study of structural action of superstructures on ships 48
TRANSCRIPT
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 1/58
. .
1,
1
,
1.
‘
-,,,
.,
(
1
1
,,,
,,g
,,
‘!’!<,
!“,’
,. .. , SERIA~ NO. SSG481
,, .,-’ ..’ ,
,“ #
COPY NO,25’
!,’,,
,,, .’ ,.,,. ‘1,
I /. REPORT ‘I ,,
,,,.,. ‘, ON ”’.””.”
A STUDY ON THE STRUCTURAL ACTION OF—--——- ———- .—. . -— . . . . -.--—-
‘1
,,
SUlWlWTMJCTUlWt5 ON SHlk’S,.
,-‘,
BY ,-
DF. HANs H. 13LtiICH‘ ““““
Columbia University
,Under Bureau of S~p# Contract NObq5053S ,
(Index, No. NS.791-034) ,
... /
,-J
1’
>’ ,,
1“FOR
SHIP STRUCTURE COMMITTEE ,,
The
Conven+ by
Secretnry of the Tremury
I,
,“
,,,.
MEMBER AG~NCIES‘--KLP STRUCTURE WMMI~EE
BUREAU o? SHIPS, DEPT. OF ‘NAVY
MILITARY ==A TRA!’J sPORTAT1ON SERVICE. DEPT. OF MAVY ,!!..
UNITED STATES COAST 13UARD, TREASURY DWPT. ,,
MARITIME ADMIUWRATION,’ DkPTt . cm COMMERCE.
,’,
AucRICAH. BUREAU OF, SHHWNU,.
)’
I,!,‘,
,. I ,’
AnORKS* CORRIKWO’ND=NCE TOI
EECRET A R Y,,
,.
,, ,’
SHIP STRUCTURE COMMl~EK
I!,
,,
,! U. S. kOABT GUARD HEADQUARTERS,. 1-
WAEWINGTON 25, 0. ., C,
,,‘,
,!)
,’ ,’,,,
1,
DECEMSER21,1951,.,
/,1“,’ ‘,,,
,,,,1
— II‘.—.—- ,
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 2/58
SHIP STRUCTURE COMMITTEE
December 21, 1951
MEMBER AGENCIES: ADDRESS CORRESPONDENCE TO:
BUREAU 0!s SHIPS. DEPT. OP NAVY SECRETARY
MILITARY SEA TRANSPORTATION SERVICE. DEPT. OF t 4AVY SHIP STRUCTUF5 COMMITTEE
UNITED STATES COAST GUARD, TREASURY DEPT. U. S. COAST GUARD HEADQUARTERS
MARITIME ADMINISTRATION. DEPT. OF COMMERCE WASHINGTON =.. D. C.
AMERICAN BUREAU OF SHIPPING
Dear Sir:
Her@with is a copy of the report covering a
pl%ltitiary study Of I~Thestructural ACtiO1’1 of &lperstrW-
tures on Shipsllby Dr. Hans H. Bleich. This investigation
was conducted by Dr. BleiCh of Columbia University for the
Ship btructure Committee.
Any questions, comments, criticism or other
matters pertaining to the Report should be addressed
to the Secretary, &hip Structure Committee.
This Report is betig distributed to those
individuals and agencies associated with and interested
in the work of the Ship ~tructure Committee.
Yours sticerely,
Rear Admiral, U. s. Coast Guard
Chairman, Ship Structure
Committee
. ... .. .—
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 3/58
REPORT
on
A Study on the Stmcturel Action of
Superstmctures on Ships
Dr. Harm H. Bleich
Associate Rrofessor of Civil Engineer-
Columbia University
Bureau of Shim
Contract NOba 50538
Index No. M-731434
June 1950...,
-.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 4/58
PREFACE
The Navy Department through the Bureau of Ships is distributing this
report for tile”HIP STRUCTURE COMMITTEE to those agencies and individuals
who were actively associa+xd with the research work. This reqort repre-
sents results cf part of ++e research program conducted under the Ship
Structure Committees directive lltoimprove the hull structures af shigsby an extefislonof knowledge pertainfrl~‘-Loesign, materials ard methcds
of fabrication”.,...,.
The distribution of this repd~t 5.sas follows:
Copy No” 1 -
Copy No. 2 -
copy No. 3,-Copy No. 4 -
copy No. 5 -
Copy No. 6 -
copy NO* 7’-
COPY ~QF ..,~z
copy No; “9 -
Copy i~om10 -
copy No. 11 -
copy 1$0.12 -
copy No. 13 -
Copy No. u -
copy No* 15 -
Copy No. 16 -Copy i!o.17 -
copy NO. 18 -
copy”No. 19 -
copy NO. 20 -copy No. 21 -
Copy i~o.22 -
COpy ~~0.23 -
copy NO. 24 -
copy NO. 25 -
COpy NO. 26 -
Copy NO. 27 -copy No. 28 .
COpY NO. 29 -
copy No. 30 -
Ship Strvc+mre Committee ;.“
‘iRear Admiral K. K. ,C’owa@yUSCG - Chairman”
Rear Admiral,R. L. Ijj:cks,SN (Ret.), Maritime Administration
Rear Admiral E; W. S~l”vesterjT13NJBureau of Ships
Capt. ?. N. Mansfield$”USNR, Military Sea Transportation’SerTfice
D. P. Brdwn,American Bureau of Shipping
Ship Structure Subcomr$ttee
Capt. E: A. Wright, USNY Hm’eau.of Ships - Chairman
Col. Jo’hnKilpatr~.c’k5SA, TransportationCorps
Comdr. E. A. Grarrbham$‘USN,Military Sea TransportationService
Comdr. D. B. Henderson, USCG, U.S. Coast Guard Headquart~~p
Lt. Comdr. M. N. P. Hinkamp5 USN9 13ursauof SkipsLt. Comdr. E. L. Per~, USCG2 U.S. Coast Guard Headquarters
W. G. Frederick7 Ma@time Admi~i@tration
Hubert Kempel, Military Sea T~ansportationService
J. M. Croviey, Office of ~a,valResearch “
M. J. Letich, American Bureau ’bfShipping
L. C. Host, American 13ureauof Shipping
E. M. ItiacCutchean3r., Eureau of’Ships
V. L. Russof Maritime Administration
Firm Jonassen, Liaison Repiesetitative,hLRC
E. H. DavidsonY Liaison R.epre,sentative,LSI
W. P. Gerhartj Ltaison Represe”ntative,AISI
Wm. Spraragen, Liaison Representative,WRC
Julius Harward, Office of Na’valResearch9 Alternate Member
Charles Hochl Military Sea TransportationService,
Alternate Member
W. E. Magee$ U. S. Coast Gtirdj Alternate Menber
J. B. Robertson, Jr., U. S. Coast Guardz Alternate MemberJohn Vasta7 Bureau of Ships? Alternate Member
Edward WenkY David Taylor Model Basin, Alternate Member
R. E. Wileyl Bureau of Ships, Alternate Uember
U. S. Army
Watertown Arsenal Labomtory, A13xu S. V. Arnold
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 5/58
-ii-
~opy lTo.31 -copy No. 32 -
copy NO* 33 -
Copy No. 34 -
Copy No. 35 -
Copy No. 36 -
Copy No. 37 -
Copy No. 3S -
Copy No. 39 -Copy No. @ -
Copy No. 41 -
Copy No. 42. -Copyi?o. 43 -Copy No. ~ -
Copy No. 45 -
Copy No. 46 -
copy Noa 47 -
Copy No. 48 -Copy No. 49 -
copy No. 50 -Copy No. 51 -
copy NO. 52 -Copy No..53 -Copy No. 54 -
Chie2j Bmeav. of Ships, Navy Departmerlt
Capt. R. };.Larnbert,IEN~ Philde3phi_a Naval Shipyard
Capt. A. G. Mumma$ USN, David Taylor Model Basin
Capt. C. M. Tookej USN, Long Beach Naval Shipyard
Comdr. H..G. Bowen, USN5 Code Re=.3,B~eau of Ordnance
Comdr.R S. Mandelkorn, USN~ Portsmouth Naval Shipyard
A. ArihxU@aU~Bureau of Yards and Dock6
A. G. Bissell, Bureau of Ships
Joseph llrock~Da~ld Taylor Model Basin
J. W. Je&ins, Bureau of ships
Carl Hartbower, Naval Research Laboratory
Noah Kahn~ New York Naval Shipyard
A. S. Martkens, Bureau of Ships
O. T. Marzke, Naval Research Laboratory
J. E. .McCambyidge,Industrial Testing Laboratory,
Philadelphia Naval Shipyard
W. E. McKenziej Metallurgical Branch, Naval Gun FactoryWm. S. Pe115ni5 Metallwgy Division, Naval Research Laboratory
N. E. Promisel, Bureau af AeronauticsDr. Wm. J. Sette~ I?avidTaylor Model Basin
Theodore L. Soo+oo, Bureau of Ships
Naval Research Laboratory
Naval Research Laboratory$ NechanicalSection
Naval Research Laboratory$.MetallurgicalSection
Post Graduate School, U. S. Naval i+ca dem y ~
Copies NOS. 55 and 56 - U. S. Navai Engineering ExperimentSection
~~Copy No. 57 - New York Naval Shipyard, Material Laboratory
Copy No. 5$ - Industrial Testing Laboratory
Copy IJoa59 - Philadelphia Naval Shipyard
Copy No, 60 - San Francisco,Naval Shipyard
Copy No. 62 - David Taylor..ModelBasin,Attn~ Library
Copies Nos. 62 and 63 - Technical,Library5Bureau of Ships, Code 364
copy No. 64 -COPY IiOa 65 -Copy No. 66 -
copy lto.67 -
Copy No. @ -
Copy No. 69 -
Copy No. 70 -copy No. 71 -
,,,“
U.S. .CoastGuard . .
Capt. R. B. Lank, Jr., USCG
Capt. H. C. Moorej USCG
Capt. G.A. Tyler5 USGG,,,,,
Comdr. C. P. Murphy, USCG
Testing andDevelopment Ditision
U. S. Coast Guard Academy, New London, Con.n. ‘ ‘
U. S. Maritime Administration
Vice Admiral E. L..G@&ns,,I!EN (Ret.).E. E.,Marti,nsky .
.. . . .
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 6/58
- iii -
Gtimmitteeon Sh,ipS-heel
cWYcOpycOpy
cOpycOpy
co~y
cOyy
cOp’y
copy
cOpyCoj?y
cOpy
cQpycopycopy
cOpy
cOpy
co~y
cOpy
cOpy
cOpy
cOpycOpycmy
cOpycCq3ycopy
cOp ycq)ycOpy
cO“pycOpycOpy
copy
cOpycOpycopy
No. 72 - P. E. Kyle , - Chairman
No. 73 M C. 3. Herty$ Jr. - Vie@ CkairmanTJoo 74 - Wm. n!.Baldwin, Jr.
No., 75 - C. S. Barrett
No. 76 - R. M. Brick
No. ‘~:, J. M. CrOWhY
No. 77 - s. L. ~0~.
No. 78 - IL W. Lightner
No. 79 --R. F.‘Mehl
No. 80 - T. s. Washburn
No. 19 - Finn Jonassen - Technical Director
No. 81 - A. Nuller . Consultant
‘@rnbersof Project Ad,tiJso~C6mittee~ SR-98, SR-99, SR--~OO,
SR-10$, SR-109, SR-11O and SR=311 (not I.istedelsew?~e~e)
NO= 82 - R. HA Aborn’
No. 83 -E. A. Anderson
No.:84 - L. C. Bibb&r ,..
No;” 85 - Morris Cohen,.
NO. 86 -W. C. Ellis
No. 87- M. Gensamer,,
No. 8$ - M. F. Hawkes
Noa 89 -W. F. HeSS
Nom 90 -l”!OR, Hibbardj Jr.
No. 91 - C.,E. Ja.cksofi ‘,..
No. 92’-,J+ Low, Jr.
No.. 93:’-H. IVePierce ‘
No;;’”9&-“W. A. ReichNO~;’~~”=C,oE~ Sims : ‘. .
No. ,,9&-~o D. stout
Ng~ 97’-LJ. G. Thompson!,
NOA “’98- B. G. Johnston, Welding Research Councilj Ltaison “’Noa 99”A’-TT.. Woodingj Pb31a~elphi’aXaval Shipyard
Commi t t e e on Res3dual Stresses “
No. 100 - J. T> Norton - Chairman ,
No. ’74- Wm& M. fia~dl~iny.Jr@
NO. 101 - Paui Ffi@ld
No. 102 - LeVan,Griffis ‘
No. 103’- K. lietnd~hoferNo. 104 - Daniel Rosen-thal
No. 19 - Finn Jonassen - Te@nZcal Director,.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 7/58
-TV-
Representatiws of Amrics,n Iron and Steel Institute
copy NO* 105 - c. M.Copy Noo $4- L. C.
copy No. 73 - c. H.
copy No. 106 - E. c.
copy No. lo7 - c. A.Copy No. 108 - Harg
Committee on hhnzi%cturing Prchlems,.. ,‘:~<<;,$
Parker, Secretary, General Technical ConmiLttee5)A~S1
Bibber, U. S. Steel Co. ,,,,
Herty, Jr.~ Bethlehem Steel Co, ,,!!’,
Smith~ Republic Steel Corp. ,, ~~ , , ‘,,.
... : ~{
‘WsldingResearch Council ,!.,i.l,,,,,
Adams copy No. 109 -,LaMotte GroWr , .~
Boardman Copy ito. 22,- Wm. Spraragen I:>
,. l’,,!,,,1,
copy NO. llo -,~Dr,D.S. Bronk$ President,,,!Na$ional’A~ademyf Sciences
copy No. m“- ,D~.C. R. So~erberg, Cha@a,n~,Di,vn. En@eering & Industri&l
Researeh~’National Research Council “
Copy No. 19- Finn Jonassen, Technical Directory;fiommi~teeon Ship Step)lGopy No. 112 - H. H. Bleich, Columbia University “
copy No. 113 - S. T. Carpenter, Investigators Resed~chProje@s SR~i~a~~~SR-llS
Copy No. 74 - Wm. M. Baldwin, Jr., Investigator,,Research Projects $R@9and
SR-111 i ‘. ., ,,,.,Copy No. XL/$- C. B.
Copy No. 115 - P. J.
Copy Nc. 1.16- M. L.
Copy No. 117 -C. A.
Copy No. 79 -R. F.
copy No. 7.6 - R. h!,
COFy }(0.11#- C. H.
Copy No. 119 -E. W.
Copy No. 120 - Carlocopy No. 30 -s* v.Copy No. 18 - V. L.
Copy No. 1.21,-B. A.
Copy No. 122 M G..S.
Rieppel,’Investig;tor,”ReseaTehLr6,j;ctSRWIOO: , , !
Williams, Investigator, Research ~&eject SR*106
ElZinger, Investigator; Resei;ch ProjFct SR-10~ .”,,
Mehl? Investigators ReseatichProject,SR-108 ., ,,
Brick, Investigator, ResearCh.p.rojeCtiisR-1~9 .
Lorig~ Investigator, Research Project SR=+~10 :;.,,,
Suppiger3 Investigator, Resear{h Project SR-113 ~,,,.
Riparbelli, Investigator, Res,e~rchl?~~jectSR-1~3~~,,.,Arnoldj Investigator, Research,P~oj@ $~~1~ ‘ :
RUSS09 Investigator, Researcb P,rojpct~SR-!>7‘I, :
Hec,htman,Investigator, Resear$h.pro~9c-~SR-U9 ,
ll@halapovY Investigator,Res.edrch~r9jqctSR-120
Co;+ No. 123 - L. Crawford, Investi~ator. Research Pro.lectSR*121
Co~>es Nos. 124 and 125 -~rmyAir Material Command, W;ight Field
Copy No, 126 - Clarence Altenburger, Great Lakes steel Co.
copy I?o.u? “ J. G. Althouse, Lukens Steel Co. ~~~~ ,,-,,
CoFy No. X28 - T. N. Armstrong, The I&t6inationall{ickelC., Inc. ‘;r
Copy No. 129 - R. Archibald, Bureau of Publi6 Roads ,
Copy No. 130”- F. Baron, Northwestern University “,
Copy No, 131 - British Shipbuilding Research Association, Attn: J. C. Asher, Sec.,’
copy NO. 132 - J. P. Comstock~ Newport News shipbuilding &Drydock Co.Copy No. 133 -1!. J. Durelli, A,~our Research Foundation
copy NO* 134 -W. J. Eney, Fritz Engr. Laboratoryj Lehigh University “
copy No. 135 - S. Epstein, Bethlehem Steel Co.
Copy NO. 136- J. H. Evans, Massachusetts Institute of Technology
-- --
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 8/58
-...
copy No. 137 - A. E. Flanigan~ University of ca~ifo~~~a ~,,
Copy No. 138 - M. G. Forrest, Gibbs & Cox, Inc. ,,
Copy No. 139 - L. E. Grinter, Illinois Institute of Technolo~y ‘ .
copy No, 140 - W. J. Harris, Jr., Metallurgical Adviso~ Board l~C. :
Copy No. u - T. R. Higgins, American Institute of Steel Cons{rtiction
Copies Nos. 142 through 166 - E. G. Hill, Bri+ish Joint Services Mission
(Navy Staff) ,“
Copy No. 167 - 0, J. Horger, TirikenRoller Bearing Co.
Copy No. 168 - L. R. Jackson, Battelle Memorial Institute ‘“
Copy No. 169 - B. J. Johnston, University of Michigan
Copy No. 170 - J. Jones, BethlehemSteel Co.
Copy JTo.171 - K. V. King, Standard Oil Co. of California
Copy }?0.172 - E. P. Klier, University of Maryland
Copy No. 173 -W. J. Krefelds Columbia University ‘:
Copy No. 174 - J. B. Letherbury, New York Shipbuilding Corp.
Copy No. 175 - E. V. Lewis, George G. Sharp
Copy No. 176 - R. S. Little, American Bureau ’ofShipping
CQpy No. 177 - C. W. Macgregor, Massachusetts Institute of Techno?.ogy
COpy NO. 178 - R. C. Madden, Kaiser Co., Inc.Copy No. 179 - G. M. Magee, Association of American Railroads
Copy No. 180 - J. H. McDonald, Bethlehem Steel Co., Shipbuilding Divn.
copy No* 181 - NACA, Attru Materials Research Coordination, NaWDepartm@nt
Copy No. 182 - N. M. Newmark, University of Illinois
Copy No. 183 - Charles H. Norris, Massachusetts Institute of Technology
Copy No. 184- E. Orowan, Massachusetts Institute of Technology
Copy No. 185 - W. G. Perry, RN, British Joint Services Mission (Navy Staff)
copy NO* 186 - Walter Ramberg, National Bureau of Standards
copy No. 1~7 - L. J. Rohl, U. S. Steel Co.
Copy No. 188 - W. P. Roop, Swarthmore College
Copy No. 189 - H. A. Schade, University of California
Copy No. 190 - Saylor Snyderg U. S. Steel Go.
copy Now 191 - E. G. Stewart3 Standard Oil Co. of New JerseyCopy No. 192 - R. G. Sturm, Purdue University
Copy No. 193 - K. C. Thornton, The American Shipbuilding Co.
Copy lsro.9/!+ A. R. T~ano, Case Institute of Technology
copy No. 195 - R. W. Vanderbeck, U. S. Steel Co.
COpy NO. 196 - T. T. Watson, Lukens Steel Co.
Copy No. 197 - Webb Institute of Naval Architecture
Copy No. 19S!- Georges Welter, Ecole Polytechnique Institute
Copy No. 199 - G. Winter, Cornell University
copy No. 200 - L. T. Wyly, Purdue University
Copy No. 201 - Division of Metallurgy, National Bureau of Standards
copy No. 202 - Transportation Corps Board, Brooklyn, N. Y.
Copies NOS. 203 through 207 - Library of Congress via Bureau of Ships, code 324
Copy No. 208 - C. A. Zapffee, Carl A. Zapffe Laboratorycopy NO. 209 - File Copy, Committee on Ship Steel
Copies Nos. 210 through 212 - Bureau of Ships
copy No. 213 -
Copy No. 214 -
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 9/58
cOpy No.
Copy No.
Gopy No.
cOpy No.
cOpy No.
Copy }Jo.cOpy No.
cOpy No q
cOpy No.
cOpy No.
cOpy No q
215 - , ,..,: #,-216 -
,.h .~17m ‘ .: ,.. ,’ .:
2u3 -219- .,, ., .,, !’, ”
220-’ “ ‘,., ,
; ,’221 -222 - . .....),1,.223 - .,, .
224 - ,., . .,,225 - ,,. :! .,
,,.. ,:..’ ,,
(Total Copies _ 225) : ,,
,-
.“ ,, ,”,, .,.
. . ’ , .. . ,,,.! . . ‘y
.-, ,, ,,. ,
,’:.
,.
. . . .
,:. ”,,
,.,, . . .
,’ ” ’,
:.
~,,- . .
,,,,
:’.
,.
‘4.<
,+
‘,, .
.
,. .,,‘.
. ..
.. . . ..
... ,
.-;”
..4.
,.
,.
..,’
;,
,1’
...
,,
‘, # .;
.< ”::,,
,,
,
., .,,:
,,
.,,
,, ,:
,.
.,,,,
—
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 10/58
Codxmts
-,,
3. Gaerd analysis of?wo-cdl trhactum
ha Sdutbn d’ the Mi%mn%$al @quatlm for constant
mmmt M
@sndix
M - M
20 “2’4
25 - 31
32 * 40
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 11/58
1. INTRODUCTION
l?@m model tests, as well as from measurements on ships, it has reamtly
become apparmt that Nsvier~s
ship section cf a ship with a
that the strain distributions
hypothesis does not always hold tme for the mid-
1)
long deck house. The tests made byHolt indicate
in the hull and in the deck house, each are straight
lines, but that +.hereis a break at the deck level, where the superstmcture is
2)
offset. A similar result was found in the tests on the S. S. ‘tPresidentWilson~~
where a very pronounced break in the strain distribution occurs at the promenade
deck level where the deck house is offset. It is further significant that god
agreement with Navierts theory was found in tests by Holt on a different mtiel,
where the superstncture was not offset at
It appears fr~m these tests that
combined section of hull and/leekhouse, if
set from the sides of the hull. The tests
the strength deck.
Navierls theoq is not valid for the
the sides of the deck house are off-
seem to indicate that, instead, tb
hull and thedmck house act as two separate beams, for eaoh of which Naviar~s
hypothesis applies; these two beams nre, of course, not i.ndepmdant of each
other, but forced to act together to a certain extent by horizontal shear ferces
and by vertical forces which act between the hull and the deck housa.
Starting from the assumption that the hull and the deck house may b
considered as beams, to each of which, separately, Navierls hypothesis is ap
plicable, this~psrhwill derive expressions for the deflections and stresses in
the hull and deck house assuming constant section of hull and house; it will h
seen that the theoretical stress distributions found are of the type observed in
the tests.
1)
2)
M. Holt, Structural Tests of Models Representing a Steel Ship Hull with @minum
Alloy and Steel Superstructures, paperpresented at the Il!arch949 Meeting, New
England.Sec~lon of the Sot. of lkval Arch. and Marine Engrs., D. 13 and 14~ Figs.
8 and 9.
J. Vasti, Structural Tests on the Passenger Ship S. S. President Wilson - Inte~
action Between Superstmcture and Main Hull Girder, paper presented at the Nov.
1949 Meeting, Sot. of Naval Architects and Marine Engrs., p. 17, Fig. 27.
—.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 12/58
2-
The results of this theory concerning the stresses in the mid-ship
section can be arranged in simple tables which permit the prediction of the de-
viation from the conventionally
method is equally applicable if
num.
assumed straight-line stress distribution. The
a prt or all of the deck house consists of alumi-
Before analyzing the full problemwe will consider in Section 2 ashpli-
fied shipts structure in which the action of wertical forces between hull and deck
house is neglected. This simplified structure does not descrik the actual con-
ditions in a ship, “fiutecause of its
the play of forces; the understandtig
problem in Section 3. The simplified
relative simplicity it is easier to study
gained is of value in treating the full
apnroach in Section 2 may be considered a
generalization ofllavgaati% theo~ of a beam attached to a deck plate subjected1)
to tension or compression due to the bending of the hull.
tial
mate
8ectioti4 and 5 are devoted to the solution of the general differen-
equation for two special loading cases, and SectIon 6 derives an approxi-
methcd for using the results of the preceding sections for any type of loading.
Section 7 contains a table of coefficients and a list of the formulas
required for the determination of the stresses in the mid-ship section, together
with a numerical example.
The final Section $ contains a review of the theoretical results OIP
tained, discusses a test program to check these theoretical results, and consi-
ders futbar research necessary to fo~ul~te design stindafis for ship auw-
structures.
l)ti,kvgaarcl~,Trans. Inst. Naval Architects, Vol. 73, 1931-
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 13/58
-3-
2* ANALYSIS OF A SIMPLIFIED TWO-CELL STRUCTURE.
We consider the structure shown in Fig. la, a hollow box beam with two
calls. The lower box, the hull, is of length L, while the upper box, the deck
andhouse, is shorte~~of length I ; both boxes are assumed to be of constant cross
section. The cross sectional area and the monent of inertia of the uppr section,
Fig. lc, and of the lower section? Fig. ld, are A,, 11, and Az, 12, respectively;
the distances of the respctive center~ of gravity from each other and from the
deck are a, u,~ and IX=& , respectively see Figs. IbJ c and d.
In this Section we make the important simplifying assumption that the
its supports have
deck A B, Fig. lb, andhno stiffness,and ~will not resist any relative ve~
tical movements between hull and deck house. This assumption is of course not
justified ix any real ship and we will abandon this assumption in the next Section.
We consider the structure just described under the action of vertical
lmds and buoyancy acting on the hull only, producing bending moments M inthe
externalvessel. We do not assume any~loads to act on the deck house.
Take a section at the distince z from the center of the deck house, and
consider the free bmly diagrams for
and direct forces in the deck house
Moments M, and M= are positive if
the deck house and hull,
and hull are Ml, N, and
they produce compression
Fig. 2a. The moment
M~, N2 respectively.
on top of deck house
or hull, and direct forces N, and N are positive if they create tension. The2
external lo~,dsand lxbyancy acting on the hull to the left of the section have a
moment M; further, a shear force T of unknown magnitude will act on the underside
of the deck house, and a
hull. The shear force T
brium of the portions of
similar force T will act in the opposite direction on the
is counted positive if it acts as shown in Fig. 2a. Equili
deck house and hull in Fig. 2a requires the relations
N,=-T,
/v== 7,
Due to the assumption that
h-’7,=- a=,T, (la)
~== k f -aKeT [r& )
Navierts hypothesis is valid for the clackhouse
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 14/58
4-
and hull seprately~ we can determine the stress at any pofit at a distance x or/
Xa from the respctive center of gravity. Counting tension stresses & as posi-
tive, we have in the deck house
T0-.—-+
ati, T—x 9 (2a)
1
and in the hull
-1 A, ~
M~= f - — ~z
2 12
The stresses q and q at the
and Eqs. (2) furniaht with =,= - ~M, ,
Eq. (3) dmterminas
junction of house and hull must be alike,
Introducing this value of T in~o Eqs. (1) and {2) we can determ~e
at any
end of
to the
rivets
point.
T was defined as the total horizontal shear force acting
<4)
moments and stnsses
between the left
the deck house and the section at z.According to Eq. (4) T is ~oportional
dTmoment h!,and the unit horiz~tal shear ~Z which w fll ~ trans fe~ed W
or welds from the hull to the deck house, will be
dT .QW2 It
Y (.5)Tz Zq + ~-z= + at (~21,-k q212)
dMxx
where V* ~ is the shear in the structure. However it will lm noticed that at the
end C or the deck house the shear T is not zero, but equal to
a ofp ~,
~ = == IIMC (6 (z )
~z + t 2 ~ a~(w221,+ %2Q
A, >2
At a point slightly to the left of @nt C in Fig. za there is no deck house and
therefon T =o; this meana that in addition to the distri~~utedshear ~~ according
to Eq. (5) there must be a concentrated horizontal shear force T accofiing ‘o ‘q”
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 15/58
direction of positive shear forces in Fig. 2a.
It % obvious that the conce:ltratedshear
in any actual structure t:heiroccurrence is due to
were neglected when wc assumed liaviertshypothesis
Ien<th of the deck house. In real~t;ythe forces T=
forces T= and T. can not exist
tinefact that
to he correct
and T= will
seives over a fintte distfi::sajresumably of the magnitude of the
house. This
represent~:lg
Of the iiOUSe
#
shear lag effects
for the .%11
distribute them-
depth of the deck
2c, the shdded are
T= and T= . This means that the stresses In the vicinity of the end
found fram Eq. (2) are incomect; but according to St. Venantls Theorem
the effect of the
of the st-ructure.
l[Rwill
direc~ forces N,
simplificationwill not affect tilestresses in the center portions
now proceed and ohtaim expressions for the moments U,, M2 and
and lip. Ti:eseexpressions become somewhat simpler if we intr~
(7Q)
(7L)
y will h~ c.alleZtke size fact~~; it is a measure of the size Of the deck house
in relation to the hall. We will also require the moment of inertia I of the total1)
s~ct,ionconsisting of hull and deck house; I can be expressed by the moments of
1) Sse Appendixy Eq,.(a).
(8]
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 16/58
-6
Making use of these notations the following expressions are derived in
the appendti:
These expressions could be used to determine
(%)
(?C>
the stresses ~ and ~a
in the deck house and hull; each of the three expressions consists of two terms,
the first term being the value of the respctive N or M if Navierls Theo~ would
1)be applicable to the entire sectiow Instead of using Eq. ($), we CaU therefore
express the the actual stresses as the sum or the stresses q according to Navieri
and a correction a~P
e= cN+nm (Io)
Navierls stresses &M can be found frcm the conventional equation
MWN= - — x
z((?)
where x counts from the centroid of the entire section, Fig. 3. Tinecorrective
stresses AT , and ACE in the deck house and hull, respectively, are
where AN and AM are corrections of N and H given by the second terns of Eqs. fq):
1) See Appeu3.lx,Eqs. (b) and (c).
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 17/58
7-
The computation of the stresses ~M and
is a simple matter. Fig. 4 shows these stresses,
stress computed for the example in Section 7.
A& from Eqa. (11), (12) and (13)
separately, and also the total
Due to the fact that the structure analyzed in this section was simpli-
fied by omitting vertical forces acting between hull and deck house the results an
of limited significance; the typical %reak h the stress distribution at deck level
is, however, already there. The value of the above analysis lies in the fact that
the more ac:urate anau~sis presented in the next Section shows that the actual
stresses, can he expressed in the form ~= ON+~Aflwhere ~ is a numerical fa~
tor depmding on the various dimensions of hull and deck house and on tho tiBffness
of the bulkheads.
It might be added at this point that the reasoning presented would be
fully applicable also if hull anddeck house would not he of constant section. All
formulas derived in this Section remain unchanged, except Eq. (5) for the unit
dT
‘hear 7Z ;
on the right
in Eq. (5).
expressed as
rection AU
when derlvlng this equation by differentiation of Eq. (4) the fraction
hand side would no longer be a constant, resulting in an added term
The important result, that the stress at the midship section can be
tinesum of the stress uN
? according to Navierts theo~~ and the cor-
remains valid.
..
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 18/58
3. GENERAL
We consider again
AN&LYSIS OF TW@2ELL STPJJCTUBE.
the structure indicated in Fig. 1; hull and deck house
are assumed to be of constint section as in Section 2, but we now want to take ac-
count of the fact that the deck house cannot move freely in the vertical direction
in rel~tion to the hull; instead, we introduce the more realistic assumption that
any rehtive displacement of deck house will be resisted by internal vertical
forces required
In other wotis~
Under
shown in Fig. 5
tions yl and yz
ta deflect bulkheads or transverse beams supporting the deck house;
we consider ths deck house as beam on elastic supports.
the action of external vertical loads and buoyancy the structure
will deflects and we can describe the deformations 3Y t~hedeflec-
of the center lines of the deck house and hull. respectively.
Fig. 5. In o~er to exclude motions of the entire vessel as a rigid I@y, we de-
fine y, and yz not as the absolute displace~ents} but as the~lative displace-
ments measured from a straight line CD rigidily coniected to the Ix1lI. As result
of this definition the displacement yz of the centrofd of the hull at points C
and D must always be zero.
We assume further that the stiffness of bulkheads or deck beams resisting
relative vertical displacements of the deck house is constant for the full length
of the deck housej the magnitude of the stiffness being given bjJa spring constant
K. K is defined as the force per unit length of deck house required to prddce a
relative deflection equal to one unit of length, Fig. 6J the vertical reaction be-
tween hull and deck house will therefore be K(y,-yZ) per unit of length. In an
actual ship the bulk heads or deck barns will have a spacing s, and the constant K
will be the force required to deflect one of the bulk heads or deck beams, divided
by this spactig S.
The structure analysed here consists therefore of two beams having areas
A A21S
and moments of inertia I, and 12 ; the two beams are connected along CD
in such a way that both, horizontal shear forces and vertical reactions can be
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 19/58
-9-
transferred. Fig. 7 indicates a general typ of loading for the vessel, including
the shear and moment diagrams. We assume Navieris hypothesis to kw valid for the
lhilland for the deck house separately, and there is no moblem concerning the
determination of stresses fowand aft of the deck house; we can, therefore, restric
our analysis to the center portion CD of the stmctme. This center portion, in-
dicated in Fig. 8, will be under the action of vertical loads p, on the deck house,
P2 on the hull (which includes buoyancy), shear forces SC ~S3)
and moments Me and
Mn ./
We will now proceed to obtain the differential equation for the two de-
flections y, &nd ya describing the deformation of the structure; these diffmen-
tial equations can conveniently k obtained from the Theorem of Stationary PotentUll\
Ener&’
the total
potential
this theorem states that the deformations of ariystructure are such that
potential energy U of the system is a minimum. In the present cage the
energy U consists of the internal strain energy V, and the potential UW
of the external foroes p , pz , S= , S= , Mc and id=. The total potential eners~2)
I
U=v+uw is,7
J[u=; ELy,V2
“z+ E12y2 + EIA (R,y,”+ ti2y= “y+ K (y, -y=)z- 2p, y / -~p, Y,]c&
2-2
+ @z’J~ - [s Y,]iz (f4-— -—.? 2 “3)
Using the symbol ~ , and the rules of the calculus of -~ariation, u will
be a minimum if
[U=o (M
4)from which we obtain Eulerts equations, which are in this case two simultaneous
differential equations of the fourth order for the unhewn functions y, and YZ :
E (z, + ti,21Ax’” ~cr, ocz EXA Y,’v - Kz +
1] F. Bleich, The Buckling
2) See appsndu, Eqs. (1) and (w).
3) The Buckling Strength of i!etalStruct~.res,Apmendlx to Chapter IV contains a pre
sentation of the calculus of variation.
4] See Appen2ix, Article 5.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 20/58
-1o-
1)Tha procms of variation also furnishes the following boundmy conditions, which
are required to d.etercin~the 8 arbitrary constants which will appear in the general
1solutions of trhedifferential equations (16). For R= +: andz=-~ ;
=0(f“7q)
.%s
E(16++4J y“ + ot, d #T4 ~“~ O [17b>
(!7C)G$ +ZAX” + E(I= + t i2 2 z& J ~“= -M
E@ h r& 4 ]2 y ’” i- m1W2EIA y=’” * o c17d)
The meaning of the ~irst of these boundary conditions is obvious, but the
other three require physical interpretation. The moments M! and U= of the longi-
tudinal stresses in the deck house and hull can be expressed by the usual relation-
/1
ships M,=-EX,~ and Me= -E~p Y=”; Eqs. (17%) and (17c) can tharefore & re-arrangeci:
M, = {ruq], EIA (=,y,”+ =.. Y2”)
M2 = Ad+ =2EZ4 (a, ~’t+ti=yz”> <18&>
These equations indicate that the moments M, and Me at the end of the deck house
are not equal to zero,
2)
appendix it is shown
actirlgas indicated in
and U , respectively, as might have been expected. In ths
that the horizontal shear T between deck house and hull,
Fig. 2a, is
Eqs. (13) become therefore
A#f= -a=t~, C20aj
M== M- acc=T , CZob>
and these equations are identical with Eqs. (1) for the simplified structure consi-
dered in Section 2. It must be remembered that Eqs. (20) ap~ly only at the end-
points of the deck house; hut at these po~nts the moments and stress~s as determined
in SectIon 2 occur and the finding that concentrated horizontal shear forces must
be presumed to act at these pbints is valid. Th~ magnitude of thase forces is
1) See Appendix, Article 5.
2) Art,icle6.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 21/58
-11-
!given by Eq. (19) for z= + /2 .
The fourth bourLdaw condition, Eq. {l?d) expresses the fact that the
shear force at the end of the deok house must vanish; this can be seen by cow
Prison with the expression for ~,heshear force, Eq. (t), derived inthe 6P
pmlti.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 22/58
-12-
4* SOLUTION OF THE DIFFERENTIAL EQUATION FOR CONSTANT MOMENT M.
We consider first the simple case that the loads ~,, ~, and the shears
se and S= are zero, the only loads being fi~== Mm’~ . From Eq. (16) we ob
tain the two simultaneous homogeneous differential equations of the fourth order:
The general
satisfy the
solution of these aquations contains eight arbitrary constants to
eight boundary conditions (17). The problem considered intils section
is symmetrical with respect to the orig!n of the coordinate z, and using only
symmetrical functions, the general symmetrical solution will contain only four ar-
bitrary constants. This general solution is
y==q+c2z2 +C3ti~z&~z+C4r-~Z~ ~z (~2a)
(~zb]&=q+ q+q+~z+~p-p gl -J -z hp
whsra
~== 7 ’23)
whi.leY
is tha size factor previously defined, Eq. (7b). The fact that Eqs. (22)
are soi~lt,ionsf (21) can be established by substitution.
Introduction of Eq. (22) into the bounda~~ conditions (17) leads to 4 linear
equetions for the constants C, to C4 . The values of cz,c~,czare:
cz=-~ (24a)
2 Ez
C3= ~,/dtM
(?46]
2X(’~+/LL)E~z~ +-w--l,)
I Jo!v--f (Z4C)q’ 272
(I+>) E (c@, +-=, ~p)
The value of C, is not.listed as it will not be required for the cmrnoses of this
paper. @ and ~ in Eqs. (24) are defined by/ I
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 23/58
where
The stresses in the mid ship section can be computed from the expres-
sions for the moments N,, ~zand direct forces ~,, W2 t
E q
q=- N2=— (YM
{f‘( + w2y2”)
Q
(for derivation of Eq. (27%) see appendti, Eq. (7).
At the midship section, we have z=O , and differentiation of Eq. (22)
furnishes
By substitution ofEqs. (24) into (28)3 and Eqs. (28) into (~7)? the following
Comparing Eqs. (29) and Eqs. (9), found in Section 2, we see as only dif-
ference the factor ~, apparing in the second term Of each of the Eqs. (29). In
Section 2 we had found that the first term of each Eq. {9) represented the re-
sult of Navierts theory, while the second term was a correction. The refined
theory in this Section furnishes a similar result, but the “correctionf’found in
Section 2 is to be multiplied by a factor @ . To compute the stresse~ accordingr
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 24/58
-3,4-
to the refined theory we can use the relationship.
(7”
where Navieris stresses ~N ~
forces AN, , AN2 and moments
and (13) in Section 2.
C7-J’, AC , (30)N
the corrective stresses ~~, A% and the corrective
~M,>d M’p are to be computed from Eqs. (11), (12)
This result , Eq. (30), is
deviation 0~ the stress distribution
non-dimensional factor # which we
~, as function of the ~rameter u
surprisingly simnle; it indicates that the
from Navierts is indicated ~he value of the
will call “Deviation F~ctor”. Fig. 9 shows
defined byEq. (26). U is a function of
the dimensions of hull and deck house and of the stiffness factor K of the bulkheads.
u is proportional to the length of the deckhouse, and increases with rising
value of K. According to Fig. 9, ~= I for u=Q and decreases for rising
values of u ; for u>2 the factor z, is a small positive or even negative
wamberj indicating that for such values Navierts stress distribution at the mid-
ship section is approximately corre$t.
Eq. (30) was derived for the mid-ship section, Z=O . The solution of
the differential equations
any other section too, and
found above permits the computation of the stressns at
a similar relationship
U’o--+$,(z).w (31)
exists all along the deckhouse; howevery the value of the deviation factor is not
the same as at the mid-ship section z=Q ; $ (z) fs a function of U and alsoi
/f the ratio z / , defining the location of the cross section. Fig. 10 shows
@ (z) as function of the ratio z/1 for several values of u . The value ofI
the deviation factor at the end of the deck house z j=o.sis 1 for all values of/
u;
of the
of the
for large values of u ,@{z)is veryI
deckhouse, indicating the validity of
deck houme.
small everywhere, except near the end
Navieris theory in the center portions
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 25/58
-15-
5. SOLUTION OF DIFFERENTIAL EQUATION FOR EQUALLY DISTRIBUTED LOADS.
In this section we consider the case of equally distributed loads
and pz , acting on deck house and hulll respectively, while the moments at
end of the deck house are A4C=LI= = CI . Equilibrium requires external
shear foroes
: (~+p?)c=-sD=–
at the ends C and D. The moment in the mid-ship section due tothe loads ~, and
pz is
A+l% ~’Mp=—
8(32)
The loading bing symmetrical, the general symmetrical solutions of Eqs.
(16) are:
(334
where ~ is defined in Eq. (23).
1)The boundary conditions (17) furnish
stants C in Eqs. (33). TO com~te the stresses only Cz, C3 and ca are required:
the values of the arbitra~ con-
&ifpc=-—z 2EZ ‘
1) It should be noted that M In Eq. (17c) is in this case zero.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 26/58
$ .2 t%$uwdu
2 =/ 4-L71u w uu+/ 4AALmd . A u ‘
~1u .= is defined by Eq. (26).
(3SL)
EqS,(34) are quite similar to Eq5,[24) and the further comoutition fol-
lows the pattern of the preceding section; the only difference is that instead of
the deviation factor *, a factor da apnears. The strasses at the mid-ship sec-
tion 2=0 arez
c= WN+$2AU (36)
Eq. (11), (12) and (13) are to M usad to compute C; the moment M in theseCO*
putations is given ~Eq. (32).
Fig. 9 shows ~ and 52 as functions of the prameter u . In the
important range uc 3, *=is larger than ~, indicating that equally distributed
load produces larger deviations from Navierts stresses than a constant moment.
There is only a quantitative difference between the two loading cases considered
in this and the precedimg section; the spmwise variation of the deviation factor
for distributed load will be simiiar to the one shown in Fig. 10 for constant moment.
One result of the computations in this section desemes attention. While
the expressions (33) for the deflecti~s c~ta~ te~s dewnding on the l~ds p,
Qand pz , separately, the stresses apparently only depend on the sum p,+pe ,
which alone is requlrd to compute the moment M according to Eq. (32). The distri-
bution of the equally distributed load IxAween deck house and hull does not affect
the stress distribution; it does, haever, influence the values of the deflections
~ and )!2 . A transfer of equally distributed load from deckhouse to hull pro=
duces onlya change of the relative deflection ~-ye of the hull in relationship
to the deck house without anycknge in Wing stresses in deck house or hull.
1) The loads p, ~d Pz act on the deck house and hull, reswctively~
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 27/58
.~.
6. DETERKINATION OF STRESSES AT MIDSHIP-SECTION FOR ANY LOADING.
In the precading sections we have determined the stresses for two simple
loading cases; we are now gotig to show how a combination of these two cases can
be used for approximate determination of the stresses for any ty~ of loading.Fig. llb
Fig. 11 a shows hull and deck house of a ship, an~a general tyne of
moment diagram due to dernal loads and buoyancy. We assume that hull and deck
house are of constant section htween points G and D, while the section of the hull
outside these points may vary.
The law of superposition being
loading of the ship in three parts which
applicable, we may divide the total
produce moments in the shi~stmcture
as shown in Figs. llc, d, and et respectively. The first part shall produce a
constant moment ~ = ~ (~c+~=)for the full length of the deck house; the secondI
part shall be such that the moment diagram is a straight line between points C and
D, the moments at these points being # (Ale-MD )and - # (~c -~f$) , respectively;
and the thiti part shall be the remainder of the loading, such that the sum of the
moment diagrams in Fig. II c, d and e is equal to the actual moment diagram,
Fig. llb. Because of the choice of the moments in Fig. 11 c and d, the moments
in the last diagram,Fig. 11 e,at points C and D must always be zero.
We can now determine the stresses at the midship section for each of the
three parts separately, and add the results to get the total stresses.
For the first part of the loading the moment Wtween C and D is of con-
~;this being just the loading case considered intant value ~
can obtain the stresses ~1 due to this loading fron Eq. (30).
0=” r~z+$,AQ1 7
where the subcripts ~ indicate that m~ and ~~ Eve determ~ned
(12) and (13) using a value Iflx for the moment N.appearing in
Section 4, we
(37)
from Eqs. (11),
these equations.
Proceeding to t!i~second moment diagram Fig. lld, we notice that the mome
qt the m!dship section is actually zero, and from the fact that the moment curve is
antisynmetric we can conclude that the stresses at this section must vanish. This
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 28/58
“-l+
part of the load gives no contribution to the stresses at the
The third part of the moment diagram is of the ty~
midship section.
produced by the
equally distributed load considered in Section 5, where the moment diagram would
be a parabola between points G and D. If we approximate the actual moment dia-
gram @ a parabola with the same moment ~= ‘~-~’ at the midship section, we
can find the stresses at the midship section from Eq. (36)~
lrz”o- N=+42=QZ “ (38)
where the subscript II indicates that the moment ~= is to be used when comput!ng
~N and A~ from Eq. (11), {12) and (13).
The entire stress at the midship section will be
We can simplify this expression by introducing the values ~~ andbfl due to the
total moment M at the midship=ction. It is obvious from Eqs. (l-l),(12] and (13)
thet
Obsening M’ MI+%, substitution in Eq. (39) leads to
e= * + q~t’%w=N
AcM
The values of ~= “and ~= were defined at the beginning of this section~
~_ MC+M=(4 / Q)
I- 2
Mx= M-L’?= = M- “:A4D (4[h)
Substituting Eqs. (Q) into Eq. (40) we obtain f~ally
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 29/58
-19-
where the deviation factor ~ is given by
The values of ~, and 12 are defined W Eqs. (25a) and (35a); their numerical
values are also given in Table I in
deviation
and ye ,
procedure
The approximation used in
factor # for a geneml
the next section.
this section permits the determination
type of loading from two basic factors
resulting h a very simple computation procedure. An example of
is shown in the following Section 7.
From Table 1
of approximation to be
table shows the values
in the next section we can draw a conclusion on the
of the
5,
thfs
degree
expected ~rom the procedure leading to I@. (43). This
of the deviation facto~ for two distinctly different
~ for constant moment, and ~ for parabolic moments.oment diagrams, ,2
spite of this pronounced difference, the numerical difference titween #>
In
and
5 never exceeds 0.11; in the expression for the final stresses,2
~“wN+$ duN,
such a variation may produce variations of possibly 2Q-25% of the total stress,
see Fig. 4. Considering the fact that the difference between the actual moment
diagram and a parabola will h very much smaller than the difference between a
parabola and a straight line, we can conclude that the error due to the approxi-
mation can & expected to be less than 5%.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 30/58
A,
z,
A2
12
z
a
d ,a
tit a
z
K
M
MC
MD
The
area of deck house
inomerltf inertia of deck house
area of hull
moment of inertia of hull
total moment of inertia of hull and deck house togetl:er
distance between centroids of hull anddeck house, see F-!J:.J
distance of centroid of deck house frm deck, sec Fti;.le.
distance of centroid of hull from deck, see I&g. M.
length of deck house
spring constant expressing rigidity of ‘oulkheads or deck hea~s. K
is the force per umit length of deck house required to produce a
relative deTlect!on d me unit o.!lei@.},l)~twee-.i~lland deck house.
moment ir~shiprs st,~wctmreat center of deck house
mom.wt at forwtirdend of deck house
moment at aft end of deck house
;tresses g according b Navier~s theory BWN
~
lqN. “ — =
I (1)
where x is the distance of any fiber from the centroM of the section:,F-1::.,,
After determining thewlues
A,&1A . a2
A,+ Ae
/’I+C4,1A
12+ a21A
~e~e~-.l~tlefcorrective!monents and Tortes in deck house and hIIll
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 31/58
..
(im
where x and x= are the distaficesof the ffluersfrom the centroids of deck hauwI
and hull, Figs. lC and d.
After computing +he constant u,
(E )
the factors ~ aricl~ can ‘W read from Table I. The 1i3eviationfactorl~I 2 5
can thm be computed
The stress distribution at the midship section (a&’center of’deck house) i
(izzz)
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 32/58
.22.
TABLE l? VALUES OF DEVIATION FACTORS @, AND ~z .
T1.000
0.2 1.000
O*4 0.996
0.6 0.979
0.$ 0.935
1.0 I 0.8521.2 O.m1.4 0.573z. 6 o*Ql1=8 0.264
5$2
1.0001*000
o*9970.9&0.944
0.8720.7640.6290.4870.357
u
2.02*22.42.62*8
3*o3.5&o4*55*o
3,
0.144
0.054“o. m9-0.050-0.074
-o* 084-0.078-0.052-a 026-0. m 9
$
0.2490.1650.1030.0590.029
o*009-0.012-O*U-Q. olo-o* 00$
Numerical Example
We consider the model of a shipts structure shown in Fig. 12, sup-
ported at the ends~ and loaded by two concentrated loads near the center. The
1)
section proprties are:
Deck house: A,= 2.75 in: z,= 1 1.4 ;n .:
w , = 0.372 am = 3.48 k’.,? t= 7flk?f
Hull: Ap = %aq i~: 12= fGo.9 ;~.4
d2 =0.Gz%, a=. = 5.88 ,h ., K= 20;000 lbs . / io?
General:/
= 2TX 106 lbs.h: a=9 .3Grh ., A+= 375,000 ~Q.j&J-
M== 150,000 ;0.!4s., MD= 22s,000 IAL6S.
Centroid of hull and deck ho-usecombined: e = 3.33 in. (See Fig. 12)
FromEqs. [II): lA= 18S.1 in: I = 360.4 “tfi.4
Eq. (I) furntshes with x=9.33, 3.33, - 6.67 the
Table 2. From Eq. (III) we obtain
1) The pro~rties used in this example agree with those of
one Of the models Ilseci HoIt, except for the vali~eofascertained.
~ l:sted intresses w
the center portion of
K which co’~ldnot be
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 33/58
/=11.4 + 0.37Px f88.1
z o.25f.z’160.9 + 0.G28x 188.1
Introducing these values in Eq. (V), and using
-&i2, respectively, the values A a in Table 2
Using Eq. (VI) we find the constant
u=?
=.=
and with this values we obtain from Table 1$
xl= 2.52, -3.48,
were computed.
u,
Referring to Eq. (VIII), the fourth column of Table 2 contains the values
$A d= 0. wm f~~d t~~ la st c~lu f i l~l the final computed stresses.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 34/58
1Fig. 13 shows the computed stresses and the stresses measured by Holt
on a model of similar cross section. The close agreement,btween the cam!~>~tefi
and measured stresses should not be construed as quantitative confirmatfcm of
the theory presented, because unfortunately the value K= 2~0190 k//#
usad in the computation could not ?w obtained accurately. The stiff~ess ofdla-
phragms usad in the test could not ‘bedetermined, and the velue K nsed i~ an
a-mrage value, estimated fron th~ measured vertical deflections and ‘mrtl.~al1)
direct stresses. It should also be noted that the cross section of the deck
house of the test model c~!angednear the ends, and that the length Z= Win.
used In th~ comp~tation is a median value only. However, one need no”’d%umtis
!2 ;~~~+d, that the theory furnishes the type of strnss distrl%t.~-orlctuallv
-TmUKI h the tests, particularly, the theory shows +,hecharacterls~,~,cfik
i::the stress distribution on the deck level.
1) F!.gs.7 and 12 of Holtfs Faper quoted on page 1.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 35/58
-25-
& CONCLUSION AND OUTLOOK.
The thau.-;-w-wented in this report is based on the coricemtthat
the hdl and the deck houie act as individual beams whfch are forced to act
tagather @ their connectitms at the deck level. These connections trans-
fer shear stresses such that the longitudinal stresses in deck house and hull
at deck level are alike; these connections also transfer vertical reactions,
the flexibility of the bulkheads the vertical deflections of
hull will riotbe alike. Denending on the elasticity of the
extremes are possible: For infinitely rigid bulk heads hull
mill deflect a s a unit resulting in Navier’s stress distribution.
For very flexible bulk heads only horizontal shear forces are transferred from
hull to deck house, the deflections of hull and deck house will be dtfferent,
tindth~ stress distribution found in Seetfon 2 will occur. The actual con-
ditian will lie btween these extremes.
h. Rasults obtained.
deffnad by the value C@
the non-dimensional para-
[A )
u characterizes the
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 36/58
%-L
The stresaies ~ and d a on ly deuend on the bending moment and on the pr-
perties of the cross section of the vessel at the noint where the stresses
are to be found, while the deviation factor @ contains all other effects
in a single package, it expresses!
1. The
2* The
3. The
length / of the deck home.
stlffiess of bulk heads or dsck beams.
type of moment diagram and loading.
4. It is also dependent on the elastic uro~rties of the cross
sec t i on of the vessel.
The existence of relation (B)
the longitudinal stress dtstrfbution in
factor * . Test results on different
is of great value, because it exnresses
a non-dimensional manner by the sfngle
stmctures with various sizes of deck
house and hull can be interpreted by the @ -concept on a common basis.
The lengthwise distribution of the deviation factor ~ as indf-
cated in Fig. 10 defines the longitudinal stres~es and it can be shown that
an equation similar to Eq. (B) defines the shear forces T acttig between hull
and deck house.
The above results where obtained under the following assumptions:
a) Navferls hypothesis of Straight line stress and strdndistri-
bution 5s assumed to be valid for the hull and the deck house
sepantely? tit not for the entire section.
b) The cross section of tha deck house
section of khe hull is constant for the
but not foreand aft of the deck house.
c) The stiffening eff’eetof bulk heads
fa constant, and the c ross
length of the deck house$
the spring constant K is assu:d
buted for the full length of the
and equally dis%,r$-
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 37/58
-27-
d) The effect of shear lag has been neglected. This is indirectly
implied in assumption a), as Navierts hypothesis is never satisfied
if the shear defamations are substantial.
The stress and strain distribution given byEq. (B) cmsists of two
straight lines with a break on the level of the deck, as shown in Fig. ~. This
agrees qualitatively with
of assumption d), Eq. (B)
the results of model and full scale tests. Because
cannot be valid and should not be apnlied in the
vicinity of the ends of the deck house
factor.
Table 1 in Section 7, can be
where shear lag must be a controlling
used for the quick numerical deter-
mination of the deviation factor ~ (JEW!unction of the parameter u) for t he
stresses at the center of the deck house. Similar tables can be computed for
other points, e.g. thifi or quarte~points of the deck house.
The theory is fully applicable if wrts of the deck house are of
aluminum instead of steel.
B* Apnlicution of the theory to actual shipts stmctures.
The simpli~tng assumptions b) and c) stated above are not satisfied
in a actual Shipis stmcture, and the q$ij~tionarises whether and to~at ex-
tent the results obtained ap~ly or can & extended to vesselshaving neither
constant cross section nor
The prfnctple of
pllcable if asswrm t i cms b)
cated mathematical methtis
equally distributed bulk heads.
the theozy presentid in this report is fully a~
and c) are not made, kt different and more comnll-
for solving the differential equation may have tm
be used. It is of considerable icmortance that the stress distribution can1)
agsim be exnressed by an equation of the form
(B)
1] %mAPPmdiXp %x!%icm 7.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 38/58
containinga
M dffferent
deviation factor ~ ,
from before. ASShOWII
suit of assumnthn a).
Thus we know that even in
a3-
the numerical value of which will of coume
in theappndlx equation (B) is a direct re-
the more general case of variable sections
and stlf’fhessthe state of stress in the vessel can be descrfhed by a curve
tndfcating the values of $ for
cumes shown in Fig. 10. We can
again denend on the parameter
.
the length of the deck house similar to the
exoect that the shane of the & cumes will
where the values of 1,, Iz?p,etc. are avenge values of these nrmerties.
The effect of variations in cross section of hull and deek hcusq will
express ttself in the shaw of the deflection tunes of the hull andthe deck
house. Deflection cumes generally being not very sensitive to variations of
the cross sections it is to be expected that a reasonable apmoximation cd’the
actual case of variable sections cm be obtained W using constant ~werage or
median values for A and I.
The fact that the bulk heads and deck beams act at certiin nol~ts in-
stead of providing a continuous effect, as assumed in assumption e)~ wIII not,
affect the overall stress distribution as long as the bulk heads are gnre=d
reasonably equal over the length of the deck houses and as long m there. are
at least five In number. TMs is concluded from the fact that the moments
and deflections of any beam due to 5 or more equidistant concentrated loads,
and due to equally distributed l-d of equal mawitude are nearly alike.
To provide for the possibility of a conceniratlon of stiffening
bulk heads near the ends of the deck houses an additional analys is mm be
made , and it is believed that this affect can be treated as a corrmtion to
the analysis made in Sections 3 - 7.
The emphasis In tha preceding paragraphs was on just~fyingt!heuse
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 39/58
-29-
of the analysis presented in this report as theoretical bnsis for actual
design. It an~am that this analysis contatis all the essential features
affeeting the stress distribution in a real ship and if suitably emplopd
should furnish a~proximations of sufficient accuracy. It must be kept in
mind that Navier’s theory gives stresses in the deck house which may be con-
siderably in error, and that M%her refinement is not naceasary if the pr~
posed theory cuts this error to say lm of the actual stresses. After all,
if the experiments quoted aarlier had shown only differences of this magnitude
with the conventional theory this thecq wuuld have been considered god for
all practical purposes.
The starting point for the analysis m-esented in this rmort wen
page 1 which indicated that
comnleted an analysis which
the essential narametirs of
the conventional theorg is not
anpears to be rational and to
the croblem, it is proner to
stop and confirm this analysis by tests.
The prima~ purpese of these tests &ing to confirm the analysis
presented, the tests should be made with models of constant cross section and of
equally distributed bulk head stiffness K in accord with the assumption on which
the theory f~ baaed. The medals could b somewhat similar h size and section
to thcm used h Holtis tests,except that the bu l k heads would have
d@s&ned in such manner that there rfgidity can be determfriedbeyond
to be
doubt.
by the
for
two
models
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 40/58
-30-
can be made adjustable for two values K each, one model havtng u=l or 2P
the other u=2 or 4, giving altogether 4 results, of which two, for u=2,
should h identical if interpreted by the non-dimensional @ concept.
It is of importance to observe the longitudinal stresges at close
internals near the end of the deck house In order to determine the extent of
the area influenced by shear lag. At the center of the deck house sufficient
stress readings should be taken to be able to check the assumed straight-line
stress distribution. It is also necessarg to obtain the relative movements of
deck house and hull for comparison with the theory.
The test results should h evaluated by commting values for the
obsened stresses and deflections from the theo~.
D. Wowsed additional theo~tical work.
Assuming that the suggested exwriments cmf~rn the themg for the
simplified structure, the next step should & to analyze numerically a typical
vessel of variable cross section and bulk head stiffness. This will rn-ovidea .
basis for judgfig the wror to be exnected from using any simnlif’iedtieory.
It appears highly desirable to use the data for the ~assenger SMO S. S. Fresi-
dent Wflson for this theoretical investig~tion because the full scala test re-
sults on this vessel provfde a nossibllity for an ultimate check of the theow.
It would not be sensible to make tests with refined mdels simulatinga shin
stmctura having variable sections and stiffness; if the tests mroposed under
c. agree with the theoqf model tests with variable sections wIH mmssartly
again agree with the theo~ (excluding the possibility of errors
rnaticalcompetitions) because there is no fundamental difference
s-t and variable sections. On the other hand~ the analysls OF
might dfsclose effects whtch do not cmzur In the siwlffled small scale mclels*
In additfon to thfs analysis of a smcial case, the %heo~ nresenked
in this report should h extended, as discussed h B, to allow for increased
bulk head atlffness at the ends of the deck house. It is ex~ected that this
can be done without materially incremfng the Ywrwariwd work in determining
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 41/58
.31-
stresseg as demonstrated on the exarnnlein Section 7.
It is also desirable to make an analysis of the shear lag effect
near the end of the deck house. The rmr-hoseof this analysis would b to
det~rmfne how far the shem lag effect reacnes, aridto obtafi a simnle con-
C~lldOllfOr degign ~rpO!3aS.
E. Derivation of design males.
The theo~ presented in this report, togother with the proposed ad-
ditional exwrimental and theoretical work is exnected to be sufficient to
ded~ct de~ign rules and cmpute tibles or charts for use in actual design,
would be based on the non-dimensional narameter
M exnected that tables or charts wrtalnlng to
required:
u.
the following
Deviation fnctors @ at equidistant points along the deck house,
0.L?5 1 aparta permitting the determination of bending and shear
stTes9e9 in the vessel.
Reaetions of daek house on the bulk heads at center and a~ads of
deck house.
Effective moment of inertia of vessel, - required for comnutatlon
o? the deflections of the entire vessel and of its natural fre-
qlmlc+ies.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 42/58
1. Froprties of section if hull and deck house act integrally according to Navier.
Uppr part: area A,, moment of inertia It
lower pint; area AZ, moment of inertia 12
Location of center of gravity (Fig. U)
Total moment of inertia 1:
1=1,+1=+ (uM,+e)2A, + (am2-e)2.42
~,d,+CK,A2+WZA2- a ,A ,
aat+e=a Q A==—A,+4Z A,+ A=
ama - e . Qa2A,+ w=A2- XZAZ + MIA, Q A,= -.
A,+ Az A,+ A=
mld
and lower
(QJ
is def~ned byEq. (7a)
The resultants N, and M, , Nz and Mp of the stresses in the upner
portions are determined as follows$ The stress cr can be expressed
as function of x$ see Fig.u.
In the deek house the stress can be expressed alternatively as function of M, , N,and :.
Cr+-:x,I I
Because:a Az
X=X,+ati,+e=X,+ —A,+ A=
~ we have
N (u A=
)N,-%=
I — ‘~ I , ‘— ‘f+ /q ,+Ap
As this must be correct for any value of X, , we have
M,=M; *Q A,A2
-Ma+=-M ~(/q,+/4z] =
Similarly, we find for the hull
(b )
(c)
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 43/58
-3>
2i Derintion of Eqs. (~).
From Eq. (4)
Using Eq. (7a) and (8),
where
~+a , ZA .P=
1= + tizA
Considering
M, IZ - w =1 , W, (I, +%zA)- w ,(Z~%L) -1-P~2 ,I=
=z,+ a ,Za + X2+ K2 1 A ]+J&
T may be written finally
we obtiin finally
(d )
(e)
(f)
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 44/58
-34”
Substituting Eq. (d) in Eq. (lb),
Considering
r,+ ‘i, ‘A -z,u,zA P
I = zt +d, J -A+ 12+ K21A = 1+ /
we ob~ain finally,
3. Derivation of strain energy of structure.
house and
be,
Denoting hy E, and &z the average longitudinal strain in the deck
hull, respectively, the strain ener~~ of the longftl~dinalstresses will
2
~
F
in the deck house: ~ (A, =:+ It y,’”) u’Z ,
z-2
J
~
in the hullxE
‘(A2E; + I= Y=” 2) dZ .F
2—.?
The strains e are counted positive if they represent elongation.
In addition to the strain merry of the longitudinal stresses} t,herewXll
be energy stored in the bulkheads m deck beams which resist the relative vertfcal
displacements of deck house and hull; th~s part of the strain energy can k ex-
pressed by the spring constant K in the form
2
1
F
i K(~-y2)2 Ciz
1-+z
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 45/58
-35-
The strain energy of the shear stresses will be neglected; it is small because we
consider the case of long deck houses only. The total strain energy V is
The stresses in the deck house and hull can be expressed by the average
strains ~, and E2 , and by the second derivatives ~“ and ya”,
in Deck house: c. EE,+ EY,”Z, ,
in Hunt C= E=z +EY;x= ,
where ~, and X2 are as shown in Fig. 1. Because hull and deck house are con-
nected by rivets or welds, the stresses
have for x, . - Uq, ,-2= ~~2 >
E G, - Eam,~”
Further,the
the average
on main deck level must be alike and we
= E6=+Eam 2y2° . (i )
longitudinal resultant of all stresses in the Deck house,Nl, must h
strain G, times EA , ; similarly N..=EA2GZ. The resultant of all
longitudinal forces in the structure consisting of hull and deck house will be
obviously N, + A/z ; as the structure is in bending only this resultant must
vanish,
By means of the two eqs. (i) and (j) e, and Ez can be expressed by the curvatures
u AtE,=—
( )-1”M=y=” y
Q AtA,+AE “Z
E2=-—A , +A2 (
ti,~”+ w2y*”)-
Sulastitutingthese values into eq. (h), we obtain7
k
v= : ][ z y,2
‘t=+ I2 Ye“ z+ 1A (a,~ “ + w2yz”1
1
+ : (X-Y,>’ dz
z-2
~. Potential UW of external forces.
The potential Uwof any load P is equal to the negative of the work
(k)
(z)
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 46/58
-34-
don~ by this force when the structure deflects. Applying this to the forces p
acting on the structure we find that we have to make a distinction htween forces
p, which act on the deck house and forces Fe which act on the hull. Counting
p, and p= positive if acting downwards, and assuming p, and p2 to he func-
tions of the longitudinal cootiinate z
$
, their potential energy will &
J (Ply, %Y, ) ‘z (m)1-
The ~hea~ fone~ S= and S= &d the moments M= and Ma act immediately outside
points C and D, and their potential enargy will depend on the vertical displ~cements
y== and Ya= and on the rotations of the end surfaces of the hull,y~= and Y;= .
Taking into account the direction of the shears and moments shown in Fig. 8 ,
their potential energy will be
Noting kg S and M the shear snd moment in the stmcture, both
of the longitudinal cofiinate z ~ we can mite this e~pession
form
and (n)
The total potential energy of the external load is the
being functinns
in the abbreviated
(72)
sum of eqs. ( m )
[0)
5. Derivation of Eulerts equations and boundary conditions, eqs. (17).
From Eq. (14) we derive by the process of variation
Performing integration @
results in the expression
1) Seen0tJ3 3 On pgO 9.
- + K(~-y2)(L&+yz) -p,~X ‘&8Yz]dz
parts twice on the first three terms under the integrals
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 47/58
“37-
8(J= ~1,y ,’’d y,’ + E1,
Rearranging this expression, we obtain.
The equation ~ U = O will be satisfied if each of the six terms vanishes.
The two integrals will vanish if the terms in brackets are zero, whfch furnishes
Eulerts equations (16). The vanishin~ of the four other terms, at both boundaries
z = //2and z= -2/2,furnishes 8 more conditions which are the boundary conditins of
the problem.
Due to the definition of y and ya as relative displacements, we have
y=: o forz’ &2/z; y..having a definite value at z.* 2/2means that the varia-
tion &yz at this point will be zero~ and the fourth term in eq. (0) vanishes.three term
The values of ~y, ,~Y,’ and ~yz’ at z=&l/zdo not vanish; tho fi#stAin ~.(p)
will vanish only if each of the expressions by which
plied will be zero at the boundaries z=&Z/Z ; after
ditions,+Eqs. (17)~ are obtained.
$X , $x’ and 8Y;
rearrangementfi.s
are multi-
b&@a~ con-
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 48/58
-3fL
6. Expressions for longitudinal forces and horizontal and vertical shears.
The resultants N, and N= of the longitudinal stresses in deck house
and hull areN,=EA,~, and N2=.E4=G= , or with reference to Eqs. (k) and (7a)
Defining by T
to any point Iaving the
E 1AN,=~ (M,~“+ ae y=”)
/y= - =“ (=, y:’+ -, ~“t)Q
the total shear force from the left end of the deck house
coordinate z , equilibrium requires
E I.7=-N=–—
(lxfy,”+ Ma y r”).
(r)I a
The unit horizontal shear will be
(&7 . _ % “’+~=yz’”).
~!Y, (s)d z Q
To obtain the expression for the vertical shear ~ in the deck house,
consider an element of the
of moments with respect to
deck house of length dz , Fig. 15. Equilibrium
the centroid requires
Introducing Nl,=-E.,Y,’’andEq. (s), we obtain
y= -E~,y)’”- EW,Z4 (a , y ,’”+ ~z y z ‘“) = - E (Z , + H :IA ) ~ ‘“- w , Up EZ4 ~:’ (t )
Similarly the vertical shear in the hull is
V!=-M,<2E1AX’” - E(Z=+ -=214) ~’” (u)
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 49/58
-39-
7. TYW Of Stress Distrikti~ if Cross Sections are not Constant.
It is possible to derive Eq. (B), (page 27 ) without any lengthy
analysis,demomtnting that it applies even if the aross sections of hull and
desk house vaxy and the hlk head stiffnass is not constant. The one and ~ly
assumption which mu~t k made is that Navier’a theo~ is applicable to the hull
and deck house aepmately.
Accofiing to this assumption the strains in the hull and in the clack
house must vary linearly-, and the stress distribution must therefore aonsiat
of stmight lims, Fig. 16, btween the values ~ , C2 and &3 of the stresses
at the top of the deck house, at deck level, and at the Imttmn of the hull,
respectively. The internal stresses a must h in equilibrium with the exte-
nal loads, which are at any particular sectfon$ the moment M, and the longi-
tudinal force N= o. VWJhave therefore two conditions
f
o-dA=o, JxodA=M (>)
A A
The stresses r in eqs. (v) can bs exmessed by the three values ~ , ~
and C, and without actually making this computation, we how that the two eqs.
~ , 5s given.
We want to
equations (v) can k
where &N and AON
the two atressea m2 and ~ as functions of the
therefore, that equilibrium alone restricts the pos-
in such a way that all stresses are defined if one,
demonstmte that all stress distrlbuti~whiah satisfy
expressed in the fom
~=~~+$AC (w)
are defined in Section 2, and ~ is a numerical faator
whfch may have any value.
It is obvious that the stmases given by equation (w) must satisfy
the equilibrium cmditions (v) &cause the stress distribution (w) was dgtemd.neda~
_Hwactual one for some structure. @ the other hand & varying the value of
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 50/58
-@-
~ in eq. (w) the stress at the top of the deck house may
qru given value~ and as only one stress distribution exists
Eqs. (v) and for which the stress is ~ on top of the deck
be mada equal b
which satisfies
housa, thfs Str’ms
distribution can be expressed by oquatkm (w). TMs conclusion is val!d
whethar the cross saetion of the vesssl remains constant or not. me actual
numerical value of @ cant of course~ not h obtained by this simple mnsidera-
tion.
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 51/58
z
‘-l
L’ D
a)
l’- /4L.—— —.———.— —__
—-1
“% 1 +
‘C.y
5
-C.g
%y. 1
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 52/58
2)’2’
1
I
Cj 1
r T
I
- .-+- —----
Me
3--2
/ M
A’ M-
I w, a
D )
D
).
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 53/58
r
I
L%’j?
I
,920
Au-
%“. 4
I
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 54/58
I I
I
I 1
I
I
II I
I ~hes;III -.
+-1=
I
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 55/58
o
U.l. o
—
$- curvef
I2 3
/’
o.E’s o.Z% 0. 75 0.
-End of Deckhouse
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 56/58
I
1
II
I
I
M=
1
I 1
I I
—- —” . —..
I ====+
I
X’g II
8/2/2019 Study of Structural Action of Superstructures on Ships 48
http://slidepdf.com/reader/full/study-of-structural-action-of-superstructures-on-ships-48 57/58
-. .-
Q—
~\N
J$ I
v
Q
.—..—
—
—-
——
—
t
— ---------
2
Pb*
————————
I II I
U_