In the United States Commercial Shipbuilding Industry

SSC-273

SURVEY OF STRUCTURAL TOLERANCES IN THE UNITED STATES COMMERCIAL SHIPBUILDING INDUSTRY

This document has been approved for public release and sale; its distribution is unlimited.

SHIP STRUCTURE COMMITTEE 1978 SHIP STRUCTURE COMMITTEE AN INTERAGENCY ADVISORY COMMITTEE DEDICATED TO IMPROVING THE STRUCTURE OF SHIPS

MEMBER AGENCIES ADDRESS CORRESPONDENCE TO:

UNITED STATES COAST GUARD SECRETARY NAVAl SHIP SYSTEMS COMMAND SHIP STRUCTURE COMMITTEE MII hART SEALIFT COMMAND U.S. COAST GUARD HEADQUARTERS MARITIME ADMINISTRATION WASHINGTON. D.C. 20591 AMFRICAN BUREAU OF SHIPPING SR-l233

With, the vast increase in ship size during the past two decades, great emphasis has been placed on reducing hull scantlings through rational determination of loads, working stresses and material properties. To support an extended use of rational analysis in ship design, it is necessary to determine the deviations from 'ideal'1 design that can be expected in construction and their effects during the vessel's service life.

The Ship Structure Committee initiated a project to determine the factors leading to and the extent of deviation from theoretical design that can be expected.

This is the final report of that project and is being published to assist in developing a rational approach to ship design0 £/ ,&.J:i W. M. Benkert Rear Admiral, U.S. Coast Guard Chairman, Ship Structure Committee SsC.23

FINAL REPORT

Project SR-1233

'TStructural Tolerance Survey"

SURVEY OF STRUCTURAL TOLERANCES IN THE UNITED

STATES CO1MERCIAL SHIPBUILDING INDUSTRY

N. S. Basar R.F. Stanley

M. Rosenblatt & Son, Inc.

under

Department of the Navy Naval Ship Engineering Center Contract No. N00024-76-C-4059

This document has beenapproved forpublicrelease andsale:itsdistribution is unlimited.

U. S. Coast Guard Headquarters Washington, D.C. 1978 ABSTRACT

Deviations from ideal structural design of differenttypes of vessels during construction and serviceare investigated. Selected U.S. commercial shipyards, ship owner/operators, steel mills, and foreignclassification societies are surveyed or interviewed with thepurpose of documenting major deviations and recurring structuralimperfections, and determining the factors lead incj to these deviations. An effort is also made to determine the extent of deviations fromtheoretical design and to establish, wherever possible, structural tolerance limitswhich are most commonly used ¡n U.S. yards and whichcan therefore be considered representative of U.S. shipbuilding practice. These are compared to published internationalstruc- turai tolerance standards, and recommendationsare given for further study. CONTENTS

ABS TRA C T List of Tables List of Figures

Section 1. INTRODUCTION

1 Background 1.2 Objective and Scope 1.3 Limitations of Surveys

Section 2. SURVEY METHODOLOGY 2-1

2.1 General 2l 2.2 Standard Survey Format 2-2 2.3 Scope of Vessels for Actual Surveys 2-2

Section 3. STRUCTURAL SYSTEMS/SUBSYSTEMS AND DEVIATIONS 3-1

3.1 Structural Imperfections/Deviations 3-4 3.2 Extent of Consideration for Each Deviation 3-6

Section 4. DISCUSSION OF SURVEY RESULTS AND TRENDS 4-1

4.1 Shipowners and Operators 4-2 4.2 Shipyards and Steel Fabricating Facilities 4-5 4.3 Steel Mills 4-23 4.4 Classification Societies 424 4.5 Foreign Institutions 4-29 4.6 Overview 4-29

Section 5. FOLLOW THROUGH OF A TYPICAL STRUCTURAL DEVIATION 5-1 5.1 Factors Related to Ideal Design 5-1 5.2 Influence of Structural Deviations on Strength

Section 6. CONCLUSIONS 6-1

Section 7. RECOMMENDATIONS 7-1 Shipyard Practice in 7.1 Recommended Guide to U.S. 7-1 Structural Tolerances 7-1 7.2 Relation of Tolerances toRational Design Quality Assurance and InspectionRequirements 7.3 7-4 in Shipyards

Section 8. REFERENCES 8-1

ACKNOWLEDGEMENT 8-3

Section 9. APPENDIXES 9-1

9.1 Bibliography 9-1 9.2 U.S. Commercia] Shipyard Standards 9-8 9.3 Existing International Standards 9-63

iii LIST OF TABLES

Table N Descriytion Pase Nc.

4.1 Structural Tolerance Standards as Reported 4-4 by Shipowners/Operators

4.2 Deviations and Tolerances at Shpyards 4-8

6.1 Comparison of USA Practice in Structural Tolerances with Published International 6-3 S tanda rds

7.1 Structural Tolerances in United States 7-2 Shipyards

iv LIST OF FIGURES

Fiure No. Descr t ion Pase No.

4.1 Bulkhead Misalignment 14_3

4.2 Damage Occurence Rate 4-6 by Age of Vessel

43 Example of Poor Detail ing 4-6 of Design

1414 Graphical Representation of Reported Struc- 14-18 tural Deviations and Tolerances

14 5. A. Full Penetration Weld 14-27

Alternate Detail 14-27

Explosion Bonding k-27

5.1 General Relationship of Cost to Dimen- 5-5 sional Accuracy

5.2 Alignment of Subassemblies 5-5

5.3 Detail Design of a Beam Bracket 5_9

5.14 Effects of Weld Undercut 5-11

7.1 Critical Quality Areas 7-3

V SHIP STRUCTURECOIiTTEE

The SHIP STRUCTURE CMITTEE is constituted to prosecute a research programto improve the hull structures of ships by an extension of kno;ledge pertaining to design, materials and methods of fabrication.

RAUM W. M. Berikert, USCG (Chairman) Chief, Office of Merchant Marine Safety U.S. Coast Guard Headquarters

Nr. P. M. Palermo Mr. N. Pitkin Asst. for Structures Asst. Administrator for Naval Ship Engineering Center Corraiercial Development Naval Ship Systems Command Maritime Administration

Nr. John L. Foley Nr. C. J. Whitestone Vice President Engineer Officer American Bureau of Shipping Military Sealift Coiriiand

SHIP STRUCTURE SUBCOMMITrEE

The SHIP STRUCTURE SUBCOMMITTEE acts for the Ship Structure Committee on techn-ical matters by providing technical coordination for thedetermination of goals and objectives of the program, and by evaluating andinterpreting the results in terms of ship structural design, construction and operation.

NAVAL SEA SYSTEMS COMMAND NATIONAL ACADEMY OF SCIENCES SHIP RESEARCH CONMITTEE Nr. R. Johnson - Member tir. J. B. O'Brien - Contract Administrator Nr. O. H. Oakley - Liaison Nr. C. Pohier - Member Mr. R. W. Rumke - L{aison Mr. G. Sorkin Member SOCIETY OF NAVAL ARCHITECTS & GUARD U.S. COAST MARINE ENGINEERS LCDR T. H. Robinson - Secretary Mr. A. B. Stavovy - Liaison LCDR S. H. Davis - Member CPT C. B. Glass - Member WELDING RESEAPSCH COUNCIL Dr.W. C. Dietz -Member Mr. K. H. Koopman - Liaison MARITIMEADMINISTRATION INTERNATIONAL SHIP STRUCTURES Mr. F. Dashnaw - Member CONGRESS Mr. N. Harner - Member Prof. J. H. Evans - Liaison Mr. R.K. Kiss - Member U.S. COAST GUARD ACADEMY Mr. F. Seibold - Member CAPT W. C. Nolan - Liaison MILITARY SEALIFT COMMAND STATE UNIV. OF N.Y.MARITIME COLLEGE Mr. T. W. Chapman - Member COR J. L. Simmons - Member Dr. W. R. Porter - Liaison Mr. A. B. Stavovy - Member N'IERICAU IRON & STEEL INSTITUTE Mr. D. Stein - Member R. H. Sterne - Liaison ERICAN BUREAuOF SHIPPING tir.

Mr. S. C. Stiansen - Chairman U.S. NAVAL ACADEN'! D.. H.Y. Jan - Member Dr. R. Bhattacharyya -Liaison Mr. I.L. Stern - Member U.S. MERCHANT MARINE ACADEMY Oc. Chin-Bea Kirn - Liaison

vi 1.0 INTRODUCTION

.1 Back9 round

A ship or any vessel, like any other complex structure,is subjected to certain imperfections or deviationsfrom their ideal structural design during construction. The deviation can be avoidable or un- avoidable depending on its location, ease ofinspection, and the pos- sbility of accomplishing corrective action.

For purposes of clarity, it may be desirable tofirst attempt a definition of the''ideal design.''

The structural design of ships is based on strengthcalculations performed using the dimensional characteristicsof the vessel, the loading criteria for the service the vesselis to be employed in, and the prevailing sea-states. Assumptions are made in carrying out the design, and safety factors are used to compensatefor unknown or un- predictable parameters. The structural model of ships developed in this fashion is expected to perform its intended service underall conditions. This is labelled the''ideal design.''

The ideal design assumes that the finished constructionwill represent accurately the configuration shown onthe theoretical structural drawings. Even though there have been cases where allowance wasmade in the ideal design for certain major deviations, ingeneral, a great majority of newly constructed vessels do not have any suchallowance as- sociated with their design except for what is intrinsicallyallowed in the classification society rules.

Yet in everyday practice, it is impossible to maintain an exact duplication of the geometric configuration depicted onideal design drawings on the physical ship being constructed. The ideal design is deviated from during the production of shipbuilding materials, duringfabrication and assembly operations, and during erection on thebuilding ways. These deviations may consist of flaws in base material, errors¡n fit-up and alignment work, unfairness of plating, errors originatingfrom the manufacturing processes used, and errors in the detail designof structures.

A ship may develop additional imperfections or deviationsfrom the ideal design during its service life. These ''in-service' deviations may originate from the actual service conditions of the vessel. Impact loads experienced during operations in heavy seas, or mechanical damageduring operations in port or at sea, may result in deviations such asunfairness of the plating, distortion or deflection of structural members, orreduc- tion of steel thickness due to corrosion, etc. If an initial imperfection exists on a newly built vessel, the service conditions may cause a worsen- ing of an otherwise tolerable deviation and lead to brittlefracture or fatigue cracks. Even with today's technology, which allows the use of improved quality shipbuilding materials, much improved manual or automatic fabrication and assembly procedures, sophisticated welding techniques, and new non destructive testing methods, some shipbuilders may not be, for varying reasons,in a position to fully utilize these improvements and provide a finished product reflecting the available technology. Furthermore, even when all available technology ¡s Fully utilized, it ¡s still impossible to eliminate all structural imperfections due to the inherent errors ¡n the automatic fabrication equipment and the human factors involved. The apparent result of this situation ¡s that the ships built by one shipbuilder, even ¡f the same ideal design drawings are followed, may be and almost always are not equivalent to each other from a structural accuracy viewpoint.

1.2 Objective and Scope

The overall objective of the present study ¡s to determine and document the present-day hull construction and inspection procedures to determine the factors leading to and the extent of structural deviations from the ideal theoretical design ¡n U.S. shipyards.

The original requirements for the study, as specified by the Ship Structure Committee, were the following:

Approximately twelve U.S. shipyards and representative steel producers supplying material for, constructing, or repairing ocean-going vessels should be surveyed.

Shipowners/Operators and classification agencies should also be interviewed.

The study should cover the range from unmanned and/or un- powered ocean going barges to fully powered vessels.

. The surveys should consider deviations from ideal design cc- curring during construction and service including:

Poor detailing of design Flaws in base material and thickness variations Fit-up alignment Welding flaws Unfairness and deflection Forming and strengthening practices

The "in-service" deviations should exclude deviations due to damage from collision, grounding or similar accidents.

Major deviations and recurring items are to be explored and documented.

The study should identify the normal deviations experienced for the factors involved as well as the maximum deviations expected. The findings should be correlated by ship type, in-shop or on-ship work, and the type of shipyard faciltMes, (e.g. repair versus new construction).

l-2 instrumentation work was envisioned. 8. No experimental or ship

As described in greater detailin Section 2, the scope ofthe the investigation to survey was expandedduring actual performance of than the cover nineteen shipyards orsteel fabricating plants rather twelve required in order toobtain a more representativecross-section ship types of the U.S. Shipbuildingindustry, and also to cover as many as possible.

1.3 Limitations of Surveys

As touched upon brieflyin Section 2,it would have been desirable to conduct detailed andin-depth surveys, especially in shipyards, to enable the project team todevelop distribution curves of the deviations measured in a quantitysufficient to permit statistical analysis. This, however, was notpossible. The quantity of structural deviations data obtained was ratherlimited partly due to the fact that due to the the yards did not maintain astatistical record and partly be difficult to carry out in fact that actual measurements proved to that it interfered with the yard'swork in progress. not enough As far as ''in service''deviations are concerned, again societies due to the simple data were available from the classification fact that this type of data is notbeing recorded and sometimes not they are not reported due even reported. lt is probable however that to following reasons:

The causes are difficult todetermine. Measurements of deviations areoften impossible. Even if the causes could bedetermined, the surveyors may still be reluctant toreport these because they may be subject to libel suits.

This last reason led the InternationalShip Structures Congress in their research the establishment 1976 report (i) to recommend for future Recording System". The report citesthe of a comprehensive "Damage need for all parties concerned,ie the Classification Societies, Shipowners, and Ship repairers totake a more liberal view of the subject and to release informationof this typefOEthe benefit of the industry.

The range of vessel types specifiedfor the surveys could not be were not being fully covered due to the fact that some vessel types constructed in U.S. commercialshipyards during the period of surveys. This limitation is further discussedin Section 2.3.

identified at the time of The data on structural deviations were documenting them during surveys as towhether they represented shop referred to. In or field work andalso as to what ship type they compiling and analyzing thedata however,it was decided that, since the shop deviations are eithereventually eliminated by corrective action or they become field typedeviations when assembled in place

Numbers in parenthesis denote similarlynumbered references in Section 8.

1-3 as-is, only ''on-ship' type de'iations would be considered for tabulations. Still, however, because of their very nature, some deviations such as cutting line accuracy, edge preparations, groove depth etc., necessarily reflect ''in-shop'' operations.

The results of surveys,in the form of structural deviation and tolerance data are reported in Section 24 separately for owners, yards, class societies, and steel mills, and theemerginçgeneral trends are discussed.

Additionally, in Section 5.0, a typical structural deviation is individually considered throughout all phases of ship design, construction, and service and the importance of maintaining tolerances isin- vest iga ted. 2.0 SURVEY METHODOLOGY

2.1 General

A number of internationalagencies/institutions have developed and published ship structuraltolerance standards and/or compiled listings of same in use in their respectivecountries at the time of pub] ication.

The most widely known structura]tolerance standards are those developed in Japan by the joint effortsof the Society of Naval Architects of Japan and the Universityof Tokyo. The "Japanese Shipbuilding Qua] ity Standards-Hull Part'' asit ¡s referred to(2) and re-edited was first published in1965 and was subsequently revised in 197], 1973, and 1975, to reflectthe changing shipbuilding technology.

The Japanese approach in developing thesestandards ¡s described in (3): Briefly, the approach consistsof taking actual measurements of structural deviations ¡n a number of Japaneseshipyards, developing histograms of the measured deviations and,from these distributions, establishing the mean standard (range) andthe maximum allowable value (tolerance) for each structural deviationconsidered. A similar but more limited approach wasfound desirable for the present project.

As a first step all referencematerial compiled was carefully reviewed, and various ways of listing thestructural deviations were Swedish shipbuilding noted. The1 istings in the Japanese, German and tolerance standards were used but rearranged toconform to the following secuence as specified by the ShipStructure Committee:

Fit-up and alignment Unfairness and deflection Forming and straightening practices Welding flaws - butt, fillet, laps, and corners Flaws in base material and tHcknessvariations Poor detailing of design

The list of deviations developed was usedin pilot surveys conducted with two shipyards, two shipowners,four classification societies, and one steel mill for the purpose of testingits usefulness. In the pilot surveys, the scope of the surveys weredefined and revised after each survey to reflect the experience gainedand also to make t easier to extract relevant structural toleranceinformation from the institutions visited.

The following conclusions were made uponcompletion of the pilot surveys: revision, could be used I. The list of deviations, with minor in surveys at shipyards. Each yard could be given a copyof the list and asked to fill in the appropriate columnsfor normally experienced and allowable maximum deviations to the extent thatthis information exists and is being used.

2-1 The ship owners do not normally have as detailed informaticn on structural deviations and tolerances; therefore, the list of deviations could be used as a guide in obtaining whatever information the owners! operators may have available to them.

For purposes of facilitating the data evaluation work, it is desirable to list all the probable questions that the shipyards may be asked to answer in connection with their quality assurance capabil ¡ties, inspection criteria, and statistical or other deviation/tolerance records.

1. Visits to various institutions, especially shipyards, have to be of short duration and take a minimum of time away from the yard personnel due to their pressing every day type work responsibilities.

It would be desirable, and mostly possible, to contact the regulatory body resident surveyors and Owners' representatives stationed in each shipyard, and to obtain their input regarding structural tolerances and quality control requirements and procedures.

Informal talks with the yard's engineering department, and especially with the hull design group, would be necessary in order to explore the yard's approach to recommended corrective action for any excessive deviations noted and in order to document their procedure for detailed design review and checking of the original structural design against any deviations/deficiencies/deformations that may be left uncorrected in the vessel under construction.

2.2 Standard Survey Format

Utilizing the experience gained from pilot surveys, and considering the general shipyard response to be expected during the visits, a format was developed for use as a standard procedure during final surveys.

The format contained:

Definition of the scope of survey Questionnarie List of deviations

2.3 Scope of Vessels for Actual Surveys

Fina] surveys were conducted at eighteen shipbuilding yards and one steel fabricating plant on the following types of vessels:

Oil Tankers (33,000 to 265,000 DWT) Roll On/Roll Off Vessels (]k,500 to 17,000 DWT) LNG Carriers (63,600 DWT) Barges (250 to 4OO Feet) Drilling Rigs (Jack-up & Semi-submersible) Drilling Ships

2-2 3.0 STRUCTURALSYSTEMS/SUBSYSTEMS AND DEVIATIONS

The complete cycle of design, fabrication, subassembly, assembly, erection, and operation of a ship is considered phase by phase for determining and listing the structural systems and subsystems to be investigated in the present study. The underlying thought is the effect of each phase or process on the ship's structural systems and the possibility of creating or causing a structural deviation or imperfection.

Specifically, the following stages and processes are considered:

Contract Design -to determine any inferior quality details or arrangements affecting the quality of detail design.

Detail Design - Insufficient or inferior quality details or manufacturing processes specified in the detail working drawings.

Base Materials - Any deviations in the actual materials delivered to the yard from the ideal materials as specified in the plans and specifications and the effect of these deviations on the structural quality of the ship being constructed.

Fabrication Methods and Processes - Lofting, cutting, forming, straightening, and welding methods, and equipment and tools used in the yard during fabrication, and the structural deviations originating from errors or lack of quality in these operations.

Assembly and Erection Procedures -Inaccuracies or imperfections in the assembly methods and erection processes follow- edin the yard, and their role in causing structural deviations in the finished product.

Inspection and Quality Assurance Procedures - Lack or insufficiency of inspection and quality control operations during vrious stages of vessel construction and structural deviations caused by these factors as well as by temperature fluctuations, improper or insufficient staging for larger vessels, and the misalignment, deflection and sinkage of building ways, basins or docks.

Service Factors - Effects of corrosion, coatings, and overall main- tenance procedures on causing structural deviations; and also any deflections or similar imperfections which may be developed due to impact and shock loadings during a ship's service life.

A listing of all possible structural deviations affecting the structural systems and subsystems existing on a vesselis developed as a result of the above consideration. This listing is compared to some of the compilations in existing international shipbuilding standards (2,15,16) and

3-1 is also reviewed to ensure that ¡t meets the guidelnesset forth by the Ship Structure Committee (SSc), The original list as developed to satisfy the sequence specified by the SSC had the following outline: i.Fit-up Alignment Li Marking .1 .1Accuracy of Cutting Line 1.1 .2Panel Block Marking Compared with Correct Location 1 .2 Edge Preparaf ion 1.2.1 Roughness of Free Edge 1 .2.2Roughness of Weld Groove 1 .2 .3Notches on Free Edge 1 .2 .4Notches on Weld Groove 1 .2 .5Dimensional Accuracy (including bevels for welding) 1 .3 Component Parts Fabrication 1 .3 .1 Longitudinal Flanges & Flanged Brackets 1 .3 .2 Angles arid Built-up Plates 1 .3 .3Plates 1.4 Alignment 1 .4 .1Minimum Distance of Weld to Ad lacent Weld 1 .4 .2 Gap Between Members 1 .4 .3Fitting Accuracy 1 .5 Subassembly 1 .5 .1 Permissible Distortion of Beams 1 .5 .2Dimensional Accuracy of Subassembly 1 .5 .3Alignment 0f Subassembly Unfairness and Deflection 2.1 Accuracy of Hull Forni 2.1.1Principal Dimensions 2.1.2 Deformation of Hull Form 2.2 Deformation of Main Structural Members 2.2.1Unfairness 2.2.2 Miscellaneous Deviations Final Work & Finishing Practices 3.1 Finishing up Traces of Temporary Pieces 3.2 Surface Defects 3 .3 Treatment of Openings Cut for Temporary Purposes or by Error 3.4 Hatch Coarnings 3.5 Access Openings 3.6Miscellaneous Pieces 3 .7 Tightness Tests 3.8 Painting of Joint at Tightness Test or Inspection

3-2 Lf Flaws in Welding Geometry 4.l Shape of Bead (including size, undercut,reinforcement) i4.2 Distortion (Angular) of Welding Joint L13 Short Bead i4.i4 Arc Strike 14.5 Welding at Low Ambient Temperatures 14.6 Weld Spatter

Flaws in Base Material 5.1 Surface Flaws 5.2 Laminations 5.3 Steel Castings

6. Poor Detail ing of Design 6.1 Deficiencies in Contract Design 6.2 Deficiencies in Detail Design

This was labelled the ''List of Deviations'' and wasused in the surveys conducted in shipyards. lt was also used as a guide in sol icitingstructural tolerance information from shipowners, classsocieties, and steel mills. lt was slìghtly modified as results were obtained,however the general sequence remained unchanged.

When analyzing and evaluating the dataobtained from the surveys, it was deemed appropriate todrop those deviations from the list for which no responses were given. One of the deviationsdroppedwas ''1.14.2 Gap Between Members". Originally, this was adopted asit existedin the Japanese 2 ), and it appeared that there was Shipbuilding Quality Standards, JSQS ( an overlap between this deviationand "Gap Before Welding, Fillet Weld". The latter was selected for use in tabulatingthe results.

The order of the remaining deviationsfor which responses have been obtained has been rearranged to better reflectthe chronology of handl ing and treatment of material as it progressesfrom raw mill product to finished components on the ship.

The resulting listing of structural deviations can berelated to the original sequence by their respective numbers asfollows:

Flaws and Size Deviations in Base Material (1) Cutting, Forming and Straightening Practices(2-8) Fit-up Alignment (9-11) Welding Flaws and Restrictions (12) Accuracy of Subassembly and Erection (13-23)

The structural imperfections and/or deviations consideredin the final analysis are listed below and a brief descriptionof the extent being considered for each is given immediately following the list.

3-3 3.1 Structural Imperfections/Deviations Considered 1. Receipt Inspection Waviness Thickness & Pits Laminations 2. Cutting Line Accuracy (Comparison with correct line) 3. Edge Preparation (Roughness) 4. Edge Straightness for Automatic Welding Semi-automatic Welding Manual Welding 5. Groove Depth

6. Taper Angle

7. Fabricated Shapes Flange Breadth

Angle

Straightness Flange Plane Web Plane

Rolled Shapes, Flange Angle Gap Before Welding a.Fillet III b.Butt / ///\\\\ c.Lap MMMA I E \_\ \ f\X Beam and Frame Gap

Butt Joint Misalignment

3- ,. 12. Weld Geometry o.Reinforcement

Throat or Leg Length

Undercut

13. Intercostal Misalgnrnent /

14. Profile Warp

15. Stiffener Deviation fromStraight Line

16. Adjacent WeldSpacing Butt to Butt

Butt to Fillet

17. Cylinder Diameter 18. Curved Shell Accurccy

19. Subassembly Accuracy Length & Width Squareness

Hatch Coaming Dimensions X 20. 21. Access Openings Dimensions Deformation 22. Unfairness Bottom Shell Side Shell Deck Superstructure Side &End

3-5 23. Overall Dimensions Length Beam Depth Keel Flatness (Deviation from Straight Line) Forebody Rise

Afterbody Rise

Deadrise

Draft Marks ¡. Freeboard Marks

3.2 Extent of Consideration For Each Deviation

Receipt Inspection covers those material defects that should be checkedto determine the acceptability of the material. Waviness can be corrected in the yard, but it must be detected before ¡t can be corrected. Uncorrected, it leads to measurement and fit-up difficulties later.

Pits up to certain size can be faired by grinding; deeper pits must be filled with weld material. The tolerance limits are those for which even welding is not sufficient.

Laminationsin rolled steel plate are produced by oblate shaped inclusions or fibers, of suiphide or oxide (slag). Either type causes the plate to act like several thinner plates stacked together to form a thicker one. Receipt inspection should reject plates with extensive laminations that are far in excess of limits.

Cutting Line Accuracy ¡s the end product of several sep, including establishment of the guideline as well as the cutting operation itself.

Edge Preparation was to cover both welded edges and free edges, but only data for welded edges was reported.

Edge Straightness is similar to cutting line accuracy, except that the relationship to original design is not included. Its importance is mostly relevant to the production of good welds.

5 & 6. Groove Depth and Taper Angle are self-explanatory, but they are relevant only to those yards that do such preparation for weld joints.

7, 8, 9,10 & 11 are self explanatory.

3-6 Weld Geometry: Undercut is a functionof the digging effect of a welding arc, which melts a portion of the basematerial. If the arc is too long, the molten weld metal from the electrode may fall shortand flot completely fill this melted in a butt weld, at Zone, leaving an undercutalong the upper leg of the weld, or, eithèr or both sides of the weld.

l/l6H Reinforcement helps to prevent undercut.A nominal weld reinforcement of above flush ¡s recomended in weldinghandbooks.

Intercostal (or Cruciform Joint) Misalignment,is the classic tolerance problem.

functions of fit-up problems, both 1l & 15. Profile and Stiffener Deviations are dependent on previous work andinfluentialin following work.

Adjacent Weld Spacing is dependent onoriginal design and on welding practices, but ¡t can lead to especially pre-heat. It ¡s not a tolerance problem, perse, distortions and locked in stresses.

Cylinder Diameter is relevant mostly todrill-rigs and similar exotic marine vehicles with large cylindrical structures.

Curved Shell Accuracy is a functionof forming practices and ¡s a major factor in shell unfairness.

19, 20, 21, 22 & 23 are self explanatory, exceptthat some shipyards do not make change are any check on L, B, or D. Forebody Rise, Afterbody Rise, and Deadríse functions of welding-induced shrinkageand 4iurnal temperature chanqes, a problem that has been partly reduced but noteliminated by extra careful design and welding sequence.

3-7 Lo DISCUSSION OF SURVEY RESULTS AND TRENDS

Resu]ts obtained from surveys are discussed in the following subsections. In the tabulations for reporting the responses to the survey questionnaire and the data supplied on deviations and tolerances, symbolic numbers and letters are usedin place of the names of institutions. L] SHIP OWNERS AND OPERATORS

Seventeen ship owner/operator executivesor yard inspectors were interviewed to discuss their experiences with deviations in construction and in serviceon their ships. Some of the comments and discussions are given below, but sincethey had quite diverse operations,most of their comments do not fit neatly lar form. into tabu-

The most commonly suggested toleranceproblem was misalignment, especially misalignment of intercostals at cruciformintersections. Three ship owners/ operators have approached the problemon an analytical basis, utilizing finite element analysis either to determinethe maximum misalignment allowable before design joint efficiencies, stresses, and safety factors were exceeded,or to determine the mechanics ofknown failures. In one case, the owner/operator found that t/3 was the maximum acceptablemisalignment for high stress, high-cycled joints. This result agrees with thatpresented ¡n the background material for the Japanese Shipbuilding Quality Standards (3). Another found that many small discontinuities and misal ignments that ordinarily were not bothersome to inspectorswere the causes of small cracks in container shipbox girders. This same owner/operator also found that misal ¡gnments of up to I' in longitudinal bulkheads had causedfatigue cracking at cruciform joints withtransverse bulkheads.

Two executives discussed the problem ofplate panel distortion (unfairness) due to welding. One said that deflectionsup tolit4''in shell plating were tolerable because they indicate that the weldments have "pulled" properly. The other stated that shipyard practicesfor straightening these deviations build up large residual sometimes stresses.He urged care and proper sequencingto prevent problems. At least three other owners/operatorsutilize curves for assessing permissible unfairness, basedon NAVSHIPS 0900-000-1000 ( t4 ).

Various opinions were givenon the subject of coatings to maintain structural strength and to reduce tabular corrosion allowances for plate thickness. Two oil tanker owners/operators found thatthe initial expenses plus theexpense of re- coating after ten years were much more than the initial cost plus serviceexpenses of simply having thicker steel plate. Two other tanker owners/operators said that deficiencies in coatings result in wastage, and that reducing thiswastage would prolong ship life. The owners of a fleet of LNGtankers use inorganic zinc coatings in the ballast tanks, but urged that someone should analyzethe trade-off between reduced scantlings forcoatings and plate buckling. A dry. cargo ship owner/operator executive, citing thecorrosive environment as diferent from that on oil tankers, said that there were benefits from coatings. He related the fortu i tous situation where some ships original ly bu iI t with ful 1scant] i ngs were jumboized, yet the plating was adequate when special coating allowanceswere utilized.

Several interesting commentswere received on the subject of weld defects. One inspector for a drill rig owner/operator said that he required about lO of overall weld footage and lOO° of criticalarea welds to be inspected by non-destructive testing. An executive for an oil tankercompany said that at one time the firm investigated the defective welds by X-raying allwelds in three ships. The result was thatl5 to 2Oof the welds X-rayed had some defects. An inspector told of one shipyard's procedure thateliminates the possibility of doing a statistical analysis of weld defects. The yard would not show the "in-process' X-rays to the owner if repairs were required,but would show only the X-rays of the structure in question after repairs were made. The yard evidently felt that if repairs were made, then the original X-rays were irrelevant. This displays a fundamental mis- conception about NDT as a means of quality control. Since less than lOO° sampling ¡s made, itis important to retain and analyze the originaldata to know what quality of work ¡s being done.

Shipyards and classification societies occasionally havefound novel ways of rectifying structural deviations in fit-up. One example involved a bulkhead that did not align with a floor at the tank top. Instead of breaking the connections, the yard welded a large bar to bridge the misalignment. This may seem like a makeshift solution, but it probably did not develop large residual stressesin the joint, as the process of breaking, force-fitting, and rewelding wouldhave (Fig. 14.1).

J''nner Bottom

FI oo r Fig. 14.1: Bulkhead 11 Misalignment

relates The graph on Figure 14.2, reproducedhere with special permission, first four damage occurrence rate to vessel age. The rate rises sharply in the should years., then falls off to alower level. The shipping lines interviewed by now, since have seen most of the tolerance-relatedproblems that would show up, ¡t would not be fair to most of their ships were morethan four years old. Hence, tolerance-related problems was due tothe state that the lack of information on This fact that such problems are just lurkingin the ships, Waiting to show up. deviations will does not mean that deviations have notcaused problems, nor that occurred already. In not cause problems in later years¡f such problems have not regarding lack of at least some cases, the statementsmade by owners/operators such oc- tolerance-related structural problems stemfrom the lack of reporting of have been reported as seawaydamage currences. Often, cracks or structural failures which reinforcing was the recommendedremedy, when ¡n fact or as design problems for Unfortunately for this the problem mght have beenmisalignment or faulty welding. suspected study, but fortunately for the shipsinvolved, the repairs made to correct deviations. design faults usually relieved theproblems posed by construction

14-3 ''- Owner/ per. I 2 #3 # 14 # 5 Item mm inch inch mm inch mm inch mm inch mir

FI an ge Breadth 1/8 3.2 7b Angle 1/8 3.2

9a Gap FlUet t 3/16 [4.8 1/4 6.4 9b min min Butt 1/3 3.2 1/16 1.6 1/4 6.4

9c Lap '1/16 1.6 0

10 Beam and Frame Gap 1/8 3.2 0 11 Butt MIsalIgnment 1/8 3.21/8 3.21/ 3.2 1/8 3.2 t/2 i 2a Reinforcement 1/8 3.21/8 3.2

12b Throat or Leg -1/16-1 .6 -1/16-1 .6-1/16 -1 .6 l2c Undercut 1/161 1.6.03 0.81/16 1.6 1/16 1.6 13 Intercostal MIsalignmentt/2 t/2 t/2 t/2 14 Profi I e Warp 1 /8 3. 2

15 Stiffener Bend 1/2 12.7 lóa Butt - Butt Spacng 6 150 6 150 1 6b Butt - Fillet Spacing 2 51 2 51

TABLE 14.1 Structural Tolerance Standards as Reported by Shipowners/operators

¡4-14 The numerical values of structura]tolerances reported by various shipowners! and metric operators are listed ¡n Table t.l , in both the English units of inches plate thickness. Necessarily, units of millimeters, or¡n terms of a fraction of the only those tolerance standards suppliedby the owners/operators are reported.

As far as deviations on the actualships built are concerned, very little numer- Much of this ical data was obtained from theowners' representatives at shipyards. One other source was discussed above alongwith "in-service" deviations reported. of deviations, "poor detailing ofdesign", was discussed by a few owners/operators. However, this One of them gave a most explicit exampleinvolving access ladder rungs. is hardly a structural tolerance problembut concerns more the design of structural details (see Figures 14.3A, B, and C).

142 SHIPYARDS AND STEEL FABRICATING FACILITIES

14.2.2 Analysis of Responses to the List ofDeviations

The responses obtained fromshipyards in the form of numerical data on structural deviations experienced duringship construction and repair activities were quite varied in completeness and depth. Nevertheless, numerical values were accumulated, and these are listed in Table14.2 both in Engl ish and metric units of measure- the left hand column of the table re- ment. It should be noted that the letters on flect the source shipyard and that bothdeviations and tolerances are listed for each such information was type of imperfection shown acrossthe top of the table wherever furn i shed.

Where numbers do not appear in Table14.2, it means that either no information was made available by the yardin question or that the information supplied was descriptive in nature and contained nonumerical data. Where reference is seen to in connection with the AWS, ABS, USCG, NAVSHIPS, Table,Curve, this denotes that, structural deviation under which these arelisted, the yard in question follows the regulations of the respective regulatory agencies orthat the yard has supplied tables or curves to represent thecriteria they follow. These are listed and included ¡n Appendix 9.2.

In the tabular listing of deviationsand tolerances reported by the shipyards, limited space prevents full explanation of sometolerances. Most listings are straight- forward, but the following qualifications are necessaryparts of the data from several shipyards.

2. Cutting Line Accuracy 1/8'' Where1/8''or 1/16-1/8'' is listed,1/16'' accuracy applies to straight lines and applies to curved lines.

9a. Gap Before Welding, Fillet Weld Where 3/16' is listed as the tolerance,O-1/16'' gap requires the specified weld size, while 1/16-3/16'' gap requires an increasein weld size by the quantity(gap-l/l6''). Also, 2 shipyards require1/8"maximum gap for flat plate and 3/16" gap only for curved plate.

11. Butt Joint Misalignment The tolerance for thinner Where 1/8'' is listed, it¡s applicable only to thick plate. plate is as follows: 25 100

20 80 cumulative rate

Q) occurrence rate 40 Q) >

4-d 20

E L) O 2 14 3 5 6 9 VESSEL AGE Figure 4.2 Damage Occurrence Rate by Age ofVessel (fatigue failures)

(Source: Reference 20)

Figure4.3.A Original Requirement

Full penetration weldswere not being accomplished, the unsealedcuter ends RUNG allowed water to enter the joints, and rungs eventually broke loose due to corros ion.

Figure 4.3.B: Full Penetration Weld

Required to keepa hazardous condition from developing. But shipyards objected to this joint.

Figure14.3.C: As-strong, Easier, and Problem-Free Solution

Shipyards objecting to the fullpenetration weld claimed that thiswas equally strong and easier to do. it finally was adopted and served well.

Figure4.3: Example of Poor Detailing of Design

4-6 Tolerance Plate Thickness 1/32" t < 3/8" 1/16" 3/8"

A quick analysis of the distribution of deviations andtolerances ¡s made in Figs. 14 even though ¡tis realized that for most items the number of data points obtained is simply not conducive to astatistical analysis. Even for the small amount of data points however, the representationsin Figs 14.L4 may still be considered usefulin that they report the minimum and maximum values for each item, as well as the ranges. In these representations the hollow bars denote the tolerance limits and the solid bars denote the deviationvalues reported by the yards.

t+.2.2 Review of Structural Tolerances Supplied by Shipyards

A review of Table-2 as well as Fig. 4J+ reveals that the situation for tolerance data is slightly better. In a few cases such as the fillet weld gap, butt weld misalignment, butt weldreinforcement, weld undercut, intercostal misalignment, and overall length of the vessel, singular spikes areobserved in the graphical representations of Fig. 414 in the tolerance frequency distribution. This shows that the majority of shipyards do indeed try to work tothe tolerances indicated.

In fourteen cases (3, 7a, 9b, 10, 13, 18, 20, 22a, 22b, 22c,22d, 23b, 23h, and 23i) the worst deviations lay at least 50 beyond the most liberal tolerances for the same cases. Of these fourteen, the nine underlined were reported by shipyards that had standards for the relevant cases.

A surrrnary of the structural tolerance specifications contained in NAVSHIPS 0900-000-1000, dated 10/68 (1f), follows:

8.3.1.1 FilletWeldSize

O - 3/8": shall not vary below specified size by more than 1/16" for more than l/ of joint length nor for more than 6" at any one location. 7/16" and up: shall not vary below specified size.

12. contains tb1es of permissible unfairness ¡n welded structure. (See Appendix 9.2.8)

12.1 Butt-type Joints ¡n Plating Thickness Maximum deviation allowable t 318" 1/16"

t > 3/8" 1/8"

lL.5.3 Butt Welds in Outer Hull Surface O - 1/16" reinforcement (for hydrodynamic reasons).

11.8 Buttering or Buildup

shall not exceed 3/8"thickness on each joint edge.

Li-7 I NNITEH Receipt Inspection 2Cutting Line Edge SHIPN Waviiess \Thicknes &Pits/ Lamirations Accuracy Preparation YARD N\inci-i mm \inch mm inch mm inch mm mm A DEV

TO L B DEV

TOL 1/8 3.2 AWS 3.2.3.1 1/8 3.2 ABS 30.3.1 C DEV TOL 1/14 j 6.14 D DEV

TO L E 0EV

TOL .lt 1/8 3.2 F DEV 3/16 14.8 TOL G 0EV 1/8 3.2 11614- '.14 - .8 TOL 3/16 +8 1/25 1. H 0EV 1/8 3.2 1/8 3.2 1/8 3.2 1/16-1/8 1.6 TOL 1/32 0.8 1/32 0.8 1/ii 6.4 1/8 3.2

I DEV JH TOL 25.14 K 0EV

TOLTable 1/614 0.14 3/8 9.5 L DEV 3/16 .5

TOL . 1/16 1.6 1/8 3.2 M DEV

TO L N DEV

TOL O 0EV

TOL P DEV

TO L Q 0EV 1/8 3.2 TOL 1/8 3.2 1/8 3.2 1/16 1.6 R 0EV

TOL 1/8 3.2 S 0EV

TOL Table TABLE 142rn Deviations and Tolerances at Shipyards

14-8 Edge Groove 6îaper Angle \.NITEM a Straightness Automatic Semj-atomatic Manual Depth SHIPN mm mm inch mm inch mm Tch YARD Tf 1mm inch A DEV -

TOL

B DEV ±7° TOL C DEV

TO L

D DEV

TO L

E DEV TOL 1/8 3.2 F DEV 3.2 1/32 0.8±2° 1/8 3.2 1/8 3.2 1/8 I TOL

G DEV o-1/8 3.2±5° 4.8±8° TOL 3/16 4.83/16 4.8 3/16 4.8 3/16 H DEV 3.2 1/8 3.21/8 3.2 1/8 3.2 1/8 1.6 TOL 1/16 1.6 1/16 1.6 1/16 1.6 1/16

t DEV TOL K DEV TOL o-3/1J '4.8 L DEV TOL 1/32 0.8 1/16 1.6 3/32 2.4o-1/8 3.2 ±5 M 0EV

TOL N DEV

TO L

O DEV

TOL P 0EV

TOL 1.65/64 2.0 Q 0EV 1/8 3.23/16 4.8 1/Li 6.4 1/16 2.4 TOL 1/16 1.6 1/8 3.2 1/8 3.2 3/32 2.43/32 R 0EV

TO L

S 0EV

TOL TABLE 4.2 Conti riued

'4-9 a b FabricatedShapes 'TMFlange Flange 8 N c(i) Straightness c(ii) RolledShape SHIPN, Breadth Angle Flange PlaneWeb Plane FlangeAngle YARD 'Ninch mm inch mm inch mm inch mm inch mm A DEV

TUL B 0EV

'TUL ±jJ14 ¶ ±l.° 1.5° C 0EV

TUL 1/8 3.2 ±111+ 6.1+ D 0EV TUL E 0EV TOL ±2° 20 F 0EV 1/8 I 3.2 1/4 6.1+ 1/1+ 6.4 TUL G DEV 25.4±10 1/4 6.1+ 2° TUL ±3° 1/1+710' 6.1+ H DEV 1/273' 12.7 1/2 12.73/4t TUL 3/873' 9.53/8 9.5 t/2 I DEV 1/8 3.2 TUL K 0EV

TUL 1/8 3.2 1/1+ 6.4l/87U' 3.21/8 3.21/4 6.4 L 0EV TOL l.5 1/8 3.2 1/8 3.22° M 0EV 1/4 6.4 TUL N 0EV

TUL O 0EV

TOI P 0EV

TOI ±1/2 12.7 0EV Q 1/8 3.2 1/874e 3.2 l/8H/3OI 3.2 1/8730'3.2 TOL 3/16 4.83/16/1+' 4.83/1630' 4.83/16'/3O' 4.8 3/8 9.5 R 0EV

TO L S DEV

TOI TABLE4.2: Continued

1+-10 IOBeamnd Butt Joint ITEM a GapBefore Welding C Frame Gap Misalignment Fillet b Butt Lap sHIN. mm inch mm mm inch mm inch mm inch YARD inch A 8EV

TUL B 8EV 6.4 1/8 3.2 3/16 14.8 1/8 3.2 TOL<3/16 14.8 '1/4 3.2 1/8 3.2 C 8EV 1/8 3.2 3/16 14.8 1/8 1/16 1.6 1/8 3.2 TOL

TUL - 1" 25.1+ E DEV 1/1+-1/2 6.1+- 1/16-1/8 1.6- TUL<3/16 4.8 3/16 14.8 F 8EV 4.8 3/l6 4.8 <3/16 4.8 3/16 I 4.8 3/l6 14.8 <3/16 TOL 14.8 '1/8 3.2 G DEV 3/16 14.8 3/16 14.8 '3/16 14.8 3/16 14.8 3/16 14.8 1/8 3.2 TUL '3/16 4.8 '3/16 I H DEV 12.7 1/8 3.2 <3/16 14.8 1/2 1/8 3.2 1/1+ 6.1+ t/4 TOL 1/32-3/320.1+- t/2

I 8EV 3.2 TUL '3/16 4.8 1/16 1.6 '1/8

K 8EV 1" 1 25.1+ 3/16-3/1+ 4.8- 3/13/1+ 4.8- '1/8 3.2 TUL 3/16 14.8 1/8 I 3.2 L DEV 3/1+ 19.1 1.6 1/16 1.6 l/8 3.2 TOL<3/16 14.8 1/16 J M 8EV 1/16-3/16 1.6 3.2 TUL 1/4 6.4 1/8 N 0EV

TO L 0 0EV TUL P 0EV

¡ 1/8 3.2 TUL <3/16 4.8 3.2 3/32 2.4 Q DEV iia 3.2 1/8 3.2 3/32 2.1+ 1/8 3.2 3/16 14.8 1/8 3.2 TOL 3/16 14.8 3/8 9.5 1/8 R 0EV

1 3.2 TUL <3/16 14.8 3/11/4 14.8- 1/16 1.6 1/8 S 0EV 1/8 3.2 TOL TABLE 4.2. Continued 12 ITEÌ-i a Weld eometry Intercostal Profile Reinforcement Throaor Leg Undercut Misalignment Warp YARD inch mm inch rim inch mm inch mm ¡n5E mm A 0EV

TO L B DEV t/-t/2 TUL 1/8 3.2 1/16 1.6 1/32 0.8 t/4 C 0EV 1/8' 3.2 TUL 1/16 1.6 t/2 D 0EV 1/32 0.8 TUL t/2 E 0EV TOL t/2 <3/8 9.5 F DEV 3" 76. TOL t G 0EV o t/2 2nt. TUL /8 3.2 «, t 2° H DEV -1/32,+1/8-o81/2 0.8 TUL 1/32 0.8 1/32 0.8 t/2 I DEV t/2 TUL 1/8 3.2 t12 K 0EV 1/2" 12.7 TOL 1/32"c 0.8j/2 L 0EV 3/II: 19.1 TUL 1/32 0.8 t/2 M 0EV

TOL 1/32 O.& t/2 N 0EV 1" 25.4 TUL 0 0EV

TOL P 0EV

TOL 1/32-1/8 0.8- 1/32 0.8 t/2 Q 0EV L116-3132 1.6- TOL 1/8 3.2 1/16 1.6 t/2 1/876" 3.2 1 R DEV TOL 1/32-1/8 0.8-+1/16 1.6 1/32 0.8 t/2 S 0EV

TUL 1/16 1.6 j /2 TABLE 4.2: Continued

4-12 NITEM 5Stiffener Adjacent WeldSping ylinder urved Shell SH!N. Bend Butt Butt Butt - Fillet Diameter Accuracy YARD inch mm inch mm inch mm inch mm inch mm A DEV

TOL B 0EV

TOL 38. l-1/2 1/2 I 12.7 1/8 3.23/16 4.8 C 0EV 1/8 3.2 TOL 6 152. D DEV

TO L E 0EV 2mal 51. TO L F DEV /]6 8.0 12 305. 2 51. 1/4 6.1+ TOL G 0EV 6 152. 6 152. 1/8 3.21/8 3.2 TOL ili+7io' 6.1+ 1/8 3.21/4 I 6.4 H 0EV 2 51. 2° TOL 61+. 3/873' 9.52-1/2 1 25.1+ 1.5° I DEV TOL K DEV 3 76. TOI 1/8710' 3.2 6 152. L DEV

TOI5/16'/3' 8.0 3 76. 1-1/2 I 38. 1/1+ 6.4 M 0EV

TOI N 8EV

TOI 1/16 1.6 0 0EV

TOI t/2 P 8EV

TO L 0EV 3 76. 3 76. 3/32 2.1+ TOI 3/8730" 9.5 1-1/2 38. 1-1/2 38. 3/16-3/89.5 R 0EV TUL S 0EV

TOI

TABLE L2 Continued

4-13 19 SubassemblyAccuracy 20Hatch Coaming21 ITEM b a Access Openings b Lengthaand Squareness Dimensions SHI Dimensions Deformation Width YARDN mm inch mm inch mm inch mm inch mm A DEV

TO L B DEV

TOL 1/8 3.2 7/8 22.2 C DEV

TOL 1/4 6.4 D DEV

TO L

E DEV 1/4 6.4 1/2 12.7 TOL .1 1/4 6.4 F 0EV 1/2 12.7 1/2 12.7 3/37 2 4 3/32 2.4 TOL G DEV .l 1/2 12.7 TOL 3/16 '+.8 1/4 6.4 .1? 1/2 12.7 H 0EV

6.4 TOL 1/4 i 6.4 1/4 1/ii 6.4 ¡/8 3.2 1/8 E 3.2

I DEV

TO L K DEyT

TOL 1/4 6.4 L DEV 3/16 4.8 TOL M DEV

TO L N 0EV 3.2 TOL 1/4 1 6.4 1/8 O DEV

TOL P DEV 1/2 12.7

TOL 1/2 12.7 3/4 19.1 Q DEV

TOL 1/4 6.4 5/16 8.0 1/4 6.4 USCG 1.15CC R 0EV

TO L S DEV

TO L TABLE 42: Continued

14_]14 22a b C ITEMBottom Unfairness Superstructure Shell Deck Sides and Plate Erds YARD 'N inch mm inch mm inch m inch mm A 0EV

TO L B 0EV

TUL 3/8t 3/8t t/2 t17 C 0EV

TO L D DEV

TO L E 0EV 19.1 TUL Curves Curves Curves Curves F DEV

TUL 1/16 1.6 1/16 1.6 1/16 1.6 G 0EV

TOL 1/4'Y3 6.14 1/14 i 6.1+1/14 6.4 1/4 6.14 H 0EV

l-1/2 38. 1-1/4t 1' 1 25.14 3/1+ 19.1 TUL "I I t/6 t 1/2/3 12.7 1/2 12.]

I 0EV 5/8 15.9 TUL 3/4 K 0EV 1/14 6.14

TUL Curves Curves Curves Curves L 0EV

TUL Table Table Table Table M 0EV TOL 5/16j 8.0 N 0EV TUL O DEV

TOL NAVSHIPS P DEV

TO L Q 0EV

TOL 3/8u/30' 9.5 3/87301 9.5 318/36' 9.53/8/30" 95 R 0EV TUL S 0EV

TUL 1/8 3.21/8 3.21/8

TABLE 4.2: Continued

14_15 NITEM23 0vea11Dimensions c d e Forebody Length Beam Depth Keel SH''N jtness Rise inch mm YARD "N inch mm inch mm inch mm inch mm A 0EV -2-1/k1 TOL B DEV

TUL 1 25.4 1/4 I 6.4 C 0EV

TO L D 0EV TOL E DEV 51. -2 51 1-1/2 38. 15.9 2 I 1 Y8 TUL 1 1/2 12.7 1 25.4 F 0EV .1c; TOL G 0EV

TUL .15 .12 .1 .1 1/4'725'I 6.4 H 0EV

TOL 1/2 12.7 1/2 12.7 1/2 12.7 2t

I DEV 7-1/2 191. 5/8 15.9 TOL .1 K DEV TUL 1 I 25.4 1 25.4 L DEV 4 102. 1/2 12.t TOL .1 3/8 9.5 M DEV

TUL N 0EV

TUL . O DEV

TUL P 0EV 1-1/k 32. 1/2 12.7 5/8 is.q 5/8

TOL .1 1/2 12.7 112 12.7 Q 0EV 1/2 12.7 3U TOL 12.7 1/2 I ]fl 76. 1/2's I 12.7 2 .4 /4 1.1 R 0EV

TU L S 0EV -1 102.

TUL .1 1 25.4 TABLE +.2: Continued

k-16 ITEM23 f Overall Dimensions g h Afterbody Deadrise Draft Freeboard 'N. Rise _flarks YARD N. inch mm inch mm inch mm ncn mm A DEV

TUL B 0EV TOL C DEV 1/16 1.6 1/16 1.6

TO L D DEV

TO L E DEV

TUL 1/8 3.2 DEV F 1/8 3.2 TUL G DEV

TOL 1/Lf725' 6.4 .1 H 0EV 1/2 12.7 1/14 6.14

TUL 1/4 6.14 1/8 3.2

I DEV TUL K 0EV TUL 1/16 1.6

L DEV 2 51. 3/8 95 TUL M 0EV

TUL N 0EV TOL O DEV

TUL 1/4 6.14 P 0EV 1/8 3.2 1/8 3.2

TO L 0EV Q 1/2 12.7 1/2 12.7 1/16 1.6 1/16 1.6 3/14 TOL 19.1 5/8 15.9 3/32 2.4 3/32 2.14 R 0EV TUL S 0EV

TUL TABLE 14.2 : Continued

4_17 LEGEND

Absc i ssae: As indicated O rd i na tes: Number of Data Points Solid Bar: Deviation Hollow Bar: Tolerance Limit

Ib Pits 2 Cutting Line 3Edge Roughness Accu racy o

i inch ¡nch inch

Edge Straightness for Welding

4a Automatic b Semi-autonatic 4c Manual

I- 11V 4 inch inch inch

5Groove Depth 6Taper Angle 7a Flange Breadth zo

1111 L 1 11 ¿ 5 78 inch degrees inch

Fig.!4i1: Graphical Representation of Reported Structural Deviations and Tolerances

4 - 18 LEG ND

Abscissae: As indicated Ord ¡na tes: Number of Data Points Solid Bar: Deviation Hollow Bar: Tolèrance Limit o. 7b Flange 7c Straightness z 8 Flange Angle Angle

11111 IT 123 2 degrees deg rees

Ir! r! IT I rl a i (4 ¡ nch ¡nch inch

Gap Before Welding 9a Fillet 9b Butt Weld 9cLap Weld Weld

Min.

Max.

fi' ir ¡4 34 inch inch i nch

10 Beam and li Butt 12a Reinforcement Frame Gap Misal ¡gnment

¼» ¡ nch inch inch Fig. 4i4: Continued

4-19 LEGEND

Abscissae: As indicated Ordinates: Number of Data Points Solid ßar: Dey ¡ a t i on Hollow Bar: Tolerance Limit

13Intercostal

i Misalignment 12b Weld Size 12c Undercut I-1 iii I o J inches

f I nl Ji /32 6 t inch inch

14 Profile Warp 15 Stiffener Bend 16a Butt to zo But t

y6 '4 23 ¿2 inch inch ¡nch

lb Butt to 17 Cylinder 18 Curved Shell Fil let D ¡ a me t e r Accuracy

i 2 J VHi inches inch i nches Fig.4.4: Continued

4-20 LEGEND

Abscssae: As irtdicated Ordinates: Number of Data Points Solid Bar: Deviation Hollow Bar: Tolerance Limit Subassembly

19a Dimensions 19b Squareness 20 Hatch Coaniirìg o Dimensions

lì ii Il'!' 11 11 Fl I Y6i4

inch ¡n ch inch

Access Opening Unfairness

21a Dimensions 21b Deformation 22a Bottom Shell zo

E rl lì'ìIlì I y2 ¡nch inch inch

22b Side Shell 22c Deck 22d Superstructure Verticals zo

lui I 11 I III]lì lì lìÍJ' I ¡/4 .8 3/4 j '4 1.(t '/4 inch inch inch

Fig. t.4: Continued

4-21 LEGEND

Abscissae: As indcated Ordinates: Number of Data Points Solid Bar: Deviation Hollow Bar: Tolèrance Limit 23 Overall Dimensions

a Length o- b Beam c Depth

fi irìch/lOO'

¡nch/lOO'

j: ¡ 2 3 4 7 j inches inches inches

d Keel Flatness o e Forebody Rise f Afterbody Rise

Y . ¡

i iìt [J 6 ¡ '2 3/4 2/4 inch ¡nch inches

g Deadrise h Draft Marks I ¡ Freeboard Marks zo

II i! ' i/a inch i nch inch

Fig.4.4: Continued

A '1 These standards, except for the tables in theirsection12,are quite similar to the American Bureauof Shipping requirements as set forth ¡n the Rules for Building and Classing Steel Vessels(12),(seeAppendix 9.2).

Most yards follow some sort of tolerance standards asfar as the unfairness of plating and misalignment of intercostalmembers are concerned. For the remaining structural deviations, veryfew yards have established written standards, except for the following, whosestandards are reproduced with special permission in Appendix 9.2:

Bath !ron Works Corpration Litton Industries Ingalls Shipbuilding Division Levingston Shipbuilding Corporation Newport News Shipbuilding & Dry DockCorporation Seatrain Shipbuilding Corporation Sun Shipbuilding & Dry Dock Company

14.3 STEEL MILLS

Two steel producers surveyed werepresently meeting and sometimes even exceeding the ABS requirements for materials.

14.3.1 General

The ABS ''Rules for Building and Classing SteelVessels'' has an entire specific chapter (Section143,Materials for Hull Construction and Equipment) in which requirements for mill practices and tests are given. The process of manufacture probably is most important to the ABS. Chemical composition, including ladle analysis, product analysis, and fine grain analysis(when applicable),¡s monitored at the mill also is by the Surveyor. Subsequent heat treatment at the mill or at the shipyard monitored by the Surveyor. For each ''heat', tension tests on carefully selected specimens are performed. Material for castings and forgings is subject to bend tests. Grade D,E, DH, and EH materials are subject to impact tests.

No measurements are required by theABS for plate or shape dimensions. These measurements are supposed to be handled by the millsthemselves, under the American Institute of Steel Construction Specification, the American Society for Testing Materials standards, and the American Iron and Steel Institute rules.

AISC's section on Standard Mill Practice is a summaryof the ASTM Designation A6, which itself is a group of common requirementsfor rolled steel plates and shapes:

Plates are to be checked for thickness and weight, width andlength, camber and flatness.

Structural-size shapes are limited to2.5 variation from theoretical cross-section area or weight, and dimensions are to be checkedfor variations ¡n cross-section, ends out of square, straightness, and length.''

ASTM permits1/14inch variation under ordered length and width.

14-23 14.3.2 Results of Surveys at Mills

Both steel mills stated that they work basically to the standard tolerances of the AISC (American Institute of Steel Construction), ASTM (American Society for Testing Materials) and MSI (American Iron and Steel Institute). Closer tolerances are generally observed for alloy and higher strength steels, particularly armor plate and steel for nuclear plants. One mill said they can work to 1/2 standard tolerances or any other tolerance level required by the purchaser but at extra cost.

The following numerical values for actual deviationsexperienced on steel material produced were furnished by one mill:

The allowance for thickness during the rolling process is generally about 10 mils over on the edge of the plate, sothat when the plate cools, the thickness will be closer to the nominal value. For a nominal plate thicknessof1/14", the variance on the finished productwill be .23811 to .255'. On the average, the finished plate is 3 to 14 overweight from the standpoint of thickness.

Plates, after torch-cutting and when they are ready forshipment, are approximately50/e overweight from an overall viewpoint, i.e., considering both their thickness and size against nominal values.

Laminations are occasionally found on rolled plates. It was stated that the laminations could never be totallyeliminated; however, they are reduced to less than l/2 of the plate production. Frequency of finding laminations depends on the thickness of the plate.

Corrosion and pitting also may be present. However, the more serious problems are scabs, slivers, etc., which require con- dit ioning of the plate by grinding. Alloy steels are subjected to a greater degree of conditioning.

¡414 CLASSiFICATION SOCIETIES

General

The American Bureau of Shipping ¡s the dominant classification society for building in United States yards. The U.S. Coast Guard accepts the current standards established by the ABS and designated "Rules for Building and Classing Steel Vessels" regarding material and construction of hulls, boilers, and machinery, except that their standards generally are compared to ABS standards to determire their acceptability by the Coast Guard.

Specific parts of the ABS rules are excerpted below:

" 214. Vessels Intended to Carry Liquified Gases

214.27 Non-destructive Testing

14-24 All butts and seams of welded primary containersfor cargoes at -73°C (-100°F) or colder are to be completely radiographed unless they are tested by ari alternate approved procedurefor nondestructive for cargoes testing. The butts and seams of welded primary containers above -73°C are to be radiographed at allintersections and at random locations of the butts and seams to the satisfactionof the attending Surveyor. The method of nondestructive testing of nonstructuralpri- mary containers or secondary barriers¡s to be specially considered.

30. Welding

30.3.1 Edge Preparation and Fitting

Weld build up should not exceed t12 or 1/2"(12/5 mm) on each plate edge.

Where sections to be joined differ in thickness and have an offset on any side of more than 1/8" (3 mm), a transitionhaving a length not less than three times the offset is to beprovided. The transition may be formed by tapering the thicker plate orby specifying a weld joint designwhich will provide the required transition.

30.3.3 Cleanliness

Slag and scale are to be removed not only from the edqes to be welded but also from each pass or layer before the deposition of subsequent passes or layers.

30.5.7 Fairing and Flame Shrinkage

Fairing by heating or flame shrinking and other methods of correcting distortion or defective workmanship in fabricationof main strength members within the midships portion of the vesseland other plating which may be subject to high stresses is to becarried out only with the express approval of the Surveyor.

30.5.9 Inspection of Welds

Radiographic or ultrasonic inspection or both is to be used when the overall soundness of the weld cross section ¡s to be evaluated. Magnetic particle or dye-penetrant inspection or both is to be used when investigating the outer surface of welds or may be used to check back chipped, ground or gouged joints prior to depositing subsequent passes. Some steels, expecially higher-strength steels, exhibit a tendency to delayed cracking. When welding these materials, consideration is to be given to delaying the final nondestructivetesting to accomodate occurrence and detection of such defects.

30.9 Fillet Welds 30.9.1 General

The weld throat size t is not to be less than 0.7 times the weld leg size w . Where the gap between the faying surfaces of members exceeds 2 mm or 1/16 and ¡s not greater than 5 mm (3/16") the weld leg size is to be increased by the amount of the opening. The gap between members is nct to be greater than 5 mm (3/16").'

4.2.2 Results from Classification Societies

American Bureau of Shipping (ABS)

ABS cited a structural detail where a plate was being welded to another plate situated perpendicular to the first plate. At first, the vertical plate was welded to the horizontal plate by a full penetration weld from thetop and by a fillet weld from the bottom. (See Fig. 4.5A) . In this configuration the vertical plate had a tendency to crack at about the middle of it. In one case where this happened, the plate was thought to be deficient and a new plate was fitted and welded in the same manner, and sure enough, the same crack occurred. It was then decided that the reason for this crackwas not a deficiency of the plate itself, but rather, a result of the shrinkage of the full penetration weld. To prevent this, the configuration shown in Fig. 14.5B was used withsuccess. This sort of failure was perhaps justly called ''lamellar tearing''; however, thereason for failure was not the plate deficiency, but the shrinkage of weld. The same type of failure was seen to occur in the case of explosion bonding of aluminum to steel, however this detailis usually located in areas above the main deck level so that there is only a compressive loading on the explosion-bonded joint and the compressive force wiH not present any danger to these normally failure-proneareas

(see Fig. 14.5C) . It follows, therefore, that no remedial action is necessary.

ABS allows mill tolerances on the minus side. However, when it comes to the owner's tolerances they do not allow this because this would cause a loss in the corrosion allowance for the owner.

Bureau Ventas (By)

BV has no publ ished standards for tolerances, however, they do have a compilation which is reportedly treated as confidential.

BV has another publication,in addition to the ''Rules'', which is titled ''Instructions for Surveys of New Construction Steel Vessels'' which also is confidential. This publication does not include tolerances in the form of a tabulation, but gives some guidelines.

BV had statistical information on actually measured tolerances on ships built in French yards, but they would not release this information due to insurance premium complications.

Nippon Kaiji Kyokai (NK)

NK does have a set of standards on ship structural tolerances ''Japanese Shipbuilding Quality Standards - Hull Part''(2) and the most recent edition is dated 1975. LAME L LAR TEAR T NG

Fig. Lf.5A: Full Pene- tration Weld

Fig. 14.5B: Alternate Deta i

Fig. k.5C: Explosion Bond i ng NK has based these standards on a large research effort conducted by the Society of Naval Architects of Japan (SNJ)in a number of Japanese shipyards. Deviations were actually measured on many structures, histograms of deviations were made and from these the standard ranges and maximum tolerance levels were established for each and every structural deviation. The results of this full-scale effort is published by SNJ in the Japanese language.One chapter of this publication, Chapter IX,is included in Appendix 9.3.

NK X-ray examination for the hull envelope of the ship, includinq strength deck and the shell plating and the number of X-rays are established on a random sampling method and the maximum number is about 150, depending on the length of the ship; this number can be reduced to 100. However, the actual number of X-rays to be taken on any one ship depends on the accuracy of the results obtained from the X-rays already performed. If some of the results are not satisfactory, then this number can be increased.

The NK criteria to accept or reject a weld, perform visual examina- tions of the welds, and evaluate X-rays of the welds, follow the Japanese indus- trial standards methods. The butt welds in the vertical areas, the faceplates of frames, beams and girders are required almost exclusively to be X-rayed. How- ever, when the physical location of any faceplate or flange or girder makes ¡t difficult to X-ray, then UTS may be substituted, depending on the aforementioned cond i t ions.

Lloyds Register of Shipping (LR)

LR has periodicals and papers that are published for the purpose of delegating senior surveyors' experiences to other surveyors in the staff. They have another publication, ''Instructions to Surveyors'' (confidential) which treats structures, deviations, and methods of repair.

LR has accumulated a great deal of statistical data ori failures of and damages to ships since lOL2. Also included ¡n this accumulation is an analysis of the failure and the methods of repair. This information is fed into a computer program and the dataare processed and evaluated for future reference. Det norske Ventas (DnV)

DnV cited their Rules, Chapter 1, Section 6C; Chapter II, Section 3E; and Chapter X, Section 3A, as being applicable to workmanship, minimum tolerances, and repair of defects. These sections were found to be written in general terms, no more specific than similar specifications by other classification societies.

Registro Italiano Navale (RIN)

RIN excerpted the RINA rules on structural steels, welding (processes, joints and edge preparation, workmanship, NOT, and repairs), special welding processes, hull structures (material, welding, repairs, and inspection), design and fabrication of welded structures, and control procedures for weld defects and included recommendations for fit-up alignment, deflection, straightening practice, welding flaws, flaws in base material and thickness variations. Research papers on quality standard/quality control and NDT of shipboard welds, especially with UTS, were appended (8,26).

4-28 i+.5 FOREIGN INSTITUTIONS Deutschen The Association of the GermanShipbuilding Industry (Verband der Schiffbau-Industrie e.V.) supplied the'Production Standard of the German A number of West German yards Shipbuilding Industry" (15),(Appendix 9.3.2). and worked in preparing thisstandard. It is being revised continually, contains, as of this date, thefollowing subjects:

Surface defects and laminations Edge Preparation Welds

14 Component part fabrication Sub-assembly Fairing work Final work Tightness test Hull-main dimensions

Standardcentral') The Swedish ShipbuildingStandards Center (Varvsindustrins supplied VIS 530, Accuracy in HullConstruction (16).

AB laboratory provided a paper onfillet The Kochums Mekaniska Verkstads No other papers were available welds (17) as compared to fullpenetration welds. in English translations. Association has recentlycommenced collecting The British Ship Research under information on structural tolerances, onbehalf of the British industry Technology (ATS). a government supportedproject, Advanced Shipbuilding

.6 OVERVIEW in the course of A complete review of alldata and documents accumulated identify widely used values surveys and interviewsshows that it is possible to in Section 3. in connection with most of thestructural deviatiors listed widely used Using the same numbering systemfor various deviations, the most values in United States shipyards appearto be as follows:

Pits in incoming material, up to1/16:(1.6mm) re normally (3.2mm) or perhaps ground smooth. Deeper pits up to 1/8" slightly larger are filled withweld and then ground smooth. In thick material, 1" or more,deeper pits may be tolerable in mild steel. However, a large numberof pits (deeper than in incoming material areconsidered reason for rejection. 1/8") with In some instances, deep pits maybe repaired in accordance sect ionI3.3.7 (b) of ABS Rules.

equipment and Cutting Line Accuracy isgreatly dependent on shop is not uncommon, and controls. Deviations up to 3/16"(L1.8mm) this appears to be the presentlyidentifiable United States shipyard practice.

4-29 Th range ofedgepreparation tolerances found in thesurveys is large, running from .04 inches to.125 inches. This indicates that the tolerance ¡s more a function of the platecutting equipment than of the necessity of obtaining satisfactory welds. Ship- yards that use relatively crude flame-cutting equipment,or which allow cutting in place after mounting materialon the ship, may have to accept an edge roughness ofup to 1/8". Yards that use the newer plasma-arc cutting equipment may be ableto achieve edge roughness as low as 1/25".

Edge Straightness affects thecare that must be taken to ensure that good welds are made. As the degreeof automation ¡n weld- ¡ng procedure increases,so does the need for ed9e straightness. Otherwise, manual control of automatic weldingmachines ¡s necessary to produce uniform welds ¡fhe groove varies in width and wanders from side to side of the machine'strack. Such manual control would defeata rnaor purpose of having automatic welding machines:labor reduction.For manual welding the (ini tedStates practice seem&-to be + 1/8" (3.2mm), andfor automated welding- ¡t -isin the order or+ 1/16'' (1.6mm)or better.

5,6 Groove Dept ndTcperngJe, values most commonlyfound are + 1/8" (3.2mm) and + 5 respectively.

7. Flange Breadth and Angleon Fabricatec 5hopes are important where the shapes must pass through bulkheadsor floors, especially when the intersections are to be made watertight.Tight tolerances reduce theamount of field work to make inserts fit andfoweld the intersections.These two dimensions also are important where shapes must be butt weldedat their ends. The representative valuesappear to be:

O. Flange Breadth: + I/ti" (6.4mm).

Flange Angle: ±l/4"/l0 (6.4mm/3m) or + 1.5°

Straightness of Fabricated Shapes is dependenton assembly procedure.If adequate allowance is made for adjusting beforewelding, this is not critical. But if force-fitting must be done too shape, with consequent residual stresses which will be high because the section modulus is high both forx-x and y-y axes, then straightness is critical, and the most widely used limit in such cases appears to be better than t/2 bend over the length.

4-30 for the Flange Angle ofRolled Shapes 3. The attainable level 1.50. appears to be the same asfor fabricated shapes, +

Accuracy. Gap Before Welding isdependent on Cutting Line Enough inforrtation existsin literature on the treatment for various gap sizes andwelding techniques.

of fabricated Beam and Frame Gap issimilar to the problem maximum gap shape straightness. Before forcing to fit, the welding, the gap for used is l/2 (12.7 mm), while before lap welds is considered toprevail.

for pleasing The maximum misalignmentnormally allowed ¡s t/ appearance and forstrength and fatigue. In out-of-sight locations that are not strengthcritical, up to t/2 is allowed.

(1.6mm) above flush is all Weld reinforcement up to1/161 Any more weld reinforce- that is normally considerednecessary. weld cost, and ment ¡s looked at as simply increasing the is required, leading to problems whenhigh fatigue strength bead since the notch formed atthe edge of a reinforcement is a surface stressconcentrator. welds is + 0, and Weld size used fornominal l/L" or smaller for larger welds it is +1/16'' (1.6mm). is 1/32'' Weld Undercut allowed in mostUnited States shipyards parallel to the weld, and moreeffort is (0.Smm) or less when stress is weld. However, made to eliminate it when stressis perpendicular to the for the welder to examinehis any undercut isconsidered to be cause equipment and technique.

the Intercostal Misalignment atcruciform joints should be levels for misalignment subject of more testing. The presently attainable appear to be:

for non-strength members t non-critical locations t/2 for strength members in and bottom of t/3 for strength members at top hull girder in midship 1/2length

L-3l The presently attainable limit for profile warpseems tobe about + I/k" (6.4mm).

Stiffener Bend is similar to straightness of fabricated shapes.

Minimum Distances between Butt Welds and between Butt Weld and Fillet Weld need more research to determine the effects of Heat Affected Zone as well as shrinkage. Dependence on base material and on fixity of surrounding structure is important.

For cyl inder diameter the limits normally used for subassembly and overall dimensions are applied unless special design requirements require better tolerances.

Curved Shell Accuracy directly influences unfairness of the finished plate panels. The most widely used limit, for most shell plating,is +l/4''(6.4mm).

Subassembly Accuracy of dimensions is maintained in most shipyards to a limit of + O.O5. Out-of-squareness is considered more tolerable than over or under size, and the widely used limit is O.l°/.

22 Unfairness of all hull plating need not be uniform for bottom, side, and deck because the need for fairness varies, e.g., bottom and lower side shell should be fair for hydrodynamic reasons, upper side shell should be fair for aesthetic reasons, and deck plate should be fair to prevent puddle formation and to prevent damage by large cargo items. Unfairness of plating, especially large spans of plate, may bea function of improperly cut or fitted support structure as well as a function of cooling after welding (hungry- horse effect). If the plate were forced to make up a gap between itself and frames, the result would be unfair shell and high residual stress in the plate and frames. The practice Tn the shipyards surveyed is generally to follow the unfairness limits given ¡n reference (4

23. Overall Dimensions of L,B,D, keel flatness, and deadrise are normally kept within O.l accuracy. Draft and Freeboard Marks are maintained within 1/8'' (3.2mm).

The numerical values for those of the above structural deviations which appear to be the most widely found and which can therefore be considered to be the presently attainable levels in United States shipyarcs with modernized equipment and controls are listed in Table 7.1.

4-32 5.0 FOLLOW THROUGH OF A TYPICALSTRUCTURAL DEVIATION of Any individual structuralimperfection, as categorized ¡n the listing structural deviations in Section 3 of thisreport, ¡s the result of amultitude the of factors and/or other structuralimperfections existing ¡n the material or structural deviations are inter- assembly. This is equivalent to saying that most dependent on each other. contributing to the The objective of this section isto review all factors misalignment of two opposing creation of one typical structuraldeviation: of ship design members attached to a through member. In doing this, all phases imper- and construction are considered, the causesand effects of interdependent fections are investigated from a practicalviewpoint.

5.1 FACTORS RELATED TO IDEAL DESIGN

The ideal structural design requirementfor a ship is for it to stand external seaway excitation up to the loads thatinternal weight distribution and Other loadings, such as machineryvibration, may serve to complicate generate. analyses of the matterS Using modern techniques,engineers can produce accurate structural loadings and responsesprovided the input¡s accurate. In many engineering structures the majority ofloads can be determined accurately, so produced and that designs no more than adequatefor the service intended can be the loads cannot be pre- used safely. However,in ocean engineering structures ship may have one or dicted accurately. Over a 20-year service life, a cargo more fundameRtal changes in typeof cargo carried, with resultant changesin the the unknowns in the internal weight distribution. More important, however, are the uncertainties seaway environment. Changes ¡n service route only make worse about such factors as maximum expectable waveheight and the frequency of occurrence of various lesser waveheights. Safety factors are necessary to cover the designers have gaps in knowledgethat such changes and uncertainties produce.Some used sophisticated ship motion studies toprovide a more accurate determination safety of seaway excitation and ship response,hence to allow the use of reduced rely on the experience-derived factors. The majority of naval architects still formulae ¡n the rules of the classificationsocieties. This practice ¡s not totally inaccurate because the experienceof the past provides a reasonable pre- diction of the future as long as other factorsremain relatively constant.

Granted that structural loadings onships are known only approximately, still the levels of arbitrariness and factorsof safety in most design criteria are quite high. This is because the execution of ideal designsin actual construction is subject to unknown deviations from thedesigners' plans. The experience- derived formulae try to account, implicitly,for all unknowns. If a designer could know that joints would line up to acertain degree and that the material and welds would have a certain level ofconformity with specifications, he could combine ship motion analysis withsophisticated strength and fatigue analysis techniques to determine the scantlingsand arrangements. Most designers cannot obtain the requisite information, sothey must fall back on the experience-derived the stakes are so criteria in the rules. In a few cases, notably the LNG ships, high and the experience is so scant that anextensive effort has been made to ideal design details. assess and control theeffects of deviations from nominally

5-1 Even here the process has not been direct. The designers applied arbitrary material and weld safety factors and then used tests and mathematical analyses to determine the tolerance levels that would be required to keep stresses within "safefl levels. Instead, the process of achieving accuracy in the shipyards could have been assessed, and the designs made with safety factors appropriate to the ascertained accuracy level.

5.1.1 Deviations Originating from the Base Materials

The deviations from idealized design begin with the condition of the material, which is a function of mill practice and inspection procedures at the mill and the shipyard. Not all shipyards bother with receipt inspection. Indeed, some depend on the classification society and mill inspections to ensure delivery of acceptable material. The ABS (and presumably other classification societies) perform surface inspections only when specially requested to do so. The societies do ensure that the basic chemistry and material properties are correct, and this of course is fundamental to having the actual structure conform to the design. But surface-detectable flaws are important as well.

The most obvious material flaws are pits and corrosion, both of which have been related to undercuts in welds. Corrosion products generate additional slag during welding. Most welders know enough to remove oxide films from the weld area before welding, so that the problem is somewhat reduced. Pits, however, cannot be removed by wire brushing. A welder making an in-tolerance weld might cause short duration or large, harmful undercuts ¡f he were not careful to notice that pits start undercuts that require special attention to stop. It ¡s shown in Section 5.2 that undercuts can be harmful or inconsequential depending on their location, but they are generally undesirable. Hence, anything that adds to a welder's problems in avoiding undercuts is undesirable. Those shipyards that do perform receipt inspection reported pit tolerances ranging from 1/64" to 1/8", while their undercut tolerances ranged from 1/32'to 1/16".

A less easily detected flaw ¡s lamination. As pointed out in Sec- tions 3 and 5.2, laminations, in special cases, cause serious problems in welds. Small laminations often grow when heated by welding. A lengthwise stressed member would not suffer from this. But a thickness stressed member would suffer a significant loss of effectiveness. For example, at a perfectly aligned cruciform joint where the non-continuous members were lengthwise stressed, varying tensile loads eventually would open up the outer layers of a laminated continuous member. Since a laminated plate acts as a layered set of thin plates, withre- duced resistance to shear between layers, the plate would bend more easily. The problems of fatigue at cruciform joints that Reference 21 discusses have been examined by some U.S. shipowners using finite element methods;it has been found that fatigue propagation would be accelerated by the laminated condition of the plate.

Fortunately, most laminations occur at or near the edges of rolled plates, so they are visually detectable. Ascertaining the extent of lamination requires sophisticated techniques, of which ultrasonic inspection probably is best. Some shipyards already use UTS to check for laminations in receipt inspection. AH material need not be examined, only that for those areas that will be subject to thickness direction loading.

5-2 l .2 Deviations Originating from Processing ¡nShipyards

After material has been received inthe yard, and perhaps blasted and primed to reduce wastage during storage,the deviations due to processing ranging from lofting inaccuracies commence. The various manufacturing processes, to erection and fairing ofprefabricated subassemb]ies, are subject toinherent inaccuracies.

Two British Ship Research Associationreports (2t,25)show the results of investigations on the accuracyof using 1/lo scale drawings to guideflame- the path of a flame-cutting cutters. The initial full-size error imparted to head was shown to lie in the range ±0.18". u.s. shipyards have reported tolerances on accuracy of cutting line inthe range of 1/16" tol/4". This the ship, error ¡s important becauseif two plates, intended to butt together on reduced were cut by the same cutter,the gap before welding could be increased or kept at the by a maximum of 1/2". Alternatively, ¡f the gap before welding were nominal value, the dimension of a subassemblycould be changed. One shipyard specifically pointed Out that plates are measured asthey come from cutting and that ¡f one comes out undersize, the nextis deliberately cut oversize.

Accuracy of cutting line ¡s importantbecause it influences the fitting together of components, hence directlyaffects the effort necessary to make Upon examin- the fits correct as well as the costof other than nominal welding. ing the graph "General Relationship of Cost toDimensional Accuracy," Fig. 5.1 low level of dimensional accuracy it is easy to agree that, after a relatively has been achieved, the fabrication shop costswill rise. Because a fabrication shop deals with individual subassemblies, acertain level of accuracy is necessary be superfluous to the to put them together. But beyond that, more accuracy may dimensional accuracy is immediate task. The behavior of the berth cost versus for slightly more complicated. Again, a certain level of accuracy is necessary the assembly, or ¡n this case, the erectionand joining of subassemblies. But ease of erection is dependent onthe accuracy of work produced at earlieroptimal accuracy level; the berth workers mayhave to rework the subassemblies. Hence, berth work requires that a higher level of dimensional accuracybe maintained from the beginning so that the workers can use theirtime for primary production and not for correction of faults in the elementsproduced at previous stages.

After cutting and shaping, the fabrication processencounters the inaccuracies inherent ¡n fit-up operations. These cover a wide spectrum of situations, and only the selected typical deviationof constructing cruciform joints will be considered here.

Misalignment of intercostals where they are welded tothe through member, forming a cruciform joint, often is cited asthe classic structural tolerance problem. The misalignmentwould seem to be the direct result of

5-3 inaccuracies in fit-up, or the consequence of the difficulty ¡n checking alignment when the through member ¡s a deep transverse or a bulkhead. The cure for inaccuracies obviously is better workmanship, includingcare- ful measurement, precise positioning, and adequate clampingor tacking. The cure for the second cause could be to drill holes to make accurate measurements on both sides of the through member, or to use a non-destruc-

tive device such as an ultrasonic "Locatron' for determining the exactpo- sitions of opposite members.

If the intercostals were individually installed, aligned, tack welded, and finish welded, al ignment would be an independent matter at each cruciform joint. However,in many modern shipyards the problem is not as simple as the preceding discussion suggests. Today, prefabrication of panels and even 200 ton subassemblies is common, and other sources of alignment problems have arisen. For example, one shipyard reported that structural shapes as received from the steel mills are not suitable for shipyard work. If some angle beams received from the mill or fabricated in the yard were bent in tie plane of flange, then even careful measurement and workmanship in constructing egg crate assemblies might not prevent misalignment of intercostals whee the subassemblies join together.

The reported tolerance for straightness in flange plane rangedover 1/8'' in 10', 1/Li»' in 10', and 3/8' ¡n 3'. The first two tolerances should not cause much problem in joining subassemblies. But the third could weil

lead to substantial misal ignment if the beam protruded from Subassembly i into Subassembly 2, as shown in Figure 5.2. The beam could be aligned perfectly at Transverse A and at the edge of Subassembly 1,but at Transverse B it could be off by 3/8" for every 3'from the subassembly edge to Transverse B. Additionally, the beam might be misaligned at A and at the edge. The best case would be for these two errors to combine with the beam curvature to result in zero error at B. The worst case, assuming an alignment tolerance of t/2, would be an error of 3t/2 at B, added to the error caused b beam curvature. These of course are pre-welding errors, but unless they were detected, theywouldbecome welded-in misa] ignments. And ¡f they were detected, the correction procedure of force-fit would generate large locked-in stresses.

The locally generated problems with cruciform joints are augmented by more global problems such as alignment of subassembi ies relative to each other. Erection and fairing of subassemblies to form a ship on the building ways is compl icated by several factors. First, it ¡s difficult to pos ition the units exactly because the reference lines and methods of location have inierent inaccuracy. As one shipyard noted, they have no problem with shifting building ways, but constructional inaccuracies in the ways themselves stil] disrupt the accuracy of the reference grid they use. This is aggravated by the inaccuracies of the tools used to locate the units relative to the reference lines. The tapes stretch, and plumb lines deflect in any breeze. This inaccuracy is being reduced by use of surveyors' transits and, more modern yet, laser beam devices, but such exotic devices are not commonly used. Second, the dimensional accuracy of each subassembly unit is not well known. Distortion due to temperature effects and lifting forces confuses the situation. Third, the accuracy of the part of the ship already erected ¡s affected by welding

5-Li Fabrication Shop Cost

Optimum Quality Level

Berth Cost

Dimensional Accuracy

Fig. 5.1: General Relationship of Cost toDimensional Accuracy (Reproduced from Reference 25)

Transve rse Transve rse Edge B A

Subassembly i Subassembly 2

Fig. 5.2 Alignment of Subassemblies

5-5 and by settling of keel blocks and the berth itself, all tending to ¡nhibit precise knowledge of the relationship between the designed and actual situation of the ship during construction. As evidence, most shipyards have reported that ships grow by a maximum of 1" per 100' in length, and that bows cock-up by 1/14" to 1/2" per 25'.

Another misal ignment problem, that apparently few shipyards mairtain any control over, involves the angle between the flanges and webs of intercostal beams. If, at a cruciform joint, one beam had an up-tilted flange and the other had a down-tilted flange, the shipyard tolerance of 1/14" to 1/2'' tilt could lead to 1/2'' to1'' misalignment of the outboard end flange fibers. Fortunately, this misalignment decreases with proximity to the beam web, so only bending loads in the flange plane cause appreciable twisting stresses due to the misalignment. Since most loads on intercostals are lengthwise, this misalignment problem probably is of little importance.

Also to be considered at cruciform joints are allthe weld-qual ity related tolerance problems such as slag inclusions, porosity, gap before welding, and heat distortion. Some shipyards have complete specifica- tians on allowable weld defects, while others leave the matter entirely to "good workmanship". The classification societies have their own inspection tolerances, e.g. ABS has a publication, "Nan-destructive Inspec- tian of Hull Welds'' (9). There are two important aspects of all these specifications: NOT is, as discussed in Section 14.6 of this report, a process control mechanism to ensure that good workmanship is practiced. Second, combinations of problems must be considered. A misalignment of t/l combined with perfect welding could produce a better cruciform joint than a combination of t/2 misalignment, 1/16'' undercut, slag inclusions and porosity up to the ABS limit, and large residual stresses. This illustrates that complete tolerance specifications still need to be tem- pered with fabricators' and inspectors' judgernent.

A Bath Iron Works study (18) and a Royal Institution of Naval Archi- tects paper (19) have shown that, regarding gap before welding For fillet welds, shipyards would do well on cost basis alone to regulate their processes to produce accurate gaps. "if the fitting is poor, the cost of fillet welding will mount rapidly." In Ref.18,it can be seen that cost increases with oversize gap.

5.1.3 Precautions to Reduce Deviations at Source

The question naturally arises of what to do to reduce the problem at cruciform joints and to reduce other tolerance related problems.

Receipt inspection can reduce wasted effort at later stages by reject- ing or sending for repair those incoming materials that are unsuitable. Immediately following inspection, blasting and priming can prevent suitable material from becoming unsuitable, especially regarding welding, during storage and exposure in unprotected subassembly stages.

5-6 Following receipt inspection and processing,the control of fabrication processes becomes paramount ¡n makingthe actual product conform to the design. Cutting line accuracy has been shown to beonly fair when 1/10 scale drawings are used as guides forflame cutter head. One of the B.S.R.A. reports (25) favored,in 1968, the replacement of manual lofting techniques by computer-aided design systemsfrom which could be derived control tapes for numerically controlledflame cutters, frame benders, However, ship- etc. Numerous shipyards are presently using such systems. yards not equipped with N/C equipment canstill improve cutting and forming quality by rigorous monitoring andcorrecting, and even compensating for deviations.

The construction of subassembly unitsshould be seen as a middle step in the construction process rather than as a stepto be optimized for the immediate problem (sub-optimized). Referring again to Fig.5.1, ¡t isim- portant that the subassemblies beconstructed with what appears to be too much accuracy, to save re-work later.

The problems with prefabricated subassemblies canbe reduced by making adequate allowances for adjustment of theinevitable imperfections in material and workmanship. One shipyard learned the hard way thatcompleted subassemblies are very difficult to mate togetherexactly, even when extreme care is exercised. In the example in Figure5.2,the intercostals should not be welded at or near theedge, so that adjustment at C would be relatively easy. As long as the beams are within straightnesstolerance, it makes no difference in structural strength¡f the beams are not on the mark at the edges after joining of thesubassemblies.

and join- Ref.23advocates the use of a grid system so that erection ing of subassembl ¡es can be done relative tothe grid rather than relative to the subassemblies already inplace. At least one of the shipyards surveyed ¡n this project uses a similar system. The virtue of such a system is that it prevents an error ¡n onesubassembly from cascading through subsequent work. lt also helps to isolate and expose deviationsbefore they become firmly built in. The grid system of course must be very accurate and must be readily accessible for reference measurements. The problem with such a system is that it might unduly restrictthe whole process of erecting and joining completed subassemblies. As indicated ¡n Section 5.1.2,most shipyards have found that theirfinished product is longer than This has been its nominal length by as much as 1 inch per lOO feet. accepted as part of the price for the virtues ofsubassembly methods. A pre-established grid system would not allow such acascading deviation, but would require greatly improved subassembly constructionand joining methods, and hence might become an impediment to overallproduction rate.

5.-7 5.2 INFLUENCE OF STRUCTURAL DEVIATIONS ON STRENGTH

Very few ships that were reportedly inspected in accordance with previous

or current structural and weld tolerance standards have failed in service. t has been difficult to establish that the ships that have experienced failures did so because of structural deviations, or that ships thatwere known to have substantial misalignments, etc., were failure prone. Only otir substantive examples were obtained:

One shipowner/operator reported that misalignment of intercostallongi- tudinal bulkheads on passenger ships caused fatigue cracking. Misalignments of up to I" had to be corrected by breaking out and rewelding the bulkheads or by attaching brackets to provide continuity.

In some cargo ships converted to container ships, hair-line cracks developed at the butts in the longitudinals of the box girders.A finite element analysis showed that many discontinuities that ordinarily wouldnot bother an inspector were the cause of the cracks. Both poor design details and slag inclusions in manual welds were found to be the problemsources. This example illustrates a situation where stresses were high, hence tolerances (or perfection) were critical. However,it also may illustrate a case where the design was not adequate for unexpectedly high stresses.

Designers and builders of LNG ships used finite element analyses to investigate the tolerance allowed for alignment of butt and cruciform joints. They found, as did the researchers for the JSQS, that 'critical'' joints required a limit of t/3 rather than t/2 to ensure that stress levels would not be exceeded and fatigue problems would not be encountered.

A vessel constructed in one shipyard developed serious structural deficiencies shortly after ¡t entered service. The yard was called in and an extensive alignment survey was conducted which revealed many sources of failure due to misalignments of various types. All deficiencies were corrected by the shipyard, at considerable cost, and this incident was sufficient reason for the yard to incorporate into the yard organization a comprehensive qualityassurance department.

Reference (20) contains the detail shown in Figure 5.3as a detail design problem (poor detailing of design). However, it is also a structura] tolerance or deviation problem, since the design in A arose because B was difficultto achieve during construction. If close tolerances were followed, the design ¡n B would be achievable. It should be noted that two detail problemsare involved here:

(I) The bracket ¡n A must transfer deck load from beam to frame, while the bracket in B has less load on it,

(2) The scallops in Aare sharp-cornered stress concentrators, while B has no scallops at all.

5-8 Bean Continued Shell to Frame Deck 1_I

Bracket

Scallops Crack DETAIL B DETAILA

Frame

Bracket Fig. 5.3: Detail Design of a Beam (Source: Reference 20)

influence exists ¡n publishedliterature regarding A wealth of information engineering as well asships' structures: of deviations on thestrength of general oxide inclusions, reduce Laminations in plates,produced by suiphide or direction and reducethe strength ¡n the strength intension in the thickness ¡s fabricated 5.0. While steel usually bending, as discussedin Sections 3 and fibers, stress design stresses areparallel to the lamination so that major entirely. perpendicular to the fiberscannot be avoided phenomenon is lamellartearing, which sometimes A very closely related due to the complexity construction at majorstructural intersections occurs during Its which results in highthrough-thickness stresses. of weld connections normally related to the presenceof non-metallic inclusions occurrence has been tearing range from reducing present in structuralsteel. The cures for larnellar these inclusions toredesigning the weld connections. cruciform joints has beeninvestigated Misalignment of intercostals at USSR, and by finiteelement analysis ¡n by destructivetesting in Japan and the strength caused bymisalignment is significant, the U.S. The decrease in joint through Fatigue problems that arisedue to bending of the but not as much as the investigate presents a summaryof worldwide efforts to Reference 1 member. other type joints. the effects of misalignment in cniciform and of cruciform joints, theauthors of a Soviet 1. Regarding the misaflynment paper (21 )reach the followingconclusions: occurring during assemblyof a Displacement and non-straightness, The number joints, considerably increasethe number of fit-up jobs. reduced with the aid oftechnological and of fit-up jobs can be section construction administrative neasures toincrease the accuracy of for the dis- and mounting on the ship,and to increase tolerances placement in cruciform jointsand for non-straightness.

of shape errors on thestressed condition b The study of the influence strength allowed arecommendation for widen- and on the welded joint the range of ing the tolerances forjoint planes displacement to O.5t to l.Ot and fornon-straightness to Lût.

5.9 c. ltis also to be noted however that:

L The calculations show that under loadings in the plane of the plates and ranging from 15°/ to 5O of yield stress, the sum of bending stress and loading ¡n butt joints 'does not exceeds' the yield stress in the presence of O.5t displacement and LOt non-straightness.

In cruciform joints the same stresses develop if displacement or non-straightness values do not exceed Ì.Ot.

This ¡s another way of stating that such deviations from ¡deal design may lead to a doubling of stresses at certain spots in the joints.

Additional local stresses on the surfaces of the plates were not included in the sums of stresses above.

y. Fatigue tests of specimens showed that welded joint fatigue strength is decreased especially when butt joint misalignment is O.5t or more, when cruciform jnint misalignment is l.5t or more, and when non-straightness ¡s l.25t or more.

Few classification societies have their own well defined structural tolerance standards; therefore ¡t ¡s difficult to assess their safety factors es to the allowances for structures that deviate from the designs. By documenting the typical tolerances and accomplished results ¡na number of U.S. shipyards, a rough idea can be obtained as to the tolerances that may prove atleast adequately safe when current design criteria are applied. Some maximum tolerances should be rigidly defined because in the context of present design criteria and inspection techniques, they have been found to provide adequate safety, are easy to check, and are generally accepted. For instance, t/2 misalignment at cruciform joints has been listed by most shipyards. Ref.2l proposes t/l misalignment as acceptable, but the fatigue characteristics of such a joint are questionable. Also, this figure was developed for otherwise perfect joints, free from weld defects and material flaws.

5. Weld defects are the subject of most NDT, hence ¡t ¡s plain that they are considered as major degradations to theideal,"as-designed"joint. Visual inspection can be used to find undercuts. As seen ¡n section 3, undercut is mostly a function of proper arc and welder skill, although pits can aggrevate the situation. Undercuts can be harmfulin two ways: if a double undercut occurs in a plate girder web, an appreciable loss in plate thicknessresults. Second, if a force must be transferred transversely to an undercut, the undercut may act as a stress raiser. This second effect is more critical. If a girder web has alocal thickness reduction, plastic strain may result but this probab'y will not lead to failure. However, a stress raiser has high probability of leading to cracking.

5-10 Cover plate ofrolled beam

Undercut along Double undercut of cover platewould plate girder web not represent any would represent an appreciable loss appreciable loss in area; ould not in web thickness; be harmful effect is harmful

Shear force Tensile orce Tens i le force Undercut not harmful undercut harmful, acts as stressraiser Undercut ( Source: Reference 28) Fig. 5.L: Effects of Weld

other tolerances,trade-offs are needed between 6. in establishing many safety workmanship versus the costsinvolved with increased the costs of careful Sugisaki less than perfect fits. In 1965, Shimizu and factors to accommodate classification societies are nowready to approve reported that, "Norwegian The (22 ) raising grade of workmanshipin the yard...". reduction of weight through however, made no allowance and 1975 issues of Detnorske Ventas rules, 1967, 1969 Only corrosion pro- due to raised gradeof iorkmanship. for weight reductions There is also a need tection was used as a criterionfor reduced scantlings. relationships between working accuracyand cost of fit up at to establish the that overall costs versusaccuracy can beevaluated. various levels of assembly, so determine. relationships are mattersfor each shipyard to These trade-offs and cruciform showed one opinion of how toestablish tolerances for Ref.21 construction, via The obvious motivatingfactor was ease of and butt joints. the definition of goodfit, as good accuracy of partsand liberal allowances on requirements. long as ¡t was adequate tosatisfy strength

5-1 6.0 CONCLUSIONS

Itis believed that the surveys and interviews conducted with nineteen commercial shipyards, eighteen shipowner/operators, four classification societies, and two steel mills have provided sufficient data to assess the state-of-the-art of the U.S. shipbuilding industry's attitude toward, and performance regarding, structural tolerances.

It is further believed that the present-day status in most major U.S. commercial shipyards of providing'good shipbuilding practice" in their constructions has been studied and documented. This is true des- pite the fact that the amount of data collected for formal shipyard struc- turai tolerances is small and that the amount for actual structural deviations from the "ideal design" is even smaller.

There appear to be two dominant reasons for this shortage of factual data.

The majority of shipyards have no formal approach to regulating structural tolerances; they also lack a consistent data gather- ing system for recording actual deviations.

The determination of a sufficient quantity of actual structural deviations would have required an extensive program of measurements and non-destructive testing at each shipyard during the course of the present survey project. This would have required much longer survey periods to be reserved for each yard and would also have required an extensive input of time and manpower by the yards.

This short study has nevertheless shown that there is a wide spread in the structural deviation values as reported by different shipyards, and even within individual yards.

Even though a small quantity of structural deviation data was collected based on consistent records,it was still possible to obtain some in- dication as to the maximum and minimum values experienced in most shipyards.

The collected data is reported ¡n Section+, Table k.2, and represents the structural deviations normally experienced in shipyards during ship building and repair activities, along with the allowable tolerances being used.

The deviation and tolerance levels reported in the survey have been subjected to an averaging process, with the purpose of obtaining a compilation which can be considered to represent a cross-section of the U.S. shipbuild- ¡ng yards. The resultant tolerances are listed in Table 6-1, and labelled "USA Practice", as was done in Reference (1) based on what information was then available. Comparison of these tolerances with the published international standards reveals that there are no great overall differences. How- ever, the U.S. practice appears to be generally more liberal than other standards. lt must be noted that the values reported here represent only a subjective averaging of values obtained in various shipyards and that as such, they do not represent the actual workmanship of any one shipyard.

6- 1 As far as "in-service" deviations reported by shipowners are concerned, the data collection is very limited. This is mostly due to the fact that no records exist of structural deviations developing after a vessel enters service. Based on their recollections, however, some shipowner/operators did report a few cases of such deviations. It appeared that in a number of vessels structural deficiencies have developed in service which could be traced back to initial structural deviations. These are discussed in Section . I.

The existence or the lack of bui lt-in initial structural deviations on completed vessels will depend, primarily, on the quality of workmanship provided by the shipyard. Secondarily,it will depend on the structural inspection and quality assurance procedures in the yard because these will help discover and eliminate deviations.

Most U.S. commercial shipyards do not enforce written structural tolerances. As stated in Section L.2, most yards rely upon the experience and know-how of their own production supervisors as well as that of regulatory body inspectors and owner's representatives. The dominant factors in assessing most structural deviations appear to be such abstract opinions as 'good marine procedure'', or ''pleasing to the eye'. However, as pointed out earlier, a few shipyards do maintain written tolerances. A compilation of such tolerance standards is included in Appendix 9.2.

6-2 J a pane s e German Swedish ITEM Standard Tolerance Range Limits Standards Standards Practice mm inch mm inch mm inch mm inch mm inch lb - Pits <3 1/3 < 3 1/8 < 3 <1/8 3.2 2- Cutting Line 3/32 ± 2 1/8 ± 3 ± 3/16 4.3 Accuracy 3- Edge Rough- 1/3 ' 3 <1/16 1 .6 ness shop Nefd Groove) 1/8 3 .2 Field 4- Edge Straight- ness a Automatic 1/64 ±.4 1/64 ±.5 1/8 3.2 bSemi-auto- 1/32 ±1.0 3/32 ±2.5 ±1/8 3.2 matic c Manual ±3/16 4.8 5- Groove Depth 1/16 ±1.53/32 ±2.0 ± 1/8 3.2 6- Taper Ang'e ± .5d ±1 Od ± 5% 7 - Fabricated Shapes a Flange Breodt1/3 ±3 3/16 ±5 5%,± no limit ± 1/4 6.4 b Flange AngIe 2.5% 4.5% ±5% 5% c Straightness 0.1% 0.25% .25% B - Rolled Shape Flange Angle1/8 ±3 3/16 ±5 ± 1/4 6.4 9- Welding Gap a Fillet 3/32 <2 /8 <3 3

Table 6.1Comparison OF U.S.A. Practice on StructuralTolerances with Publ ished Internaticnal Standards

6-3 Japanese German Swedish U.S.A ITEM Standard Tolerance Range Lirnits Standards StandardE Practice inch mm irch mm inch mm inch mm inch mn

13 - Intercostal Strength < t/ < t/4 +3 < < t/2 Msalign- Members ment Others< t/3 < f72 < t/4 +3 14 - Profile Warp 200mm 3/8 <10 500mm 11/16 < 18 <1/4 6.4 1000mm 1. < 25 15 - Stiffener Bend f L c 1000 mm 3/16 5 5/16 < 8 5/16 < 8 L 3500 mm 3/8 < 101/2 < 13 1/2 < 13 5/16 3.0 1000

16 - Weld Spacing a Butt - Butt 1-1/4 '30 2 '50 + 4t >3 76 b Butt - Fillet 3/8 '101-1/4 >30+2t >3 76

17-Cylinder 3/16 ±5 5/16 ±7.5 for DlO00mn ±1/8 3.2 Diameter 4-(.005 D+1) 1/4 ±6 18-CurvedShell3/32 ±2.5 3/16 ±5 ±1/4 6.4 Accuracy 19 - Subassembly a Dimensions 5432 ±4 1/4 ±6 ± 1/4 6.4 bSquareness 5432 4 5/16 8 3/8 9.5 20 - Hatch Coom- 7ió ±5 3/3 ±10 ¡ng Dimensions ± 3/8 9.5 21 - Access Openings oDimensions 5432 ±4 9/32 ±7 ± 1/4 64 b Deformation .2% .3% 1/3 3 .2 22 - Unfairness oBottom 5432 4 1/4 6 7/16 11 1/2 12.7 bSide 3/16 5 9/32 7 3/8 j 9 1/4 6 3/3 9.5 cDeck 1/4 6 3/8 9 3/3 10 1/4 6 1/2 12.7 dSuperstructure 5/32 4 1/4 6 1/4 6 1/4 6 1/2 12.7

Table 6.1 continued

6-4 Japanese U.S.A Standard Tolerance German Swedish ITEM Standards Practice Range Limits Standards mm mm inch mm inch mm inch mm inch inch

23 - Overall Dim ens ions o Length 0.05 4 100 0.1% 0.1% b Beam 9/16 ±15 0.1% 0.1% cDepth 3/8 ±10 3/810 max -1 .% 0. 1% d Keel Flatness9/16 ±15 i I+25rnm/ 1 25.4 100m 5/8 15.9 e Forebody 1 1/4 ±30 +2 +50 Rise -1 -25 fAfterbody 3/4 ±20 +2 +50 1 25 .4 Rise -1 -25 g Decidrise 9/16 ±15 +1 ±25 1/2 12.7 6.4 h DraftMarks 1/32 i ±1 1/161 ±2 1/16 ±2 1/4 Freeboard 1/64 :0.5 1/64j ±0.5 1/8 3.2 Marks

Table 6.1 continued

6-5 7.0 RECOMMENDATIONS

7.1 Recommended Guide to United States Shipyard Practice in Structural Tolerances

Table 7J ¡s a compilation of what appears to be the United States practice ¡n structural tolerances, based oninputs from allinstitutions surveyed. The tolerance values shown reflect the capabilities oflarge, well equipped, and relatively modern shipyards inthe United States. It will be seen in Table 7.1 that the following basic tolerances arecovered:

Cutting Line Accuracy Dimensions of Fabricated Shapes Misal ignment Weld Geometry Accuracy of Curved Shell Accuracy of Subassemblies Unfairness of Plating Accuracy of Hull Form

Additional descriptive information on these and afew other candidate structural tolerances can be found in Section 1+.6.

It was not the intention of this project to produce any results other than a statistical representation of the current United States shipyard tolerance practices. The tabulation offered in Table 7.1 as the American shipyards' practices is included in this report as the investigator's opinion.

7.2 Relation of Structural Tolerances to Rational Design

The present day structura] tolerances reflect an assessment of shipyard capabilities within a reasonable cost framework, the yard's past experience, and the results of some basic research into material and welding properties.

It is felt that more research work is needed to develop bases for determining tolerances. Finite element analyses of small assembl ¡es and intersections, along with some destructive testing, have already been used to show the strength and fatigue characteristics of cruciform and butt type joints (17,21). More research work of this type will no doubt help increase the understanding of the stress phenomena in critical joints and will therefce shed light on the establishment of relevant and attainable tolerances and ways of considering these tolerances ¡n the rational design efforts.

The advantage of a rational design accounting for imperfections which are known to be occurring on the final product is that the semi-empirical formulae of the classification society rules will not have to be followed. The rational design may be based on investigative methods such as the finite element analysis technique, and may study the structural model of the complete vessel, especially in the strength-critical areas, from the viewpoint of structural imperfections.

7-1 Cutting Line Accuracy ± 1/16" ] ± 1.6 Fabricated Shapes Dimensions (Note 1) 1/4' ± 6. io ¡usaI iqnrnent Butt Joints t/4 for strength and/or appearance t/2 where neither is required

Cruciform Joints t13for critical strength rr,embers t/2for less critical strength members t for non-strength members

WeId s Reinforcement < 1/16" < 1.6 rre Undercut < l/32' 0.3 c Size (nominally 1/4") ± O (nominally> 1/4") * 1/16" * 1.6 rin ± 1/4u Curved Shell Accuracy (Note 2) ± 6.' rrrl - Subassembly Dimensions ± O.O5

Squareness J1ote 3) - e ü.î - Plating unfairness (Note 4) ± 3/8u ± 9.5 rin Overall Dimensions, including L, B, D, keel flatness, and deadrise ± O.1 Hull Markings (draft, freebd.) (Note 5) ± 1/8" ± 3.2 rin

Table 7.1: Structural Tolerances- in United States Shipyards

NOTES: I. Tolerances given are for all sizes of fabricated structural shapes normally used ¡n ship construction (see page 40, Item #7). Tolerance for Curved Shell Accuracy'is for deviation of actual line of shell from the design molded line. Subassembly tolerances may be established as follows:

a) Dimensional Accuracy Deviation from design d ¡mens ¡on

Measured or Calculated Diagonal

Cor rec t Diagonal Out of Squareness

b) Out of Squareness

Tolerance values given are maximum for normally used thickness of shell, deck, and other strength structure plating. For larger thickness, unfairness, curves are used as reference basis.

Draft and freeboard tolerances are maximum allowable deviations of markings from design or assigned values.

In general, all of the tolerances shown in this table represent the levels of imperfection that are attainable with modernized equipment and controls. -

7-2 Cargo tanks

O4L midship

A B Cj

r A

different critical quality areas. Fie: 7.1: A 210,000 DWT tanker divided into

(Source: Reference 6)

Some differentiation on the basis of location,allowable stresses, materials, and types ofproduction equipment must therefore be made in establishing achievablestructural tolerances. The approach in Fig. 7.1 covers the differentiationwith regard to location and allowable Stresses. Further work is required to take into consideration the effects of materials andproduction equipment.

After all factors have been considered, andpractically attainable levels have been established fordifferent strength critical areas of the vessel's structure, theproblem would be reduced to one of introducing these established tolerances intothe detailed engineering analysis of the structure.

7-3 7.3 Quality Assurance and Inspection Requirements in Shipyards

In order to ensure that allowable tolerancesare not exceeded cn the finished product and that no unacceptableor unaccounted for initial structural deviations are permitted, consistent inspection and quality assurance procedures are necessary.

For verification of adherence to allowable structural tolerances, the combined efforts of the structural inspection and qualityassurance group must cover the following general areas:

Receipt Inspection of Incoming Material

Non-destructive testing

Visual structural inspections

Inspection and Measurements for:

- Misal ignments - Deflections - Distortions - Gaps before welding - Out-of-Squareness - Curvature

Dimensional accuracy checks for:

- Deformations of hull form - Over-all dimensions - Draft and Freeboard marks

Final Finishing Practices

Tightness Tests

7_14 8.0 REFERENCES

I. "Fabrication Factors Affecting Structural Capability of Ships and Other Marine Structures'' Report of Committee Il I.3, 6th Interna- tional Ship Structures Congress, Boston, 1976.

''Japanese Shipbuilding Quality Standard (J.S.Q.S.) (Hull Part),'' SNAJ Publication No. 82, Tokyo, 1975.

"Japanese Shipbuilding Quality Standards - BackgroundDocument." SNAJ Publication No. 8-3, Tokyo, July 1976.

''Fabrication, Welding and Inspection of Ship Hulls, ''NAVSHIP 0900- 000-1000. Naval Ship Systems Command, Washington, D.C.,Oct. 1968.

"Alternative Methods of Non-Destructive Testing" ResearchReport by Lockheed Shipbuilding Corp., SNAME.

"The Acceptability of Weld Defects", Harrison and Young,The Naval Architect, London. AprM 1975

"Registro Italiano Navale" letter No. 20663 to Mr.Rosenblatt & Son, Inc. dated Nov.12, 1975.

''Qua] ity Standards and Quality Controlin Shipbuilding: A Joint Task of Shipyard and Classification Society'', W. Santini,RINA, 1975.

''Rules for Non-Destructive Inspection of Hull Welds'', AmericanBureau of Shipping, New York, 1975.

''Guide for Interpretation of Non-Destructive Tests ofWelds in Ship Hull Structures,'' Weld Flaw Evaluation Committee, SSC-177,Ship Struc- ture Committee, Washington, D.C., Sept.1966.

''A Guide for Ultrasonic Testing and Evaluation ofWeld Flaws'' R.A. Youshaw, SSC-213, Ship Structure Committee, Washington,D.C.,1970.

''Rules for Building and Classing Steel Vessels'', American Bureauof Shipping, New York, 1976.

"Radiographic Standards for Production and RepairWelds", NAVSHIPS 0900- 00 3-9000.

8-1 114."Code of Federal Regulations Title f46, Shipping" Subchapter D of Chapter 1, October 1975.

''Production Standard of the German Shipbuilding Industry,'' Association of the German Shipbuilding Industry, Hamburg, Novem- ber 1974.

''Accuracy in Hull Construction, VIS 530'', Varvindustrins Standard- central, Stockholm, 1976.

Full Penetration K-Welds or Double Fillet Welds in Heavy Ship- buMding", N.G. Leide, Structural Design and Fabrication in Shipbuilding, London, November 1975.

''Fillet Weld Costs versus Fit-Up Gaps''- Bath Iron Works Memo- randum dated Feb.19, 1976.

''Tolerances in the PrefabricationofShips,'' Carl-Erik Fredriksson, Trans. RINA 1962, pp. 201.

''Structural Details Design Review'', R. Glasfeld, Ship Structure Committee SSC-226, 1976

"Investigation of Technological Errors in Welded Ship Hulls Assembly Joints'', Alferov & Matskevich, International Institute of Welding. Document XIII-761-714 (USSR), October 19714.

''Total Quality Control in Shipbuilding'', Shimizu & Sugisaki, Nippon-Kokan Technical Report Overseas, Sept. 1965.

''A System of Dimensional Control for Ship's Structure'' J.R. Saizer, MARAD Ship Producibility Program, April 19714.

''The Accuracy of 1/10th Scale Plate Drawings,'' J. S.Collìnson, D. Adams and P.C. McConnell, BSRA 1967. Technical Memorandum No. 300

''Quality Controlin Shipbuilding. 1st Report. Dimensional Accuracy of Steelwork," D.H.S. Bunn and W.G. Smith, BSRA 1968. Report NS238.

"AspectsofNDT InspectionofWelds in Shipbuilding with Particular Regard to Ultrasonic Testing", M. Papponetti, RINA, 19714.

''Welding Metallurgy -Iron and Steel'', Claussen & Henry, American Welding Society, New York, 19149.

''DesignofWelded Structures'', Blodgett, James F.Lincoln, Arc Welding Foundation, 1966.

8-2 ACKNOWLEDGEMENT

During the many visits to shipowners/operators,shipyards, classification societies, and steel mills, the project investigators have received excellent cooperation and assistance from all. The names are too numerous to mention here; however, we would like to acknowledge allcontributions to the project by each of the following organizations:

Shipyards and Steel Fabricating Facilities

American Bridge Div.(U.S. Steel) Avondale Shipyards BathI ron Works Bethlehem Steel Shipbuilding, - Beaumont, Texas - Sparrows Point, Maryland Campbell Industries FMC Corporation General Dynamics Corporation, Quincy Yard Galveston Shipbuilding (Kelso) Levingston Shipbuilding Litton Industries, Ingalls Shipbuilding Marathon Le Tourneau, Gulf Marine Div. National Steel Shipbuilding Newport News Shipbuilding Seatrain Shipbuilding Sun Shipbuilding Todd Shipyards - Houston, Texas - Los Angeles, California Seattle, Washington

Chantiers de LAtlantique, Paris, France G8tawerken, Gothenburg, Sweden Howaldtswerke-Deutsche Werft, Hamburg-Kiel, Germany Kockums Mekaniska Werkstad, Malm, Sweden Swan Hunter, Wallsend, Northumberland, England Verolme United, Rotterdam, Holland

8-3 Ship Owners/Operators

Aries Marine British Petroleum Company Burmah Oil Tankers (Energy Transport) Chevron Shipping El Paso LNG Corporation Exxon Corporation, Marine Division Farrell Lines Global Marine, Inc. Gulf Oil Corporation Lykes Brothers Shipping Marine Transport Lines Maritime Overseas Corporation Mobil Shipping and Transportation Company Prudential Lines Sea-Land Services States Steamship Company United States Lines Zapata Technical Services Corporation

Steel Producers

Armco Steel Corporation, Houston, Texas United States Steel Corporation, Pittsburgh, Pennsylvania

Classification Societies

American Bureau of Shipping Bureau Ventas Lloyd's Register of Shipping Nippon Kaiji Kyokai Det Norske Ventas Germanischer Lloyd Registro Italiano Navale

Research Institutions

British Ship Research Association Verband de Deutschen Schiffbau-Industrie AB Svensk Varvsindustri

8- '+ The authors would like to extend special thanks to thefollowing institutions which have given official permission to the publicationof their tolerance standards in this report:

Bath Iron Works Ingalls Shipbuilding Levingston Shipbuilding Newport News Shipbuilding Seatrain Shipbuilding Sun Shipbuilding Nippon Kaiji Kyokai (and The Society of Naval Architects of Japan Verband der Deutschen Schiffbau Industrie e.V. AB Svensk Varvsindustrins Standardcentral

We appreciate the contribution Mr. sao Takeuchi of Nippon Kaiji Kyokai provided by partially translating the backgrounddocument for the Japanese Shipbuilding Quality Standards.

We would also like to express our gratitude toall persons who have so graciously given us their time andattention.

Thanks are due Messrs. John Johnson and Frank Mei whohave taken part in some of the surveys, and to Mr. NareshManiar both for his taking part in a number of surveys and for the generalguidance he has provided in planning and executing the project as well as inpreparing the final report.

8-5 9.0 APPENDIXES

9. BIBLIOGRAPHY

V.E. Braid, ''Dimensional Control of Steeiworkin Merchant Shipbuilding,'' B SRA. Part)" SNAJ ''Japanese Shipbuilding QualityStandard (J.S.Q.S.) (Hull Publication No. 8-2, Tokyo. 1975.

Trans.. ''Tolerances in the Prefabricationof Ships,'' Carl-Erik Fredriksson, RINA 1962, pp. 201.

ist Report. Dimensional Accuracyof Steel- k. ''Quality Control ¡n Shipbuilding, Report NS238 work," D.H.S. Bunn and W.G. Smith, BSRA1968. An investigation into ''Quality Controlin Shipbuilding. 2nd Report. Report NS266 Erection and Fairing on theBerth," V.E. Braid, BSRA 1969. M. Oda and Y.J. "The Deformation of a Ship's HullUnder Construction,'' 18924, 2073, 2139, 2272, 2375 Kaziwara, SNAJ 1963-9, (BSRATranslations). Nos. & 22496. D. Adams and "The Accuracy of 1/10th ScalePlate Drawings," J.S. Collinson, P.C. McConnel I, BSRA 1967. Technical Memorandum No. 300 SNAHE 1963 ''Maritime Administration Researchand Development, MacCutcheon, Transact ions. Lowe, SNAME Hampton Rd. ''Non Destructive Testing inShipyard Environment," 1971 McQuade, Marine Technology Oct. 1969. IO. "Japanese Shipbuilding Practice," SNAME Gulf Coast, "The Marine Industry Measuredfor Size, Shape and Position," February 1969.

"Welding Problem in Shipbuilding,Masubuchi," Marine Technology, Jan.1969. 1961. ''High Tensile Steel and theFabrication,'' Grubb, SNAME Philadelphia, Dooley, SNAME Philadelphia, 1k. "Reflection on 35 Years of ShipyardWelding," 1971. of Welds on "A Review of the PracticalApplication for Ultrasonic Inspection Ship," Frank, SNAME S.E., October1968.

"Recent Development in ShipyardWelding," McDermott, SNAME, Canadian Maritime, April 1965.

9-1 ''A Rational Approach to Inspection and Testing of Engineering Weidments," Clough, SNAME, Pacific Northwest, October 1961+.

''Thermal Analysis for Complex Structural Systems,'' Cox and Carneal, Marine Technology, October 1971.

''On the Flow Line of Assembly Blocks,'' Fujil and Yagi, SNAJ.

"The Continuous Flow Production System,'' Terai and Kurika, SNAJ.

''The Development of Quality Controlin Shipbuilding,'' NECIE&S trans. 1967-1971. 1970.

"Effect of Flame and Mechanical Straightening on Material Properties of Weld- ments,'' SNAME, Ship Structure Committee SSC-207, 1970.

''Effect of Temperature and Strain Upon Ship Steel ,'' SNAME, Ship Structure Committee SSC-235, 1973.

21+. "Fabrication, Welding and Inspection of Ship Hulls," NAVSHIP 0900-000-1000. Naval Ship Systems Command, Washington, D.C., Oct.1968 "Variation of Product Analysis and Tensile Properties of Carbon Steel Plates and Wide Flange Shapes,'' AIS] 197+.

"Bonded Joints and Non-Destructive Testing. NDT of Resistance Spots, Roll- Spot, Stitch and Seam Welds,'' Hall and Crecraft,IPC Science and Technology Press, Ltd., Surrey, England, 1971.

''Influence of Thermal Cutting and Its Quality on the Fatigue Strength of Steel,'' F. Goldberg, AWS 1973.

''Control of Distortion in Welded Fabrication,'' Welding Institute, Cambridge, England, 1968.

''Specifications of Permissible Defect Sizes in Welded Metal Structures,'' A.A. Wells, Queen's University, Belfast, Brighton, England, 1969.

''Weld Imperfections. Identification and Interpretation,'' R.P. Meister, Symposium on NDT of Welds and Material Joining, 1968.

''A Plan for NDT,'' D. Birchon, Brit. Jour. of NDT, Essex, England, 1970.

''Production Standard of the German Shipbuilding Industry,'' Associationofthe German Shipbuilding Industry, Hamburg, November 1974.

"Alignment Investigation Following a Crankshaft Failure in a Medium-Speed Diesel Engine," Caste], SW&S October 1970.

''An Investigation of Major Variables in Conventional Tailshaft Assembly,'' Calamari, Marine Technology, October 19614.

9-2 "Measurement of Bull Gear and Line Shaft Bearing Alignment During Operation,'' Lerro, SNAME Philadelphia,1963.

Considerations in the Design of Marine PropulsionShafting Systems," Lehr, SNAME N. England1961 -

"A Study of Accuracy Controlin Hull Construction Works," Takeshi Yokota & Taizo Shimoura, BSRA Translation No.1861(1963)

''Vibration Analysis & Deviation Concept, VIDEC,''Dickinson, SNAME Gulf Section. Sept.1971.

''Ultimate Strength of SquarePlates Subjected to Compression; Welding Re- First Report on the Effectsof Initial Deflection and June sidual Stresses'', Y, Ueda etal., Journal of SNAJ, Vol.137, 1975. with Imperfect ions; SecondReport'' 240.''On the Strength of Structures Buckling Collapse of Girderswith Initially DeflectedCorners''. Fujita, Yoshida, and Takazawa,JSNAJ, Vol.137,June1975.

Cracks Emanating from SideNotches in 1+1. ''Stress Intensity Factors of Plates'', Tamamoto, Sumi, and Ao,JSNAJ, Vol.137,June1975. Combined Tensile "A Preliminary Study onSurface Crack Growth in a and Bending Fatigue Process",Kawahara & Kurihara, JSNAJ, Vol.137, June1975. Axial ''Difference ¡n Low Cycle FatigueStrength by Strain Controlled Load and Deflection ControlledBending Load", lida, Matsumoto, & Nagal, JSNAJ, Vol.137,June1975.

Stiggeners at Cutouts of BottomTransverses", 2424. ''On Fatigue Damage of Web H. Mano et al., JSNAJ, Vol.137,June1975.

''Strength Analysis of Shell Structurewith Stiffeners by Finite Strip Method - Second Report", N. Fukuchi,JSNAJ, Vol.137,June1975.

"Bending of a Wide and ThickMild Steel Plate - ThirdReport", Yagi, Funaki, & Kada, JSNAJ,Vol.137, June 1975. Second "Studies ori Deformation andCracking in One-Sided Welding - June Report'',K. Satak, et al., JSNAJ,Vol.137, 1975. Naval "The Acceptability of WeldDefects", Harrison and Young, The Architect, April1975.

''Structural Problems In MethaneCarriers'' B. Javelle and J. Raynaud (of Bureau Ventas), ShippingWorld & Shipbuilder, Sept.1975, pp. 831.

9-3 'Third Decade of Research Under the Ship Structure Committee" Chazal, Goldberg et al. SHAME Ship Structure Symposium Paper. October1975

"Classification Society Experiences in Today's Ships" William M. Hannan. SHAME SSSymposiumOctober1975

''Structural Considerations In the Design of the Polar Class of Coast Guard icebreakers", Barker et al., SNAME SS Symposium, October1975

"Unusual Hull Design Requirements, Construction and Operating Experience of the Barge Carriers". Thayer & Schwendtner, SHAME SS Symposium, October1975

"Yesterday's Technology -Today's Ships - Some Tanker Experience", Szostak, SNAME SS Symposium October1975

"Structural Design Criteria for the Safe and Economical Trans- portation of LHG" - Shumaker & Hay, SHAME, October1975

''Observation of Ship Damage over the Past Quarter Century" - Townsend, SNAME SS Symposium, October1975

"Dynamic Loadings Due to Waves and Ship Motions'- Lewis & Zubaly, SHAME SS Symposium - October1975

"Structural Response & Computer - Aided Design Procedure", Stiansen, SHAME SS Symposium, October1975

"An Assessment of Current Shipboard Vibration Technology", Noonan. SHAME SS Symposium, October1975

'Fracture Mechanics, Fracture Criteria and Fracture Control for Welded Steel Ship Hulls", Rolfe, SHAME SS Symposium, October1975

6l."An Overview of Structural integrity Technology", Palermo, SNAME SS Symposium, October1975

"Fundamental Considerations of Fatigue, Stress-Corrosion Cracking, and Fracture in Advanced Ship Structures". Crooker et al., SNAME SS Symposium - October1975

"Analysis and Design Requirements", - Wilson, SNAMESS Symposium, October1975.

"Experimental Methods in Ship Structural Evaluation", Dinsenbacher, SNAME SS Symposium October1975

"Joining Technology and Quality Control" - Manley, SHAMESS Symposium October1975. ''Ship Structural Analysis in Lloyd's Register ofShipping''. The Naval Architect, July1975.

"The Statistical Approach to Hull Design",Prof. D.Faulkner, The Naval Architect, July1975.

''A Study of Sub-Critical Crack Growth in ShipSteels'', Francis, Lankford & Lyle,SSC-251, 1975.

Full Penetration K-Welds or Double Fillet Welds inHeavy Ship- building'', N.G. Leide, Structural Design and Fabrication in Shipbuilding, London, November1975.

"Finding Flaws by Ultrasonics", Marine Engineering/Log,November1974.

"Materials for Ocean Engineering", K. Masubuchi ,MIT Press, Cambridge 1970.

''Design of Welded Structures", Blodgett, James F.Lincoln Arc Welding Foundation,1966.

''The World's Largest Offshore Mobile Drilling Unit'',Surveyor, American Bureau of Shipping, August1976.

"Welding Metallurgy -Iron and Steel", Claussen & Henry, American Weld- ¡ng Society, New York,1949.

"Ship Design and Construction'', D'Arcangelo, SNAME, New York,1969.

''A Guide to Sound Ship Structures'', D'Arcangelo, CornellMaritime Press. Cambridge, Maryland,1964.

''Steel Construction Manual'', American Institute ofSteel Construction, 1975.

"Application of Probabilistic Concepts for Determining Limits of InitialImperfection of Ship Plating", Yasukawa, Ikegami, &Ominami, Journal SNAJ, Vol.138,December1975.

''Numerical Control of High Speed Cutting Boosts Productivity'', R. L. Bullard, Marine Engineering/Log, March1975.

"General Specifications for Ships of the United StatesNavy - Section 100: General Requirements for Hull Structure" January1973.

"Dimension Accuracy Control¡n the Construction of Welded Ships". Morten Ringard, PH.D Thesis, MIT, January1974.

9-5 ''Fabrication Factors Affecting tructural Capability of Ships and Other Marine Structures''Report of Committee 111.3, 6th International Ship Structures Congress, Boston, 1976.

'Japanese Shipbuilding Quality Standards - Background Document.'' SNAJ Publication No. 8-3, Tokyo, July 1976.

81+. ''Accuracy in Hull Construction, VIS 530'', Varvindustrins, Standard- central, Stockholm, 1976..

"Alternative Methods of Non-Destructive Testing" Research Report by Lockheed Shipbuilding Corp., SNAME.

''Building and Operating Experience of Spherical Tank LNG Carriers'' Howard, Kvamsdal, & Naesheim, SNAME New York Section Paper, Sept. 1976.

"Investigation of Technological Errors in Welded Ship Hulls Assembly Joints'', Alferov & Matskevich, InternationalInstitute of Welding-

Document XII I-761-7k (USRR), October 1974.

"Registro Italiano Navale'' letter No. 20663 to M. Rosenblatt & Son, Inc. dated Nov.12, 1975.

''Quality Standards and Quality Controlin Shipbuilding: A Joint Task of Shipyard and Classification Society'', W. Santini, RINA, 1975.

"Rules for Non-Destructive Inspection of Hull Welds", American Bureau of Shipping, New York, 1975.

''Guide for Interpretation of Non-Destructive Tests of Welds in Ship- Hull Structures,'' Weld Flaw Evaluation Committee, SSC-177, Ship Struc- ture Committee, Washington, D.C., Sept. 1966.

''A Guide for Ultrasonic Testing and Evaluation of Weld Flaws" R.A. Youshaw, SSC-213, Ship Structure Committee, Washington, D.C., 1970.

''Rules for Building and Classing Steel Vessels'', American Bureau of Shipping, New York, 1976.

9k. ''Aspects of NDT Inspection of Weldsin Shipbuilding with Particular Regard to Ultrasonic Testing'', M. Papponetti, RINA, 197k.

''Code of Federal Regulations Title 1+6, Shipping'', Subchapter D of Chapter 1, October 1975.

''A System of Dimensional Control for Ship's Structure'' J.R. Saizer, MARAD Ship Producibility Program, April1971+.

9-6 "Fillet Weld Costs versus Fit-Up Gaps" -Bath Iron Works Memorandum dated Feb. 19, 1976.

"Structural Details Design Review'', R. Glasfeld, ShipStructure Committee Project SR-216, Preliminary Report, 1975.

''Total Quality Control ¡n Shipbuilding'', Shimizu &Sugisaki Nippon- Kokan Technical Report Overseas, Sept. 1965. lOO.''Photograrnmetry in Shipbuilding'', J.F. Kenefick. First Draft of Final Report to MARAD Photogrammetry Project, Feb. 1976.

''Detail Design in Ships'' Lloyd's Register of Shipping, London, Octo- ber 1967.

"Radiographic Standards for Production and Repair Welds",NAVSHIPS 0900- 00 3-9000. 9.2: U.S. COMMERCIAL SHIPYARD STANDARDS

CONTENTS

APPENDIX DESCRIPTION PAGE

9.2.1 : Bath Iron Works

9.2.1.1: B.I.W. Inspection Guidelines 9-9 9.2.1.2: Ro/Ro Dimensional Control Guidelines 9-12

9.2.2 : Litton Industries (Ingalls ShipbuHding) 9-15 Sample ''Manufacturing StandardProcess''

9.2.3 Levingston Shipbuilding 9-20 Special Tolerances for Drill Rig

9.2i3 : Newport News SB&DD 9-23 Dimensional Tolerances for LNG Tanker

9.2.5 : Seatrain Shipbuilding 924 Structural Tolerances

9.2.6 : Sun Shipbuilding and DD 9-26 Shipbuilding Production Standard (Hull Division)

9.2.7 : Tables of Permissible Unfairness in Welded 9-62 Structure (from NAVSHIPS 0900-000-1001)

Note: The fol lowing are reprinted with specialpermission from the respective organizations.

9-8 APPENDIX: 9.2.1J

BEW INSPEOTION GUI DEUNES

FITTING MIERIAL A.MISALIGANENT ANO FIT-UP A. SURFACE CONDITIONS 1. ALI. PLATES ARO SHAPES MiST MEET SURFACE CON- I. MESALIG*IONT NEASGJREMEPXT - DITIONS OF ABS RULES. SCARS IMPERFECTICTS ARE TO 8E AVOIDED IN ALL AREAS OF SNIPS STRUC- ¿ .1 TURE. SPECIAL. CONSIDERATIONS OF ThESE RESOlED- IL MENTO SHALL BE DIRECTED TO LONGITADISAL MID T T1 Cn!fl-,,.r. ThInk.sS Datation Cr.,- Mo1d.d Iin. TRANSVERSE STRENGTH STRUCTURE AS ISDIATED

BELOW. NOTCHES SHALL BE AVOIDED. 't±..ltc.,..,nt & t'.fC.r.nt. L4GITUDINAL PLATES AND SHAPES IN MIDSHIPS It 3/5 LENGTH. ¿If t UPPER 'A' DECK P.ATNG. TANE TOP " DECK (U,,,.t. tH* aid,) PLATING SHELL PLATING.

BOX GIRDERS. MISAUGXQXENT NID FIT-OP BUTTS - i.LONGITUDINALS AND ALL DECK STRINGER PLATES. TRANSVERSE WEN FRAMES. LONGITUDINALS ANO TRANSVERSE EXLKHEVDS HAD ATTAOHIENTS. '7r IX. PILLARS. 7T i. ALL DECK CUTOUTS. I1\tD,(t 2, REPAIRS TO SCARS AND IMPERFECTIOTS OF FEBEHS INCLUDED IN PARA. A-1 ABOVE MUST DE MADE BY GRINDING. CHIPPING OR WELDIJIG DEPULDING 0X4 T.L...Th... Ta T . E'o.r KAGNITUDE OF IMPERFECTION OUTLINEC AS FCLLOWS IN GENERAL. MINOR SCARS MAO BE REPAIRED BY MRSIIRJRI MISALIGNMENT AP WEBS, FLANGES AND FACE GRINDING. PLATES - SCARS WHICH EXCEED 3/32" IN DEPTH ROO G IM 1, LENGTH SHALL WC REPAIRED BY CHIPPING. GRIXID- z ING AND WELDING.

REPAIR WELDS WOICH ARE GENERALLY LOU '4 S ¡__; ij- PROPILE (3/32') NID ARE DOT M) APPEARANCE FACTOR SEED NOT BE GROUND. ditoa.,bV - j t, A10...'Il. b - 3. SCARS IN NON-STRENGTH AREAS WHERE APPEARANCE IS IHPORTAI4T. SCARS MAY BE REPAIRED AND DRESSED BY ASIlO UN t APPROVED EPDXY COMPOUB4D. 2 T. 1N b. WHERE APPEARANCE IS NOT IMPORTANT MINOR -1k H-N SCARS WILL NOT RROUIRE REPAIR OR TREAT- *lla,t.bL. . { T SENT. n.a. t i.aai. b t T

MY MISALIGFQRENT EXCEEDING THE TOLERANCES SHOWN BILI. SE INSPECTED AND RESOLVED ON R CASA ARSIS SO 144E BIB INSPECTION DEPARTMENT. WHEN A MISALIGRIENT EXISTS ADDITIONAL WELD REINFORCEMENT SHALL BE APPLIED TO MISALIGNED ME1.fRS FOR ADDITIONAL STRENGTH OF MISALIGNED SEMBERS AS NECESSARY. 9-9 FITTING WELDING FAXING SURFACES WELDING MATERIALS SHALL BE DISPENSED WITH UTMOST I. CLEANANCE BETWEEN PAYING SURFACES OF LAP JOINTS CARE. ONLY THOSE ELECTRODES WHICH ARE COERPATIBLE MC PEMYANENT INSTALLED BACKING BAR Burr JOINTS WITH DESOGNATED MATERIALS SHALL RE USED IN ACCORD- SHALL NOT EXCEED l/16 EXCEPT AS SPECIFIED ON MCE WITH APPLICABLE WELDING PROCEDJRE. SUBSTIrU- PLANS. DOES NOT APPLY TO RIVETED AliCE BOLTED TIDN OF WELDING MATERIAL IS NOT PERNITTED WITHOUT JOINTS. PRIOR APPROVAL OF WELDING ENGINEER. PENETRATIONS ERROR IN USE OF WELDING MATERIALS IS CAUSE FOR 1. PLANS ARE TO BE CROSS CHECKED WITH CONCERNED REJECTION OF UNIDENTIFIED WORK IN PROCESS. DEPARTMENTS PRIOR TO MAKING CUTS SO ELECTRICAL, PIPE AND VENTILATION PENETRATIONS CAN BE PlIS-. POSITIVE RELATION SHALL BE ESTARLISRED BETWEEN IMIZED. PARENT METALS AND FILLER WELD METALS ON ALL IN O. FILLET GAP PROCESS WORK. ONLY APPROVED ELECTRUDES WILL BE USED FOR TACK OR BLOCK WELDING. R /,6.00pt- bU.. (IF, IN ERROR. WELD METAL IS UEPOSITED WHICH IS /,6 R 3/,A CUNTRARY TO APPROVED WELDING PROCEDURE. WELD riAl.,t.ib METAL SHALL BE REMOVED IN ITS ENTIRETY AND THE 3/tE' r.ij ptrtii,. CONCERNED AREA INSPECTED PRIOR TO REWELDING.) d?q..Ar. C or corr.c t by WELDS SHALL BE FREE OF CRACKS DR CRACK-LIKE (..,t b tpp..fb øX In- INDICATIONS OR LINEAR INDICATIONS.

I; E. THE HEIGHT DF REINFORCEMENT OF A BUTT WELD OR T SENI SHALL BE KEPT TO A MINIMUM IN THE FOLLOW- ING AREAS: S. EXTERIOR SHELL. T.> T > EXPOSED AREAS OF WEATHER DECKS. 3. T- 2R EXTERIOR SIDES UF DECK HOUSES. RN DECKS THAT HAVE A COVERING:WELD RUIN- FORCEMENT SHOULD BE 1/16 NOT TO EXCEED 3/32'.

F. SIDE OF WELDS SHALL BE UNIFOAI TO REQUIRED SIZE AND CHECKED WITH A WELD GAUGE. PLATE EBUE BUILD-UP G. FILLET WELDS FOR STRUCTURE 1. PLATE EDGE BUILD-UP WHERE PEAlYITTED WILL BE 1. UNDERCUT AT WELD EDGE IN EDCESS DF 1/32' ZN STRICT ACCORDANCE WITH ABS RULE REQUIRE- WILL BE REPAIRED BY WELDING. UNEERCUT NESTS ADD WITH COUCURRESCE CF OWNERS ANC REG- TO 8E MINIMIZED EV PROPER WELDING TECH- ULATORY BODIES. BUILD-UP TO BE WITH TYPE OF MItJE. FILLER METAL SPECIFIED BY WELDING PROCEDURE. H. Z. WHERE PLATE EDGE BUILD-UP IS EMPLOYED FOR A WHERE APPEARANCE 15 A CONCERN OR CRITERIA FUR FIX. THE JOINT IS TO BE FULLY PREPARED ASO ACCEPTANCE OR REJECTION THAN ATTACHIEBT ArID

INSPECTED PRIOR TO RELEASE FOR FINAL PRO- PLACEMENT OF RESTRAINTS (WELDING OFSTRONG- )JCTION WELDING.TO COMPLY WITH ABS RULES. BACKS) SHOULD BE CONSIDERED SO THE LEAST

ARC STRIKES ARE TO BE AVOIDED. MOUNT DF COSMETIC WORK WILL BE REQUIREDAFTER FIT-UP RESOLUTION THE REMOVAL DF TVE RESTRAINTS.

U. IN ALL CASES WHERE STRUCTURE MAKE-UP CLEARANCE I.WELDING POROSITY EXCEED PLATE THICKNESS. WORK SHALL NOT PROCEED D. VISIBLE WELDING POROSITY SHALL NOT BE UNTIL RESOLVED BP BZW INSPECTION DEPARTMENT. ACCEPTABLE AT OIL TIGHT OR WATER TIGHT THE ACREED FIX WILL BE PERFDRNED WITHOUT BOUNDARIES. NON-TIGHT BOUNDARIES MAY DEVIATION. RE CORRECTED BY FILLING WITH FORTIFIED

EPDXY COYIPOIJND PRIOR TO COATING. IN OTHER ARCAS POROSITV SHALL 6E ACCEPT- ABLE PROVIDED THERE ARE SOT INOICATIOSS GREATER THAN 3/32 DIAMETER, WITH NO MORE THIRD (4) INDICATIONS IN ANY 6 LENGTH OF WELD.

9-10 FAIRNESS WELDING B. STRUCTURES NOT IN CONFORMANCE WITH FIGURE 7V WELDING POROSITY (COST.) SHALL BE STRAIGHTENED BY APPROVED METHODS TO 2. ELONGATED GAS HOLES LESS THAN 1/2 LENGTH MEET FAIRNESS CRITERIA. MD 1/16' IN WIDTH ARE ACCEPTABLE 1M NON- WHERE HEATING ANO SHRINKING IS EMPLOFED WATER TIGHT & NON-STRUCTURAL ATTACFBYENT F ILLET FOR STRAIGHTENING EXCESSIVE TEMPERATURES WELDS. SHOULD A GENERAL POROUS CONDITION EOIST ARE TO BE AVOIDED. IN ANY AREA. TRE CONDITION SHALL BE CORRECTED. 1400°F - MAX. MED. CARBON STEEL (DULL OVERLAP RED COLOR) OVERLAP AT WELD EDGES SHALL BE REPAIRED BY 1. 1250°F - MAO. HTS -MUY. 0H AND EH WELDING OR GRINDING TO CREATE A SMOOTHLY MINOR DANAGES INCURRED TO PLATE EDGES. PAIRED WELD EDGE. ETC. WNICH REQUIRE STRAIGHTENIVG BY

SNIPES LOCAL HEATING SHALL NOT BE QUENCHED. 1.BALLAST TANKS, WATER TANKS, BILGE ARDAS AND ViSIBLE DEFORMITIES SHALL BE DEALT WITH WEATHER DECK AREAS SHALL BE COMPLETELY SEAL AS REQUIRED PRIOR TO ASSEMBLY.

WELDED. SNIPES REQUIRED FOR DRAINAGE SHALL STRAIGHTENING BY THE USE OF HEAT SHALL BE SIDED SUITABLY TO EFFECT COMPLETE SEAU. IT BE EMPLOYED ON STRINGER ASO SHEAR

WELDING. STRAKE PLATING WITHIN 3/5 MIDSHIP LENGTH. IM GENERAL STRAIGHTENING UF HTS BY USE OF WELDING OVAL ITS IAT SHALL BE KEPT TO A MINIMUM. LEAOINGREN SHALL INSPECT RACK GOUGING PRIOR

TO AUTHORIZING BACK WELDING. C. STRUCTURAL DISTORTIONS MEMBERS TO BE WELDED SHALL BE INSPECTED FOR SURVEYORS ARE TO BE USED TO ESTABLISH WORK- ACCEPTABLE FIT UP PRIOR TO CORMESCEMENT OF ING LINES MO WILL VERIFY PEVIODICALLY THAT

WELDING. FAULTS SHALL BE CORRECTED PRIOR TO SHAPE AND SIZE IS BEING MAINTAINED. PRODUCTION WELDISO. DURING THE COURSE OF RANUFACTURE WHEN DIS- APPROVED SEQUENCE WELDING AS OUTLINED IN TORTIONS OCCUR AS A RESULT OF WELDING OR APPRONED WELDING PROCEDURE SHALL BE STRICTLY UNDUE FORCE REINO APPLIED TO ESTABLISH.

ADHERED TO. SHAPE: FABRICATION SHALL NOT PROCEED UNTIL REJECTABLE WELDING SHALL BE PROMPTLY DEALT NECESSARY CORRECTIONS ARE ACCOMPLISHED. WITH AS NECESSARY TO PRODUCE A FIr1ISHED USE OF WELDING SEQUENCE. FITTICO RESTRAINTS PRODUCT WHICH MEETS APPLICABLE RELES. HAD RESHAPING WILL 8E EMPLOYEE TO ENSURE RECORDS OF WELDOR QUALIFICATIONS AGO JOINT THAT UNITS/SUB-UNITS WILL VOT HAVE DISTOR- PENETRATIONS FOR HULL WELDING SHALL 0E MAIN- TIONS PRIOR TU LEAVING FURRDINGS PANEL SHOP TAINED AND MAGE AVAILABLE TO CONCERNED BR ASSEMBLY BIJILDING. PARTIES.

FAIRUIESS A. FAIRNESS OF ALL WELDED STRUCTURE SHALL CONFOBI TO FIGURE 7*.

/A 1/, 3/0 /2 Y/A 7 'V P!Tt T,tCKNtXV ((C.L) 9-11 RO/RO APPENDIX: 9.2.1.2

DIME P4SIO&4L BATH IRON WORKS CO TR DL

GUIDELiNES

Nov;:1914-

GENERAL NOTES

2.1 RESPONSIBILITY: NARDINOS A. THE SHIPFITTERS SHALL ACCOOPLISH ALL WORK INGENERAL 3.1ALL MEASURING TYPES USED BY LAYOUT PERSONNEL SAAL.. BE ACCORDANCE WITH THESE SUIOELINES; SHALL BE HERRE OF CHECKED BIMONTHLY. MD TAKE CORRECTIVE ACTION FOR UNSATISFACTORYITETS; MD SHALL ACCOMPLISH ALL LAYOUT AND DIMENSIONAL CHECKS 3.2 ALL F,B. MD SHAPE FABRICATION SHALL RE ACCOMPLISHED TO THAT CAN REASONABLY BE DONE WITHOUT SURVEYORS. ti/a'. THE SURVEYORS SHALL ASSIST THE SHIPFITTERS III LAYOUT 3.3 ALL FLARE PLANNER PLATESAALL BE CHECKED YFTER BURNING ANO DIMENSIONAL CONTROL AS REOUIRED: SPILLREPORT 0.11 TO ±1/16'. UNSATISFACTORY ITEMS FOUND ro SHOP SUPERVISION FUR CORRECTIVE ACTIOS TO TAJAR UNITS SAALL RECORD ALL 3.4 TELEREO OUTPUT SHALL RE CHECKED 1510E A SHIFT FOR WIDTH CRITICAL DIMENSIONAL CHECKS. SHALL TRANSFER CRITICAL AND LENGTH. RECTANGUL.AY PLATES SHALL BE HELD TO tU/a'. REFERENCE LINES TO TVE TOP SIDE OF THE UNIT FOR EREC- TION, ANO INSURE COMPLIAOCE WITH THESE GGIOELIRES PRIOR 3.5 THE FOLLOWING CAECES SHALL RE MADE AT SUB-ASSEMBLY: TO CSHtIENCEMENT OFMAJOR WELOING OR MOVING FRON THE ASSEMBLY POSITION. LOCATION OF DIAPHRAGTS AND CHOCKS IN A BOO G IRDERS 2.2 IWO PILLARS PRIOR TO INSTALLING THE CLOSING PLATE. ALL 'WORKING DIMENSIONS" SHALL BE TAREN FROM MOLOLOFT SKETCHES. WHERE IT IS NECESSARY TO USE O,VENSIOOS FROM ALL RADS. DR ASSEMBLIES BUILT TO LOFT SKETCHES SHALL THE 'BOOK OF OFFSETS OR THE 'MOLO LUFT INFO BOAR" A VEUT BE CHECKED AND OVERALL DIMENSIONS HELD TO ±1/4'. STOCK ALLDWASCE SHLL SE ADDED IN ACCORDANCE WITH ENCLO- SiRE (1). NOTE: THE TRANSVERSE DECK OEVIRS ATTER ASSEMBLY FOR CUTOUTS. CHOCKS AND FACE PLATE EEYELS. A. TJNITS211.300,2, AFTER WELDING STRAIGHTEN 3, 4 AND 5 HAVE i' STOCK TO BE CUT 1F NECESSARY TO HDLD o3/8" OF CARRER. ro 1/4" NEAT STOCK ILE MARRIED. PJ4EE BRACKETS ON SHELL WEBS SHOULD BE INSTALLED AND UNITS221AND222 TOHOVE THE BOTTOM OF THE SHELL SET AI 1/4' NEAT STOCK. HELD 0/4' HIGH TO MOLD LOFT TEMPLATES. 3.6ALL SHAPED SHELL PLATES SHALL RE CHECKED FOR SACK SET 2.3THE 3-O' BTK SMALL BE USER IN LIEU 0F4 FOR THE AND MASTER TWIST AFTER FOSIING TO MOLD LOFT COMMON BASE TEMPLATE. REFERENCE AT PANEL. ASSEF'ULY ARO ERECTION FOR UNITS FR. 38-072 AS SHOWN ON ENCLDSAP.E (2). WHERE THE OR 3-O' PANEL SHOP RIE CANNOT BE USED THE STR USEU SHOULD BE CLEARLYINDU.- TIPlES ON THE STRUCTURE. 4.1 DECK ASSEMBLIES 2.4 ALL CRITICAL REFERENCES íA, 3-O' STE. MASTER FRTIES.ETC.) SMALL BE CLEARLY CEYTER POUCHED AND OUTLINED ADOIDENTIFIED ON THE STRUCTURE WITH BLACK PAINT OR MARKINGPEN.

2.5 HARDINGS THE PANEL SHOP AND HYDE SHALL RECOMO 0EV iHTION FRON 'WORKING DIMENSIONS' TN THE STRUCTURE WITH CLACK Loft Sk,ttft i/U' PAINT OR MARKING PENS. CRITICAL DIYENSIONS SUCH AS HALF WIDTHS SHALL 8E RECORDED O:; THE STRUCTURE AS 'UCTOAL' VS. WORKING DIMENSIONS' IA BLACK PUlItI OR MARKING PEN.

2.6 OHMAJOR UNITS THE SURVEYORS ASSISTED BY THE SHIPFITTERS SHALL RECORS ALL CRITICAL DITENSIOIIAL CAECES Of; THE"UNIT DIMENSIONAL RECORD F084, ENCLOSURES (3), (4) AND (5) OR A SPECIAL PO4 FOR HOME COMPLEX UNITS.

2.7 ALL SIB-ASSEMBLIES. PANEL ASSEMBLIES MID UNITS DATtiLO BE BUILT HO(DII6 ALL STRUCTURE TO THE INI WER FEHlE AR BULE- HEAD FROH THE NEAT ENO. (THIS IS THE MASTER FREME.) fYoLd C Roo Ctrd.r to Btk. La ALL STRUCTURE SHALL BEHELD TO THE MOLD LOFT LAYOUT ON R.cot IB SO Ao S'qt4..d to Hold THE PERIPHERY DF THE UNIT AS POULDWS: M.I.t hdtho to 1/A' Loft Nk.tcf,.w.- / HOlE: A. PRIOR TO) INSTALLATION. ESTA3LIYH THE MENI OF TI4E BOX GIRDER - ANY DEVIATIONS FRON LUFT DIMENSIONS SHOULD BE HOTED FOR ADJUSTMENT OF THE LAYOUT ON P/S PANELS. FR. L. C/D. & SP Itt ff.nor. PRIOR TO INSTALLATION OF THEA BOO BIRDER. THE 3-O' 81K SHALL BE ESTABLISHED FROM THE IB SEAR ALLOWING FOR THE BOX GIRDER CONDITION. THE HALF WIDTHS SHALL BE CHECKED ONTHE FED/AFT ESOS AND IF NOT WITHIN ±1/4' THE 3-A" BTI( SHALL BE ADJUSTED ANO THE IB SEAN RECUT.

THE NEAT END SHOULD BE RECUT IF OUT OF SIRRE FRON THE 3'-OBTK IN EXCESS UF ±1/4.

THE STOCK ENO SHOULD BE REChT IF MORE lOMAN Pt" OF STOCK EX ISIS.

THE LAYOUT FOR GRIDS SHOULD INCLUDE ALL BHES. WEBS, ETC. FOR HYDE OR MAIN ASSEMBLY. G.id.ro, R.b. & StitRo THE OB. EXE SHOULD RC REChT ORLY IF BECUTTIRS THE IB SENI WILL MOT HOLO HALF WIDTHS TO ±1/4".

SPECIAL HOTS:

UNIT DUlA. P/S SHALL BE MARRIED IN THE INVERTED POSITION IB THE PANEL SHOP NID THE 0" 0F STOCK LEFT BY THE LOFT ROOSTTI 9-12 HOLD 1/4' 'NEAT STOCK' OVER LOFT DIMENSIONS. PANEL SHOP (COEIT.) HYDE (CONT.)

4.2 INNER8OTTR ASSEMBLIES 5.1SHELL ASSEMBLIES )CONT.) PRIOR TO SETTING DIAPHRAGMS LAYOUT ALL SEAN AND VER FRANE H']. I *11I.tS°hc -lOft LOCATIONS ON PLATEN OR FLUCH: DURING ASSEMBLY CHECK THE i Y L'. T__5SCC+ - AT1 tt.,,,- - FOLLOWING: B-3/o' B- 3/h' UPPER AND LOWER SEANS TO LOFT LAYOUT.

tort Sk.,,hol/'_' - FORE/AFT BUTTS TO LOFT SKETCH - RECUl NEAT AND IF OUT AL'. to 10CC SketCh .1',' MORE THA,NI/4".

lIttle, Fr,,. WEB ERNIE (AND ANY TRANSVERSE FRANE) LOCATIOTS TO LOFT LAYOUT USING REVEL AVGLE HELD SQUARE TO DIAPA.GGMS NID FLODR - HOLD ALL WEBS (OR TRP,ISVERSE FERIES) TO THE BEVEL ANGLE.

HOLD WIPE ACROSS TOP OF WEBS TO HOLD DECK CUTS IN PLANE. SB SSbd TRUE FRAIlE SPACING SQUARE TO WEB PLATVG. LAYOUT FORE/AFT ENDS USING GIRTH TAPES FR011 OECE (OR THE MOLD LOFT INFO EOOK" - dOLO ALL LUNGITUCI VALS NO4AL EOCEPT US NOTED ON THE PLAN. Loft S k.tch BO NOT WELD THE TOP AND BOTTOM CHOCKS IN WEB FIfillES. Ch.Cito.Cit Fron. U- 0.-h Loft Sketch 0,/2" - I 5.2WING TANK ASSEMBLIES 5,2,1WING TANKS NOT BUILT IN SHELL ROCKS SHALL BE BUILT «r ON LEVEL NOCES ADJUSTED FOR PLATE THICKNESS VARIA- Eni 0'. 1.3. 4 & S lolo. TIONS IN EXCESS OF 1/4". CURE MUST EU TAKES IN - 0Ch.r C,,tt. 5colcrLOl.okt SETTING WEBS TO BEVEL ANGLE 1F MOCK BASE IS rOT PARALLEL TO 4.. IIERBOTTI ASSEMBLIES TO BE MANGLED SIMILAR TODECKS EX- LAYOUT SHELL LO1VGITOYIIALS FROM DECK USING GIRTH CEPT AS FOLLOWS: 5,2.2 TAPES (OR THE "MOLO LOFT INFO DOCK"). WITH 1" OF A.UNITS 301. 2, 3. 4 ANO 5 VAHE BEEN LOFTED STOCK P/S ON FLOORS AND 1" STOCK PORT ON IT. THE 5.2.3 SHIPFITTERS SHALL REVERSE CRITICAL LIVES (AGOTEN LAYOLT IS TO AK ADJJSTES PER THE ABOVESKETCH TO UDO ERNIE, DRIP LOCATIOn. ETC.) PRIOR TO TORITO FR011 1/4V NEAT STOCK P/S AFTER ASSEMBLY. THE 1" 0F STOCK THE MOCK. WILL BE REMOVED ST SCRIBING THEÇ DOEGIRDER ANO THE P/S UNITS IN TVE V.B. 5.2.4UNITS SHALL BE WELDED TO THE MUXIMUM EXTENT PRAC- TICABLE AND IN ALL CASES BLOCK TACKED ON THE OVER- IS HEAD SIDE PRIOR TI YIOVINU OR TURNIXG. AFTER MOW- E. UNITS 301.2.3 fIT 4 VBYKP! THE TANK TOP OB. QUTOOT BETWEEN FLOORS HALE WIDTHS SHALL BECHECKED ING FOR 108815G OR FURTHER ASSEMBLY VOlTS SHUALL EU AT EACH FEHltE ArID HELD TO j/2- D" TO THE LOFT CHECKED FOR LEVEL. SKETCH (AIM TO LOFT DIMENSION +1/4°'). 44 BTK FOR C.ON UNIT 201, FR. 172½ - 188½. USE THE THE MASTER 8TK PORT.

4.3 MISC. PANEL ASSEMBLY 4.3.jMISC. PANELS SHALL BE RESOGARED ArlOCHECKED TO LOFT DIMENSIONS PIOIOR TO LAYOUT. FICAT TOGES SEIALL BE ANO REcUT WHEN THEY EXCEED LOFT DIrTENSIONS BY 1/4°' ASSEMBLY SHOP STOCK KOOKS SHALL BE RECDT WHEN THEY EXCEEDLOFT DIMENSIONS 8V 1/2°'. 6.1 GENERAL NOTES 6.1.1 IT IS MOST IMPORTANT THAT UNITS RE ASSEVIBI EN IN H HOOK LEVEL CONDITION. ERRORS RESULTING FR014 MAJOR FRAY'IIlWD INTERSECTIONS ('SBOA GIRDER. L.S T. 6H05.. 5.1SHELL ASSEMBLIES DECKJSHELL WEB I1ITERSELTIOSS, ETC.) BEING OOT UF Do 'p, i.1f Top 4 LEVEL CAN RE VS SERIOUS AS FAILURE TO HOLD REMESAS lit.1 C hoc o, Sop Cf W.V. PLOMB OR CHECK HALF WIOTHS. LEVEL PLATES OR CHIES SHALL BE USED TA ADJUST FOR PLATE ThICKnESS VARIA- )4o1.t to f.1rlh TIONS IN EOCESS OF 1/4". VIlO WEIGHTS OR PIJLLISG Iron Doch GEAR SHALL BE USED TO HOLD TUE MAJOR FR021155 INTFR- SECTIONS WITHIN A ONIT TOHl/U'FR011 A MEAN ELUSE. IN HEAVILY RESTRAINED UNITS OCCHI AS INYIEYUOTTOI-TS WHERE IT IS NOT POLVIULE TO HOLD +1/4" 2 MEAr] EVADITIOTI 15 TO RE ESTABLISHED NID THE CORNERS OF THE UNIT SHALL BE RESTRAINED WIllI WELDED BRACES OR STE,IOIBOAT RATCHETS. 7 i Lo:°tSh.lh S Tor.Lico' H 6.1.2UNITS HOU TV DE ASSET'IRLEI ON SUBSTANTIAL ANO RIGIO MOCKS. DIAPHRAUM ANO POST ROCKS ARE TO SE REPAIRED H.r.11.t$L 4 j j AND AODITIVNAL TEMDERS ADDED IF NECESSARY TO SUPPORT MAJOR FRAMING INTERSECTIOSS. WHERE STEEL HORSES UNE L. ASEO THEY SHALL RE POSITIONED TO SUPPORT FRANTIC IN- TERSECTIONS AND LEVELED WITH LEHEL PLATES PRIOR TO UNIT ASSEMBLY.

6.1.3 WITH THE UNIT LEVEL ON THE MOCK, THE SURVEYORS SMALL REVERSE CRITICAL REFERENCE LINES (t. MASTER UIL AND Checo Tr' F-.0. Spott \ Sç. To "oNt MASTER FRAIlE) ID TVE TOP OF THE UNIT FOR ESECTION. 6.1.4IN FITTING SHELL ASSEMBLIES THE FORE/AFT POSITION OF THE FIRST WEB FR011 THE SEAT END SY!OULD BE TAKEN AT ITS MID-HEIGHT Ir] YACER TO SPLIT PilO ERROR IN WEB LOCATION BETWEEN ASSEMBLY AND ERECTION.

6.1.5BULKHEADS SHALL RE HELD TO THE LAYOUT TO Al/B" UND SHALL BE HELD PLUMB AT INTERSECTIONS TO ±1/4" IN L I THEIR HEIGHT.

f Pc14 O-col AtSo Aq. o, DpTc. R'' floor

9-3 ASSEMBLY SHOP (CaNT.) ASSEMBLY SHOP (COAT.

6.1GENERAL NOTES (CaNT.) 6.3 INNERBOTTOM UNITS (CONT.)

6.1.6 'A' DECK UNITS 241. 042 and 141 VASE A SHEER OF 6.3.3 (CORT.) .3679' PER FOOT P/ID 'B" AND "C' DECK UNITS 231. 232. 131, 221. 222 AND 121 HAVE A SHEER OF IN FITTING SHELL HOLD PLVTE GIRDERS VOID SHELL .5482' PEA FOOT. THEY WILL BE BUILT ON LEVEL LONGITGDINALS IB PLOMB TO THE TANK TOP LAYOUT. MOCKS AND CARE MOST BE TAKEN TO SET ALL BULK- THE LONGINTUDNOLN V.8. OF THE TA' -R' GIRDER HEADS, SHELL ASSEMBLIES. PILLARS AND FOUNDA- UNITS 3GO. 2, 3 AlIO 4 AND ALL L000IIUOITALS TIONS TO THE PROPER DECLIVITY. UNITS 307, 401 UND 402 ARE COAXIAL ASO SHALL BE LAID OUT WITH GIRTH TP°ES PR:ED TO FITTING. 6.2OECE/SHELL ASSEMBLIES ON LONGITUDINALLY FRAMED UNITS CHECK HEIGHT 0F (SEE "UNIT DIMENSIONAL RECURS" FOIRI. ENCLOSURES (3) EVERY 2ND LONGITUDINAL FROM THE TANK TOP (AFTER AND (4)) END ONLY) TO LOFT OFFSETS U/IS TCHPOAARILY RUADE IF NECESSARY PRIOR TO SHELL INSTALLATION. 6.2.1PILLARS SHALL BE SET TO HOLD TVE BOTTOM TO A Al/A" OF POSITION RELATIVE TU THE 3-O' 8TH 6.4 SPECIAL UNITS AND TIlE MASTER FRUJ0. THIS StALL 8E MONITORED DERING WELDING AND WELS SEQUENCIUG SHALL 8E 6,4.1 C'IPLEX '3-0" UNITS SHALL 8E BUILT IN GENERAL ACCORD- USED TO CONTROL OR CORRECT 08V IVT 10113. ANCE WITH THE GUIDELINES OUTLISED FOR SIYILAY UNITS. PARTICULAR ATTENTION SHOULD BE PAID TO HOLDING MAJOR 6.2.2SHELL ASSEMBLIES SHALL BE SET TO LOFT OFFSETS STRUCTURE (DECKS. NAO.. SHELL. 411.0 TANES, STRIUGEAS. (PLUS NEAT STOCK) USING V BTK ESTABLISHED BR ETC.) TO PROPER HEIGHTS. HALF WIDTHS 4140 TORE/AFT THE SURVEYORS. CURE RUST BE TAKER TO MEASURE POSITION AT THE ERECTION PLANES: HALF AOÛTAS FROM A NEAT CUT LINE ESTABLIG DO SR CHECKING HEIGHTS FROM TAC DECK AS IN SEVETLY EX: A. BOTT'OM OF RHOS. FAD/AFT AND IB/OB Oil UNIT SHAPED AREAS ERRORS IN HEIGHT CAN RESULT IN 321. APPROXIMATELY EQUAL ERRORS IN HALF WIDTHS AS S. DUCE AND STRINGER HEIGHTS 011 AFT EBD OF SHOWN BELOW: UNITS 404 A/ID 433. £rrr'r in M*tNi Ai,i.UinX Error io Holt Width 6.4.2UNIQUE DIMENSIONAL RECORD FOP/IS SHALL BE USED OS THE 1.1 FOLLOWING U:IITS: worAir.i HUr t.',.dth - - Mold lo i/N" Pia. 101 6 111 HORIZ. INTERFACE ArID FUS ERO O" io for' Shr'Urdn.. 201 TOP 202 & 212 HORI2, INTERFACE AND AFT END 102 & 122 HORIZ. INTERFACE 121 BOTTOM Error. jo ).1f oidOR r..o,UtthS 142 BOTTOM fro., U000rr.00h.Wgbi 321 & 301 4ORIZ. INTERFACE AND F40 END 201 402 FAD NID AFT END HALF WIDTHS (INCLUDINO NEAT STOCK) SHALL ER HELD TO 404 AFT END ±1/4' BITA AN ADDITIONAL D TO I/4" ADDED FOR SHRINK- 433 BOITCPI AND AFT END AGE DEPENDING UPON THE SHAPE AND EOPERIENCE.

6.2.3WING TURK ASSEMBLIES SHALL 8E SET TO THE L000ITUOISRL BULKHEAD IN L lEO OF TVE SHELL PLATIOG DUE TO THE POSS- IBLE ERRORS OUTLINED ABOVE, THE LOVER EDGE OF THE ERECTION RAD, SHALL 8E HELD TO ±1/4". 7.1THE MASTER REFERENCES ESTABLISHED IN THE SHOP SMALL RE USES 6.2.4 SLOPING LONGITUDINAL BULKHEADS AND ROMPS SHALL RE SET FOR REGULATING UNITS UN THE UVAS. IF MASTER LISES ARE ROT TO THE "LOTI IOFOP/IATIO1I BOOK" (PLUS IlESO STOCK). AVAILABLE. THEY SHOIJLD VE ESTABLISHED FROM DECK FONITYG, THE ENDS OF RP/1P5 SHALL BE FITTED TO *1/4' VSA LEFT HEIGHTS OF DECKS AND FLATS AND FORE/AFT POSITION OF IST ENUELDED FROM THE ERECTION BUTT FOR U-O" OR THE NEAR- MAJOR WEB FROM THE NEAT END. (THE SHELL HALF WIDTH SHOULD EST WEB FRANE, WHICHEVER IS LESS. NOT BE USED AS THIS INCLUDES THE ERROR RESULTENG FROTI PLATE BURNING. WELDING ANO TRIMMING.) 6.3 INNERBOTTAFI UNITS 7.2THE SURVEYORS SHALL ESTARLISII THE MASTER OR 3-O" 8TH ORI 6.3.1P/S INNERBOTTOM ASSEMBLIES. FR. 73½ TO 202½. SHALL BE EACH DECK AS FITTING AND WELDING PROGRESSES TV HOLD IT. LEVELED USING AND THE V.B. GIRDER. UNITS WD TIlO AFT Of THIN (UNITS 307.401 and 100) SHALL BE LEVELED 7.3 DURING THE INITIAL ERECTION AND WELDING OF THENIIIRROTTON USING THE T.T. 0.6. THE KEEL CONDITION P/ID THE TANK TOP AT 21-6" AlIO 39-6" OFF SHALL BE CVECKED ANO PLOTTED BIWEEKLY. 6.3.2ON UNITS 301. 2. 3, 4 AND S THE P/S ASSEMBLIES VYD BOX GIPSER SHALL BE SCRIBED ¡N USING THE 3-O" 01K 7.4THE CONDITION OF TRE RAMP RECESS IN "C" 3240 "D" DECK SHALL ESTABLISHED IN THE PANEL SHOP. (SEE SECTION C.D FOR BE MONITORED DURIllO ERECTION LAND HELD TO A MEAN PLANE nl/4, ADJUSTMENT TO UDO 1/4"(EUT STOCK.) (IF THE P/S UNITS ARE NOT RURRIED IN TVE V.B. LEAVE THE STOCK ON 7.5THE DECK OR TANK TOP CONDITIONS 8ELOW CARGO DOORS SHALL RE THE PORT SEAN TO 8E CUT UN THE WAYS.) MONITORES DURING ERECTION. UNITS CONTAINING CARGO DOORS 6.3.3 SHALL RE REGULATED IN ACCORDANCE WITH REFERENCE (3). 7.6CARE MUST BE TAKEN TO ROLO DECK TO DECK HEIGHTS IN ERECTING UNITS 111. 411. 422. 432 AND 442 IS ORDER TO INSURE ALIGN- MENT WITH 1-TSLTI-LEVEL UNITS FaS OR AFT.

ti-

9-14 MSP 909-002 APPENDIX: 9.2.2

ACCEPTANCE CRITERIA INGALLS SHIPBUILDING MANUFACTURING STANDARD Module Integration PROCESS NO. 909-002

ITEM DESCR I PT ION ACCEPTABLE TOLERANCES

DEVIATIONS FROM THE MOLDED FORM, FOR THE COMPLETED HULL (AFTER WELDING MODULES TOGETHER) NOT E THE FOLLOWING TOLERANCES ARE FROM NAVSHIPS 0900-000-1000 AND ARE ONLY TO BE USED FOR FINAL BUY OFF OF THE COMPLETED HULL: - + 1" RFAN OF HULL O-1000-O'' BEAM OF HULL OVER lOO-O" + 2, -1"

LENGTH PER 100-O" + 1"

HALF BREADTH O-50-0" + 1/2'' HALF BREADTH OVER 50-0" i',, - 1/2"

TWEEN DECK HEIGHTS + 3/8" MAX. ACCUMULATED DEVIATION; - FOR DK. HEIGHTS FROM BASELINE TO 50-O" + 1" FOR DK. HEIGHTS ABOVE 50-0" 1 1/2", -1"

THE DISTANCE BETWEEN THE END FRAMES, BULKHEADS OR FLOORS OF ANY TWO ADJACENT MODULES ii. 1/2''

ALIGNMENT 0F MATING ENDS OF 1/2 THICKNESS OF THE STIFFENERS, LONGITUDINALS AND THINNER MEMBER G I RDERS

1/2 THICKNESS OF THE 14 ALIGNMENT 0F DISCONTINUOUS MEMBERS ON OPPOSITE SIDES OF A THROUGH MEMBER. FOR STRUCTURAL SHAPES BOTH THROUGH MEMBER. FLANGE AND WEB ARE TO FALL IN THIS LIMIT.

SEE APPENDIX 'A' 5. PLATING FAIRNESS

9-15 Module Assembly (Assembly to Assembly Interface)

ITEM DESCRIPTION ACCEPTABLE TOLERANCE

MODULE LENGTH +1''¡n ÌOO'-O'' EE APPENDIX 'B'

MODULE HALF BREADTHS 1/k''

BULKHEADS BETWEEN DECKS, TOP TO BOTTOM OFFSET FROM VERTICAL 1/k"

14 PLATE EDGE ALIGNMENT PLATE THICKNESS O to 3/8'' 1/16" PLATE THICKNESS OVER 3/8" +1/8"

PLATING FAIRNESS SEE APPENDIX 'A'

ALIGNMENT OF MATING ENDS OF 1/2 THICKNESS OF THE STIFFENERS, LONGITUDINALS, THINNER MEMBER FLOORS, ETC.

ALIGNMENT OF DISCONTINUOUS 1/2 THICKNESS OF THE MEMBERS ON OPPOSITE SIDES OF A THROUGH MEMBER. FOR THROUGH MEMBER STRUCTURAL SHAPES BOTH FLANGE AND WEB TO FALL IN THIS LIMIT

9-16 Structural Assemblies

ITEM DESCRIPTION ACCEPTABLE TOLERANCE

OVERALL LENGTH AND WIDTH +1/2''in 50-O'' USE PPENDIX 'B'FOR LENGTHS AND WIDTHS ABOVE 50-O''

HEIGHT-INNER BOTTOM +1/14" OVERALL HEIGHT - ELSEWHERE +1/2" (INCLUDES INN. BTM.

PLATE EDGE ALIGNMENT PLATE THICKNESS O to3/8" +1/16" PLATE THICKNESS OVER3/8 +1/8"

ALIGNMENT OF MATING ENDS OF 1/2 THICKNESS OF THE STIFFENERS, LONGITUDINALS, THINNER MEMBER FLOORS, ETC.

BULKHEADS BETWEEN DECKS, TOP TO BOTTOM OFFSET FROM VERTICAL

PLATING FAIRNESS SEE APPENDIX 'A'

BOWSIN PRIMARY STRUCTURE SPAN (FEET) (FRAMES, BEAMS & STIFFENERS) DEPTH (INCHES) x 14 WHERE SPANIS DISTANCE BETWEEN THE SUPPORTS & DEPTH IS DEPTH OF MEMBER FROM UNDERSIDE 0F FLANGE

9-17 MSP 909-002 APPENDIX 'A' 43 1! 1 / 44 it U 4Z _L_'I ;;-- // I ____L_ fi/ f f, ì2 ' 3' 1'AF 7 /8 .V ¿ 7/ ,.,I ,,,/I j 4" 34 - /1 J .- J 7/ ,." _____ 'I 1 t/ , 4' .: j_ - L ¿f ,/ ra // /1 J' ,' .,I .'

7/_I - 1 i/ 7/ 7/ Li7/ -'-i -

18 FIS.2-6

48 4'.- I fit 44'-a 42 I y ii 4c. .ri i /

34.____j,,- _-___-1 30 t. -

I f 22 .7

FIS. 12-e

Lo YL. Y4 '/i ve/4re'o yß V4 !/ s,'8 /.L, Plate Thickness (Inches) PERMISSIBLE UNFAIRNESS IN STEEL WELDED STRUCTURES NOTES: The tolerances specified above are plus or minus dimensions from afair line. Figures 12-7 and 12-9 apply as fol lows: a) entire shell b) uppermost strength deck c)longitudinal strength structure within midships 3/5 length including inner bottoni tank top continous decks below the uppermost strength deck d) bulwarks and inteior superstructure BHDS. For other structural BHDS and decks the unfairness as shown byFIG. 12-6 or 12-8 may be increased by 1/8". 1'' material. '4) For internal thickness greater thenI'' use the tolerances for 5) FIGS. 12-6 and 12-8 apply as follows: a) structural ends forming living space boundary and passagewayscontiuous to such spaces b) decks in way of living spaces c) decks exposed tothe inner bottom platelongitudinals. weather d) tank and nain transverseBRDS e)

9-18 APPENDIX 'B'

MSP 909-002

3/14."

N / 1/2"L_ 4

0 10 20 30 L0 0 60 70 60 90 .100

Lertth in Çeet

PERMISSIBLE DIMENSIONAL TOLERANCE (Pius or Minus)

9-19 APPENDIX 9.2.3: SPECIAL TOLERANCES IN USE AT LEVINGSTON SHIPBUILDING FOR A DRILLING RIG

WELDING NOTES Maximum gap for fillet weld = 3/16'. Gaps > 1/16" shall have the fillet we i d size increased by an amount equa] to the size of the gap.

Permanent backing bars shall be used only where shown on the drawings or spec i - fically approved by the Owner's representative. They shall be of steel equai ¡n grade to the material being welded. Splices in backing bars shall be welded with full penetration welds.

Temporary backing bars shall be removed and the weld root gouged to sound metal before finishing welding.

All faying surfaces shall be seal welded and all welds shall be continuous except as noted. Seal welds shall be a minimum size of V8" and large enough for fus ¡on to the plates joined without cracking.

T1C/. A'I j .5Th.ß. L')t. s ± P.F1 L 5 = 4-_¡'I

TI C.- L

'i Ir

T'(PIòL fz-i

ERECTION TOLERANCES DETAIL

n,, EQ M4JhfO GiER 7-rEPE4r

i \wìvMEA 5JD 4CPOS TI-lE V4LEY ALLOv4LE /4QPj,'./ A-7A/s&urr5

C/SE VALUE 5QW,V ...Cp- Ccut4;v CI EQE'r4L 3E4M. .41JLL 4 T//vG, 3C.aS'-'Q.JSET1 W4? 7,G,iTF4$, 84 A".Q GQCEQ 'VE&5 4ir'-'ìe 5ij.

TUB ULAPS ' STA 8/L/rY CGLUM/VS L_ O WEP /-'UL L S ./4.ATC,V% 1vo: D,O,'CQ2 0/ c 02 ,4rsrAa,L/ry COL cUMAIS L OWR /-JLT4VS/A'SE 7JEuL4PsTrQ3 A ìf5, TIE 7UEULARS 7Q 3"TO54 r T/AL %_,A-'cr,q Q- CtJE STA&LJrY COL LíA4A/S r ÁYSEE NO TE) .S Ti-I -S fr'4 L. L. aE w''JA' '' z'CF 7 ACJACVTLE/v5Tn LCviER HUL L AJO 7E) -

9-21 ,_/5'J._-. 5.c-.- I;' GAC.F5C.iC -

CL I - s WOELD - -_.-:4._ ;. j TE1 j : 'O'.i&.5E4.-1.

,144)(AVLE C/-,4NGE AT 5,'L/CE

.7'4,'i5VEQ55 55A,d -SE

h ,'

&Lg -r TcEZ. -

C Q,'NCf.-''3 c.'w-' ¿ i_a rAws,r,C1S ?Ç...75 Cc'TOL

'\' UPi/TC AiVGL c A A / -I ¼_IL_L:i ,ç.'- / 'L_.Z- \'/L. CoÑ 7.0 . F.i t

9-22 APPENDIX9.2i+

NEWPORT NEWS SB & DO DIMENSIONAL TOLERANCES FOR LNG TANKERS ± I" oW

INNETgLN D

ft4N&Tc$ OF tt SUI?

I1OZi ZONTAL -ATl»1 PLAÌJ.E ow

J 'a

I NIIEP- TRANSV SECIIOM TRU4 DK T --T COFF /0v?LANEAT4 ht

HQRJ ZCNTL 90* L Ç&TUI PLANE

±1/2 ----t 1/2.

roJ LOJL. SECT IONi

9-23 APPENDIX: 9.2.5 SEATRAIN SHIPBUILDING STRUCTURAL TOLERANCES

QUALITY CONr1ROL INSTTWcT iou FAIRNESS

'\ppcndix(A) is to be used to determine acceptable fairness with-

in welded structure. Areas of applicability are as follows:

Entire Shell

Uppermost strength deck

Bulwarks and exterior superstructure bulkheads Tank and main transverse bulkheads

For other structure bulkheads and decks, the unfairness

as permitted by Appendix (A) as applicable, may be increased by 1/8 inch.

¡lETIIOD OF INSPECTION - FAIRNESS CHECK

Departure from a plane surface on flat plating or geometric form of curved plating shall not be greater than specified in Appendix

(A) by measuring with a straight batten on straight plating and a curved batten on curved plating.

The above noted method is to be used it aid is necessary in determining the acceptability of structure.

9- 2 APP. 9.2.5

DIMTNSIONAL TOLERANCES

TOLE PAN CE "lJ\T'L TK STIFFNER SPACING 3/8" flt1TO 3/8k O"To16" 17"To70" 7/16" 21"To36" 5/8" 37"'ro48' 3 / 4" 49"To72" 1" Over72" 1½" /16" To 5/8' O" To 16" 5/16" 17" To 22' 3/8" ')3" To 26" 9/16" 27" To 36" 11/16" 37" To 48" 7 / 8" 1" 4')" To 72" iL. " Over72" L." 11/16" To 1" 0" To 24' 25" To 32" 3/8" 33" To 42" 1." 43" To 48" 11/16" 49" To 72" S3 / 16" Over72" 1" L." r, n r' -4 Q .)O 1 1/16" To lo 37" To 46" 3/8" 47" To 60" 7/16" 61" To 72" 1. 'I Over 72" 5/8"

9-25 APPENDIX: 9.2.6

SUN SHIPBUILDING AND DRYDOCK COMPANY

SHIPBUILDING PRODUCTION STANDARD

(HULL DIVISION)

1976

9-26 Section Division Sub-section Item MATERtAL Remarks IJnit Inches Grade of Pit 1. Grade A Graderepairis B to containsis be unnecessary. repaired a medium if necessary. degree of pitting and is to be considered so slight that any -o e AreaN ratio a (percentage) r 2. Boundary lnes of grade B areGraderequires included C contains some In graderepair. an extreme degree of pitting and A B I 3. The area ratio axis,A Ar respectively. means the percentage of pitted areas where of pit denoted r on the horizontal o- .- I - a) . I c surface appearance is unsatisfactory for practical use. For skin plate Areaof pit £ j ¡4, Repair method of surfaceDepth of flaw defects Percentage of pitted area d Total area of a plate w. on Note:the d This indicatesgraph and not steelthe area dPlate of thickness O.07t(but in no case d t 1/8) removed by grinding L quality.Grade of surface flaking I. Grade A is to be considered so slightO.07t that any repair d O.2t welding.grinding followed by - o Area ratio A I (percentage) r GracieGradeis unnecessary, BC containsandcontains a medium degree of surface is to be repaired ¡f necessary. n extreme degree of surface flaking flaking C u B 2. Boundary lines of grade B areorand C included rc4uiresrespectively. insome grade repair. A a) 3. Repair methodDepthPlate of ofsurface Inicknessdefects flaw td L I d I O.G)t ...... moved by grinding 1/8) t. (but in no case d d O.2t . . grinding followed by welding. Section Division Subsection over In20% case of thickness,where defect or is Item MATERIAL In the case where cavity cracks and other injurious tema ks Unit: Inches ,;c 6over inches of length, 1 inch of depth and andbedefects to checked be repairedare found, by afterapproved removing welding the procedi.re. defects, it by magnetic inspection, or ultrasonic inspection1 is to cO (a) ( in itIn(a). cdnthe becase gouged where Out the and range built-up of lamination by welding is aslimited. shown - (b) / ) thebutIn built-up the case welding where theas shownrange ¡nof (b).lamination is limited,near the plate surface, it ¡s preferable to make Thecaseit standard where the minimum range breadthof lamination of plateis is recommended tofairly be exchanged: extensive. to locally exchange the plate, in the o Shell and strength decknotlarge underunder constraint large constraint . . . . 36"60' J O extent.Thedegree whole of plate lamination must be is exchanged very severe in casesand wide where in theits Other structural members 12" ros)o) o ofThe befraiiiincj. direction the same asof therolling original. for the Butts replacement are to be plate6'' clear must ,i roero>-eeQLO Division MARKING Unit: Inches Section Subsection ComparedSize and withshape correct ones. Item Standard Range Tolerance limits -+ 1/8» Remarks ComparedCorner angle. with correct ones. O" ± 1/8" +0' ± 1/8" I § Curvature o» ± 1,8 Locationfor fitting of member compared and markwith position.Reloft to correct ' m ° csiL correct one. euO)CL o ComparedBlock marking with (panelcorrect block) one, . - position.Reioft to correct L)4-J withfittingLocat correct ion to of one.block member compared for Reioftposition. to correct Division Item GAS CUTTING unit: Inches Section Subsection I) Upper edge strake.of sheer Standard range Tolerance limits grindMachine edge cut smooth 900 and to Remarks 3)2) Strength deck between strengthMainof (3/5)Lopening longitudinal member. of shell plate. and free edge 0 1/16 eliminate notches. strengthLongitudinal members and transverse 0 Butt Shell plate & upper 1/16 Grindif and present. weld notches e> Eoe Weld OEhersdeck between (3/5)L 0 1/16 (notch unnotchesGrind edge and if weldpreparation)present. Straightness Automatic welding Fillet weld O0 ± 1/321/8 max. NO root opening edgeof plate Manual welding)emi-autoniatic welding) -+ 3/16 1/8 1/16 allowed. of Groove Depth --y>-- - ± 1/16 ± 1/8 Division Subsection Item GAS Standard range CUTTING Tolerance limits UnIt: Inches Remarks Section of Angle Taper L'3a 600 AngleLength ofof tapertaper not to d t.. L-.j j a requirement,anybe lesscase than Rules600 in for . (See ABS T I SectionSteciBuilding1975,letter Vessels, 30.3.1,byB.Aniia,of& Classing 1975, and prin- + August, .2 c Size General members + 1/16 + 1/8 A.B.S.cipal surveyor of Member of ofEspeciallycompared floor withand for girder correctthe depth of sizes ±- 1/16 ± 1/8 comparedBreadthcorrectdouble bottomsizes. ofwith face correctcompared bar, size. with ± 1/16 Section Division Subsection Item GAS Standard CUTTINGrange Tolerance limits Unit: Inches Remarks Preparation Edge Automatic welding Bevel Angle No bevel OrangePlate thickness t 5/8 100150 5/8 tL I 1/4"i 1/4" ° ManualSemi-automatic welding weldIng 30022k° t I 1/1+' Enec ConsumableFab welding Nozzle (one-sided(Electric Slag) welding) 3/4',' No bevel 3/4 to 7/8 root No bevel 200 bevel SeaDwg. Sun (SS-7O4standard Division FABRICATION llnit. Inches Section BreadthSubsection ofFlange Item Comparedcorrect with size Standard range ± I/B Tolerance limit ± 1/4 Remarks Mill allowance ' ' correctCompared size with ± i/8 ± 1/4 Mill allowance Flange & AngleWeb between Comparedtemplate with ±411 1/8 ± 1/4 Per LH breadth Curvature or ç 4is c straightnessthe flangeplane ofin Per 400 Inches in length ± 3/8 ± 1 Standard mill al lowance ac - = straightnessin the planeCurvature or ± 3/8 ± 1 Standard millallowance of theflange web Breadth Per 400 inches in length correctcompared size with t 1/8 ± 1/4 cot;) . Angie between flange and web 'Ff flangetemplateinch breadth percompared 4 of with + l/8 ± l/L Samelongitudinal as for Division Section SubsectionboxTemplate shape in comparedLocation of plate edge, with correct Itemone Standard range ± FABRICATION 1/16 Tolerance limit Unit: Inches Remarks LocationcomparedShape of ofwithcurved check correct surface, line onefor ± 1/16 ± 1/8 1/8 C Q. Sectiontemplate with(forlevelling correct transverse) byone. sight, compared ± 1/8 1/4 orUse furnace of roll sets. sets .0 .0L (for longitudinal ± 1/8 ± 1/4 machine.methodStraightfranl9 using bendingline cold moEas) C o-L Shape,correct compared One with ± 1/16 ± 1,'8 Per 2OO' length - templatesStringer Other correctShape, compared with angle one. ± 1/16 ± 1/8 a ang!e d[ fl( gaugewith angle compared ± O1/16 ± D 1/8 Miii tolerance s) _._L - ±1/8 Co) compared with template - 4o 40 ectlon Division Subsection Item FABRICATION Standard range Tolerance limit RemarksUnit: Inches Longitudinals Frame and templateCurvature or compared check line with i' lengthPer 35 -0" Deviationform from correct fo rrned Openingto formed of fitsection. up 15 ------. section ± 1/8 ± 1/4 4-J4) Correct form inscribed ru Do- -'th-' '1 Deviation in -t,- D correct form flange angle + 1/8 ±1/1+ Ccn4)ru -' \\ k- Deviation of face plate -+ 1/8 ± 1/1+ weidAs the distortion resultper of 4" length -z- - + 1/8 ± 1/4 Per 4" length ii Section Division Subsection item FABRC/TION Standard range Tolerance limit Unit: Inches Remarks 5tCylindrical ructu(mast, res post, etc.) Diameters But ± 5/32.± ___UO But ± 1/4 + 150 diameterDepends on Curved shell plate In regard to check line (for longitudinal) (max.) (max.)± 1/2 Per 30'-O" length -o-s) " (for transverse) + l/4 + onlyOpen to template Gapand between sect ion shell template plate 1/1+ (max.) mustMot permitted, either fit set or 1/4open inch at max.sides to 11 se pl aT>" (max.)

i Division SUB-ASSEMBLY Unit: Inches Section Subsection Breadth of subassembly Item Standard range ± 1/4" Tolerance limits 5/16" (40' cut when tooDepends lonq. on size Remarks x 60' typical) .0 > SquarenessLength of sub of assemblysub-assembly ± im ± 1/4 5/16 MeasuredCut when differencetoo long -e De of atdiagonal linesfinal markinglength OEcs) e0. Distortion of sub assembly ± 1/4 Measuredgirder.of length.web on beamDepend5 the or face On 9- euO> u.. membersDeviation from ofskin interior plating. ± 1/16 ±1/4 when interiorExcluding the members case uDL arelapped connected joint. Frame by over Accuracy of this dimens ionSkin plate >. Breadth of sub assembly ± 3/16 ± 5/16 girthMeasured cut along whenOnq thetoo L) -0..e viV Length of sub as-embly t 3/16 ± 5/16 Cut when too long Division SUB ASSEMBLY Unit Inches Section Subsection Distortion of sub assembly Item Standard ranqe 3/8' Tolerance limits 3/14" markingCorrectwebMeasured or girder. theline on final facewhen ofthe Remarks -o >- thedistortion limits. exceeds 141 Squareness of sub assembly 3/8 9/16 of Measureddiagonal differencelength InoC mD at lines.final marking EuC >g) membersDeviation from of skin interior platIng 1/4 The same as for 'I- >-o L)DL the f lat assembly. ç,DLluu - Breadth of each panel flatThe sameolate as assembly.sub- for the 04.1Uv% -a >- Length of each panel Same as above. (ti4-4-°VIlu Distortion of each panel Same as above. 0 D 14, membersDlation from ofskin interior plating Same as above. Section Division Subsection Item Standard range ACCURACY OF HULL FORM Tolerance limit Remarks Unit: Inches perpendicu'arsLength between ± 1/2" per 330 feet ± I AppliedmeasurementForlenUth tothe andships convenience theabove. of point ofwhere the 330 ft. oCI, Lenqth per 330 ft. be thesubstituted keelcurve of forthe thestern may is connected to 0eC . measurementfore perpendicular of the length. in the - Cua. Lengthbulkheadperpendicular between of engineaft and fwd. - 1/2+ - 1+ 1/2 Shaft(Stock length. allowed on flanges.)shaft L Breadth Moldedroom breadthamidships + ± 1/2 breadthApplied toand ships above. of 149 ft. weatherMeasuredApplied deck. onto theships main of or 33 ft. Depth Molded depthamidships ± 1/8 ± 1/2 )ps depth and above. (- and downs (+) 9- o Flatness of wholeDeformation length for the 1/2 - 1 to + 3/14 keelagainstlength. sighting. the check line of Per 600' O00e Keel adjacentdistanceDeformation bhds.between for thetwo ± 1/8 ± 3/8 Sightinglaser beam,by the or transit,water level. Division RIVETING unit: Inches Section Subsection Diameter item C0f11 Standard range Tolerance limits Relation between rivet Rerirks Hoie correctcompared size. with dia (d d 5/8and hole dia Ød + 1/16 ( 0 from standard -Difference 3/k4Id i/Sd5/8 -I/3201/16 1 1/k -1/I6A1/i6 l/64øl/16 d3'4Jl 1/8 I l/+ ød0d + +1/16 1/16 () correctparedDepth corn-with size. V Counter sunk standardfrom HDlfference(H) -1/32AN4l/l6 .-1/32 AH 1/16 Section Division Subsection Item Standard RIVETING range Toleronce limits Unit: Inches Remarks Counter- sunk inclination F1 ,H,-H1P M1.1/l6 ¿1H' 1/16 SurfaceFaying Clearance atTest feelercaulking knife gauge tightedge. not contact Discrepancy C & 1/61+ AC i/6. oppositeprmttedrivetßG hole side.to> 1/16reachon To be reamed ThroughUnfairness i' I G ¿ 1/32 G 1/16 '.To be I 4Jo holes L weld.buildingre-drilled up byaft.r > Pitch markingDeviation point from p. P, p. 'o-o-o- P, - 4P ± 1/16 ± 1/8 e Counter- head sunk Edge height I h h O h L I/B O L h 7/16 Section DIViSiOn Subsection Item Standard range RIVETING Tolerance limits Unit: Inches Remarks Deformation D dd > 4 7/83/1 d 1/81/16 dd 7/83/14 ¿ 3/161/8 PointOverlap height "HDD- D - d 4 5/87/83/14 1/164 j 5/161/14 1/8 d 3/145/87/8 1/16 3/85/16 H 3/16 PoInt Edge height :h d d 3/43/,7/8 I h1/81/84h ±H 1/8 4 H 1/8 3/161/14 ddz 4 3/14,7/8 3/14 1 1/8l/8 hh 4 1/8 1/141/4 N 1 . 7/8 7/8 1/8 Division WELDING Unit: Inches Section Subsection Undercut Skin plate and face Item Tolerance d 1/32 no repair necessary. limits To be repairedRemarks by (butt weld) plate between (3/5)L d 1/32 dthanin continuous lengths greaterL 1/16 repairL4i'd where found 1/16 repair where found using small e1ctrode. -o oe (fillet weld) Undercut d 1/16 equal to 1/2". for 't" greater than or exceeds toleranceRepair when defect eo- Length Leg onesCompared (L, t)with correct d 1/32 t:L: throatleg lengthlengthfor ''t" less than 1/2". limits.sizeNo allowance, and repair gauge if to Angular under,length. in continuous distortion of welding joint Skin plate(3/5) between L Span of frameor beam - overby itIntoleranceline cases where limits, it Is to be repaired heating or to is platingFore and andaft transverseshell ' 5/16 W i/ cuttingbe re-welded & refitting. after '- co°oc ' -' strength member.Others WW - 5/16 o.-L 5)C Section DIvision - SubsectionButt weld item ALIGNMENT AND FINISHING Tolerance limit Remarks UnIt: inches >4.. La.- Butt weld tO a ho rivetle a 2» No set limit, ABS rule. v-04)-00n 4-' 4) Butt weld 4 2" Same as above. CE00 4) u to r- 4 a 3/n" z.'-' o o - Fillet weld ------cut,Rat hole is not to be if "a'' Isfrom at leastedge of butt smallerIn special distances cases caneven Gap between Stiffening member location weweld I3/l+i d. to edge of fillet be acceptable. C 4) memberstiffeningplate and (without edge preparation) - C 3/16 avI) L platePerpendicular to cU '0E B (not limited) No (natural1/2"edge plate preparation opening) or below. on Oblique to plate Section Division Subsection Item ALIGNMENT AND Tolerance limit FINISHING RemarksUnit: Inches Through tightpiece plateand 2 C C1 < C2 C2 1/2 4- ab 1/2 max. Maximum off center -Q eLV' t -web thickness La s 3/16 max. a Tee'if overmax. 1/4". 1/4" install liner 1-Je t - web j / a 2' collarsApplied only.to clips and e thickness j j I a 2t + I VesselsStandardBurule lappedildin9 30.9.31975'' for joint.& effectiveClassing''Rules See forA.B.S.Steel I iEi b = 2" Same as above. Division ALIGNMENT AND FINISHING Section Subsectionpermissible Deviation from Item the straight Tolerance limit d 5/16' For length of up to 30 Remarks Unit: Inches stiffenersgirdersbeams,distortion frames and of support.lengthline ¡n between reference 2 points of to the d + L For125 lengths inches. ofinches. up to i . d s 1/2 Forinches. lengths over 125 i. L warpingPermissible Profile to plate d d wIth profilesof deviation Max. be IntermediateInterpolated. values must L : ,, : :': Flange In Line up error Ii Ji ab -< 40"5/16, but with a d 1" a T longitudlnals b maximum oir .04b Refit.04b, if exceeds H-: \ Section Division Subsection Item ALIGNMENT AND FINISHING Standard range Tolerance limit RemarksUnit: Inches Alignmentfillet joint of Srength member a 2 Maximumfilletcase1/2 of more offsetweldt, thanbut size. to in thebe no adit1t-thickness fference t2 Other a 2t Maximum offset L beambetweenDifferences & frame the Beam . a Beamknee a: difference 3 a 2t 1/2 web or beam thickness u - Frame I) C Lapp weld 1 1/8Increase weld leg a 3/16 L. cIIia a a a 2) a> 3/16 refit length, rule leg Alignmentbutt join of Strength member O a 8 a Ref i t 1 adifferencetth (thinner aT plate)i ckness Other max.a .2t 1/8 a ) .2t or a Reflt > 1/8 Section Division SubsectionGap before Fillet weld Item Standard range ALIGNMENT AND Tolerance limit FINISHING Remarks Unit: Inches (fillet welding weld) tl H1t2 ---t-i O a 1/4 t2 Weldingpreparation with bevel to make bevel >. t, a . tl weldremoveattachedge the of afteroppositeaweb backing to welding,30..1450, side. strip thenWeld and out.. b t2 b 1/8 Usesize when114". to openingincrease exceeds by ''a'. "T" piece to extend c 4., C?2.5xt 2t2than1/8 t2 beyond but notweld, more than t2 t 2t2 t not less t2 - d d 6 UsePartialliner) when renew. texceeds 4 (t2). (thickness of ('L\ ----- reduceBuild upseparation with weld gap. to Section Division Subsection Item ALIGNMENT Standard range AND FINISHING Tolerance limit RemarksUnit: Inches weldingbefore Gap Butt weld (manual welding) a: Gap a'3/i6+ 1/16Gap opening a ( t, 1) After welding withweldmovebacking afterit and st back finishrip, re- 2) chipping. a , t1 (' \L a ofproduceAdd 3/16. weld requiredto edgel/4 togap u>' Standard gap opening. a3/l6 3) Trimbevel edge to andaa-3/16 < re-1/8 oDt. Butt weld (automatic 14) a > t, plate renew. Partial 4-J c weldingbefore Gap a: minimum separation gap welding) I) Both sides submerged arcbetween welding plates O a t 1/16 0 t a t 3/16 burnth rough mayIn theOccur, case where 2) Flux with manual or core welding CO2 o a t 1/8 0 s a 3/16 use a seaiinqoverIn bead. 3/16,case where see a is Gap 3) One sided submerged arcwelding with flux copper O a t 1/16 0 t a 1/8 burnthoughweldingIn the manual case may where occur before welding 14) One sided submerged arc asbestosweldingbackng with backing.or fluxfiber backing. O a 1/14 0 t a 5/16 use a sealingSame bead. as above. Division ALIGNMENT AND FINISHING Ren'arks Unit Inches Section Subsection Item Tolerance limits I) j, Build up with Edge X-bevel (/LL a 3/16 3/l6at+3/l6 T7/ weldmax.on eachA.B.S.tor l/2tplate. toinspec- be Preparation 2) a > t + 3/16 'Z14 not i fRenew ¡eci. plate sect ion. K-bevel }\\ ('Z Z4 a 3/16 I) 3/l6 t + 3/16 notified.Renew platesect ion Division tern ALIGNMENF Tolerance limits AND FINISHING Remarks Unit: Inches Section SubsectIon NN a 3/16 I) -j -t/2 side.upBevel with 1450,A.B.S. weld build in-on One Edge 1-bevel weldingUsed for only, submerged maximum arcN''J 2) 3/16

IJ I liii III II I.i I Ifi III I il i/s I/lI 3/54 3/8 1/2 5/8 7/3 1 .0

PLATE TH'CKNESS (INci-Es) Figure 12-7.Permissible unfairness in steel welded structure.I

Note: Applicable to - - Entire Shell - Uppermost Strength Deck - Longitudinal Strength Structure NAVSHI PS 0900-000-1001 I JUNE f969

9-62 9.3: EXISTING INTERNATIONAL STANDARDS

CONTENTS

9.3.1 Japanese Shipbuilding Qual ity Standards, JSQS, Hull Part, 1975

9.3.2 Production Standard of the German Shipbuilding Industry, Nov. l97+