DISTIUTED BY:- - Defense Technical Information · PDF fileAD/A-006 827 A STUDY OF PLANING CATAMARAN HULL AND TUNNEL INTERACTIONS T. Jeff Sherman, et al Michigan University Prepared

Embed Size (px)

Citation preview

  • AD/A-006 827

    A STUDY OF PLANING CATAMARAN HULL ANDTUNNEL INTERACTIONS

    T. Jeff Sherman, et al

    Michigan University

    Prepared for:

    Office of Naval ResearchNaval Ship Systems Command

    February 1975

    DISTIUTED BY:-

    N. L TMUT IU I-O

    L Ln MII OF

  • SICUrI11 tVLASIFICATIOrt OF TWIi PAGE (A;ten O.ws I`Nt.t d)

    REPORT DOCUMkENTATION PACE RUA0^DIN ar',,UCT(ONS/1. nPORT HU'0EH. GOVT ACCESSION NO, I. fLCIPIBfit'" CATAt'00 Nw'-41f.Mt

    011073-1-r- ,9h/ "az eW~74. TITLE (and Subtitle) S.-"rVAIE 01 REPORT 6 P,,Roo dbvan.o

    A STUDY OF PLANING CATAMARAN FinalHULL AND TUNNEL INTERACTIONS 1 Apr 72 - 30 Aug. 74S. PER7OftMIAG 0110. REPORT NLdI.AOR

    7. AUTrO. Shm . CoNTSCT)jT OR GRANT NUMiEmR~T. Jeff Sherman N0004-67-A-0181-0050"Peter A. FisherRichard B. Couch

    S, PIERORMING ORGANIZATION NA14I AND AOORESS ""O. POCRAM, .[LEMCNT PROJECT. TASKt The University of Michigan AREA 6 WORK UNIT NUMea[tsNaval Architecture & Marine Eng. Dept.550 E. Univ. St, Room 126 NR 062-473Ann Arbor, Michigan 48104

    M CONTROLLING OFFICE NAM. AN' ADMZESS 12. RPORT OATS'

    Office of Naval Research February 1975,800 N. Quincy St. 13. NUMSEROF PAGESArlington, VA 22217 37

    14. MONITORING AGENCY NAME G AOORESS(it EUi..t for& tm Controlling Oafla.J Is. sEcumr-Y CLASS. (of this report)

    S(same) UnclassifiedIS&. OI CL ASSI FICATION OO',et,4.RAO1NG

    SCtICOUL.E

    16i. DISTRISUTION STATEMENT (of this Report)

    as per contract agreement

    7,. DISTRIiUTION STATEMENT (at th- abstroct ent.,ed in Stock 20. 11 different from R.port

    12. SUPPLEMENTA,74Y NOTES

    12. KEY WOROS (Continue on ,v,-.,aide i pceaary and identify by block nutnbr)

    high speed, low displacement, catamaran, hull separationhull interaction, resistance, tunnel height, model testprismatic planing boats, computer prediction

    I20. AaSTnACT (Continue on roverse, side of necessary~ and Identity by block numnbec)A high speed, low displacement set of catamaran hulls hasbeen model tested with various hull separations and tunnelheights. Symmetric, axisymmetric and unsvmmetric hull formshavym been tested and cohipared in terms of resistance todetermiiie the interaction effects of the sponsons. A com-puter program for the prediction of power for prismaticplaning boats has been modified to include catamarans.DD , il 1473 EoTo lerdvsl yTo WDD , 3N, 3 NATIONAL TECHNICAL

    INFORMATION SERVICE iccuoTy cL A,$IIc ATIO% OF TrWs PAGE rNo-re Soet Entee.-,US O alme4 d CmegmM

    Spri.glI4d. VA. 221t1

  • IIII COLLEGE OF ENGINEERING

    Department of Naval Architecture and Marine Engineering

    Ship Hydrodynamics Laboratory

    IA STUDY OF PLNING CATAMARAN

    I HULL AND TUNNEL INTERACTIONS

    IFinal Report

    byT. Jeff Sherman

    Peter Fisher

    Project Director

    IR. B. Couch'I

    DRDA Project No. 011073under contract with:

    Naval Ship Systems CommandI Contract No. N00014-67-A-0181-0050S

    I Department of the NavyOffice of Naval Research

    Arlington, Virginia 22217

    February 1975 i7? II

  • I ,

    Abstract

    Little doubt exists that the catamaran hull form offers

    a considerable operational advantage over the conventional

    monohedron hull form under certain specified constraints.

    There has been a renawed interest in the application of the

    j catamaran for high speed ,limited displacemenK. service.However, in many instances, model tests have indicated con-

    Iflicting results in the evaluation of resistance data.

    Three pairs of synmmetric, assymmetric, and unsymmetric

    1 hulls have been tested at the Ship Hydrodynamics Laboratory

    of The University of Michigan to determine the effects of

    hull separation, hull form and tunnel height. Data has been

    presented comparatively in each case and expanded to a full

    scale corresponding to a displacement of I00,OOO pounds.I ,

    II,!'U

    -* $

    U.|

  • ( NOMENCLATURE*SAp Projected planing-bottom area, excluding

    area of external spray strip, sq. ft.

    IB Beam on breadth over chines, excludingexternal spray strip, ft.

    B PA : Mean breadth over chines: Ap/LP, ft.

    I BPT Breadth over chines at transom, excludingoxternal spray strip, ft.

    Bpx Maximum breadth over chines, excludingexternal spray strip, ft.

    BL : Base Line

    b : Breadth over spray strips at longitudinal[ center of gravity, ft.CL * Center Line

    I" CG : Center of gravityCT : Total resistance coefficient

    CR : Residuary resistance coefficient

    Ii h : Finite water depth, ft.rFN: Froude number based on length =V/\I-

    I FNL Froude number based on depth m V/1'4" *Fv Froude number based on volume V

    ai T Acceleration of gravity, ft/sec2

    i LAV Average ,vetted length, ft.LCG Longitudinal center of gravity

    L: Projected chine length, ft.

    LID * Lift- dracg ratio

    P: Effective horsepower

    SRT Total model resistance, lb f

    1 -2-I!

  • RTS : Total ship resistance, lbf

    RR/& : Residuary resistance - displacement ratio

    " TS/: Total ship resistance - displacement ratio

    Rise/V1 / 3 : CG rise coefficient

    S I Wetted surface, sq. ft.

    S/v2/ 3 : Wetted surface coefficient

    VW I Velocity of wave propagation, ft/sec.

    SK: V elocity in knots

    VM : Velocity of the model, ft/sec.

    V/4tI- z Speed-length ratio

    : Angle of attack at after portion ofplaring bottom, degrees

    :z Scale ratio, ship to model

    x W : Wave length, ft.

    9 Deadrise angle of planing bottom

    P : Mass density of water

    V : Kinematiu viscosity

    V : Volumetric displacement, cubic ft.

    Displacement, lbf

    V/ApH : Mean draft-water depth rat!.

    W : Same as

    *Nomenclature used is ITTC Standard Symb'l and that rec-

    ommended in SNAME T & R Bulletin 1-23.

    -3-

  • Introduction and Background

    A significant amount of interest has been shown in the

    possible application of the catamaran hull as an alternative

    to the standard monohedron hull form. Isolated model tests

    have been conducted to evaluate individual designs with re-

    spect to resistance performance. However, only a limited

    amount of actual experimental work has been done to dater-mine the hydrodynamic effects of hull interference.

    [ In the 1960's the U.S. Navy limited investigationsshowed that one specific catamaran design had greater ro-

    sistance than the equivalent mono hull forms. However,

    theoretical investigations and model tests have shown that

    a correctly designid catamaran can actually, have less re-

    sistance in addition to its other operational advantages.

    The theoretical work of Eggers concerning wave interference

    effects revealed the strong possibility of reducing signifi-

    cantly the wave drag below that of the single hulls. This

    was accomplished by phase relitionships in the wave pattern.

    Work at the National Physical Laboratory [3] has indicated,

    however, that the interference effects on viscous resistance,

    could in fact, be the opposite, resulting in an increase in

    resistance.

    There are various methods available for predicting the

    performance of planing catamarans. Stevens Institute has

    don% a significant amount of planing boat work both on the

    theoretical and experimental levels. Savitsky of the

    --4--

  • Davidson Laboratory (8] has developed a computer program for

    the prediction of liower for prismatic planing craft. This

    has been modified for catamarans but does not include inter-

    ference effects on drag, trim and flow characteristics on

    t sponsons and the connecting tunnel.Planing catamaran studies made by the U.S. Navy have

    indicated that the catamaran is inferior at low speeds, only

    performing well at high speeds, i.e. F - 5.0. However, a

    study of this work revealed that the tunnel of the model was

    wetted with solid water. This in effect decreased theLp/Bpv ratio of 6.2/1 (for each of the sponsons) to 2/1, in-

    creasing the wetted surface significantly.

    To gain an understanding of why this leads to a hull form

    of poor resistance characteristicb and what can be done to

    correct this particular aspect of catamaran hull forms,

    Figure 1 is provided. For illustrative purposes, a catamaran

    hull form can be approximated by a summation of two monohedron

    hull forms. This is true only as long as the tunnel of the

    catamaran hull form, hull form B, has a high, dry tunnel and

    thereby sponsons with a 6/1 L P/B ratio. However, hull formC, with a lou wetted tunnel, acts on a monohedron hull form

    with an L p/Bp, ratio of 2/I with bottom discontinuity. This

    obviously leads to a hull form of poor resistance character-

    istics. However, as was discussed in the first paragraphs as

    the hull picks up speed, approximate Fv > 3.5, the tunnel is

    no longer wetted with solid water and the hull becomes a

    catamaran.

    -5I

    - -- " - . .. .. .. i l liI i

  • HULL FORM COMPARISON

    A. MONOHEDP.ON OIULL FORM

    I.

    B. CAT-AMARA.N HULL FORM (HIGH TUNNEL)

    I ___T (each side)

    C. CATAMARAN HULL FORM (L.W TUN=,EW

    -. 6 * F 3.5

    EL TI FIgureI Figure I

    1 -6-

  • I

    TADU I

    ?MODEL CIIARNCTRISTICS

    I LOA 36"

    Beam 6.0" (per Sporicn)Dopth 5.625"Displacement 8.C.6" (per Sponadn )

    lba. 0 700 F

    Volume .129 PT 3

    LCG 9.0" Af t Of FP

    Tunnel Hiiqhtlow 4.31 Off Basc Iinohigh 5.3" Off Base Linn

    Sponson Spacing 0"6"

    12 "

    I

    I

    I -7 -

  • Throe pairs of mdels were constructed at tho Sh