered to be quite unlikely from
 the preceeding  analysis and only small roll angles due to wave effects
 are expected.

          Instability No.  2 (rotation of the boom with respect to the
 vessel):   Two conditions are possible, either the boom is angled and
 constrained from increased rotation relative to the vessel or forces
 cause  the boom  to collapse against the side of the vessel.  Condition 1
would be the  skimming mode of operation characteristic of approximately
 45° angle to  the vessel and taunt cables.  Condition 2 is a collapsed
boom angle with loose ineffective cables.
          Condition 1                    Condition 2
The cable forces which cause the condition 1 stable configuration are due
to spray forces and drag forces.  The spray forces are a function of flow
and pressure characteristics of the nozzles only, while the force on
the boom due to drag is a function of speed -of advance and the boom-
to-vessel angle (advancing in a straight line) or boom-to-water current

                                30

-------
                                      Center of Roll
F cable = F drag + F spray
    F cable  > F drag

Cable Moment = F cable  X
                         c
Drag Moment = F drag X
  X   > X.            d
   c     d

Therefore Cable Moment > Drag Moment
                                                    F  cable
                                                   F  spray
            Figure 11

       Spray Boom Moments

               31

-------
 angle (when turning).   As  the boom-to-vessel  angle is decreased, the
 drag term is likewise  decreased.   With  significant water spray reaction
 forces,  it would be impossible to  obtain  a  condition 2 situation while
 advancing in a straight line.

           In the case  of a turning maneuver,  the support vessel will
 pivot about a point 1/4 the vessel length from  the bow.  The vessel will
 also have a drift angle, (angle of vessel center line to the direction of
 travel),  similar to a  slip angle for an automotive tire during a turn.
 The stern will drift out on the turn.   The  following Figure 12 will
 illustrate the turning maneuver.

           When spray booms are attached to  each side of a vessel, the
 boom on  the inside of  the  turn will be  pulled sideways by the vessel at
 an angle  relative to the stationary water column.  This angle is equal
 to the vessel-to-boom  angle plus the drift  angle.  The outside boom will
 also be  forced at an angle to  the  underlying  water column.  This angle
 however  is equal to the boom  to vessel  angle  minus the drift angle.  From
 these observations,  the outside boom would  be in jeopardy if the drift angle
 were to  exceed the boom-to-vessel  angle (in the absence of spray forces).
 This could be considered a conservative limit for turning maneuvers when
 spray forces  are present.
          When a boom-to-vessel angle of, say, 10°, compared to 45° is used,
 such as during high speed transport between slicks, smaller drift angles
will produce the unwanted instability.  However, high speeds result in
 smaller drift angles thus indicating that high speeds can probably be
 sustained in most situations.  Low speed maneuvers are consistent with
 large drift angles, making hazardous situations possible.  Therefore,
 during low speed cases with small boom-to-vessel angles or normal skim-
ming angles with no spray present, careful and considered vessel handling
would be required.

          Instability No. 3 (rotation in elevation view plane due to
boom end forces).  Forces along the length of a jointed boom will tend
to cause it to collapse.  The buoyancy and weight distribution of the
boom sections will counteract this tendency.  As the water level
raises and lowers on the boom, the weight and buoyancy produce restoring
forces.
ss ^<^^
fc *s.
* ^
^^^^ n —
\s
I^UIIU_LI_J-U11 X
Condition 2
                                                   Condition 3
Condition 3 will be the equilibrium position in calm waters.  Condition 1,
though exaggerated, is the correct performance boom shape in a wave
situation.  Condition 2 would result only from the endmost boom section
diving due to a combination of hydrodynamic conditions (waves and current).
                                32

-------
             Direction of Travel
                                         Right Turn
                                   Propulsion or Rudder Direction
                           Drift Angle
                    Figure 12




Skimming Configuration During a Turning Maneuver




                        33

-------
           The boom  sections  are  considered to be approximately half
 submerged  during an equilibrium  condition.  Total buoyancy will be
 approximately twice the  total weight for a symmetrical body floating
 half  submerged.  Such  a  floating device would act as a spring—a given
 vertical force applied would cause a displacement which is proportional
 to  the  force.  The  maximum force possible would be either equal to the
 weight, if the device  is lifted  from the water, or equal to the net
 buoyancy (total buoyancy minus total weight) when the device is com-
 pletely submerged.   Consider that equal vertical forces are applied to
 each  end of a boom  but in opposite directions.  This produces a list
 or  angular deflection  and a  counter moment.  By applying a summation of
 moments equal to zero  (a stable  list) and by considering a half sub-
 merged  object, a list  angle  can  be determined for any value of this
 vertical force.  The following equation is evident:

      F  = I2w B tan  9                                              (14)
             24
     where:

      F  = vertical force
      1  = length of  object

     w  = width of object
     9  = list angle of object relative to the water surface
         (in length direction)

     B  = either the net buoyancy  or the total weight per unit volume

           Forces along the length of a jointed floating device will
 produce alternating vertical force components (force up at one joint
 and down at  the next).

           It can be seen that for a given buoyancy or weight per unit
 volume, stability is enhanced by  increases in width and length.  As
 the rotation angle is  increased,  as is the case for steeper and steeper
 waves,  the instability is worsened.

          The end force on the boom is due to several contributing
 factors; boom drag,  spray reaction forces and cable forces.  The spray
 force will tend to reduce the end forces and will dominate (producing
 tension along the length of  the device rather than compression forces)
 during very slow speeds.  The cable force will definitely reduce the
 end forces—but only on the boom sections at the aft portion of the system.
 Cables attached to the outward extreme boom connections may have a
 small component which  adds to the end force due to drag.  The location
 of the"front vessel, cable mounting and the length of the floating boom
will drastically change this effect.   The major difficulties which may
 arise will be from high speed travel, with low spray forces and will be
more likely at the extreme inboard joint.
                                34

-------
          Diving of the end boom is  possible and  should be  considered
in the final boom design.   Greater buoyancy  or  hydrodynandc fins may
be reasonable preventative approaches.

          The above instability characteristics imply  that  speed
of advance in both the skimming and  speed modes of  travel should be
restricted.  Trailing of the booms will therefore become necessary
above a characteristic speed and in  combination with waves  which produce
large rotation angles.

          WAVE AND SPRAY INTERACTIONS

          Waves will obviously affect both spray  effectiveness  and  the
boom motions and stability.  The nozzle action  on a choppy  sea  is
such that the spray impingement angles  vary as  a  wave  passes through
the spray impingement area.  The significance of  this  effect is
indicated on Figure 9, which shows maximum speed  of advance as  a function
of spray impingement angle which can be sustained without submergence
of an oil slick.  The dynamic action of a floating  boom in  waves,
producing transient changes in both  the relative  nozzle elevation above
the water and impingement angles due to rolling;  surging, is also of con-
sequence.  Boom sections should be designed to  closely follow the wave
profiles which they encounter.  Rolling of boom sections  from one to the
next would  be constrained; therefore the boom-to-water surface angle
will be at some average value over the entire boom  length.   Spray
nozzle distribution angles should be such as to allow  for considerable boom
roll and surge and still produce a wide spray impingement area. Longer
period (long wave length)  waves will have little effect here because of
the shallow slopes of such waves. For example, the whole boom assembly
can follow a five foot wave which has a 100 foot wave  length.  The
larger the wave, the easier it is for the boom-to follow it.  The short
period "chop" or fresh wind waves will cause the greatest  difficulties
from both the spray effectiveness and the boom wave following aspects.
Model testing should be performed in a "short chop" wave environment.

SPRAY BOOM DEVELOPMENT TESTS

          Experimental verification  and parameterization of spray nozzle
characteristics, flows, and pressures on oil slick  motion were necessary
for a firm design basis of the model system.  A spray  boom  "simulator"
was designed and fabricated for this work.  The spray  simulator as  it
was called, Figure 13, consisted of  a cantilever structure  supporting
a manifold on which various nozzles  were mounted.  This structure was
supported by a jib crane arrangement from a truck.   A pump  and tank
located on the truck provided pressurized water to  the spray manifold.

          The truck was driven along the test basin wall with the boom
overhanging the water adjacent to the wall during the  experiment.
Water spray from the manifold was directed at floating oil  slicks with
various combinations of nozzle types, pressures,  and flows.  A portion
                                 35

-------
  Spray Manifold
Spray Simulator
After Test Run
Before Test Run
                  Figure 13




Spray Simulator For Spray Nozzle Evaluation
                      36

-------
of the basin was enclosed by a floating boom and used for testing.
The simulator provided for nozzles at six inch centers or multiples  thereof
for a length of 20 feet.   The manifold could be positioned at  any
desired angle to the direction of travel and the nozzles could be
oriented at any desired angle to the water surface or to the manifold.
The effect of the water spray on the oil during each run was recorded
photographically and the geometrical and hydraulic parameters  for
each run were recorded.  Qualitative visual observations supplemented
recorded data.

          The search for a viable combination of test parameters
resulted in a system judged to be sufficiently effective for design  of
a scale model prototype.

          Table I is a summary of nozzles evaluated using the  spray
simulator.  Figure 14 shows the nomenclature for the various operating
components of the spray simulator.

          The distance of the nozzles above the water surface  was
about 30 inches when the simulator was in a level position. Irregu-
larities in the roadway along which it was driven allowed the  manifold
to vary from ^36 inches above the water surface to within M.2  inches
of the surface.  It was observed that effective combinations of nozzles
and angles worked inspite of these height variations.

          Basic conclusions drawn from this effect pertain to  spray  boom
effectiveness in moving an oil slick without overruning it. Empirical
trials were made for various A, B and C angles (see Figure 14  for
definitions).  Nozzle types were evaluated and flows and pressures were
varied.  Runs were made up to 3.0 knots speed of advance, 24  inch
significant wave height and different oil types (Texaco Crude  Oil 42-
44° API and diesel fuel oil).

          Conclusions from this phase of the work were as follows:

          •  Nozzle height had no apparent effect upon skimming efficiency
             within the range of 12 to 36 inches above the water
             surface.

          •  The optimum boom angles appeared to be:  A = 67 1/2°,  B =  45°
             and C = 10°.

          •  The most effective nozzle type for moving surface oil
             slicks of those tested was the 2HH30350 full jet  style
             (Spraying Systems Co., Bellwood, Illinois).

          •  Twenty gpm per foot of boom and 100 psi was required to move
             oil in 24 inch waves and at a 3.0 knot speed of advance.
                                37

-------
Full Jet Nozzles
Flat Jet Nozzles




          3/4 P35200*
                                   TABLE I




                            NOZZLE CHARACTERISTICS
Nozzle No.*
3/4 H4
1 H10
1 1/4 H14
2 HH 30350
Flow
12.4 gpm
31.4 gpm
41 . 4 gpm
50 gpm
at Pressure
80 psi
80 psi
80 psi
80 psi
Spray Angles
63°
71°
73°
30°
28   gpm
80 psi
Fabricated Nozzles (flattened pipe  couplings)




          3/4 A             50   gpm          80 psi
40°, 22'
                                 20°,
        A
        r*\
      Full Jet Nozzle
 Flat Jet Nozzle
      Fabricated  Nozzle
*  Spraying Systems Co.,  Bellwood,  Illinois
                                    36

-------
                                  Cantilever Structure
Spray Nozzle
Manifold
                                                       Direction of
                                                       Travel
                     Principle Spray Direction
 Water Surface
                         PLAN VIEW
                                        Principle Spray Direction
                                                               Direction
                                                               of Travel
                    END VIEW OF MANIFOLD
                          Figure 14

         Nomenclature of Spray Simulator Variables
                               39

-------
           With two exceptions,  the nozzles  tested were not effective
 at speeds above 2.8 knots  in calm water.  The  fabricated nozzles were
 more effective than most nozzle types.  However, they produced an
 irregular spray impingement  pattern and required large flows and
 pressures.  Most variation in performance was  attributed to the char-
 acteristic spray exit  angle  for the particular nozzle.  Flatjets are
 not appropriate because of their small spray impingement area.  Wide
 angle,  fulljets fail because of energy transfer inefficiencies inherent
 in the  wide angle (60°) spray impingement on the surface.  The model
 2HH30350  has only a 30° spray exist angle in a full cone distribution
 pattern,  thus providing the  required surface current without undue
 turbulence.

           The flow and pressure requirements determined from these
 experiments  are based  on a cantilivered boom and not a wave following
 (floating)  spray boom.  Also the visual observations and photographs
 of the  skimming operation  require subjective judgment for interpre-
 tation—thus,  resulting conclusions should be  considered qualitative.
 Detailed  test data are given in Table II.

 SPRAY BOOM DESIGN

           Spray boom design  was  performed as several tasks, each rep-
 resenting a  performance aspect:   hydraulic, hydrodynamic and structural.

           HYDRAULIC

           Spray boom simulator  experiments showed that approximately 20
 gpm/foot  of  boom and 100 psi was  sufficient to obtain 80-90% spray
 effectiveness  in two foot waves  and speeds to  3.0 knots.  These criteria
 were used in sizing  pumps  and piping for the model system.

          A  total  length of  40  feet in five 8-foot sections was chosen
 for the boom model.  The water  requirements of this boom are therefore
 800 gpm total  flow with a  100 psi  capability for pressure.

           Choice  of pipe diameter was made from hydraulic pressure drop
 data for  schedule  40 steel pipe.  For 800 gpm, a pressure drop of 4.43
 psi/100 foot length  (5" pipe) or  1.75 psi/100  ft. (6" pipe) can be
 expected.  The  choice was made to  use  two 4-inch pipes interconnected
 along the boom  length.   This is equivalent in  cross-sectioned area to
 one 5.65  inch pipe and results in about 1 psi loss within the spray
boom length.  This is a conservative estimate because of the reduction
 of flow along  the boom length.  Vertical uprights at 6-inch centers
 are provided for  the spray nozzle connections, thus allowing for easy
changes  iu nozzle spacing and other  piping.  Flexible hose connections
between boom sections provide for 20  to 30° angular deflection.

          A  diesel-driven  centrifugal pump, capable of 1400 gpm at
 120 psi C2400  rpm) , was used as  the water supply.  This pump was
 mounted on a 35-foot steel hull  lifeboat to simulate a support vessel.
                                 40

-------
      TABLE II
SIMULATOR TEST DATA
Test
No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

Water
Speed Pressure
(knots) (psig)
5.0
1.6
(no data)
2.5
1.5
1.5
3.0
3.0
1.5
3.0
(no data)
(no data)
(no data)
3.5
2.8
(no data)
2.8
40
40

40
40
40
40
30
30
30



80
38

50
Spray Oil Angle
Flow Type
(gpm)
270 Crude
270 "

140
140
240
240
335
335
335 "



700
316

285 "
Thickness Nozzle A
(mm)
.7 3/4 P35200 80°
" " " 80

.6 3/4 H4 90
1.2 " " 90
1H10 90
1.8 " 90
1 1/4 H14 90
II M II gQ
M i. •• 90



.7 3/4 A 90
.6 45

1.2 3/4 P35200 45
B
35°
0

45
45
45
45
45
45
60



45
45

45
C E
20°
20

20
20
20
20
20
20
20



10
10

10
If f ectiveness
(percent)
0
0

0
50
50
0
0
90
30



90
70-90

75

-------
     TABLE II (Continued)
SIMULATOR TEST DATA
Test
No.
18
19
20
21
22
23
24
25
26
27
Speed
(knots)
2.8
2.0
1.5
1.5
1.5
3.0
3.0
3.0
3.0
3.0
Water
Pressure
(psig)
100
100
100
100
100
40
118
80
80
118
Spray Oil
Flow Type
(gpm)
400 Crude
400 "
400 "
400 "
400
250 Diesel
500 "
350 "
350
500
Angle
Thickness Nozzle A
(mm)
1.0 3/4 P35200 45
n n it 45
.6 " " 45
.4 2 HH30350 67
.2 " " 67
.8 " " 67
67
67
n n n 67
67
B
45
45
45
45
45
45
45
45
45
45
C Effectiveness
(percent)
10°
10
10
10
10
10
10
10
10
10
85
90
80
85- 90
80- 90 with
2 ' waves
50- 80
80- 90
80
80
85

-------
Water was supplied to the pump suction from the basin via a bulkhead
fitting through, the boat hull.  Discharge was through twin flexible
hoses, 4" in diameter, to connections at the inboard end of the spray
boom assembly.

          HYDRODYNAMIC

          Current and wave forces on the boom sections could only be
estimated at this stage in the development effort.   For a determination
of expected maximum forces, the following was assumed:  a single floating
boom section 8 feet long, 36 inches wide, and 8 inches deep (rounded
ends), a wave characteristic of 3.0 feet (3.9 sec.)  superimposed on a
5 knot speed of advance, no wind effect, and a cable rigging which will
withstand the forces with little elasticity.  With these assumptions,
the maximum force due to waves was calculated to be  340 pounds and due
to the current, 510 pounds.  Superposition of these  conditions results
in a total maximum force of 850 pounds per boom section due to hydro-
dynamically induced forces.

          STRUCTURAL

          Constraints on boom motion and maximum forces are important
to the ultimate boom performance and boom interface  structures.  Each
boom section must be capable of withstanding the forces which are imposed
under the worse case encountered during its operation.  The spray
reaction force was found to be 80 pounds per boom section considering
the requirement of 20 gpm per foot of boom, 100 psi  spray pressure and
a nozzle efficiency of 75%.  This in combination with hydrodynamically
induced forces must be offset by cable forces and the boom to vessel
interface pivot.

          By making the assumption that cable forces are balanced (no
moments are encountered at the connection between boom sections), forces
as shown in Figure 15 result.

          Considerable compressive force (up to 3300) pounds between boom
sections is evident.  Cable forces range up to approximately 1000 pounds
tensio'n for the balanced force case; 2500 to 3000 pounds should be the
maximum requirement of anyone cable as a worst case  (unbalanced forces).
The front and rear connections to the support vessel will also require
considerable strength in resisting the maximum forces.

          Unbalancing of the cable tensions will result in bending
moments in the affected boom sections in addition to  those created by
the drag and spray reaction forces.  If the outermost cable becomes
loose, a bending force is applied at the connection between the outboard
boom sections.  Inserting the expected maximum forces into appropriate
equations produces a maximum moment in the boom section of 660 ft Ib
at the connection.  With balanced cable forces a maximum of only 335  ft Ib
is reached.  Considering the possible problems in rigging, 130 ft Ib
for bending and 4000 Ib compression in each boom section is considered
a practical design critera.

                                 43

-------
2870#
                       2940#
                     Figure 15

   Structural Connections Between Boom Sections
   and Between the Boom and the Support Vessel;
  Including Maximum Expected Force and Reactions
                          44

-------
          Other possible forces due to wave action must be considered.
A wave could possibly lift a central portion of a boom section
resulting in both ends being lifted from the water.  The weight of the
boom would then produce a bending moment.   The buoyancy would produce an
upward force at the center and the unsupported ends would produce a
downward weight.  Forty pounds of force per foot of boom would produce
a maximum moment of 160 ft Ib in this plane.

          Since the boom sections are constraiined from relative twist
between sections, twisting torque will be randomly applied to the boom
sections due to wave action.  For design purposes, a 60 pound weight
on one stabilizer at a distance of 8 ft from the center of boom roll
was assumed to give a maximum twisting torque.  The boom length which
would be affected must sustain a torque of 240 ft Ib.

          The boom sections were designed, fabricated and tested for
loads in excess of these expected maximum values.  Figure 16 is a
photograph of the final design arrangement used in concept evaluation
tests.  Vessel to boom structural connections were made by a lever
arrangement which compensates for relative vessel to boom surges.
                                 45

-------
            Figure 16




Final Design of Model Spray Boom
                46

-------
                           SECTION V

                 COLLECTION CHAMBER DEVELOPMENT
          Collection chamber development,  as in the case of the spray
booms, was performed by analytical treatment followed by model experi-
mentation.  Concept evaluation equipment was then designed for system
evaluation.

COLLECTION CHAMBER ANALYSIS

          Development of the collection chamber concept was based on the
environmental conditions in which it must function and the program
design goals.  Performance factors were classified as effectiveness,
stability, and wave following characteristics.

COLLECTION EFFECTIVENESS

          Recovery of floating oil which is "swept" by a hydraulic
spray boom is not simple.  Retrieving a small quantity of oil while
avoiding pumping a large quantity of water requires the oil to be con-
centrated in a stagnant area or stilling chamber.  Collection or reten-
tion chambers which have open fronts to admit incoming oil and water
and then retain only the oil are subject to the phenomenon of under-
flow and draining.  This phenomenon has been extensively studied in the
laboratory and attempts have been made to predict analytically the   ,  _.
underflow behavior and provide a physical explanation for the effect.  '
A side entrance oil collection chamber scheme was employed for concept
development to circumvent the difficulty of underflow.  The rationale
is basically to minimize the quantity of water which must flow into and
out of the chamber.  For example, in calm water, the only flow of fluid
into the chamber is that induced by the spray nozzles.  In wave motion,
however, there will be a variable flow of water into and out of the
chamber as it heaves with the waves.  Figure 17 illustrates the collec-
tion chamber concept and it is further explained as follows:

     •  The concentrated oil reaching the inboard boom section is
        driven into the side entryway of the collection chamber by
        induced spray current.

     •  Once inside the chamber, the oil is out of the influence of  the
        free stream flow and is moved forward by a series of internally
        mounted spray nozzles.

     •  At the forward most location within the collection chamber  (the
        oil recovery zone), the oil stagnates until oil suction pumps
        operate to pump the oil on board the support vessel.
                                47.

-------
                        Surface
                        Oil Flo
                         Lines
Collection
 Chamber
                                Spray Boom Sections
                                         Direction of Current
                                          (Vessel Direction)
                       Figure  17

        Side Entry Collection Chamber Concept


                          48

-------
          The flow below the chamber open bottom will cause cells of
induced circulation.   It was expected that a chamber having a series of
internal baffles would provide a controlled liquid circulation pattern
within the chamber as it is towed.   Correct positioning of baffles
and the use of a short bottom plate at the front was intended to pre-
vent oil underflow.  The front portion of the chamber is closed at the
bottom thus forming a stagnant oil  recovery zone.  Figure 18 shows the
flow patterns expected in this arrangement.

          This concept was verified by developmental testing on a model
scale.  Several spray nozzles were  arranged to provide additional
impetus to thicken the oil.

          CHAMBER STABILITY AND INTEGRITY

          As in the case of the floating spray boom, the chamber is subject
to all the relative water surface displacements of any floating object.
The collection chamber was considered for each instability mode.  Figure
19 illustrates the expected induced motions of the collection chamber.

          Instability No. 1 (Rotation of Chamber with respect to vessel
axis).  The relative rotation of the collection chamber must be con-
strained by the use of an appropriate linkage system.  Connections to
the vessel are made at fore and aft locations on the chamber.  Thus no
relative rotation is possible.

          Instability No. 2 (Heaving of the Chamber).  The connection
links between the vessel and chamber will allow for considerable vari-
ation of the relative elevations of the chamber and vessel.  The relative
displacement permitted must be consistent with the expected sea state
environmental conditions.  The chamber is essentially free to follow
encountered wave profiles.  To minimize the relative heave of the chamber
with respect to the water surface a minimum weight or inertial mass and
a maximum feasible buoyancy were desired.  By the use of such a design,
displacements from equilibrium by a wave will produce large restoring
forces.  The larger the restoring force, the quicker the response and
the smaller the heave.

          Instability No. 3 (Chamber Roll).  Modeling studies produced
a configuration which is much deeper than wide.  This shape is sensitive
to rolling because there are large areas to produce hydrodynamic side
forces but a small distance between floatation cells (width of chamber)
to counteract the moments produced.  It therefore evolved that the chamber
be connected to either the vessel or the first boom section in such a
way that the chamber is constrained to roll with these more stable members.
A  linkage is feasible in producing this constraint while still allowing
for correct positioning and debris removal requirements.

          WAVE AND OIL COLLECTION RELATIONSHIPS

          The effect of waves on the collection system has been considered
and means of compliance to the wave environment have been described.

                                 49

-------
                                              Top View
          • Oil Recovery Zone
                                              Side View of
                                              Collection Chamber
                             Direction of Advance
                         Figure 18

Collection Chamber Internal Circulation Pattern Illustration

                             50

-------
            X
    Rotation with respect to vessel axis
                  Plan View
                                 1
    Heave
1
                          1 ---- 1
                                     l. Water surface
               Elevation View
                           Water surface
                  Figure 19
Collection Chamber Displacements Illustration
                      51

-------
 The effect of waves  on the collection  system is manifested both in
 the oil collection effectiveness  and in  the chamber motion (stability
 and integrity).

           First,  the potential effect of waves on collection effective-
 ness is due to  the water  level changes within  the chamber.  With the
 fast response of  the concept  chamber,  this will be of little consequence
 except  in waves containing a  large, short period energy distribution.
 This will be the  case only for locally active  storms producing choppy
 wind wave conditions.   If this situation is encountered, it will cause
 heaving of the  chamber and distortion  of the circulation patterns within
 the chamber.  These  short period  waves are characteristic of smaller waves
 (on the order of  2 foot significant wave height) and will have little
 effect  on a chamber  of greater depth.  Another effect of heaving on
 chamber effectiveness  is  the  resulting flow in and out of the entry door
 with surge.   If the  chamber bottom were  completely sealed, this effect
 would produce gross  inefficiencies; with the present concept chamber
 (roughly 80% open bottom)  no  appreciable in and out flow with heaving
 is  expected.  A quick operating control  mechanism for recovery of
 the floating oil  in  the chamber is suggested,  whereby the suction pickup
 is  automatically  kept  within  a small deviation from the surface of
 the watar at the  front of the chamber.

           Stability  and integrity in the presence of waves requires a
 system  which can  withstand the expected  forces.  The present' concept
 affords  no major  shortcoming  or limitations in this respect.

 COLLECTION CHAMBER SMALL MODEL EXPERIMENTS

           Small model  tests were  made  of the spray boom alone, the
 collection chamber alone and  the  boom  and chamber together.  Larger
 scale tests  to determine full scale nozzle requirements for the spray
 boom were  also made  prior  to  concept model design as described in
 Section  IV.

           Scaled models of collection  chamber  geometries were evaluated
 in a 24-inch wide  by 48-inch  deep flow channel.  The free stream flow
 velocity of  the channel was regulated  by a 20  hp outboard motor mounted
 on  the periphery of  a 15 foot diameter circular channel.  Fresh water
 flows past  two flow  straighteners and  through  a straight section used
 for  testing.  The honey-comb  flow straightsner reduce turbulence and
 provide reasonably uniform flow for testing.   Surface velocities were
 determined from pitot  tube pressure readings taken 3 inches below the
water surface at the center of the channel.

          CHAMBER MODEL TESTS

          Verification of  the concept  geometry was by a series of flow
channel experiments in which'  small plexiglas scale models of collection
chamber concepts were evaluated.  Figure 20 is a photograph of the flow
channel with an experimental  chamber adjacent  to the viewing window.
                               52

-------
        Figure 20




Flow Channel Test Section



           53

-------
           Both open  front and side entry chamber types were evaluated.
 It was  soon  evident  that there would be major difficulties in isolating
 the  oil slick from the large quantities of water entering the open front
 model.   The  side entry model was therefore chosen early in the development
 testing.

           Testing of this chamber concept showed that oil losses occurred
 and  were attributable to an unwanted circulation pattern within the
 chamber.   This circulation was found to be produced by high stream
 velocities beneath the open bottom chamber.  Incoming oil was entrained
 in a. downward circulation and escaped.  As was analytically predicted,
 the  installation of vertical baffles at specific locations along the
 length  and a bottom baffle at the front of the chamber significantly
 reduced the  internal turbulence and provided an essentially calm region
 at the  front for oil recovery.  Figure 21 shows the final arrangement.

           Effectiveness of the collection chamber was measured by flow
 channel experiments as the quantity of oil moved to the oil recovery
 zone divided by the quantity of oil entering the chamber.  Experiments
 were performed with crude oil in which a known quantity of oil (500 ml)
 was  administered at the opening and after equilibrium was reached, the
 quantity of  oil remaining in the recovery zone was measured.  Figure 22
 shows the  result as a function of the free stream velocity.  The rela-
 tively  high  internal chamber effectiveness suggest that once oil enters
 the  chamber  it is not likely to escape.

           COMBINED SPRAY AND CHAMBER SMALL MODEL EXPERIMENTS

          The operation of the hydraulic oil recovery system, including
 the  spray boom,  collection chamber and oil pickup device, was simulated
 in small scale model tests.   A manifold with 12 small spray nozzles
 approximated the spray boom for the purpose of visually observing the
 oil  flow patterns and streamlines in the vicinity of the collection
 chamber side entrance.   Oil slick simulation was accomplished by metering
 a quantity of oil through a linear manifold one inch above the water
 surface with outlet holes on one inch centers.  This configuration allowed
 oil  to spread to a uniform slick before entering the spray region.

          Figure 23 shows the floating oil slick being moved into the
 small model collection chamber by the model spray boom.  The view of
 the oil entrained as a result of turbulence caused by the spray and
 corresponding surface current impinging on the opposite side of the chamber
 is shown through the side window of the flow channel.  This form of
 turbulence was prevalent during all of the experiments and was the
primary reason for the observed loss of oil out the bottom of the chamber.
Once oil droplets entered the chamber, they were moved forward by
chamber spray nozzles to the stagnant region.  Figure 23 also shows the
calm region towards the front of the chamber.  The maximum depth of oil
contained within this region during experiments was 1 3/4 inches.
                                54

-------
                    Honeycomb  Flow
                    Straightener
                                                    .ection
(j_ Plexiglass  Col
         Chamber
                                  •Internal Baffles
          Flow Channel Plexiglass Window
              Figure 21
Final Collection Chamber Arrangement
                  55

-------
          100 __
Internal Effect
     Percent


           90 _
           80 _
           70 _
           60
.veness
     Lagunillas Crude Oil
        (Venezuela Heavy)
                           f
                           1
                        I
                        2
i
3
                      Free Stream Velocity
                            (Knots)
                            Figure 22

         Internal Collection Chamber Effectiveness as a
                Function of Free Stream Velocity

                                56

-------
                                            Top View of
                                            Test Chamber
                                            showing
                                            Lagunillas
                                            crude being
                                            skimmed into
                                            a simulated
                                            collection
                                            chamber.
                                               Side  View of
                                               Test  Chamber
                                               showing
                                               Lagunillas
                                               crude oil
                                               entrained at
                                               the wall  due
                                               to spray  sur-
                                               face  velocities
             Figure 23
Spray and Chamber Effectiveness Tests
                  57

-------
           Effectiveness  for the model system was  defined  as  the  percent
 of floating oil moved into  the chamber and forward  to  the stagnant
 collection area.   Several nozzles  with different  spray patterns  were
 tried with a square  pattern jet showing the greater utility.  As shown
 on Figure 24,  effectiveness at 1.2 knots with the square  jet  spray  nozzles
 was 100  percent.   At higher free stream velocities, the efficiency
 declined.   Flatjets  proved  to  be less effective.  It was  also noted
 that oil could escape before reaching the chamber if the  model spray
 boom was not properly positioned.   Also,  at the higher free stream
 velocities,  the model spray system could not attain the spray pressure
 and flow required  to move all  of the oil into the collection  chamber.

           Results  of flow channel  small model system experiments are
 summarized in  Tables III  and IV following:
                             TABLE  III

          SMALL MODEL INTERIOR  CHAMBER EFFECTIVENESS DATA
Run No.
1
2
3
4
Quantity of
Oil Added
(a*)
500
500
500
500
Quantity of
Oil Retained*
(in.)
13/16
13/16
12/16
11/16
Velocity
Knots
0.87
1.60
2.80
3.70
                            TABLE  IV

              SMALL MODEL SYSTEM EFFECTIVENESS DATA
Pitot Tube
Reading
Test No. (in. of oil)
(Flatjet)
//1/8P3504**
5
6
7
(Square Jet)
#1/8GG3004**
8
9

0.12
0.31
0.85
0.30
0.50
Velocity
Knots

0.85
1.30
2.16
1.30
1.75
Quantity of
Oil Added

500
500
500
500
500
Quantity of
Oil Retained*
(in.) %Eff

12/16
9/16
8/16
13/16
12/16

92.5
69.0
61.5
100
92
       Flow =2.5 gpm, tests 5,6 & 7; flow - 3.5, tests 8 & 9
*  13/16 inches retained is equivalent to 500 mJl
** Spraying Systems Co., Bellewood, Illinois
                                58

-------
 100
Effect3veness
System
    (percent)


         80 _
  60 -
  40
  20 -
                              1.0
                 Free Stream Velocity (knots)
                          Figure 24

        Spray and Chamber Effectiveness as a Function  of
                    Free Stream Velocity
                From Small Model Experiments
                              59

-------
 COLLECTION CHAMBER DESIGN

           A large  scale  model  collection  chamber was designed and fab-
 ricated based on the  preceding developmental work.

           HYDRAULIC

           Spray nozzles  were used  to induce surface flow toward the
 forward oil recovery  zone  of the chamber.  Manifolds were designed
 for orientation, nozzle  types,  and spray  impingement angles consistent
 with those of the  small  developmental model.  Piping arrangements pro-
 vided for  pressure and flow measurement at the  nozzles.

           Oil suction piping and recovery pumps were sized for a recovery
 rate of 260 gpm.   To  achieve this  pumping rate, 3 pumps (diaphram type)
 of  85 gpm  each were employed.   Diaphram pumps were used rather than cen-
 trifugal pumps to  avoid  excessive  emulsification of oil-water mixtures.

           Each pickup pump was  provided with an individual suction line
 (two inch  flexible neoprene-lined  hose).   Each was also attached to a
 counterweighted recovery device which pivoted at the chamber center and
 floated in the water  in  the oil recovery  zone of the chamber.  Oil could
 enter any  one of the  three floating suctions.  An arrangement including
 flotation  material  and a horizontal baffle provided vertical position
 control.   See Figure  25  for details.

           HYDRODYNAMIC

           The drag  forces on a  deep and narrow object, such as the
 collection chamber  are from waves,  currents, and the forward motion
 of  the  system.  These forces were  calculated as 765 pounds for a three
 foot wave  (3.9 sec. period) and 285 pounds from a five knot (8.45 ft/sec.)
 relative velocity  (speed of forward motion plus current).  By super-
 position,  the maximum frontal  force would be 1050 pounds.  Since large
 scale basin experiments were to involve only straight ahead travel, side
 forces were not determined.

           STRUCTURAL

           The  model test chamber was constructed of plywood with
 angle iron reinforcement at all corner  locations.  Buoyancy was pro-
 vided by closed cell styrofoam held in  place by 1/4 inch bolts and
 large plywood washers.   Buoyancy material was positioned in the front
 and rear for nearly the  full depth of the chamber.  Additional buoyant
material was  positioned at and  just above the neutral water line to
provide large  restoring  forces  against  perturbations (waves).

          Ballast  (220 pounds)   in  the form of lead bricks was attached
 to the bottom of the chamber.   This reduced the center of mass sig-
nificantly below the center of buoyancy,  therefore enhancing roll
stability.   The total chamber weight (including ballast) was 440 pounds
                                6Q

-------
                              To diaphram pump
  Pivot  for oil recovery
  piping	-^
                                            Oil recovery piping
                                                              Flotation Material
                                                                (polyurethane
                                                                   foam)
v^	^   X. ^f" -^J "'->->•
 \\    o   „•              ~   \   foam>
—^A    Suctxons	           \
                   ^i      ri        i

               Chamber Side-entry
                                       	|
	
                                Figure 25

          Schematic  Showing One of the Three Duplicate Oil Pickup
                 Arrangements and Its Relation to Other Parts
                        of the Collection Chamber
                                    61

-------
and the total buoyancy 880 pounds.  Buoyancy was equally divided between
material for restoring force and direct chamber weight compensation.

          A photograph of the large scale model test chamber as designed
and fabricated is shown on Figure 26.
                                62

-------
                 Figure 26




Large Scale of the Model Collection Chamber




                     63

-------
                           SECTION VI

                       CONCEPT  EVALUATION


          The basic system components,  previously  described, were  fab-
ricated and combined for experimental  evaluation.  A  thirty-five foot
lifeboat was used as the support vessel.   It  was attached  to a  trolley
that was supported by the test  basin wall.  A model spray boom, 40  feet
long and a 9 foot long collection chamber  were  attached  to  the  support
vessel.  The assembly was towed by a tractor  driven parallel and exter-
nal to the basin wall.  The system model simulated one-half of  a scaled
(approximately 40 percent scale)  model hydraulic oil  skimmer.   (See Figure
3 for system arrangement and construction  details  of  the major  com-
ponents—spray boom and collection chamber.)  Oil  was purposefully spilled
in a boomed off area to be swept by the system. During  testing, the
instantaneous oil thickness was monitored  as  the area was  swept, thus
enabling measurement of the amount of  oil  encountered.   Spray flows and
pressure and vessel speed were  monitored.   Recovered  oil and water were
evaluated for each test to determine system effectiveness  and oil
recovery rates.  During all tests, system  integrity and  stability  was
noted.

        The main considerations in evaluating concept performance  were
(a) structural integrity and stability and (b)  oil recovery performance.

STRUCTURAL INTEGRITY AND STABILITY

        Before oil recovery experiments were attempted,  several test
runs were made in order to identify any weak design characteristics
or system instabilities in waves or under  high  speed  conditions.   Minor
problems could then be solved before they  could impede  subsequent  per-
formance testing.

        HIGH SPEED PERFORMANCE

        Several tests were performed in which the  support  vessel,  with
the concept system attached, was towed (in the absence  of  waves)  at
progressively higher speeds.  The collection chamber  and boom were
separately evaluated.  It was initially found that at a speed of three
to four knots and the boom angled at 45° to the vessel  (no spray),
sections of the boom began to dive or "porpoise".   This affect  was
caused by the bow wave formed in front of  the boom (estimated as six
inches to one foot in height).   When these experiments  were repeated
with the spray system in operation, this effect did not become apparent
until speeds greater  than five knots were reached.  Additional boom-to-
vessel cable connections and adjustment of tension in each partially
corrected the problem.  The roll stability floats  for each boom section
were adjusted to increase the lift with forward motion.  Fifty pounds
of additional ballast (per boom section) were also added to each roll
                                 65

-------
 stability float.   Following  these modifications, tests up to and inclu-
 ding seven knots  revealed  no instability or undesired motions.

         The speed mode  of  operation, which might be used for example
 in moving from one oil  patch to another, was evaluated.  The boom-to-vessel
 angle was set  at  10°  to the  vessel centerline and a light spray action
 was applied to hold the boom in position.  One test at a speed of advance
 of 3.75  knots  was  made  without stability or other problems.  A slight
 tendency toward buckling between boom sections was noted, although it
 was clearly safe  and  stable,  showing only a few degrees of rotation from
 one boom section  to the next.

         No  collection chamber instability was noted up to speeds of
 advance  of  seven knots.  A bow wave occasionally overtopped the chamber
 front at higher speeds.
        PERFORMANCE IN WAVES

        Several structural tests were made with waves up to 36 inches
 (significant) wave height.  Initial tests were statically performed.
 Several minor changes were required in the interfaces between the
 vessel and collection chamber and between the vessel and boom.  See
 Figure  27 for photographic evaluation of the concept system performance
 in waves.  Experiments were then performed with the system in motion
 at speeds up to five knots.  With the modifications derived from earlier
 high speed runs, the presence of waves resulted in no inherent diffi-
 culties.  The boom sections followed the waves quite well even at the
 higher speeds.  The collection chamber also reacted well to waves.  The
 stilling action of the collection chamber resulted in quite small surges
 in the chamber oil recovery zone (less than six inches).  Some of the
bow waves overtopped the front of the chamber just as in the earlier tests.
No correction was made, although it was noted for change in the event
another similar device is built.

OIL RECOVERY PERFORMANCE

        Once the system was found to be structurally sound and stable,
oil recovery performance experiments were performed.  The system per-
formance was to be characterized with respect to the following criteria:
        •  oil recovery rate
        •  system effectiveness

        •  oil-to-water ratio
        •  recovered oil properties.

        The important system variables were evaluated relative to these
criteria in order to determine the system operating characteristics and
                                66

-------
                                               Spray Boom
                                               Collection
                                               Chamber and
                                               Support
                                               Vessel
                                                Spray Boom
                                                and Boom to
                                                Vessel Connection
                       Figure 27

Concept Evaluation Model Operating in a 30" to 36V Wave Condition

                           67

-------
 performance limitations.  Three types of oil were used;   diesel  (No.  2
 fuel), crude oil (Texaco,  Anacortes - 40-42° API) and Bunker  (No.  5 fuel).
 A considerable range in speeds and several tests  with waves were made
 with each oil type.

         As in the integrity and stability tests,  the system was  towed
 along the test basin wall  by a wheel tractor.  The average speed of
 advance of the model was obtained by timing the travel over the  length
 of a test run.  The test area was a rectangle,  bounded by booms, approxi-
 mately 60 feet by 360 feet in length.  The performance test length was
 203 feet, to accommodate acceleration and deceleration of the system.
 The effective skimming width was found to be 23 1/2 feet  for  the 40 foot
 boom angled at 45°  to the  vessel and correcting for end effects.   The
 oil thickness was monitored along the test length using a pulse-echo
 ultrasonic transducer.   Thickness was obtained  by using an oscilloscope
 to monitor the time between sound echoes from the oil/air interface and
 the oil/water interface.   The thickness readings  were integrated to deter-
 mine the average slick thickness encountered by the skimmer.   A  full
 description of the  oil thickness system used in evaluating the oil skimmer
 concept is appended.

         During acceleration (before the test length was entered),  the
 boom spray pressure was adjusted to the value expected to be  appropriate
 during early nozzle development tests.   It was  maintained at  a value
 which was visually  observed to move oil to the  correct position  adjacent
 to the collection chamber  inlet.  Also  within this  test length,  oil
 suction pumps were  started;  the inlets  at the pickup device in the
 collection chamber  were held manually from contacting the oil and  water
 in the chamber.   Once the  collection chamber entered the  test length,
 the pickup inlets were  dropped into the chamber to  remove the collected
 oil.   Throughout  each run,  the spray pressure remained the same  as that
 set at the beginning  and the pickup device remained operating in bringing
 the oil  and water on  board  and into a large recovery tank.  When the  end
 of the test  length  was  reached,  the system was  decelerated, the  spray
 nozzle pressure was diminished and the  suction  pickup device  allowed  to
 pick up  any residual  oil left  in the chamber.

         After  each  run, data were recorded including all  pertinent geo-
 metrical  relationships  as well as nozzle  pressures,  speeds of advance
 and  oil  thickness measurements.   An oil  sample  was  taken  of the  oil at
 the  surface  of the  recovery  tank.   This  sample  was  measured promptly
 for  oil,  emulsion and water  volumes.  Later evaluations of these samples
 indicated  the water,  oil and entrained air contents.   The depth  of
water, emulsion and oil was  also monitored in the large recovery tank
and  used  later in determining  system performance.

        When tests were made in  waves,  the significant wave height was
manually measured.  The significant  wave height was  the measured peak
to trough of the highest 1/3 waves  of a  random  sea.   This height was
estimated by visual observation  of  the water  height  on a  meter stick.
                                68

-------
        Definitions of performance measures and their determination were
as follows:

        Oil recovery rate is the volume of oil or emulsified oil  recovered
on board the support vessel (in the recovery tank)  per unit time.   This
represents the rate of continuous oil recovery from the model system under
the condition existing during the test.

        System effectiveness relates the oil encountered within the skim-
ming area (as implied by the pulse-echo ultrasonic oil thickness  measure-
ments) to the oil recovered on board the support vessel (in the recovery
tank).  A 100 percent effective system would be able to recover 100 per-
cent of the oil encountered under conditions prevailing during a  test.
Fifty percent effectiveness would imply that fifty percent of the oil
encountered is recovered.

        Oil-to-water ratio is the amount of oil (and emulsified oil)
recovered divided by the amount of free water recovered on board.

        Recovered oil properties were from samples taken directly after
each test run.  After a static holding period for coalescence, the
oil, emulsion and water volumes were measured for each sample.  Oil
concentration in emulsions and entrained air volume were also measured.

        Test results are included in Table V following.  This Table
includes all data relating to oil recovery rates and system effectiveness,
i.e., the practical system operating characteristics.  Data which relate
oil-to-water ratios recovered and recovered oil properties to system
variables are also included.  Such information is required to delineate
the fluid processes necessary for effective stream quality program goals.

        Explanation of the tabulated data is necessary for complete
understanding of these results.  First, experiments were performed at
an outdoor facility during the late fall and early winter.  In addition
to the discomfort to operating personnel, the low ambient temperature
of test basin water (32 to 50°F) prevented oil spreading at rates
typical of more moderate conditions and considerably influenced other
oil properties such as oil viscosity and pour points of Bunker fuel.  In
addition, equipment freeze-up periodically prevented prompt initiation
of experiment runs.  Also, when winds occurred, the slick was frequently
swept out of the path of the test apparatus against the basin wall.
Work was periodically delayed for up to five days.  The net result of
these environmental difficulties was to slow the rate of experimental
evaluation, although such difficulties may be quite likely in actual use
of the system.  It is believed that considerable knowledge of the
possible effects of adverse conditions was gained during this work.

        The first twelve test runs were performed with No. 2  fuel oil.
During these test runs, data acquisition techniques were developed,
operational sequences devised, equipment was run-in, and numerous
empirical adjustments were made.  The latter entailed  relocating  the
                                69

-------
 collection chamber position,  enlarging  its oil entrance aperture, and
 "tuning"  the  spray pressure and  flow  to best  correspond to the speed of
 advance.   This  tuning was not possible  during the performance of a
 test  run(which  lasted for only one  to two minutes) and had to be done
 from  run  to run.  This break-in  period  was much longer than had been
 anticipated and performance results from the  first 12 runs were poor.
 Effectiveness ranged from 1.5 to 12.8 percent and recovery rate from
 60  to 1920 gph.  Because of the  aforementioned reasons and because of
 later highly promising test results, the first twelve runs are not con-
 sidered to be representative  of  concept system capabilities.  Data from
 the subsequent  experiments (13 through  29), are analyzed in the following
 discussion.

        OIL EECOVERY RATE

        The effect of changes in slick  thickness for the three oil
 types is  shown  on Figure 28.  Bunker and diesel fuels are little
 effected,  however crude oil tests show  a marked increase in recovery
 rate with  increased oil thickness.  Figure 29 relates speed of
 advance with oil recovery rates.  Up to 1870  gph was achieved for the
 speed of  advance range of 1.2 to  4.5 knots for Bunker fuel.  Crude
 oil recovery rates as great as 8700 gph were  achieved between 1.0 and
 4.2 knots  and for diesel fuel, 7500 gph up to 3.0 knots.  System
 limitations became apparent especially  for the crude oil above 3.0 to
 3.5 knots.  Oil recovery rates are limited at low speeds by the amount
 of oil available for recovery, rather than system deficiencies.
spray pressure divided by thi
squared, or OH °'5/(T V2) ii
        A  semi-empirical  factor which  combines vessel  speed,  oil
 thickness  and spray  flow  pressure  characteristics  is used  to  evaluate  this
 system.  This performance parameter  (spray  flow  times  square  root  of
                          the  product  of  oil  thickness and vessel  speed
                   .  „_  .  . is  of similar  form to equation  (4)  or equation
 (8) in the spray boom performance  analysis.   The analysis  provided the
 basis for  selection  of  this  combination of  variables.   Data obtained
 from concept evaluation tests  shows  a  correlation  with this empirical
 factor for the range in variables  tested.

        Oil recovery rates for Bunker  and diesel fuels, are rather
 insensitive to variation  of  this factor as  shown on Figure 30.  Recovery
 rates of crude oil increased from  3000 to 8000 gph as  the  empirical
 factor increased from approximately  15 to 25.  A value of  25  would
 produce 1500, 7000 and  8000  gph  oil  recovery  rates for Bunker, diesel
 and crude  oils, respectively.

        SYSTEM EFFECTIVENESS

        Effectiveness is  defined as  the percentage of  oil  encountered
which is collected by the system and brought  onboard for processing.

        Average slick thickness  is plotted  against effectiveness in
Figure 31.  Singular relationships, "if they exist, are obscured by
                                70

-------
                                                                                                                                               TABLE V
                                                                                                                                CONCEPT SKIMMING PERFORMANCE TEST RESULTS
Test Run*
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
13
x9
20

21
22
23
24
25
26
27
28
29
Time/Date Spill
Material
11:30/12/3 Diesel Fuel (#2)
12:10/12/3
10:30/12/4 "
12:00/12/4
1:30/12/4
2:30/12/4 "
11:15/12/5
12:00/12/5
1:00/12/5
2:30/12/5
9:30/12/8 Bunker Fuel (#5)
10:30/12/8
2:00/12/8
9:00/12/9
10:00/12/9
10:00/12/10
11:00/12/10
10:00/12/11
11:30/12/11
.11:30/12/12 Crude Oil
(40-42° API)
1:00/12/12
2:30/i^/::
11:00/12/13
12:00/12/13
1:00/12/13 "
11:30/12/14 Diesel Fuel(//2)
12:00/12/14
1:00/12/14
2:00/12/14
Slick
Thickness
(mm)
.8
3.8
.8
1.3
.6
.9
2.3
1.4
.8
.8
1.5
1.9
1.1
1.3
2.4
2.1
0.5
1.1
1.4
1.2

1.7
1.7
1.0
1.6
1.0
3.0
2.5
2.5
1.5
Wave Condition
(height /period)
calm
n
n
n
n
n
n
it
18"/2.5 Sec.
24"/3.0 Sec.
calm
n
"
n
n
"
n
28"/3.0 Sec.
28"/3.0 Sec.
calm

"
n
n
14"/3.5 Sec.
14"/2.5 Sec.
calm
30"/3.5 Sec.
calm
30"/3.5 Sec.
Speed of
Advance
(knots)
.5
.9
1.3
2.0
2.9
0.8
3.3
3.6
1.4
2.6
0.8
1.4
2.1
2.7
3.0
1.4
4.5
1.2
2.8
1.6

3.0
4.2
2.0
2.0
1.0
1.5
2.0
3.0
0.9
System
Effectiveness
7.2
12.5
7.4
9.8
9.4
2.5
1.5
2.1
12.8
5.3
3.0
4.2
10,4
6
12
13
6.5
8.7
30
84

55
24
40
97
100
56
39
41
81
Oil Recovery
Rate (gph)
140
1920
520
910
590
60
430
360
530
390
130
480
1870
1170
1420
820
1500
420
1380
7700

8700
3000
4670
6800
4900
7500
6700
6640
68LO
Oil/Water
Ratio
.05
.52
.12
.40
.05
.03
.03
.03
.10
.03
.03
.05
.12
.34
.16
.07
.20
.06
.08
7.7

2.4
.6
6.8
7.5
2.3
1.5
1.2
1.2
4.9
% Water (air)
in Recovered Oil
	
—
8
22
56
60
77
38
30
50
(25), 7
(17), 7
(36)
(25)
(4)
(27)
(12)
(12)
(13)
70

67
64
13
3
3
1.5
3.1
6.9
3.2
Empirical Factor
Q^H 0.5/(T v2) gph/ ft
psi°-5/(mm kts2)
488
36.3
72.2
25.4
34.8
67.7
7.9
11.6
51.9
32.4
61.3
20.1
27.3
17.3
9.2
25.9
27
47.2
14.0
34.8

11.1
9.2
20.3
15.0
69.8
19.6
14.3
7.3
53.4
*  Test runs 1 through  12  are not considered to indicate system performance;
   Test runs 13  through 29 will be considered to indicate __ -tern
   concept.
but were made  to determine necessary modifications.
for the particular arrangement  tested.   It  does  not indicate optimum performance of basic
                                                                     71-72

-------
CO
 o
 o.
 60
 0)
 J-l
 13
 0)
 O
 a
 a
 •H
 O
      10

       9
       8  .
       5  .
O
A
O
Bunker Fuel
Crude Oil
Diesel Fuel
Darkened Forms Indicate Tests
  with Waves  (14-30")
                    0.5          1.0         1.5
            Average Slick Thickness Encountered (mm)
                                             2.0
                                                     2.5
                              Figure 28
       Oil Recovery Rate as a Function of Slick Thickness
                                 73

-------
 P.
 60
3
&
M

-------
     50
     40
 •*]
 ml
 n-it
F
H

-------
 c
 
-------
such variations as the emulsification of the oil by recycling from run
to run and nonuniformities in slick thickness.   As is  expected,  system
effectiveness decreases with increases in speed of advance (see  Figure
32).  This is a result of both inefficiencies in skimming at higher
speeds and limits in system variables such as spray flow and pressure
at increasing speed.  Bunker fuel showed relatively poor system  effec-
tiveness.  During evaluation tests, low temperatures caused this
material to congeal so that it did not physically resemble the slicks
of the other materials.  It appeared very sluggish and tended to roll
along near the surface rather than flowing on the surface as did the
other material.  Several possibilities exist for improving performance
on Bunker fuels.  The most attractive is to provide the capability for
adjusting the height of the spray nozzles so that spray water exits just
at the water surface.  Such a reduced spray impingement angle would be
expected to substantially improve effectiveness on heavy oils.  This
conclusion is supported by the spray boom performance analysis in con-
sideration of the submergence phenomenon caused by spray impingement on
heavy oils.

        Figure 33 shows system effectiveness as a function of the
previously defined empirical factor.  The correlation defined in Figure
33 is consistent for the three oil types.  An empirical factor of
approximately 25 is required for recovering 100 percent of crude oil
encountered.  Extrapolation of the diesel test data shows a value of
40 for 100 percent recovery.

        Bunker fuel tests also show a correlation, but at a lower range of
effectiveness.  Extrapolation far beyond the experimental parameter
range should be done with caution.  Considerable increase in the empirical
factor is necessary in order to increase system effectiveness for Bunker
fuel to the levels observed for the lighter materials.  Approximately
20 percent system effectiveness was obtained at an empirical factor
value of 40.  Changes in the system would be expected to change the per-
formance correlation.

        OIL-TO-WATER RATIO

        The relative quantities of oil, emulsified oil and water
recovered on board  the model test support vessel varied considerably.
Since the oil recovery pumps (3) were all operated continuously for
each full test run  at  their maximum pumping rate  (no attempt was made
to optimize the oil  to water relative pickup rates), variations in
the quantities of oil, emulsion and water recovered occurred.  Exper-
iments were performed over a broad range of speeds of advance, over
varying slick thicknesses, with different oils  (pumping Bunker  fuel
would be slower due  to viscosity effects).  With waves, some  pumping
of air occurred  (the suction device would frequently raise  above  the
surface and elementarily pull in air).  All of  these would affect
the relative quantities of oil (and  emulsion)  and water being recovered
by  the pumping system.
                                  77

-------
 c
 CD
 o
 M
 0)
 P.
 CO
 CO
c
cu
0
cu
M-t
4-1
W



-------
       50
m

o
 M


 I

 H
m
 •

o
 o
 4J
 O
 u
 •K
 cx
       40
30
       20
       10
                                    O

                                    A


                                    D
                _L
                                 Bunker Fuel



                                 Crude Oil


                                 Diesel Fuel


                                 Darkened forms  indicate

                                   tests with  waves

                                   (14-30")            /
                                                                     A
               _L
                  J_
JL
                                              J_
                                             _L
0    10     20   30    40     50     60     70


         System Effectiveness (percent)
                                                                _L
                                                          80
                                                         90   100
                             Figure 33



 Empirical Factor as a Function  of  Model System Effectiveness
                                  79

-------
         With continuous  operation of  a large  prototype system, the
 rate of pumping of the oil  recovery pump  system could be readily con-
 trolled.   Approximate maximum oil-to-water  ratios  for Bunker fuel,
 crude oil  and diesel  fuel were found  to be  1/3,  20/1 and 4/1, respec-
 tively.   It  is also probable  that frequent  and  appropriate "tuning"
 of the system would achieve oil-to-water  ratios  on a continuous basis
 as were obtained during  the best  test  runs.   Bunker fuel oil-to-water
 ratios would possibly be increased from the 1/3  value for continuous
 operation.   Greater ratios  would  be possible  if  recovery rates were
 increased  or pumping  rates  reduced.  Here,  continuous operation implies
 oil recovery over a time period on the order  of  an hour rather than
 the short  duration test  runs  of one to two  minutes.

        RECOVERED OIL PROPERTIES

        Characteristics  of  recovered oil were implied by sampling
 recovered  oil  and allowing  the constituents to coalesce.  Thus, a
 determination  of the  extent of emulsification and  air entrainment
 was  possible.

        Bunker fuel (No. 5, P.S.  300 grade) showed no tendency to
 emulsify.  Volume was found to decrease with  time, however.  This
 decrease was observed to be caused by  air entrainment in the spraying
 and  pumping  operations.  Air  entrainment varied  from 4 to 36 percent
 in  the recovered  oil  product:  for  the test runs evaluated, averaging 19
 percent over the  seven tests.   This represents a decrease in effective
 density of on  the order of  19  percent, i.e.,Bunker of density 1.0 would
 have an effective density of  0.81.

        Both diesel fuel (No.  2)  and the  crude oil (40-42° API) samples
 showed varying degrees of emulsification, primarily water-in-oil type.
 Water  concentrations  ranged from  three to 70 percent and 1.5 to 9.6
 percent for  crude and diesel  fuel, respectively.   Crude oil samples
 averaged 37  percent water,  while  diesel oil samples averaged only
 3.7  percent.   For both fuels,  it  was impossible  to distinguish free
 oil  from emulsified material  in the samples when initially examined.

        Crude  and diesel oils  showed no tendency to capture entrained
 air, probably  due to  their  relatively  low viscosities.  Bunker fuel, on
 the other hand, was quite viscous, almost to  the pour point of the mater-
 ial, and showed no emulsion with  water.

        Continued cycling of  the  oils  resulted in  added emulsification
 from one test  to  the  next,  as  was noted in  the initial 12 test runs
with diesel  fuel.

 SYSTEM PERFORMANCE ASSESSMENT

        The  following paragraphs  compare  the measured or extrapolated
 concept system evaluation test with design goals as specified in the
 request for proposal.
                               80

-------
        (1)  Skim in 5 foot waves with 20 mph winds and 2  knot  currents
             superimposed

        Skimming with the complete system at approximately 40 percent of
full prototype dimensions was performed in random waves of three feet
height with little or no degradation in system performance.  Waves  in
the area covered by surface oil slicks were suppressed to  about 30  inches
in height.  It is believed that the short chop waves encountered by the
concept system during testing are probably more detrimental to  performance
than longer waves of the 5 foot size.  This is because of  the ability of
the floating components to follow the longer wave profiles.  Component
interfaces will be designed to be compatible in displacement and strength
to the expected environmental effects of a five foot sea.

        The effect of wind on performance (structural and  stability)  was
not determinable during the high wind conditions periodically experienced
during model testing.  These winds did, however, force the oil  slick
against the test boundaries and out of the skimmer path, thus making it
impossible to test for skimming performance during such periods.  All
exposed components are of a low profile to the wind and the sea.  It is
therefore believed that winds of 20 mph will have little effect other
than to cause a surface oil movement with respect to the water  column.

        Currents of 2 knots are assumed to be relative velocity values
between the system components and the water column.  Since skimming was
quite successful at greater values than 2 knots, this was  met.

        (2)  Capable of 8 knot speeds in 10 foot waves and 38 mph winds
             and capable of 12 knot speeds under the condition of (1)
             (not while skimming)

        The configuration of the system in this mode of travel would be
such that  the booms would be towed in a trailing position.  The low pro-
file of the booms and their ability to follow the sea surface are expected
to make them quite suited to towing.  The outrigger floats are expected
to present some resistance to such speeds.  However, with  adequate design
of the connection to the booms, these floats will handle 8 or 12 knot
travel.  Problems may arise at the joints between boom sections and
between the boom sections and the support vessel.  Design  of these inter-
faces must make provision for expected environmentally induced forces.
Total  travel      for  the vessel boom interface must be suited to the
wave heights encountered.  Quick breakaways or other similar devices
might be necessary for  protection of  the vessel and personnel under
severe storm conditions.

         (3)  Storage of 1,000,000 gallons of oil

        Storage requirements were beyond  the experimental  phase of this
concept development program  although  pillow tanks  are  presently available
and capable of filling  this  demand.
                                 81

-------
           (4)   Able to recover oils ranging from light  diesel  to Bunker C

           Test runs with, diesel fuel CNo.  2),  Bunker  fuel  (No. 5) and
 with a crude oil (40-42° API)  showed capability for handling each type.
 The recovery rate for Bunker fuel was at about one sixth the rate for
 crude oil and diesel fuel recovery.  Test  runs involving Bunker fuel
 showed that  the spray jet tended to aerate this material so that it
 exhibited a  lower density than it would have otherwise.  This  enhances
 the ability  of the system to recover heavy fuels as compared to con-
 cepts involving only gravity.

           (5)   Capable of recovery 50,000  gallons per hour under design
                environmental conditions as stated in  (1)

           Recovery rates of 50,000 gph can be  accomplished by  a prototype
 recovery system based on the concept developed in this  program.  Test
 data show that 1500 gph of heavy oil or 7000 to 8000  gph of light oil
 (crude oil or  diesel fuel)  can be recovered by the concept model which
 covered a 23.5 foot sweep width.   Simple scale-up shows that 50,000 gph
 can be recovered using a skimmer boom combination which has a  140 foot
 sweep width  for light oils and 780 foot for heavy oils.  As previously
 mentioned, flexibility for optimal placement of nozzles closer to the
 water surface,  should enhance  performance  for  heavy oils.  Further tuning
 of  the spray pressure and boom-to-vessel angles may also improve on these
 scale-up predictions.   Concept evaluation  tests were  performed on contained
 slicks.   Linear extrapolation  to  the prototype system would be expected
 to  be conservatively in error  because of the boundary effects  associated
 with boomed areas  as  contrasted to large unrestricted slicks.

           (6)   Operate on slicks  of 1.5 mm in  thickness^ or less

           The  concept  system recovered oil throughout the  range of oil
 slicks  employed in evaluation  tests (0.465 to  3.75 mm).

           (7)   The recovered oil  shall not contain more than 10 percent
                sea water and the  effluent  water shall not  contain more
                than 10  mg/liter of oil

          Additional equipment, such  as centrifugal separators, coalescers,
 or  settling tanks, will  be  required to meet  these goals.   The  develop-
ment  of  these was beyond the scope of  the  present program.  Features
 of  the proposed  concept  which  tend to  enhance  the oil-water separation
process  are:    (1)  the  entrainment  of  air in  the heavier oils makes
 them more buoyant  (air  floatation  may  be a natural process in  the system);
 (2)  water effluents  from separation processes  may be  recycled  through
 the pressure spray  system and;  (3)  recovery  pumping rates  can  be "tuned"
to produce a consistently high  proportion  of oil  in the recovery product
pumped from the collection  chamber.
                                82

-------
                           SECTION VII

                         ACKNOWLEDGMENTS
          Experimental analysis and design phases of the study were
performed by a team from Battelle Memorial Institute's Pacific Northwest
Laboratories at their hydrodynamic testing facility in Richland,
Washington.  Acknowledgment must be given to those organizations  which
assisted in this effort.  The assistance of the Texaco and Shell  Oil
Companies in providing test oils is greatly appreciated.  Local material
fabrication contractors and equipment purveyors also must be given
credit:  Keltch Construction Company, Metalfab Company, R and N  Inc.,
W I and M Fuel  Company and Felton Oil Company are but a few who  made
contributions.

          The authors gratefully acknowledge the support and guidance
given by personnel from the Water Quality Office of the Environmental
Protection Agency, specifically, Mr. John J. Barich, III, who served
as project officer and Mr. Harold Bernard and Mr. Kurt Jakobson of
the Agriculture and Marine Pollution Control Branch.

                                                 Mr. John R. Blacklaw
                                                 Mr. Elaine A. Crea
                                                 Mr. Roy C. Kelley
                                                 Dr. E. Roger Simonson
                                                 Mr. P. C. Walkup
                                                   Battelle Northwest
                                 83

-------
                          SECTION  VIII

                           REFERENCES
1.  "Oil/Water Separation System with Sea  Skimmer,"  a report by  the
    Garrett Corporation to the  Environmental  Protection Agency
    (Project 15080 DJP),  October 1970.

2.  "Final Report, Sea Dragon Oil  Spill Containment  and Removal
    Systems>" Airsearch Report  70-6787, September 1970.

3.  "Final Report, Development  of  a Water  Jet Device for  the
    Collection of Floating Oil  Slicks," a  report by  Battelle
    Northwest for the Garrett Corporation, September 30,  1970.

4.  S. A. Berridge, R. A. Dean, R.  G. Fallows,  and A. Fish, "The
    Properties of Persistent Oils  at Sea," J. Inst.  Petrol, Vol. 54,
    No. 539, November 1968.

5.  P. C. Blokker, "Spreading and  Evaporation of Petroleum Products  on
    Water," Paper presented at  the Fourth  International Harbor  Con-
    ference, Antwerp, June 22-27,  1964.

6.  H. G. Schwartzberg, "The Spreading and Movement  of Oil Spills,"
    summary of report number 15080 EPL 4/70 in the Water  Pollution
    Research Series of the Environmental Protection  Agency, April
    1970.

7.  M. Wicks III, "Fluid Dynamics  of Floating Oil Contained by  Mechan-
    ical Barriers in the Presence  of Water Currents," Proceedings,
    Joint Conference on Prevention and Control of Oil Spills,  spon-
    sored by the American Petroleum Institute and the Federal Water
    Pollution Control Administration, Dec. 15-17, 1969.

8.  W. H. Trask and B. D. Bingham, "The Conduct of a Feasibility
    Study of the Inverted Weir Oil Collector Concept," prepared by
    Battelle-Northwest for the U.  S. Coast Guard, 1970.
                                85

-------
                           SECTION IX

                            APPENDIX


Ultrasonic Oil Thickness Measurement Device

          An ultrasonic system was assembled and used during concept
evaluation tests of the floating oil recovery system employing a spray
boom and side-entry collection chamber.   It was operated in waves varying
from essentially calm conditions to 3.0  foot short chop waves and in
currents up to 4.5 knots.  Considerable  difficulty was experienced with
the system as assembled at the maximum environmental conditions experienced.

          The concept involves a specially designed ultrasonic trans-
ducer which obtains oil/water and oil/air interface signals for further
processing.  Detection of the two interfaces is accomplished by using
a pulsed ultrasound signal and monitoring the return reflected pulse.
The time delay between return pulses is  an indication of the presence
and thickness of an oil slick.  All that is necessary to cause a reflected
pulse is to have a significant abrupt density change at an interface.

          The photographs shown on Figure 34 of an oscilloscope trace
of the pulse-echo signal train for diesel fuel on tap water illustrates
the ultrasonic technique.  In the first  photograph, tn denotes the trans-
mitter pulse or time 'O1, t.. represents  the reflected sound energy from
the water-oil interface, t_ represents the oil-air interface, and t«,
t,, . . .  . represents the reflected trapped sound in the oil layer.  The
time between t  and t« and t_, etc. is directly proportional to the oil
thickness.  Calibration of tne instrument would involve determination of
the velocity of sound in oil, thus, the time measurement may be con-
verted directly to oil thickness.  The second photograph represents near
the minimum thickness (near 10 mils) which can be measured with commer-
cially available ultrasonic instrumentation.

          A support must be provided to hold the transducer within a
given distance of the surface.  A floating catamaran pontoon was used as
the support principally because of the wave following capability of such
structures.

          Processing of  signals can be by many approaches depending upon
the required accuracy and the wave conditions present.  With any
appreciable waves, a time delay, a pulsing system and electronic gates
are employed so that only the time difference between signals from the
two interfaces is monitored.  This method and monitoring of  compressed
signals was used in the  work described herein.

          High accuracy  measurements or measurements of very thin slicks
require high quality signal processing equipment and appropriate trans-
ducers.  Testing was accomplished with a thickness accuracy  of  approxi-
mately 1/16 inches.  Equipment presently available may be assembled  to
measure down to 1/100 inches and  further refinements may make a resolu-
tion of 1/1000 inches possible.
                                87

-------
                                                   Diesel Fuel
                                                   (No. 2) 3.0 mm
                                                   thick, on tap
                                                   water
                                                    Diesel  Fuel
                                                    (No.  2)  0.3 mm
                                                    thick,  on  tap
                                                    water
                       Figure 34

Oscilloscope Trace from Slick Thickness Measuring Device

-------
BIBLIOGRAPHIC:
          c KortWeal Laboratories, laticlle Mrarlal Inatltate. Concept
for (b* Cnvlrooatt&tal frorec
1SOW rwr 4/71. Aft 11 1971.

ABSTRACT
                                           S1J« »ount*d
1« pnapid to an onboard separation ayites fro* which the oil la transferred to
flMtlne. tanks or bar^ea and the water IB recycled to the spray sy»tn.

      ExperlBental work vta directed toward coponcot development and evalutlon
      Mo4rl mp*rlm*nt ruult* ihowcd. for ll^-it elli, 6O to 100 pcrcccC cff*ctl*«
     1 all rtco-itrj iat*i of ttOO to 67CW gph.  fLeiulc* with luiuct fuel wer« DOC
      .Further Irvclopaent of the concept. Including •ddltlotial Mdtl t**Ct«t
•ftb heavy Bill plui cp*o •*• *T«l«atlo« ( all illck*. Nachinieal tccowrr MthodB. which do aot eai»« addltlorul
f»e«, «r« »tt*chnl to an op«» •» vorkbeat. 5** «*t«t l« f»c?*4 through »pi*T
»otll«» wounterf ot» the h»ad*r» to BO«« ca oil click toward th> boat. Sil Frotcctlon A(eac;t U*tcr Quality .01 Clce. trotte* k'i«Bv*
 IXIK FVT 4/71. April 1971.
ABSTRACT
       Cffort* ore fcelae. directed to develop effectln cei0t«r*eoaure« «|iln*t
 H»«tlB| •!! allcfca. Mechanic*! recovery Bethod*. which do aot c»u«« additional

 *•* lnv««tl|at*d.

       floating header*! providing • linear water ffitj pattern on the water »<
 lac*. *r« *ttached la an open tea workboat. Sr* water 1* pJip*4 through *puy
         rtMrtt•] work w** directed touaid co*pomat 4«v«lop«e9t aad avaliMtloa
P«rmtt e>p«rla«atal vrrlflcatlaa of aodlf leal totiB >ap«ct*d to lucre*** perfai
#* heavy olla.
       Further ocvelcp^eni
ACCESSION NO.
   KEY WORDS:

 Concept
 Equipment  Development
 Oil  Skimmer
 Open Sea
 Evaluation
 Oily Water
 Separation Techniques
 Technical  Feasibility
                               .atjpe »j»ttm VM r

-------
* A <. ves;, /on .Vumber
Q Subject Field &. Group
05D
SELECTED WATER RESOURCES ABSTRACTS
INPUT TRANSACTION FORM
        Organization

         Pacific Northwest Laboratories of Battelle Memorial Institute
        Title
         Concept Development of a Hydraulic Skimmer System for Recovery of Floating Oil
    10
Authors)
  Blacklaw,  John R.
  Crea, Elaine  A.
  Simonson,  E.  Roger
  Walkup,  Paul  C.
16
    Project Designation
                                       21
                                           Note
    22
        Citation
         Environmental  Protection Agency,  Water Quality Office, Program Number 15080 FWP
         4/71, April  1,  1971 }  88 pp.
       Descriptors (Starred First)
         Evaluation*,  Oily Water*,  Separation Techniques*, Technical Feasibility*,
         Efficiencies,  Hydraulic Systems,  Hydrodynamics, Jets, Mathematical Studies,
         Testing, Water Pollution Treatment
    25
       Identifiers (Starred First)
         Concept, Equipment  Development,  Oil Skimmer, Open Sea
   27
Abstract Efforts are being  directed  to develop effective countermeasures against floating
oil slicks.  Mechanical  recovery methods,  which do not cause additional environmental insult,
are most attractive.   Such  a  concept,  a hydraulic skimmer, was investigated.  Floating headers-,
providing a linear water spray  pattern on  the water surface, are attached to an open sea work-
boat.  Sea water  is pumped  through spray nozzles mounted on the headers to move an oil slick
toward the boat.  Side mounted  chambers are positioned to collect the concentrated floating oil.
Recovered fluid is pumped to  an onboard separation system from which the oil is transferred to
floating tanks or barges and  the water is  recycled to the spray system.  Experimental work was
directed toward component development  and  evaluation of a large system model in a simulated
environment.  A 35 foot  support vessel, a  40 foot spray header, and a 9 foot collection
chamber provided  the  23  1/2 foot model sweep width.  Speeds of advance to five knots, random
waves to 30 inches significant  height  and  three oil types were used in evaluating this system.
Other equipment such  as  an  ultrasonic  oil  thickness gauge, process pumps and tankage were also
used.  Model experiment  results showed, for light oils, 80 to 100 percent effectiveness and
oil recovery rates of 6600  to 8700 gph. Results with Bunker fuel were not as good, being on
the order of 1300 to  1800 gph and 12 to 30 percent effective in Recovering oil from the water
surface.  However, program  time constraints did not permit experimental verification of mod-
ifications expected to increase performance on heavy oils.  Further development of the con-
cept, including additional  model testing with heavy oils plus open sea evaluation of a proto-
LV »« 	 UVULl
Abstractor
ittt-
wau —
John
•t
R.
umnutiemat 	
Blacklaw
Institution
Pacific
Northwest
Laboratories,
Battelle
Memorial
Institute
    it R: 1 02 (REV JUV.V
    VR Sr C
                                         SEND TO'-  VtATER RESOURCES SCIENTIFIC INFORMATION CENTER
                                                 U.S DEPARTMENT OF THE INTERIOR
                                                 WASHINGTON, D. C 20240

                                                         i V. S. GOVERNMENT PRINTING OFFICE : 1971 O - 428- 628

-------