Guidebook For The Design of Asme Section Viii Pressure Vessel

FOR THE DESIGN OF ASME SECTION VIII
···PRESSUREVESSE·lS··
Second Edition
GUIDEBOOK
by
James R. Farr
Wadsworth, Ohio
Maan H. Jawad
Nooter Corporation St. Louis, Missouri
ASME Press
New York
2001
Copyright © 2001 The American Society of Mechanical Engineers Three Park Ave., New York, NY 10016
Library of Congress Cataloging-in-Publication Data
Farr, J:m1e~R,
Guidebook for the design of ASME Section vm pressure vesselslby James R. Farr, Maan H. Jawad.-2"d ed. p. em. Includes bibliographical references and index. ISBN 0-7918-0172-1 1. Pressure vessels-Design and construction. 2. Structural engineering. 1. Jawad, Maan HII. Title. TA660. T34 F36 2001 681'.76041-dc21 2001046096
All rights reserved. Printed in the United States of America. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of the publisher. Statement [Tom By-Laws: The Society shall not be responsible for statements or opinions advanced in papers ... or printed in its publications (B7.1.3) INFORMATION CONTAINED IN THIS WORK HAS BEEN OBTAINED BY THE AMERICAN SOCIETY OF MECHANICAL ENGfl'I.'EERS FROM SOURCES BELIEVED TO BE RELIABLE. HOWEVER, NEITHER ASME NOR ITS AUTHORS OR EDITORS GUARA.NTEE THE ACCURACY OR COMPLETENESS OF ANY INFORMATION PUBLISHED IN THIS WORK. NEITHER ASME NOR ITS AUTHORS AND EDITORS SHALL BE RESPONSIBLE FOR ANY ERRORS, OMISSIONS, OR DAMAGES ARISING OUT OF THE USE OF THIS INFORMATION. THE WORK IS PUBLISHED WITH THE UNDERSTANDING THAT ASME AND ITS AUTHORS AND EDITORS ARE SUPPLYING INFORMATION BUT ARE NOT ATIEMPTING TO RENDER ENGINEERING OR OTHER PROFESSIONAL SERVICES. IF SUCH ENGINEERING OR PROFESSIONAL SERVICES ARE REQUIRED, THE ASSISTANCE OF AN APPROPRIATE PROFESSIONAL SHOULD BE SOUGHT. For authorization to photocopy material for internal or personal use under circumstances not falling within the fair use provisions of the Copyright Act, contact the Copyright Clearance Center (Ccq, 222 Rosewood Drive, Danvers, MA 01923, Tel: 978-750-8400, www.copylight.com.
To our children, Katherine, David, Susan, Nancy, and Thomas Jennifer and Mark
Cover Photo Courtesy of Nooter Corp.
PREFACE TO
SECOND EDITION
The ASME Boiler and Pressure Vessel Code, Section VIII, is a live and progressive document. It strives to contain the latest, safe and economical rules for the design and construction of pressure vessels, pressure vessel components, and heat exchangers. A major improvement was made withln the last year by changing the design margin on tensile strength from 4.0 to 3.5. This reduction in the margin permits an increase in the allowable stress for many materials with a resulting decrease in minimum required thickness. This was the first reduction in this design margin in 50 years and was based upon the many improvements in material properties, design methods, and inspection procedures during that time. Chapters and parts of chapters have been updated to incorporate the new allowable stresses and improvements which have been made in design methods since this book was originally issued. Some of these changes are extensive and some are minor. Some of the examples in this book have changed completely and some remain unchanged. This book continues to be an easy reference for the latest methods of problem solving in Section VIII. James R. Farr Wadsworth, Ohio Maan H. Jawad St. Louis, Missouri July 2001
CKNOWLEDGMENTS
We are indebted to many people and organizations for their help in preparing this book. Special thanks are given to the Nooter Corporation, fellow Committee Members; and to former coworkers for their generous support during the preparation of the manuscript We also give thanks to Messrs. Greg L. Hollinger and George B. Komara for helping with the manuscript, and to our editor Ray Ramonas at ASME for having great patience and providing valuable suggestions.
vii
CONTENTS
Preface .Acknowledgments List of Figures List of Tables ~.;~ .~;~ ""
'C.,
__
~.;; ;"
__ ''''''
" .. ". ,'.' ' .. ;.' '.. ::.:::; ".;" .:;',',',;' ' ;; ;;, ,_.,_."_"'_"'_""'_"".;,";";'; ;';;~'..' .'. ;'
" ""
__ .
V 'Vll
__ xiii . . xvii
Chapter 1 Background Information , , 1.1 Introduction "'" 1.2 Allowable Stresses 1.3 Joint Efficiency Factors 1.4 Brittle Fracture Considerations 1.5 Fatigue Requirements 1.6 Pressure Testing of Vessels and Components 1.6.1 ASME Code Requirements 1.6.2 What Does a Hydrostatic or Pneumatic Pressure Test Do? 1.6.3 Pressure Test Requirements for Vill-1 1.6.4 Pressure Test Requirements for VIlI-2 Chapter 2 Cylindrical Shells 2.1 Introduction 2.2 Tensile Forces, VIII-l 2.2.1 Thin Cylindrical Shells 2.2.2 Thick Cylindrical Shells 2.3 Axial Compression 2.4 External Pressure 2.4.1 External Pressure for Cylinders with Dolt 2.4.2 External Pressure for Cylinders with Dolt 2.4.3 Empirical Equations 2.4.4 Stiffening Rings , 2.4.5 Attachment of Stiffening Rings 2.5 Cylindrical Shell Equations, VIIl-2 2.6 Miscellaneous Shells 2.6.1 Mitered Cylinders 2.6.2 Elliptical Shells
"'"
"""
__
__.__ . __
. . . . . . .. . . .. ..
1
1
2 3 9 19 22 22 22 23 24
__
2:
<
10 10
'''''''''''' __
.. .. .. . .. . .. . .. . . . . . . .
27 27 27 27
33 36 42 43
46
47 48 50 53 54 54 55
Chapter 3 Spherical Shells, Heads, and Transition Sections , 3.1 Introduction 3.2 Spherical Shells and Hemispherical Heads, VIII-I 3.2.1 Internal Pressure in Spherical Shells and Pressure on Concave Side of Hemispherical Heads 3.2.2 External Pressure in Spherical Shells and Pressure on Convex Side of Hemispherical Heads 3.3 Spherical Shells and Hemispherical Heads, VIlI-2 , ix
.. . .. .. . ..
57 57 57
57
61
64
x 3.4
Contents Ellipsoidal Heads, VIII -I 3.4.1 Pressure on the Concave Side 3.4.2 Pressure on the Convex Side Torispherical Heads, VIII-I 3.5.1 Pressure on the Concave Side 3.5.2 Pressure on the Convex Side Ellipsoidal and Torispherical Heads, VIII-2 Conical Sections, VIII-I 3.7.1 Internal Pressure 3.7.2 External Pressure Conical Sections, VIII-2 .. . .. . .. . .. .. .. . . 65 65
3.5
67 68 68 71
72 74 74 85 95
3.6 3.7
3.8
'"Chapter 4" Flat Plates, Covers, and Flanges

4.1 4.2 Introduction Integral Flat Plates and Covers 4.2.1 Circular Flat Plates and Covers 4.2.2 Noncircular Flat Plates and Covers Bolted Flat Plates, Covers, and Flanges 4.3.1 Gasket Requirements, Bait Sizing, and Bolt Loadings Flat Plates and Covers With Bolting 4.4.1 Blind Flanges & Circular Flat Plates and Covers 4.4.2 Noncircular Flat Plates and Covers Openings in Flat Plates and Covers 4.5.1 Opening Diameter Does Not Exceed Half the Plate Diameter 4.5.2 Opening Diameter Exceeds Half the Plate Diameter Bolted Flange Connections With Ring Type Gaskets 4.6.1 Standard Flanges 4.6.2 Special Flanges Spherically Dished Covers '" 4.7.1 Definitions and Terminology 4.7.2 Types of Dished Covers
4.3 4.4
4.5
4.6
4.7
. . . . .. .. .. .. .. .. . . . .. . . .. .. ..
101
101 101
101 104 105 105 106
106
107
107
107
108
108
109
118 124 125 125
Chapter 5
Openings 5.1 Introduction 5.2 Code Bases for Acceptability of Opening 5.3 Terms and Definitions 5.4 Reinforced Openings-General Requirements 5.4.1 Replacement Area 5.4.2 Reinforcement Limits 5.5 Reinforced Opening Rules, VlII-l 5.5.1 Openings With Inherent Compensation 5.5.2 Shape and Size of Openings 5.5.3 Area of Reinforcement Required 5.5.4 Limits of Reinforcement 5.5.5 Area of Reinforcement Available 5.5.6 Openings Exceeding Size Limits of Section 5.5.2.2 5.6 Reinforced Opening Rules, VIII-2 5.6.1 Definitions 5.6.2 Openings Not Requiring Reinforcement Calculations 5.6.3 Shape and Size of Openings 5.6.4 Area of Reinforcement Required 5.6.5 Limits of Reinforcement 5.6.6 Available Reinforcement . . .. . .. .. .. . . .. . .. .. . . . .. . .. .. .. 133 133 133 134 134 134 134 136 136
137
137 140 140
151
153
153
153 155 155 155 157
Contents 5.6.7 Strength of Reinforcement Metal 5.6.8 Alternative Rules for Nozzle Design Ligament Efficiency Rules, VIII-I 6
xi 157 157 164
5.7 Chapter 6.1 6.2
Special Components, Vlfl-I

Introduction Braced and Stayed Construction 6.2.1 Braced and Stayed Surfaces ., .,.......................................... 6.2.2 Stays and Staybolts Jacketed Vessels 6.3.1 Types of Jacketed Vessels .,.. "6.3.2 Design of Closure Member for Jacketto Vessel 6.3.3 Design of Openings in Jacketed Vessels .,... Half-Pipe Jackets 6.4.1 Maximum Allowable Internal Pressure in Half-Pipe Jacket .,.................... 6.4.2 Minimum Thickness of Half-Pipe Jacket .,........................................................................... Vessels of Noncircular Cross Section 6.5.1 Types of Vessels .... .,....................................................................................................................... 6.5.2 Basis for Allowable Stresses 6.5.3 Openings in Vessels of N oncircular Cross Section 6.5.4 Vessels of Rectangular Cross Section
6.3
6.4
6.5
169 169 169 169 172 173 174 175 179 181 181 182 186 187 187 187 197
Chapter 7 Design of Heat Exchangers '" ,.... 7.1 Introduction ,........ 7.2 Tubesheet Design in U- Tube Exchangers 7.2.1 Nomenclature 7.2.2 Design Equations for Simply Supported Tubesheets 7.2.3 Design Equations for Integral Construction 7.2.4 Design Equations for Integral Construction With Tubesheet Extended as a Flange 7.3 Fixed Tubesheets ,.............................................................. 7.3.1 Nomenclature " ,', , ,.............. 7.3.2 Design Equations 7.4 Expansion Joints .. " , , _ , "................ Chapter 8 Analysis of Components in VIII·2 8.1 Introduction , 8.2 Stress Categories , 8.3 Stress Concentration 8.4 Combinations of Stresses 8.5 Fatigue Evaluation , , References ,
201 201 201 201 207 209 212 213 213 217 230
,..........
,", " , " ,'" , , " ,.... ,."....................... ,................................................ ,......................................... ,., ,..... ,
233 233 233 239 240 245 249
Appendices Appendix A"-Guide to VIII-l Requirements Appendix B-Material Designation Appendix Cs--Tcint Efficiency Factors ,' Appendix D-Flange Calculation Sheets Appendix E-Conversion Factors
. , "., ,", ,." , ,................................................... ,.................... , ,',........... ,.,.,.................................
251 253 255 277 283 285
Index
"
"..............................
LIST OF FIGURES
Figure Number 1.1 1.2 El.1 1.3 El.2 1.4 1.5 1.6 1.7
2,1
Welded Joint Categories (ASME VIII~l) Category C Weld Some Governing Thickness Details Used for Toughness (ASME Vlll-I)
.. .. .
5
7
8
11 14
Impact-Test Exemption Curves (ASME VIII·1) . Charpy Impact-Test Requirements for Full Size Specimens for Carbon and Low Alloy Steels With Tensile Strength of Less Than 95 ksi (ASME VIII-I) ; . Reduction of MDMT Without Impact Testing (ASME VIII-I) . Fatigue Curves for Carbon, Low Alloy, Series 4XX, High Alloy Steels, and High Tensile Steels for Temperatures Not Exceeding 700"F (ASME VllI-2) .. Comparison of Equations for Hoop Stress in Cylindrical Shells Chart for Carbon and Low Alloy Steels With Yield Stress of 30 ksi and Over, and Types 405 & ..
15
16 17 20 28 29 30 38 39
2.2 2.3 2.4 2.5 E2.8 2.6 2.7 E2.13 2.8 2.9 2.10
3.1
410 Stainless Steels

C Factor as a Function of R!T (Jawad, 1994) Geometric Chart for Cylindrical Vessels Under External Pressure (Jawad and Farr, 1989) Some Lines of Support of Cylindrical Shells Under External Pressure (ASME VIII-I) Some Details for Attaching Stiffener Rings (AS ME VIII-I) Mitered Bend Elliptical Cy linder
.. .. .. .. . . .
40 43 45 49
55 56 59
5]
E3.4 3.2 3.3 3.4
62 66
69
70 73
3.5
3.6 3.7 E3.11 E3.12 E3.13 3.8 3.9 3.10 3.11 3.12 4.1 4.2
75
78
•• ~ •• ~ ••••............. , •••••••••••••••••••••••••••••••••••••••• h ••••• • '"
* •••••••••••••••••••••••••••••••••••.••••.
79
83
92
Inherent Reinforcement for Large End of Cone-to-Cylinder Junction (ASME VllI-2) Values of Q for Large End of Cone-to-Cylinder Junction (ASME VIlI-2) Inherent Reinforcement for Small End of Cone-to-Cylinder Junction (ASME VIII~2) Values of Q for Small End of Cone-to-Cylinder Junction (ASME Vill-2) Some Acceptable Types of Unstayed Flat Heads and Covers Multiple Openings in the Rim of a Flat Head or Cover With a Large Central Opening xiii . .. .. . .. . 95 96 97 98 99
103
109
xiv
E4.5 E4.6 E4.7
List of Figures Ring Flange Sample Calculation Sheet Welding Neck Flange Sample Calculation Sheet Reverse Welding Neck Flange Sample Calculation Sheet Spherically Dished Covers With Bolting Flanges (ASME VIII-I) Example Problem of Spherically Dished Cover, Div. 1 Reinforcement Limits Parallel to Shell Surface Chart for Determ.ining Value of F for Angle Determination of Special Limits for Setting t, for Use in Reinforcement Calculations Example Problem of Nozzle Reinforcement in Ellipsoidal Head, Div. 1 Example Problem of Nozzle Reinforcement of 12 in. X 16 in. Manway Opening, Div. 1 Example Problem of Nozzle Reinforcement of Hillside Nozzle, Div. 1 Example Problem of Nozzle Reinforcement of Hillside Nozzle, Div. 1 ExamplePT6blern' of N6zzieReillforcernent With Corrosion Allowance, Div .1'.. ~ . . .. . .. .
110 115
119
4.3
E4.8 5.1 5.2 5.3 E5.1 ES.2 E5.3J E5.3.2
125
128 135 138
..
. .. .. .. . .
139 141 143 146 147

149 152 152 154 158
ES.4
5.4.1 5.4.2
5.5
5.6
ES.S
E5.6 E5.7 6.1 6.2
6.3
6.4 6.5 6.6 6.7 6.8 6.9 6.10 6.11 6.12 6.13 6.14 E6.8
7.1
7.2 7.3
7.4
7.S 7.6
7.7 7.8
7.9
7.10 7.11
7.12 ES.! 8.1 ESA 8.2
Nozzle Nomenclature and Dimensions (Depicts General Configurations Only) Limits of Reinforcing Zone for Alternative Nozzle Design Example Problem of Nozzle Reinforcement in Ellipsoidal Head, Div. 2 Example Problem of Nozzle Reinforcement of 12 in. X 16 in. Manway Opening, Div. 2 Example Problem of Nozzle Reinforcement of Series of Openings, Div, 1 Typical Forms of Welded Staybolts Typical Welded Stay for Jacketed Vessel _._ Some Acceptable Types of Jacketed Vessels _ Some Acceptable Types of Closure Details Some Acceptable Types of Penetration Details Spiral Jackets, Half-Pipe and Other Shapes _ Factor K for NPS 2 Pipe Jacket ., __ , _., _ ,, Factor K for NPS 3 Pipe Jacket Factor K for NPS 4 Pipe Jacket Vessels of Rectangular Cross Section , , Vessels of Rectangular Cross Section With Stay Plates , Vessels of Obround Cross Section With and Without Stay Plates and Vessels of Circular Cross Section With a Stay Plate , ,,, Plate With Constant-Diameter Openings of Same or Different Diameters Plate With Multidiameter Openings Example Problem of Noncircular Vessel, Div. 1 , Various Heat-Exchanger Configurations (TEMA, 1999) _ Some Typical Tubesheet Details for U-Tubes (ASME, 2001) Tubcsheet Geometry Effective Poisson's Ratio and Modulus of Elasticity (ASME, 2001) Chart for Determining A (AS ME, 2001) Fixity Factor, F (ASME, 2001) Some Typical Details for Fixed Tubesheet Heat Exchangers (ASME, 1995) Zd, Z" and Z"" versus X" (ASME, 2001) _ Values of Q3 Between 0.0 and 0.8 Values of Q3 Between -0.8 and 0.0 , Bellows-Type Expansion Joints Flanged and Flued Expansion Joints Linearizing Stress Distribution Model of a Finite Element Layout in a Flat Head-to-Shell Junction Fatigue Curves for Carbon, Low AHoy, 4XX High Alloy, and High Strength Steels for Temperatures Not Exceeding 700 P (AS ME VIII-2) Cyclic Curves _
0
.. .. . .. . . .. . . ,. .. .. .. _ .. .. . .. . . .. .. .. . _ ..
.
IS9 162
165
171 171 174

176 180 182 183
184
18S 188
190 191 192 193 198

202 203
205
206 208
.. _ .. . .. .. .. .. .. .
209 214 221

222
223
231 231
238
240 243 246
8.3 A.1
247 252
List of Figures C.l C.2 C.3 C.4 C.5 C.6 C.7 C.S C.9 C.lO C.ll C.12
xv
255 256 257 258 259 260 261 262 263 264
265
•••••••••••••••••••••••••••••••• T •••••••••••••• H ••••••••••••••••••••••••• , •••••••••••••• • • -,' ~ ., .; .; -.--.-; • -,' ~ ~ ',' ~ -.--,-; -,_: .: .-~ .• -~~.'.'; ~.; ; ~ ~ ~ ~ ~-~ -., ;;"; ~ ..... '~; ~~.: ~••• '~';' •• ~.'.'~ ~:~,~; ~.~ •• ~ ... ~,;· ..
c.n
i·. ,'..
C.14 C.I5 C.16 C.l7 C.18 C.19 C.20.E
266 267 268 269 270 271

272
D.l
D.2 D.3 D.4
273 274
Fig. Fig. Fig. Fig. Fig. Fig. D.l-Ring Flange With Ring-Type Gasket D.2-Slip-On or Lap-Joint Flange With Ring-Type Gasket D.3-Welding Neck Flange With Ring-Type Gasket D.4-Reverse Welding Neck Flange With Ring-Type Gasket D.S-Slip-On Flange With Full-Face Gasket D.6-Welding Neck Flange With Full-Face Gasket . . . . . . 277
D.5
D.6
278 279 280 281

282
LIST OF
Table Number 1.1 1.2 1.3 1.4 1.5 ELl 1.6 1.7 2.1 3.1 3.2 3.3 3.4 E3.14 6.1 6.2 6.3 8.1 8.2 8.3 8.4 8.5 8.6 E8A B.1 B.2 B.3 B.4 B.5
ABLES
Criteria for Establishing Allowable Stress Values for VIII-l (ASME II-D) Criteria for Establishing Design Stress Intensity Values for VITI-2 (ASME II-D) Stress Values for SA-5I5 and SA-516 Materials Allowable Stress Values for Welded Connections " Maximum Allowable Efficiencies for Arc- and Gas-Welded Joints " """ Stress Categories " Assignment of M.aterials to Curves (ASME VIII-I) Minimum Design Metal Temperatures in High Alloy Steels Without Impact Testing Tabular Values for Fig. 2.4 Factor Ko for an Ellipsoidal Head With Pressure on the Convex Side Values of b. for Junctions at the Large Cylinder Due to Internal Pressure Values of b. for Junctions at the Smail Cylinder Due to Internal Pressure Values of t:. for Junctions at the Large Cylinder Due to External Pressure Allowable Stress and Pressure Data Example of Pressure Used for Design of Components Closure Detail Requirements for Various Types of Jacket Closures Penetration Detail Requirements ." Primary Stress Category , , Structural Discontinuity , Thermal Stress , Stress Categories and Their Limits (AS ME VlII,2) Classification of Stresses (ASME VIII-2) Some Stress Concentration Factors Used in Fatigue Summary of Finite Element Output Carbon Steel Plate Chrome-Moly Steel Plate Specifications, SA-387 Chrome-Moly Steel Forging Specifications, SA-182 Chrome-Moly Steel Forging Specifications, SA-336 Quench & Tempered Carbon and AHoy Steel Forgings, SA-508
. . . . ,,.. . . .
.
3 3 4
5 6 9
10 18 37 67 76 77 85 96 173 179 181 234 234 235 236 237 239 244 253 253 253 254 254
"
. . . .. .. . .. .. . . . .. . .. . .. . . .. ..
xvii
CHAPTER
1
BACKGROUND. INFORMATION ..
1.1 INTRODUCTION
In this chapter some general concepts and criteria pertaining to Section VIII are discussed. These include allowable stress, factors of safety, joint efficiency factors, brittle fracture, fatigue, and pressure testing. Detailed design and analysis rules for individual components are discussed in subsequent chapters. Since frequent reference will be made to ASME Section VIII Divisions 1 and 2, the following designation will be used [TOm here on to facilitate such references. ASME Section VIII, Division 1 Code will be designated by VIII-I. Similarly, VIII-2 will designate the ASME Section VIII, Division 2 Code. Other ASME code sections such as Section II Part D will be referred to as II-D. Equations and paragraphs referenced in each of these divisions will be called out as they appear in their respective Code Divisions. Many design rules in Vlfl-I and VIII-2 are identical. These include flange design and external pressure requirements. In such cases, the rules of VIII-l will be discussed with a statement indicating that the rules of VIIl-2 are the same. Appendix A at the end of this book lists the paragraph numbers in VIII-l that pertain to various components of pressure vessels. Section VIII requires the fabricator of the equipment to be responsible for its design. Paragraphs UG22 in VIII-l and AD-110 in VIII-2 are given to assist the designer in considering the most commonly encountered loads. They include pressure, wind forces, equipment loads, and thermal considerations. When the designer takes exceptions to these loads either because they are not applicable or they are unknown, then such exceptions must be stated in the calculations, Similarly, any additional loading conditions considered by the designer that are not mentioned in the Code must be documented in the design calculations. Paragraphs U-2(a) and U-2(b) of VIII-l give guidance for some design requirement'>. VIII-2, paragraph AD-1l0 and the User's Design Specifications mentioned in AG-30I provide the loading conditions to be used by the manufacturer. Many design rules in VIII-l and VIII-2 are included in the Appendices of these codes. These rules are for specific products or configurations. Rules that have been substantiated by experience and used by industry over a long period of time are in the Mandatory Appendices. New rules or rules that have limited applications are placed in the Non-Mandatory Appendices. Non-Mandatory rules may eventually be transferred to the Mandatory section of the Code after a period of use and verification of their safety and practicality. However, guidance-type appendices will remain in the Non-Mandatory section of the Code. The rules in VIII-l do not cover all applications and configurations. When rules are not available, Paragraphs U-2(d), V-2(g), and UG-lOl must be used. Paragraph V-2(g) permits the engineer to design components in the absence of rules in VIII-I. Paragraph UG-lOl is for allowing proof testing to establish maximum allowable working pressure fo~ components, In VIII-2 there are no rules similar to those in UG-101, since VIII-2 permits design by analysis as part of its requirements. This is detailed in Paragraphs AD-IOO(b), AD·140, AD-ISO, and AD-160 of VIII-2.
1
2 Chapter 1
1.2 ALLOWABLE STRESSES

The criteria for establishing allowable stress in VIII-l are detailed in Appendix P of VIII-l and Appendix 1 of IJ-D and are summarized in Table 1.1. The allowable stress at design temperature for most materials is the lessor of 1/3.5 the minimum effective tensile strength or 2/3 the minimum yield stress of the material for temperatures below the creep and rupture values. The controlling allowable stress for most bolts is 1/5 the tensile strength. The minimum effective tensile stress at elevated temperatures is obtained from the actual tensile stress curve with some adjustments. The tensile stress value obtained from the actual curve at a given temperature is multiplied by the lessor of 1.0 or the ratio of the minimum tensile stress at room temperature obtained from ASTM Specification for the given material to the actual tensile stress at room temperature obtained from the tensile strength curve. This quantity is then multiplied by the factor 1.1. The .. effective .tensile stress is then equal to the lessor of this quantity or the minimum tensile stress at .room temperature given in ASTM. This procedure is illustrated in example 4.1 of Jawad and Farr (reference] 4, found at back of book). The 1.1 factor discussed above is a constant established by the ASME Code Committee. It is based on engineering judgment that takes into consideration many factors. Some of these include increase in tensile strength for most carbon and low alloy steels between room and elevated temperature; the desire to maintain a constant allowable stress level between room temperature and 50QoF or higher for carbon steels; and the adjustment of minimum strength data to average data. Above approximately 500 P or higher the allowable stress for carbon steels is controlled by creep-rupture rather than tensile-yield criteria. Some materials may not exhibit such an increase in tensile stress, but the criterion for 1.] is still applicable to practically all materials in VIII-I. Table 1.1 also gives additional criteria for creep and rupture at elevated temperatures. The criteria are based on creep at a specified strain and rupture at 100,000 hours. The 100,000 hours criterion for rupture corresponds to about eleven years of continual use. However, VITI-l does not limit the operating life of the equipment to any specific number of hours. The allowable stress criteria in VIII-2 are given in II-D of the ASME Code. The allowable stress at the design temperature for most materials is the smaller of 113 the tensile strength or 2/3 the yield stress. The design temperature for all materials in VIII-2 is kept below the creep and rupture values. Table 1.2 summarizes the allowable stress criteria in VIII-2. A sample of the allowable stress Tables listed in Section II -D of the ASME Code is shown in Table 1.3. It lists the chemical composition of the material, its product form, specification number, grade, Unified Numbering System (UNS), size, and temper. This information, with very few exceptions, is identical to that given in ASTM for the materiaL The Table also lists the P and Group numbers of the material. The P numbers are used to cross reference the material to corresponding welding processes and procedures listed in Section IX, "Welding and Brazing Qualifications," of the ASME Code. The Table also lists the minimum yield and tensile strengths of the material at room temperature, maximum applicable temperature limit, External Pressure Chart reference, any applicable notes, and the stress values at various temperatures. The designer may interpolate between listed stress values, but is not permitted to extrapolate beyond the published values. Stress values for components in shear and bearing are given in various parts of VIII-I, VUI-2, as well as II-D. Paragraph UW-IS of VIII-l and AD-132 ofVIII-2lists the majority of these values. A summary of the allowable stress values for connections is shown in Table 1.4. Some material designations in ASTM as well as the ASME Code have been changed in the last 20 years. The change is necessitated by the introduction of subclasses of the same material or improved properties. Appendix B shows a cross reference between older and newer designations of some common materials. The maximum design temperatures allowed in VIII cannot exceed those published in Section II-D. VIJI-J defines design temperature as the mean temperature through the cross section of a component. VIII-2 defines design temperature as the mean temperature in the cross section of a component, but the surface temperature cannot exceed the highest temperature listed in II-D for the material. This difference in the definition of temperature in VIIl-l and VIII-2 can be substantial in thick cross sections subjected to elevated temperatures.
0
Background Information TABLE 1.1 ALLOWABLE STRESS VALUES FOR VIII-1 (ASME II-D)
Room Temperature and Above Tensile Strength Yi.ld Strength
CRITERIA FOR ESTABLISHING

Below Room
T esnperature Tensee
PtoductlMat-edal
Wrought or cast
Stre-ngth
Yi.lrl Strefl9th
2/J
Str ess
Rupture F._~~Sii;W'iI 0.$5."",
I
i
C,...p Rate 1.0S~
hrr01JS and
MnfI1'HOU:i-
!.L
>.5
S,.
.0.
3.5
!:!. SrR,
3.5
~/_" s,
l/J
S,.Ry Or O.9S,.Ry £NQt~ un
Welded pipe or lube, ferrous and ncnterrcus
~Sr 3.5
Va x 0.8551'
~Sr 3.5
~SrRr
~!)x G.SSSy 3.5
'/~ )( D.SS.)!"R!" or 0.9 x O,S5SrRy

[NaM l1}J
(Fa""J:.e: O.85)SR"t~o;
(0.8
O.85)Sll'mt.r.
0.855,
NOTE: (1) Tw()sets of aHbwable stress values may be provided In Table lA for austenitic ..materials .ano.inJ),ble 1B for specific nonferrous alloys .. The lower values are not specifically identified by a footnote. These lower values do not exceed two-thirds of the minimum yield strength at temperature. The higher alternative allowable stresses are identified by a footnote. These higher stresses may exceed two-thirds but do not exceed 90% of the minimum yield strength at temperature. The higher values should be used only where slightly higher deionnation is not in itself objectionable. These higher stresses are not recommended for the design of flanges or for other strain sensitive applications.
Nomenclature RT = ratio of the average temperature dependent trend curve value oftensiie strength to the room temperature tensile strength Ry = ratio of the average temperature dependent trend curve value of yield strength to the room temperature yield strength SR.VQ = average stress to cause rupture at the end of 100,000 hr S~mi" = minimum stress to cause rupture at the end of 100,000 hr Sc = average stress to produce a creep rate of 0.01%/1000 hr ST = specified minimum tensile strength at room temperature, ksi Sy = specified minimum yield strength at room temperature
CRITERIA FOR ESTABLISHING

ProducVMaterial Wrought or cast, ferrous and nonferrous
TABLE 1.2 DESIGN STRESS INTENSITY VALUES FOR VIII-2 (ASME II-D)
Tensile Strenlith Yield Strength
I;, s,
0.85 3
l:l s.s,
J
% s,
--5,..
1.5
0.85
S,Ryor
O.9SrRy(Note (I))
Welded pipe
or
tube, ferrous
and nonferrous
s,
n.;
x 0.B5j S R
rr
0.85 SrRror
1.5 (0.9 " O.85)SyRy [Note (1)]
NOTE: (ll Two sets of allowable stress values may be provided In Table lA for austenitic materials and in Table 18 The lower values are not specifically identified by a footnote. These lower values do not exceed two-thirds at temperature. The higher alternative allowab!e stresses are identified bY a footnote. These higher stresses not exceed 90% of the minimum yield strength at temperature. The higher values should be used only where not in itself cbjectionable. These higher stresses are not recommended for the design of flanges or for other
for specific nonferrous alloys. of the minimum yield strength may exceed two-thirds but do slightly higher deformation is strain sensitive applications.
1.3 JOINT E}"'FICIENCY FACTORS

In the AS ME Boiler Code, Section 1, as well as in VIII-2, all major longitudinal and circumferential butt joints must be examined by full radiography, with few exceptions. VIII-I, on the other hand, permits various levels of examination of these major joints. The examination varies from fun radiographic to visual, depending on various factors specified in VIII-l and by the user. The degree of examination influences the required thickness through the use of Joint Efficiency Factors, E. The Joint Efficiency Factors, which are sometimes referred to as Quality Factors or weld efficiencies, serve as stress multipliers applied to vessel components when some of the joints are not fully radiographed. These multipliers result in an increase in the factor of safety as well as the thickness of these components. In essence, VIII-l vessels have variable
4 Chapter 1 TABLE 1.3 STRESS VALUES FOR SA 515 AND SA-516 MATERIALS
n
Alloy Line No. 28 29 Nominal Composition CS CS Product Form Plate Plate Spec No. SA-SiS SA·516 Type/Grade 70 70 DesigJ UNS No. K03101 K02700
Class/ Cond.! Temper Size/ Thick, in. P·No Group No. 2 2
Min. Tensile Line No. 28 29 Strength ksl 70 70
Min. Yield Stress ksi 38 38 1000 850
Applic. & max. Temp. Limits (NP = Not Permitted) (SPT = Supports only)
External Pressure Chart
III
700 700
vm-:
1000 1000
No. CS-2 CS-2
Notes G10, Gl0,
sr,
sr,
T2 T2
Maximum Allowable Line No. 28 29 -20 to 100 20.0 20.0 150 20.0 20.0 200 20.0 20.0 300 20.0 20.0
Stress, ksi, for Metal Temperature, 400 20.0 20.0 500 20.0 20.0 600 19.4 19.4 information 650 18.8 18.8
OF, Not Exceeding 700 18.1 18.1 750 14.8 14.8 800 12.0 12.0 850 9.3 9.3 900 6.7 6.7
Note: Gl0, 81, T2 are described
in 11-0 and pertain to metallurgical
factors of safety, depending on the degree of radiographic examination of the main vessel joints. As an example, fully radiographed longitudinal butt-welded joints in cylindrical shells have a Joint Efficiency Factor, E, of 1.0. This factor corresponds to a safety factor of 3.5 in the parent materiaL Nonradiographed longitudinal butt -welded joints have an E value of 0.70. This reduction in Joint Efficiency Factor corresponds to a factor of safety of 5.0 in the plates. This higher factor of safety due to a nonradiographed joint results in a 43% increase in the required thickness over that of a fully radiographed joint. ASME VIII-I identifies four joint categories that require E factors. They are Categories A, B, C, and D as shown in Fig. 1.1. Category A joints consist mainly of longitudinal joints as well as circumferential joints between hemispherical heads and shells. Category B joints are the circumferential joints between various components as shown in Fig. 1.1, with the exception of circumferential joints between hemispherical heads and shells. The attachment of flanges to shells or heads is a Category C joint. The attachment of nozzle necks to heads, shells, and transition sections is categorized as a Category D joint. The four joint categories in VIII-l do not apply to items such as jacket closure bars, tubesheet attachments, and ring girders. The degree of examination of the welds attaching these components to the shell or head is not covered in VIII -1. Most designers assign an E value of 1.0 when calculating the shell or head thickness at such junctions. This is justified since in most cases the strain in the hoop direction, and hence hoop stress, is close to zero at the junction due to the restraint of tubesheet or bars.
Background TABLE 1.4 STRESS VALUES FOR WELDED CONNECTIONS VIIH Component
Fillet weld Fillet weld Groove weld Groove weld Nozzle neck Dowel bolts Any location
Information
ALLOWABLE
Type of Stress
tension shear tension shear shear shear bearing stress for Vlflc1' construction'
Stress Value
0.55S· 0.49S 0.748 0.608 0.70S O.BOS 1.60S
Reference
UW-18(d) UW-15(c} UW-15(c) UW-15(c) UG-45(c)
JI-D
II-D
'*s '~"allowable
Component
Fille! weld Fillet weld Groove weld Groove weld Nozzle neck Any location
VIII-2 Type of Stress

tension shear tension shear shear bearing
Stress Value
0.5 O,SSm O.75Sm O.7S8m 0.8Sm
e:
Reference
AD-920 AD-920 AD-920 AO-920 AD-132.2 AD-132.1
s,
'Sm = stress intensity values for VIII-2 construction
FIG. 1.1 WELDED JOINT CATEGORIES
(ASME VIII-i)
The type of construction and joint efficiency associated with each of joints A, B, C, and D is given in Table 1.5. The categories refer to a location within a vessel rather than detail of construction. Thus, a Category C weld, which identifies the attachment of a flange to a shell, can be either fillet, comer, or butt welded, as illustrated in Fig. 1.2. The Joint Efficiency Factors apply only to the butt-welded joint in sketch (c). The factors do not apply to sketches (a) and (b) since they are not butt welded. The Joint Efficiency Factors used to design a given component are dependent on the type of examination performed at the welds of the component. As an example, the Joint Efficiency Factor in a fully radiographed longitudinal seam of a shell course is E = .1.0. However, this number may have to be reduced, depending on the degree of examination of the circumferential welds at either end of the longitudinal seam. Appendix B shows some typical components and their corresponding Joint Efficiency Factors.
Chapter]
MAXIMUM ALLOWABLE
TABLE 1.5 JOINT EFFICIENCIES1,5 FOR ARC~ AND GAS-WELDED
JOINTS
Degree of Radiographic Examination Type No.

(1)
Joint Description
Butt joints as attai ned by double-welding or by other means which will obtain the same quality of deposited weld metal on the inside and outside weld slJ:riacestb agree with the requirements of UW-35. Welds usinq metal backing strips which remain in place are excluded. Single-welded butt joint with backing strip other than those included under (1)
Limitations
None
Joint Category
A, B, C & D
a
FulF
1.0
b spot'
0.85
c
None
0.70
(2)
(a) None except as shown in (b) below (b) Circumferential butt jOints with one plate offset, see UW-13(c) and Fig. UW13.1(k). Circumferential butt jolnts only, not over 5/8 in. thick and not over 24 in. outside diameter Longitudinal joi nts not over 3/8 in. thick Circumferential joints not over 5/8 in. thick
A, B, C & D A, B &C
0.90 0.90
0.80 0.80
0.65 0.65
(3)
Single-welded butt joint without use of backing strip Double full fillet lap joint
A, B& C
NA
NA
0.60
(4)
A 8 & C£ B
NA NA NA
NA NA NA
0.55 0.55 0.50
(5)
Single full fillet lap joints with plug welds conforming to UW-17
(a) Circumferential joints" for attachment of heads not ave r 24 in. outside diameter to shells not over 112 in. thick (b) Circumferential joints for the attachment to shells of jackets not over SIB in. in nominal thickness where the distance from the center 01 the plug weld to the edge of the plate is not less than 1-1/2 times the diameter of the hole for the plug.
NA
NA
0.50
(6)
Single ful! fillet lap joints without plug welds
(a) For the attachment of heads convex to pressure to shells not over 5/8 in. required thickness. Only with use of fillet weld on inside of shells, or (b) For attachment of heads having pressure on either side. To shells not over 24 in. inside diameter and not over 1/4 in. required thickness with fillet weld on outside of head flange only.
A&B
NA
NA
0.45
A&B
NA
NA
0.45
Notes: (1) The sing!e factor shown for each combination of joint category and degree of radiographic examination replaces both the stress reduction factor and the [otnt efficiency factor considerations previously used in this Division. (2) See UW-12(a) and UW-51. (3) See UW-12(b) and UW-52. (4) Joints attacninq hemispherical heads to shells are excluded. (5) E = 1.0 for butt joints in compression. (6) For Type No.4 Category C joint, limitation not applicable for bolted flange connections.
Background Information 7
rAI
,
I , I
... ,.
/~
.
I
""01111""
I
J
· , · ·
I I I
.........
(b)
, I
..
FIG. 1.2
.
I I I
CATEGORY C WELD
8 Cha ter 1 Example 1.1 Problem Determine the category and Joint Efficiency Factor of the joints in the beat exchanger shown in Fig. El.l. The channel side is spot radiographed. The longitudinal seam, b, is a single-welded butt joint with a backup
COVER
L
CHANNEL
e
PASS: PARTITION .
... i
···d_et
t,
b
~
u
C:~
SHELL
h------~~--------r~----~
TUBE
9 .. ~~~.u'_~~f
J r=~
JACKET
n.
..
HEAD
SKIRT
FIG. E1.1
Background Information
bar. The shell side is not radiographed. The longitudinal and circumferential seams m and I are singlewelded butt joints with backup bars. The jacket longitudinal seam, n, is a single-welded butt joint without backup bar. Solution The joint categories of the various joints can be tabulated as given in Table E1.1: TABLE E1.1 STRESS CATEGORIES Joint
(a) (b) (c) (d) (e) (f) (g) (h) (i) 0) (k) Longitudinal
Location
Ohannel-to-flanqeconneetton channel seam weld .
Category
A C
D 0.80
Joint Efficiency
Does not apply
Channel-to-tubesheet Nozzle-to-channel Flange-to-nozzle
Does not apply Does not apply Does not apply Does not apply. See also UW-15(c) and UW-18(d) of VIII-l Does not apply. See also UW-20 of VlIl-1 Does not apply Does not apply Does not apply Does not apply 0.65 0.65 0.60 Does not apply
weld neck weld
C
None None
Pass partition-to-tubesheet Tube-to-tubesheet Shell-to-tubesheet weld weld
C
weld weld None None
Jacket bar-to-inner-shell
Jacket to bar-to-outer-shell Nozzle-to-jacket Longitudinal Head-to-shell Longitudinal Skirt-to-head weld
0 A
B
(I)
(m) (n) (0)
shell seam seam jacket seam seam
A
None
1.4 BRITTLE FRACTURE CONSIDERATIONS

Both VIII-l and VIII-2 require the designer to consider brittle fracture rules as part of the material and design selection. The rules for carbon steels are extensive and are discussed first. VIII-l has two options regarding toughness requirements for carbon steels. The first is given in Paragraph UG-20(f) and allows the designer to exempt the material of construction from impact testing when all of the following criteria are met: 1. 2. 3. 4. 5. 6. The material is limited to P-No. 1, Gr. No. 1 or 2. Maximum thickness of 1/2 in. for materials listed in Curve A in Table 1.6. Maximum thickness of 1 in. for materials listed in Curves B, C, and D of Table 1.6. The completed vessel shall be hydrostatically tested per UG-99(b), (c), or 27-3. Design temperature is between - 20 P and 650oP. Thermal, mechanical shock, and cyclical loadings do not control the design.
0
The above requirements are intended for relatively thin carbon steel vessels operating in a service that is neither severe in thermal and pressure cycling nor in extreme cold temperatures. Vessels of low alloy
10
Cha ter 1
ASSIGNMENT
TABLE 1.6 OF MATERIALS TO CURVES (ASME VIII~1)
GENERAL NOTES ON ASSIGNME.NT OF MATE.RIALS TO CURVES: (a) Curve A applies to: (1) ail careen aod all low alloy steel plates, structural shapes, and bars not listed In Curves 8, C, and 0 below; (2) SA-21!> Grad., WeB and wec If normalized and tempered or water-quenched and tempered; SA-2l7 Grade wet, If normalized and tempered or water-quenched and tempered, [b) Curve 8 applies to: (1) SA-Zlt, Grade WCA If normalized and tempered or water-quenched and tempered SA·<l6 Grades WCB and WCC for thicknesses not exceeding 2 In., If produced to fine grain practice and water-quenched and tempered SA·217 Grade WC9 if normaUzed and tempered SA·285 Grades A and B SA-414 Grade A SA-515 Grade 60 SA·516 Grades 65 and 70 If not normalized SA.b12 If not normalized SA-662 Grade B If not normalized; (21 except for cast steels, all materials of Curve A if produced to fine grain practice and normalized which are not listed In Curves C and 0 below; (31 all pipe, fittings, forgings and tubing not Hsted for Curves C and 0 below; (4) parts permitted under UG-ll shall be induded in Curve B even when fabricated from plate that otherwise would be assigned \0 a different curve. (c) Curve C ill SA-182 Grades 21 and 22 if normalized and tempered SA-302 Grades C and [) SA-J36 f21 and f22 if norrnalired and tempered SA-387 Grades 21 and n If normalized and tempered SA-5lb Grades 55 and 60 If not normalized SA-533 Grade. 8 and C SA·b62 Grade A; (21 all material of Curve B If produced to fine grain practice and normalized and not listed for Curve 0 below.
(d) Curve D
SA·203 SA-50S Grane 1 SA-51t, if normalized SA·524 Classes 1 and 2 SA·537 Classes 1, 2, and 3 SA·612 if normalized SA-bb2 If normalized SA-73e Grade A SA'73a Grade A with Cb and V deliberarely added In accordance with the provislcns of the material specificaticn, not colder than -20'F {-29'C} SA·738 Grade B not colder than -20'F {-29'C} (e} For boltlng and nuts, the lollowlng Impact test exemption temperature shall apply: Bolting Spec. No. SA-193 SA-l?3 SA·l?3 SA-193 SA-307 SA-J20 SA-325 SA-354 SA-354 SA-449 SA-540 Grade B5 87 [2'1, in. dia. and under) ([Iver 2 'I, tn, to 7 In., inct.) 87M Blb Impact Test Exemption Temperature, OF -20 -55 -40 -55 -20 -20 Impad tested -20
L7, L7A, L7M, L43 1, 2 BC
BD
BZJ/~4 Nuts
+20 -20 +10
Spec. No. SA-194 SA-540

(I)
Grade 2, 2H, 2HM, 3, 4, 7, 7M, and 16 823IB24
Impact Test Exemption Temperature,
OF
-55 -55
When no class or grade is shown, all classes or grades are included, (g) The following shall apply to all material assignment notes. ll) Cooling rates faster than those obtained by cool iny i.n air, foUowed by tempering, as permitted by the material specification, are considered to be equlva lent to normalizing Dr normalizing and tempering heat treatments. (2) Fine grain practice is defined the procedure necessary to obtain a fine austenitic grain size as described in SA·20.
a.
NOTES; (1) Tabular values for this Figure are provided in Table UCS-66. (2) Castings not listed in General Notes (a) and (b) above shall be impact tested.
11
steel or those with carbon steel operating beyond the scope of Paragraph UG-20(f) require an evaluation for brittle fracture in accordance with the rules of UCS-66. The procedure consists of
1. Determining the governing thickness in accordance with Fig. 1.3. 2. Using Fig. 1.4 to obtain the temperature that exempts the material from impact testing. If the specified Minimum Design Metal Temperature, MDMT, is colder than that obtained from the figure, then impact testing in accordance with Fig. 1.5 is required. The specified MDMT is usually given by the user, whilethe calculated MDMT is obtained from VITI-I. The calculated MDMT is kept equal to or colder than the specified MDMT.
Section x-x
fg1 ~ fA
f(f2 = fA (seamless) or f8 (welded)
---<t,lal Butt Welded Components
q_
tg1 = the thinner
t(f2 = the thinner
fg3 = the thinner
Of!A or Ie
of fe or Ie
of fA or fe
NOTE: Using Igl' tg2, and fg3. determine the warmest MDMT and use that as the permissible MOMT for the welded assembly,
(bl Welded Connection with Reinforcement Plate Added
SOME GOVERNING THICKNESS
FIG. 1.3 DETAILS USeD FOR TOUGHNESS (ASME VIII-1)
12 Chapter 1
tgl =
4" (For ~
fA
t:\
tgt welded
tA = 4" (For
a-. welded
or nonweldedl
or nonweldedl
The governing thickness of
is the greater
of t01 or t02
Ie) Bolted flat Head or Tub8~heet and Flange
(dl Integral Flat Head or Tubesheet
tUl
="4
fA
(For
17\ .
0.;
welded
or nonweldedl t02 th inner of tA
or ta
The governing thickness

of
0
8
is the greater
of tgl or tg2 leI Flat Head or Tubesheet Corner Joint
With
FIG. 1.3 ,(CONT'D)
13
tg1 = thinner
of tA or tB Defined in UCs,s6(a}
{fl Welded Attachments
illS
FIG. 1.3 (CONT'O)
3. The temperature obtained from Fig. 1.4 may be reduced in accordance with Fig. 1.6 if the component operates at a reduced stress. This is detailed in Paragraph UCS-66(b) of VIII-I. At a ratio of 0.35 in Fig. 1.6, the permitted temperature reduction drops abruptly. At this ratio, the stress in a component is about 6000 psi. At this stress level, experience has shown that brittle fracture does not occur regardless of temperature level. 4. The rules in VIII-l also allow a 300P reduction in temperature below that obtained from Fig. 1.4 when the component is post-weld heat treated but is not otherwise required to be post-weld heat treated by VIII -1 rules. The toughness rules for ferritic steels with tensile properties enhanced by heat treatment are given in Paragraph UBT -6 of VIII-l. The rules require such steels to be impact tested regardless of temperature. The measured lateral expansion as defined by ASTM E-23 shall be above 0.015 in. The toughness rules for high alloy steels are given in Paragraph UBA-51 of VIII-I. The permissible Minimum Design Metal Temperature for base material is summarized in Table 1.7. Similar data are given in VUI-l for the weld material and weld qualifications. Thermally heated stainless steels may require impact testing per the requirements of UBA-51(c). The rules for toughness in VIlI-2 are different than those in VIII-I. However, the concepts of exemption curves and Charpy impact levels are similar in VIU-2 and VIII-l. The toughness requirements for carbon and low alloy steels are given in Paragraph AM-2I8 of VIII-2. High alloy steels are covered in Paragraph AM-213.
14 Chapter
Example 1.2 Problem Determine the Minimum Design Metal Temperature, MDMT, for the reactor shown in Fig. EL2. Let the shell, head, pad, and ring material be SA-516 Gr. 70 materiaL Flange and cover material is SA-105. Pipe material is SA-106. The required shell thickness is 1.75 in., and the required head thickness is 0.86 in. The required nozzle neck thickness is 0.08 in. Assume a joint efficiency of 1.0 and no corrosion allowance. Solution Shell .. SAc516specifications require thematerial to be normalizedwhenthe thickness exceeds 1.5 in. Thus, from Table 1.6, Curve D is to be used for normalized SA-516 Gr. 70 material. Using Fig. 1.4 and a governing thickness of 2.0 in., we get a minimum temperature of - 5"F. The ratio of required thickness to actual thickness is 1.75/2.0 = 0.88. Using Fig. 1.6 for this ratio, we obtain l2"F. Hence, MDMT = - 5 - 12 = -lrF. Head For a 1 in-thick head, SA-516 specifications permit a non-normalized material. Thus, from Table 1.6, Curve B is used. Using Fig. 1.4 and a governing thickness of 1.0 in., we get a minimum temperature of 30°F. The ratio of required thickness to actual thickness is 0.86/1.0 = 0.86. Using Fig. 1.6 for this ratio, we obtain 14°F. Hence, MDMT = 30 - 14 = 16°P.
3" X ~.
RING
4' RAD. 300" WN FLANGE SA 105

4' SCH. 40 PIPE
FIG. E1.2
15
140
I
120
I
~ I
100
I
I
U-
60
'f"'"
" e a. '" E
c
I+
60
.. .. s
r '"
OJ
'u;
40
I I ~-r
If
'" .,
E
/ V/
/
V····
A/
:..--
V
V
_/
k
~
!o"""
~
···1
...-
20
" s
'" .1'1
I I
I "1
1/ /
/ '/
./
V .,
P"
V .---V
..-
f--"
~~
----
- 20
- 40 - 55 -60
! V/ --(.~- -- -- -- ...
f
J
r--
1---
-- -- -- -- -4 6
tesiinll reQUlrr
-80
0.394
3 Nominal Thickness, in.
(Limited 10 4 in. for Welded construction}
FIG. 1.4
IMPACT-TEST EXEMPTION CURVES (ASME VIII-1)
Stiffener For a 0.75-in. stiffener, Curve B of Table 1.6 is to be used. Using Fig. 1.4 and a governing thickness of 0.75 in., we obtain a minimum temperature of lSoP. Since stresses cannot be established from VIII-l rules, the MDMT = lSOF.
Pad
The material will be normalized since it is 2.00 in. thick. Curve D of Table 1.6 is used. From Fig. 1.4 and a governing thickness of 2.0 in., we obtain a minimum temperature of - S°F. Since stresses cannot be established from VIIl-l rules, the MDMT = - SOF.
16 Chapter 1
0,394 in,
!I
I
I
50
I I
40
Minimum
. .. .
.'
specified ...
I I
I
! ... yi~ldm~r'lmh
65 ksi
" E 'u .,
£ '0
'"
"
" e
a30
I
I
01
~ .s
£
.;:: u
" E
I
20
15
I ....
::>
/. ...
_".,.
V
~
F'"
'/
~ ~
./
All'
V
~ ~
./
55 ksi
...
...
...
50 ksi
45 ksl
.. ,..
-,..
'" 38 ksi
L.
...
_... ~
10
!'""
I
I I
o
1,0 Maximum
2,0
Nominal Thickness of Material or Weld, in.
;;. 3.0
GENERAL NOTES; (a) Interpolation beIWeen yield strengths shown is permitted, (hi The minimum impact energy for one specimen shall not be jess than 2/3 of the average energy required for three Specimens, lei Materials produced and impact tested in accordance with SA·320, SA·333, SA-334, SA·35D, SA-3S2, SA-420 and SA·765 do not have to satisfy these energy values, They are acceptable for use at minimum design meta! lemperatyre not colder than the lest temperature when the energy values required by the applicable specification are satisfied. (dl For materials having a specified minimum tensile strength of 95 ksl or more, see UG-84(c)l4)(b).
FIG. 1.5 CHARPY IMPACT-TEST REQUIREMENTS FOR FULL SIZE SPECIMENS FOR CARBON AND lOW AllOY STEELS WITH, TENSILE STRENGTH OF LESS THAN 95 ksi (ASME VIII-1)
17
1.00
.g
Cl: Q.l
co
.~
>
«
(5 ....
II)
... ...
c
Q.l
0.80
'0
~ .... co
E
t: CD
...
0.60
z r.n
I
0
(!) (!)
-c
..::
* I.i.J ..:.;:;
0.40 0.35
a::
ro
0.20
O. DO
LL....'-L..."-'-4...L...a..L..<-i.........:-"-';...LL'-"-L..LJ4...L..""-L-.lL.L.""'-"~-"-':....c....'-"-"-L-4...L..L....A..<-i.....4-dL..L..<'-"-'''-'''-'-''-£l.L..4...L...tJ
20
40
60 80 of [See UCS-66(b)]
100
120
140
Nomenclature (Note references to General Notes of Fig. UC$-66.2.l tr = required thickness of the component under consideration in the corroded condition for all applicable loadings [General Note (2)), based on the applicable joint efficiency E [General Note (3)1, in. tn .. nominal thickness of the component under consideration before corrosion allowance is deducted, in. c := corrosion allowance, in. E* = as defined In General Note (3). Alternative Ratio = S* £* divided by the product of the maximum allowable stress value from Table UCS-23 times E, where S* is the applied general primary membrane tensile stress and £ and E* are as defined in General Note (3).
FIG. 1.6
REDUCTION OF MDMT WITHOUT IMPACT TESTING (ASME VIII-1)
Nozzle Neck From Table 1.6, Curve B is to be used for a nozzle neck of O.258-in. thickness. From Pig. lA, minimum temperature is - 20°F. The ratio of required thickness to actual thickness is 0.08/0.258 X 0.875 = 0.36. Using Fig. 1.6 and this ratio, we get l30op. Hence, MDMT = - 20 - -130 = - 150°F.
18
Chapter 1
TABLE 1.7 MINIMUM DESIGN METAL TEMPERATURES IN HIGH ALLOY STEELS WITHOUT IMPACT TESTING
Base Material
stainless steel
Austenitic chromiummanganese-nickel stainless steel (200 series)
.. .. js~ateriaJ304,304L,3i
6,
316L,321,347
Is material • Austenitic ferritic duplex steel with t < 3/8 in. • Ferritic chromium stainless steel with t< 1/8 in. • Martensitic chromium stainless steel with t « 1/4 in.
Flange Since the flange is ANSI B16.5, it is good to -20°P' Cover From Fig. 1.3(c), the controlling cover thickness is 2.5/4 = 0.625 in. Curve B applies for this material, and the MDMT = SOP. Therefore, the MDMT for this reactor is governed by the head with a value of 16°F. A colder value can be obtained by impact testing the various components, Thus, assuming a specified MDMT of -15°P is required, then the head, stiffener, pad, and cover need impact testing,
19
1.5 FATIGUE REQUIREMENTS

Presently, VlII-l does not list any rules for fatigue evaluation of components. When fatigue evaluation of a component is required in accordance with UG·22 or U-2(g) of VIII-I, the general practice is to use the VIJI-2 fatigue criteria as a guidance up to the temperature limits of VIII-2. At temperatures higher than those given in VIII-2, the rules of III-H are followed for general guidance. Other fatigue criteria, such as those given in other international codes and ASME B31.3, may also be considered as long as the requirements of U-2(g) of VIII-l are met. VIII-2 contains detailed rules regarding fatigue. Paragraph AD-160 gives criteria regarding the need for fatigue analysis. The first criterion is listed in Paragraph AD-160.1 and is based on experience. Vessels that have operated satisfactorily in a certain environment may be cited as the basis for constructing similar . vessels Operating under similar conditions without the need for fatigue analysis. . . The se~ond criterion for vessel components isbasecl em the rule that fatigue analysis is notrequiredif all of Condition A or all of Condition B is satisfied, as noted below.
Condition A Fatigue analysis is not required for materials with a tensile strength of less than 80 ksi when the total number of cycles in (a) through (d) below is less than 1000. a. The design number of full range pressure cycles including startup and shutdown. b. The number of pressure cycles in which the pressure fluctuation exceeds 20% of the design pressure. c. Number of changes in metal temperature between two adjacent points. These changes are multiplied by a factor obtained from the following chart in order to transform them to equivalent cycle number.
Metal Temperature 50 or 51 to 101 to 151 to 251 to 351 to Higher less 100 150 250 350 450 than 450 Differential,
OF
Factor
o
1 2 4 8 12
20
d. Number of temperature cycles in components that have two different materials where a difference in the value Cal - (2)i:.T exceeds 0.00034. Where, a is the coefficient of thermal expansion and sr is the difference in temperature. Condition B Fatigue analysis is not requited when the following items (a) through (j) are met:
20
Cha ter 1 a. The number of full range pressure cycles, including startup and shutdown, is less than the number of cycles determined from the appropriate fatigue chart, Fig. 1.7, with an S. value equal to 3 times the allowable design stress value, Sm. b. The range of pressure fluctuation cycles during operation does not exceed P(1/3)(Sal Sm), where P is the design pressure, Sa is the stress obtained from the fatigue curve for the number of significant pressure cycles, and Sm is the allowable stress. Significant pressure cycles are defined as those that exceed the quantity P(lf3)(S / Sm). S is defined as S = Sa taken at 106 cycles when the pressure cycles are -s 106• S = Sa taken at actual number of cycles when the pressure cycles are > 106• c. The temperature difference between adjacent points during startup and shutdown does not exceed Sal (2Eex), where Sa is the value obtained from the applicable design fatigue curve for the total specified number of startup and shutdown cycles. d. The temperature difference between adjacent points during operation does not exceed S./(2Ea), where Sa is the value obtained from the applicable design fatigue curve for the total number of significant fluctuations. Significant fluctuations is defined as those exceeding the quantity S/(2Ea), where S is as defined in (b) above. Adjacent points are defined in AD· 160.2, Condition A, Paragraph (c) of VIII-2. e. Range of significant temperature fluctuation in components that have materials with different coefficient of expansion or modulus of elasticity and that do not exceed the quantity Sal [2(E] a) - E2 a2)], where a is the coefficient of thermal expansion and E is the modulus of elasticity. Significant temperature fluctuation is that which exceeds the value S / [2(E) ex) - E2 0.2)], where S is as defined in (b) above. f. Range of mechanical loads does not result in stress intensities whose range exceeds the Sa value obtained from the fatigue chart.
107CC------------.-----
-.
-r
~------------~
NOTES: (1, E ~ 30 x 10' poL (2) Interpol.t. 10' UTE 80-1151<0., (3) Table 6-110. ~ contains tebujated values end a f.otmula for
811 eccurere
interpolation of
the.
CtlNSS.
106
'g_
~
;
~
106
10
. FIG. 1.7 FATIGUE CURVES FOR CARBON, LOW ALLOY, SERIES 4XX, HIGH ALLOY STEELS, AND HIGH TENSILE STEELS FOR TEMPERATURES NOT EXCEEDING 700 F (ASME VIII-2)
Q
21
The third criterion for nozzles with nonintegral reinforcement is given in Paragraph AD-160.3 of VIII-2 and is very similar to Conditions A and B detailed above.
Example Problem
1.3
A pressure vessel consisting of a shell and two hemispherical heads is constructed from SA 516-70 carbon steel material. The self-reinforced nozzles in the vessel are made from type SA 240-304 stainless steel material. The vessel is shut down six times a year for maintenance. At start-up, the full pressure of 300 psi, and full temperature of 400 F are reached in two hours. The maximum tlT between any two points during start-up is 250QP.At normal operation, the '.tlT· isnegligible ..At shutdown, the. maximum 11T is 100oP. Determine the maximum number of years that this vessel can be operated if a fatigue evaluation is not performed. Let the coefficient of expansion for carbon steel be 6.5 X 10-6 in.linPF and that for stainless steel be 9.5 X 10-6 in'/inPF.
D
Solution From Condition A, determine the number of cycles in one year. a. Number of full pressure cycles for one year is 6. b. This condition does not apply for this case. c. From the chart, the 250 P difference in temperature during start-up corresponds to 4 cycles. The lOO°F difference in temperature during shutdown corresponds to 1 cycle. Thus total equivalent cycles due to temperature in one year is (4 + 1) 6 = 30 cycles. d. At nozzle attachments, the quantity (9.5 X 10-6 ~ 6.5 X lW6) 400 is equal to 0.0012. Since this value is greater than 0.00034, the equivalent cycles per year = 6.
0
Total cycles per year due to (a), (c), and (d) "" 6 + 30 + 6 "" 42. Number of years to operate vessel if fatigue analysis is not performed
1000/42
23.8 years.
Example Problem
1.4
A pressure vessel has an inside diameter of 60 in., internal pressure of 300 psi, and design temperature of SOO°F. The shell thickness is 112 in. at an allowable stress level of 18,000 psi (material tensile stress "" 70 ksi). The thickness of the hemispherical heads is 114 in. at an allowable stress level of 18,000 psi. Integrally reinforced nozzles are welded to the shell and are also constructed of carbon steel with an allowable stress of 18,000 psi. At start-up, the full pressure of 300 psi and full temperature of SOoop are reached in eight hours. The maximum tlT between any two points during start-up is 60°F. At normal operation, the tlT is negligible. At shutdown, the maximum tlT is 50°P' Determine if the shell and heads are adequate for 100,000 cycles without the need for fatigue analysis. From II-D, the coefficient of expansion for carbon steel is 7.25 X 10-6 in.lin'!°F and the modulus of elasticity is 27.3 X 106 psi. Use Fig. 1.7 for a fatigue chart. Solution Condition B is to be used. a. Three times allowable stress at the nozzle location is 1.7 for this value gives a fatigue life of 4200 cycles. b. This condition does not apply.
= 3(18,000)
54,800 psi. Using Fig.
22
Chapter 1
c. From Fig. l.7, with 100,000 cycles, the value of Sa = 20,000 psi. The value of Sa/C2Eo.) = 20,000/(2 X 37.3 X ]06 X 7.25 X 10-6) = 51°F. Since this value is less than 60°F, the specified cycles are inadequate. The designer has two options in this situation. The first is to perform fatigue analysis, which is costly. The second option, if it is feasible, is to reduce the AT at startup to 51°F. d. This condition does not apply. e. This condition does not apply. f. This condition does not apply. Example 1.5 ·······Problem . In Example 1.4, determine the required thickness of the shell and heads for 1,000,000 cycles without the need to perform fatigue analysis. Solution From Fig. l.7, with a cycle life of 1,000,000, the value of Sa = 12,000 psi. From Condition B, subparagraph (a), the maximum stress value for the shell is (12,000/3) = 4,000 psi. The needed shell thickness = 0.5 X 18,000/4000 = 2.25 in. The required head thickness = 0.25 X 18,000/4000 = 1.13 in. The maximum AT at start-up or shutdown cannot exceed Sa/(2Eo.) = 12,000/(2 X 27.3 X 106 X 7.25 X 1O~ 6) = 31°F, otherwise a fatigue analysis is necessary.
1.6
PRESSURE TESTING OF VESSELS AND COMPONENTS
1.6.1 ASME Code Requirements

Pressure vessels that are designed and constructed to VIIl-l rules, except those tested in accordance with the requirements of UO-1OI, are required to pass either a hydrostatic test (UG-99) or a pneumatic test (UG100) of the completed vessel before the vessel is U-stamped. Pressure vessels that are designed and constructed to VIII-2 rules also are required to pass either a hydrostatic test (Article T·3) or a pneumatic test (Article T-4) before the U2-stamp is applied. Each component section of the ASME Boiler and Pressure Vessel Code has a pressure test requirement that calls for a pressure test at or above the maximum allowable working pressure indicated on the nameplate or stamping and in the Manufacturer's Data Report before the appropriate Code stamp mark may be applied. Under certain conditions, a pneumatic test may be combined with or substituted for a hydrostatic test. When testing conditions require a combination of a pneumatic test with a hydrostatic test, the requirements for the pneumatic test shall be followed. In all cases, the term hydrostatic refers not only to water being an acceptable test medium, but also to oil and other fluids that are not dangerous or flammable; likewise, pneumatic refers not only to air, but also to other nondangerous gases that may be desirable for "sniffer" detection.
1.6.2 What Does a Hydrostatic or Pneumatic Pressure Test Do?

There is always a difference of opinion as to what is desired and what is accomplished with a pressure test. Some persons believe that the pressure test is meant to detect major leaks, while others feel that there should be no leaks, large or smalL Some feel that the test is necessary to invoke loadings and stresses that are equivalent to or exceed those loadings and stresses at operating conditions. Others feel that a pressure test is needed to indicate whether a gross error has been made in calculations or fabrication. In some cases, it appears that the pressure testing may help round out comers or other undesirable wrinkles or may offer some sort of a stress relief to some components.
23
1.6.3 Pressure Test Requirements for VInwl

1.6.3.1 Hydrostatic Test Requirements. A hydrostatic pressure test is the preferred test method. A pneumatic test or a combination of pneumatic/hydrostatic test is conducted only when a hydrostatic test cannot be done. Except for certain types of vessels that are discussed later, the hydrostatic test pressure at every point in the vessel shall be at least 1.3 times the maximum allowable working pressure multiplied by the ratio of the allowable tensile stress value at test temperature divided by the maximum allowable tensile stress value at design temperature. As an alternative, a hydrostatic test pressure may be determined by calculations agreed upon by the user and the manufacturer. In this case, the MAWP (maximum allowable working pressure) of each element is determined and multiplied by 1.3 and then adjusted for the hydrostatic head. The lowest value is used for the test pressure, which is adjusted by the test temperature to design temperature ratio. In any case, . thetestpressure is jimitedtothatpressurewhich willnotcausefil1yvisible:pennanent distortion(yielding). of any element. The metal temperature of the vessel or component to be tested is recol11mended to be at least 30°F above the MDMT to be marked on the vessel but need not exceed 120°P_ Also, a liquid pressure relief valve set at 1 113 times the test pressure is recommended in the pressure test system. After the test pressure is reached, the pressure is reduced to the test pressure divided by 1.3, and welded joints, connections, and other areas are visually examined for leaks and cracks. The visual examination may be waived if a gas leak test is to be applied, if hidden welds have been examined ahead of time, and if the vessel will not contain a lethal substance. Venting shall be provided at all high locations where there is a possibility of air pockets forming during the filling of the vessel for testing. The general rules for hydrostatic testing do not call for a specific time for holding the vessel at test pressure. The length of this time may be set by the Authorized Inspector or by a contract specification.
1.6.3.2 Pneumatic Test Requirements. For some vessels, it is necessary to apply a pneumatic test in lieu of a hydrostatic test. This may be due to any number of reasons, including vessels designed and supported in such a manner that they cannot be safely filled with liquid and vessels that cannot tolerate any trace of water or other liquids. If a vessel is to be pneumatically tested, it shall first be examined according to the requirements ofUW -50. This paragraph requires that welds around openings and attachments be examined by MT or PT before testing. Except for certain vessels, the pneumatic test pressure at every point in the vessel shall be 1.1 times the maximum allowable working pressure multiplied by the ratio of the allowable tensile stress value at test temperature divided by the allowable tensile stress value at design temperature. For pneumatic testing, the metal temperature of the vessel or component shall be at least 30°F above the MDMT to be marked on the vessel. The test pressure shall be gradually increased to no more than half of the full test pressure and then increased in steps of one-tenth of the test pressure until the full test pressure is reached. After that, the pressure shall be reduced to the test pressure divided by 1.1 and all areas are to be examined. All other requirements for hydrostatic testing shall be observed, including the waiving of the visual examination, provided the same limits are met.
1.6.3.3 Test Requirements for Enameled or Glass-lined Vessels. The maximum test pressure for enameled and glass-lined vessels does not have to be any greater than 1.0 MA WP unless required by the Authorized Inspector or by a contract specification. Higher test pressure may damage the enameled or glass coating. All other rules for hydrostatic testing apply.
1.6.3.4 Test Requirements for Vessels Built to the Rules of Parts UCI or UCD. For those vessels designed and constructed to the rules of Part UCI for Cast Iron and Part UCD for Cast Ductile Iron, where the factor of safety on tensile strength to set the allowable tensile stress values is 10 and 5, respectively, the multiplier for the hydrostatic test pressure is set differently. For Part UCI, the test pressure shall be 2.5 MAWP, but is not to exceed 60 psi for a design pressure less than 30 psi and 2.0 MA WP for a design
24 Chapter 1 pressure equal to or greater than 30 psi. For Part UCD, the test pressure shall be 2.0 MA WP. With these changes, the remaining rules of UG-99 are followed. 1.6.3.5 Test Requirements for Vessels Built to the Rules of Part ULT. Alternative rules for the design and construction of vessels to operate at cold temperatures as low as - 320°F are given in Part ULT. These rules permit the use of increased allowable tensile stress values at temperatures colder than ambient temperature to as low as - 320"F for 5%, 8%, and 9% nickel steels, 5083 aluminum alloy, and Type 304 stainless steels. Other materials listed in both Section II and Subsection C may be used for vessels and parts for design at cold temperature with the allowable tensile stress values set by the value at Woop. When the vessel is designed and constructed to Part UL T rules, special hydrostatic testing requirements are necessary dueto the fact that theillaterial isstr0ngeratdesign telllperature than at ambient test temperature. The vessel shall be l1ydrostatically tested at ambient temperature with the test pressure held for 15 minutes and either of the following criteria may be applied: a. A standard hydrostatic test as described in 1.6.3.1 is used, but with the ratio of allowable stresses not applied and the test pressure shall be 1.4 MA WP, if possible, instead of 1.3 MAWP. b. In applying (a), the membrane stress in the vessel shall not exceed 0.95 of the specified minimum yield strength nor 0.5 of the specified minimum tensile strength. In complying with these stress limits, the ratio of hydrostatic test pressure divided by the MAWP may be reduced below 1.4, but it shall be not less than 1.1 MA WP. If the value comes out less than 1.1 MA WP, a pneumatic test shall be conducted using the rules of UG-100, but omitting the adjustment for the allowable tensile stress ratio. A vessel to be installed vertically may be tested in the horizontal position, provided the test pressure is applied for 15 minutes at not less than 1.4 MAWP, including a pressure equivalent to the liquid head in operating position. 1.6.3.6 Proof Testing to Establish MA WP. In addition to the hydrostatic or pneumatic pressure test of the completed vessel, a pressure proof test is permitted to establish the MA WP of vessels and vessel parts for which the strength cannot be calculated with assured accuracy. The rules for such a pressure proof test are given in UG-lOi of VIII-l and may be based on yielding oron bursting of the vessel or vessel part. Proof tests must be witnessed by the Authorized Inspector, who indicates acceptance by signing the Manufacturer's Data Report Form, Duplicate or similar parts to that part which has had its MA WP established by a proof test according to the requirements of UG-IOl(d) of VIII-l may be used without a proof test of their own, but shall be given a hydrostatic or pneumatic pressure test as part of the completed vessel pressure test.
1.6.4 Pressure Test Requirements for VIII-2

1.6.4.1 Hydrostatic Test Requirements. Except for glass-lined and enameled vessels, the hydrostatic test pressure at every point in the vessel shall be 1.25 times the design pressure (or MA WP) to be marked on the vessel multiplied by the ratio of the design stress intensity value at test temperature divided by the design stress intensity value at design temperature. For glass-lined or enameled vessels, the hydrostatic test pressure shall be at least equal to, but need not exceed, the design pressure (or MA WP). Similar to the alternative in VIII-I, the hydrostatic test pressure may be determined by calculations agreed upon between the User and the Manufacturer and shall be described in the Design Report. The hydrostatic test pressure shall not exceed a value that results in the following: a, A calculated primary membrane stress intensity F; of 90% of the tabulated yield strength S, at test temperature;
Background Information b. A calculated primary membrane plus primary bending stress intensity F; the limits given below:
2S
+ Ph not to exceed
(Ll) (1.2)
1.6.4.2 Pneumatic ing prevails:
Test Requirements.
A pneumatic test is permitted only when one of the follow-
a. Vessels cannot be safely filled with water due to their design and support system; b.· Vessels 'in. which traces of testing liquid carinorbe tolerated. When a pneumatic test is permitted in lieu of a hydrostatic test, except for glass-lined and enameled vessels, the pneumatic test pressure at every point in the vessel shall be 1.15 times the design pressure (or MA WP) to be marked on the vessel multiplied by the ratio of the design stress intensity value at test temperature divided by the design stress intensity value at design temperature. For glass-lined or enameled vessels, the pneumatic test pressure shall be at least equal to, but need not exceed, the design pressure (or MAWP). The pneumatic test pressure shall not exceed a value that results in the following: a. A calculated primary membrane stress intensity P; of 80% of the tabulated yield strength S, at test temperature; b. A calculated primary membrane plus primary bending stress intensity Pm + Ph not to exceed the following limits:
(1.3) (1.4)
CHAPTER
2
CYLINDRICAL SHELLS
2.1 INTRODUCTION
The rules for cylindrical shells in VIII-l and VIII-2 take into consideration internal pressure, external pressure, and axial loads. The rules assume a circular cross section with uniform thickness in the circumferential and longitudinal directions. Design requirements are not available for elliptic cylinders or cylinders with variable thicknesses and material properties. However, such construction is not prohibited in VIII in accordance with Paragraphs U-2eg) of VIII-l and, AG-IOO(b) and (d) of VIII-2. The design and loading conditions given in VIII-l are discussed first in this chapter, followed by the rules in VIII-2.
2.2 TENSILE FORCES, VlII~1

The governing equations and criteria for the design of cylindrical shells under tensile forces are given in several paragraphs of VIII -1. The tensile forces arise from various loads such as those listed in Paragraph UG-22 and include internal pressure, wind loads, and earthquake forces.
2.2.1 Thin Cylindrical Shells

The required thickness of a cylindrical shell due to internal pressure is determined from one of two equations listed in Paragraph UG-27. The equation for the required thickness in the circumferential direction, Fig. 2.1(a), due to internal pressure is given as t where
= PR I
(SE ~ O.6P),
when
< O.5R
or
P < 0.385SE
(2.1)
E S
t
= =
Joint Efficiency Factor
P = internal pressure R = internal radius
allowable stress in the material
= thickness of the cylinder
This equation can be rewritten to calculate the maximum pressure when the thickness is known. It takes the form
P=
SE./(R
O.6t)
(2.2)
27
28
Chapter 2
--I --I
1 1 1 1 1
I I
--' .... ~...
(b)
FIG. 2.1
It is of interest to note the similarity between Eq. (2.1) and the classical equation for circumferential membrane stress in a thin cylinder (Beer and Johnston, 1992), given by
t = PRISE (2.3)
The difference is in the additional term of O.6P in the denominator. This term was added by the ASME to take into consideration the nonlinearity in stress that develops in thick cylinders, i.e., when the thickness of a cylinder exceeds O.1R. This is demonstrated in Fig. 2.2 for circumferential stress calculated by three different methods. The first is [rom Eq. (2.3), the theoretical equation for thin cylinders; the second is from Eq. (2.1); and the third is from Lame's theoretical equation for thick cylinders and is discussed later as Eq. (2.12). Similarly, the equation for the required thickness in the longitudinal direction, Fig. 2.1(b), due to internal pressure is given as
t "" PRI(2SE
+ O.4P),
with
< O.5R
or
P < 1.25SE
(2.4)
Cylindrical Shells
29
20.0,10.0I-
i\'
SIP
!~" <,
to
~
I'~I-.
<,
~--_
<,
1--- __ .
...
__
Eq.(2.12} Eq.{2.1 )
--;
---.. r--._
-r-3.5
Eq.(2.3)
.10
to
1.5
2.0
2.5 (R+t) I R
FIG. 2.2
3.0
4.0
COMPARISON
OF EQUATIONS FOR HOOP STRESS IN CYLINDRICAL
SHELLS
or in terms of pressure,
p = 2SEt! (R - OAt)
(2.5)
Notice again the similarity between Eq. (2.4) and the classical equation for longitudinal stress in a thin cylinder given by
t = PR/2SE
(2.6)
Equations (2.1) and (2.4) are in terms of the inside radii of cylinders. In some instances, the outside radius of a shell is known instead. In this case, the governing equation for circumferential stress i~ expressed in
30 eha ter 2
terms of the outside radius Ro. This equation, which is obtained from Eq, (2.1) by substituting (Ro for R, is given in VIII·i, Appendix 1, Article 1·1, as
t = PRo/(SE
t)
OAP),
with
< Q.5Ro
or
<
0.385SE
(2.7) (2.8)
P "" SEt I (Ra - OAt)
VIII·1 does not given an equation for the thickness in the longitudinal direction in-terms of outside radius Ro. Such an expression can be obtained from Eq. (2.4) as
t "" PRo/(2SE
lAP)
...+orirr-termsofP,
P "'" 2SEtl(Ra
-
1At)
(2.10)
Equations (2.1) through (2.10) are applicable to solid wall as well as layered wall construction. Layered vessels consist of thin cylinders wrapped around each other to form a thick cylinder, Fig. 2.3. At any given cross section, a-a, the total thickness consists of individual plate material as well as weld seams. The Joint Efficiency Factor for the overall thickness of a layered vessel is calculated from the ratio E
=
C£
Eot;)!t
(2.11)
FIG. 2.3
Cylindrical Shells where
31
E = overall Joint Efficiency Factor for the layered cylinder E, = Joint Efficiency Factor in a given layer t = overall thickness of a layered cylinder ti = thickness of one layer
The rules in VIlI-l assume that the longitudinal welds in various layers are staggered in such a way that E in Eq. (2.11) is essentially equal to 1.0.
Example 2.1 Problem A pressure vessel is constructed of SA 516-70 material and has an inside diameter of 8 ft. The internal design pressure is 100 psi at 450°F. The corrosion allowance is 0.125 in., and the joint efficiency is 0.85. What is the required shell thickness if the allowable stress is 20,000 psi? Solution Refer to Paragraph UG-27 of VIII-I. The quantity 0.38SSE = 6545 psi is greater than the design pressure of lOO psi. Thus, Eq, (2.1) applies. The inside radius in the corroded condition is equal to
R = 48
=
+ 0.125
48.125 in.
t = [PR/(SE
=
0.6P)]
corrosion
[100 X (48.125)/(20,000
X 0.85 - 0.6 X 100)]
+ 0.125
= 0.41 in.
The calculated thickness is less than O.SR. Thus, Eq. (2.1) is applicable. A check of Eq. (2.4) for the required thickness in the longitudinal direction will result in a t = 0.27 in., including corrosion allowance. This is about 60% of the thickness obtained in the circumferential direction.
Example 2.2 Problem A pressure vessel with an internal diameter of 120 in. has a shell thickness of 2.0 in. Determine the maximum pressure if the allowable stress is 20 ksi. Assume E = 0.85. Solution For the circumferential direction, the maximum pressure is obtained from Eq. (2.2) as
P = 20,000 X 0.85 X 2.01 (60
=
+ 0.6 X 2.0)
556 psi
32
Chapter 2
For the longitudinal direction, the maximum pressure is obtained from Eq. (2.5) as
P = 2 x 20,000 x 0.85 x 2.0/(60
= 1149 psi
- 0.4 x 2.0)
Thus, the maximum pressure permissible in the vessel is 556 psi.
Example 2.3 Problem A vertical boiler is constructed of SA 516-70 material and built in accordance with the requirements of VIII-I. It has an outside diameter of 8 ft and an internal design pressure of 450 psi at 709°F. The corrosion allowance is 0.125 in., and the joint efficiency is 1.0. Calculate the required thickness of the shell if the allowable stress is 17,500 psi. Also, calculate the maximum allowable additional tensile force in the axial direction that the shell can withstand at the design pressure. Solution From Eq. (2.7), the required thickness is
t = 450 X 48/(17,500 1.222
X 1.0
0.4 X 450)
0.125
0.125
1.35 in.
From Eq. (2.10), the maximum allowable axial pressure is

P = 2 X 17,500 X 1.0
= 924.0 psi
1.222/(48
1.4
1.222)
Subtracting from this value the internal pressure of 450 psi results in the additional equivalent pressure pi, that can be applied to the cylinder during operation. P'
;=
924.0 - 450 = 474.0 psi
Total corroded metal area of cylinder = 'IT(R~ - R2)
= 363.9 in.'
Hence, total allowable force in cylinder during operation is

F = 474.0
363.9
= 172,5001b
Cylindrical Shells Example 2.4 Problem
33
What is the required thickness of a layered cylinder subjected to an internal pressure of 1400 psi? Let R = 72 in., S = 18 ksi, t, = 0.25 in. The longitudinal seams of the layers are staggered circumferentially so that any cross section will have only one longitudinal joint with an efficiency of 0.65. Solution This problem must be solved by trial and error. Let E
t = 1400
=
x
1.0. Then from Eq. (2.1),

1.0 0.6 X 1400)
72/(18000
= 5.87 in. Try 24 1/4 in. layers with a total thickness of 6.0 in. The joint efficiency from Eq. (2.11) for the total cross section is
E = (23 X 1.00
= 0.985
1 X 0.65)/24
Using this Joint Efficiency Factor, recalculate the required thickness:

t = 1400 X 72/(18000
X 0.985 - 0.6 X 1400)
= 5.97 in. Since this thickness is less than the assumed thickness of 6.0 in., the solution is complete. Hence, 24 1 !4 in. layers are adequate.
2.2.2 Thick Cylindrical Shells

The VIII· 1 code is routinely referenced in constructing vessels with internal pressures higher than 3000 psi. Special consideration must be given to details of construction, as specified in Paragraph U-I(d) of VIII-I. As the ratio of t! R increases beyond 0.5, the thickness given by Eq. (2.1) becomes nonconservative, as illustrated in Fig. 2.2. A more accurate equation that determines the thickness in a thick cylinder, called Lame's equation, is given by
(2.12)
where Ro and R are outside and inside radii, respectively. By substituting the relationship Ro this expression, Eq. (2.12) becomes
t = R(Zlf2 1)
into
(2.13)
where
Z = (SE
P)/(SE
P)
34
Chapter 2
Equation (2.13) is used in Appendix 1-2 of VIII-l to determine the required thickness in thick cylinders for the conditions t > O.5R or P > 0.3855£. This equation can also be written in tenus of pressure as
P = SE[(Z -
l)/(Z
1)]
(2.14)
where
Z = [(R
t)IR]2
For longitudinsJ. stre:ss,..

t = R(ZlI2 1), with
t>
O.5R
or
P>
1.25SE
(2.15)
where Z = (PiS£) + 1 Equation (2.15) can be written in tenus of pressure, P, as

P = SE(Z -
1)
(2.16)
where
Z = [(R
t)/R]2
The thick cylinder expressions given by Eqs. (2.12) through (2.16) can be expressed in terms of outside radii as follows. For circumferential stress,
with t
> O.5R
or
> 0.385SE
(2.17)
where
Z = (SE
P)/(SE
- P)
or in terms of pressure, P where

= SE[(Z 1)1 (2
1)]
(2.18)
For longitudinal stress with t t
> O.5R or P > 1.255E,

1) / ZlI2
= Ro(ZlI2
(2.19)
Cylindrical Shells 35 where

/ Z = (PIS£)
or in terms of pressure, P,
P "" SE(Z 1)
(2.20)
where
Z "" [Ro I R? = [RoJ (Ro t)F
All of the equations given so far are in terms of internal pressure only. VID-l does not give any equations for calculating stresses in cylinders resulting from wind and earthquake loads. One method of calculating these stresses is given in Section 2.3.
Example 2.5 Problem Calculate the required shell thickness of an accumulator with P psi, and E = 1.0. Assume a corrosion allowance of 0.25 in. Solution The quantity 0.385SE = 7700 psi is less than the design pressure of 10,000 psi. Thus, Eq. (2.13) is applicable.
Z = (SE
= 10,000 psi, R
ziz
18 in., S
20,000
P)/(SE - P)
= (20,000 = 3.0
1.0
10,000) / (20,000 X 1.0 -
10,000)
t "" R(ZlI2
1)
= (18.25)(3.0°-1 = 13.36 in.
1.0)
Total
13.36
0.25
13.61 in.
Example 2.6 Problem What is the required thickness in Example 2.5 if the design pressure is 7650 psi and the corrosion allowance is zero?
36
Chapter 2
Solution The quantity 0.385S£

=
7700 psi is greater than the design pressure of7650 psi. Thus, Eq. (2.1) is applicable.
0.6P) X 1.0 -
t '" PR/(SE
= 7650 X 18/(20,000 = 8.94 in.
0.6 X 7650)
It is of interest to determine the accuracy of Eq. (2.1) by comparing it with the theoretical Eq. (2.13), which gi yes
Z = (SE
P)/(SE X 1.0
P) 7650)/(20,000 X 1.0 -
= (20,000 = 2.239
7650)
I=
R(ZH2 -
1)
'" 18(2.239°.5 = 8.93 in.
1.0)
This comparison demonstrates the accuracy of the "simple-to-use" Example 2.7 Problem
Eq, (2.1) over a wide range of Rltratios.
What is the maximum stress in a layered vessel subjected to an internal pressure of 15,000 psi? The outside diameter is 24 in., and the inside diameter is II in. Solution The thickness of 6.50 in. is greater than 0.5R. Thus, either Eq, (2.17) or Eq. (2.13) may be used, since both the outside and inside diameters are given. Both of these equations are in terms of the quantity Z, which is a function of stress S. Solving for S in these equations is not easy. However, since both of these equations were derived from Eq. (2.12), we can use it directly to solve for S. Thus,
SE = 15,000 (122
= 22,980
5.52)/(122
- 5.52)
psi
2.3 AXIAL COMPRESSION

Vessel components are frequently subjected to, axial compressive stresses caused by such items as wind, dead loads, earthquake, and nozzle loads. The maximum compressive stress is limited by either the allowable tensile stress, using a Joint Efficiency Factor of l.0, or the allowable compressive stress, whichever is less. The allowable tensile stress controls thick cylinders, while the allowable compressive stress controls thin
Cylindrical Shells
37
cylinders. The procedure for calculating the allowable axial compressive stress in a cylinder is given in Paragraph UG-23 of VIII-l and is based on a theoretical equation with a large LID ratio (Jawad, 1994). It consists of calculating the quantity
A = O.125/(Ro/t) (2.21)
where A
t
strain thickness
Ro"" outside radius of the cylinder
""andthenusingastress"straindiagramfurnishedbytheASMEto"determinethe permissible.axial.compressive stress, B. The ASME plots stress-strain diagrams, called External Pressure Charts, for various materials at various temperatures on a log-log scale. One such chart for carbon steel is shown in Fig. 2.4. The strain, A, is plotted along the horizontal axis, and a stress, B, along the vertical axis. The majority of the materials listed in the stress tables of II-D or VIII-1 construction have a corresponding External Pressure Chart (EPC). Tabular values of the curves in these charts are also given in II-D, for example those shown in Table 2.1 for Fig. 2.4. If the calculated value of A falls to the left of the stress-strain line in a given External Pressure Chart, then B must be calculated from the equation
B = AEI2 (2.22)
TABLE 2.1 TABULAR VALUES FOR FIG. 2.4
OF
300 0.100 0.765 0.800 0.900 0.100 0.200 0.300 0.400 0.500 0.25 0.100 0.100 0.663 0.900 0.100 0.250 0.300 0.800 0.100 0.150 0.200 0.272 0.100
A
-04 -03
S, psi
0.145 +03 0.111 +05 0.114 0.119 0.123 0.150 0.163 0.170 0.172 0.178 0.178 0.135 +03 0.895 +04 0.965 0.101 +05 0.121 0.124 0.143
0.147
OF
700 0.100 0.559 0.100 0.300 0.100 0.250 0.100 0.100 0.499 0.100 0.150 0.200 0.300 0.300 0.100 0.100 0.247 0.100 0.150 0.200 0.300 0.800 0.300 0.100
A
-04 -03 -02 -01 +00 -04 -03 -02
B, psi
0.124 +03 0.685 +04 0.808 0.101 +05 0.122 0.139 0.139 0.114 +03 0,569 +04 0.717 0.805 0.849 0.897 0.124 +05 0.124 0.104 +03 0.444 +04 0.605 0.689 0.742 0.795 0.927 0.112 +05 0.112
-·02
800
-01 +00 -04 -03 -02
500
-01 +00 -04 -03 -02
900
-01
0.155 0.163 0.170 +00

0.170
-01 +00
38 Chapter 2
25,000
GENERAL NOTE; See Table 2.1 for tabular values.
J V __
/ V
/~
P 1; 3~
'I
'1
'/I
V
/'
/'
_...-
.-- .....
--
~ f- V f- !-'""V V
_r--
...... ---
..
'
f-.......
saOF
20,000 16,000 16,000 14,000 12,000 10,000 9,000
11 900 F
800 F_
700 F
II
8,000 7.000 6,000
""
a; 0
f-
E", 29.0- X 10"

~ H'
27.0 X lOll
X
E." 24.5 x 10~

E .. 22.8 101!:
2O.B x 106
IIi 1'""--' 1/1. ...... tlf// r--. )j~

~
, ..
I···· ....,
.,
5,000···
4,000 3,500 3.000
Ilill
2: .00001 3456789 0.0001
~V
23456789
3 0.001
I'ACTORA
4 5 6 789
0.Q1
4 5 6 7 89 0.1
2,500
FIG. 2.4 CHART FOR CARBON AND LOW ALLOY STEELS WITH YIELD STRESS OF 30 ksi AND OVER, AND TYPES 405 & 410 STAINLESS STEELS
where E "" modulus of elasticity of the material at design temperature The modulus of elasticity, E, in Eq. (2.22) is obtained from the actual stress-strain diagrams furnished by the ASME, such as those shown in Fig. 2.4. It is of interest to note that the stress B in the External Pressure Chart, Fig. 2.4, has a value of half the stress obtained from the actual stress-strain curve of the given material. This was done by the ASME in order to utilize these charts for other loading conditions, such as external pressure on cylindrical shells as well as axial compression and vacuum on heads with various shapes. Thus, the stress, B, from Fig. 2.4 for carbon steel at room temperature corresponding to a strain, A, of 0.1 in.lin. is 17.8 ksi. This is half the actual yield stress of 35.6 ksi for this material, as obtained from the actual given stress-strain curve. Also, the value of the modulus of elasticity obtained from the elastic portion of the curve by finding the slope between any two points along the curve is half the actual indicated value. If we substitute Eq. (2.21) into Eq. (2.22), we find that in the elastic range, the buckling equation for design becomes
B = E I 16(Ro I t)
This can also be written as

B
="
O,0625EI(Rolt)
(2.23)
Cylindrical Shells The theoretical equation for the critical axial buckling stress used by the ASME is given by
IT"
39
= O.6E/(Rolt)
(2.24)
A comparison of the design Eq. (2.23) and the critical axial buckling stress Eq. (2.24) indicates that a factor of safety of about ten was used by the ASME. However, experiments performed subsequently to the publication of Eq, (2.24) have shown that a more realistic critical axial buckling stress equation is of the form
IT"
= O.6CEI{Rolt)
(2.25)
whereC·is·obtained·fromFig.2.5. A comparison of Eq. (2.23) with Eq. (2.25) indicates that the factor of safety varies from a conservative value of 10.0 for small RoJ t ratios to an unconservative value of 1.0 for large Rolt ratios. This fact should be considered when designing cylinders with large diameter to thickness ratios. VIII-l allows an increase of 20% in the value of B obtained from Fig. 2.4 or calculated from Eq. (2.22) when live loads, such as wind and earthquake, are considered. Wind and earthquake loads are usually obtained from various standards, such as ASCE-7 and the Uniform Building Code.
Example 2.8 Problem The tower shown in Fig. E2.8 has an empty weight of 60 kips. The contents weigh 251 kips. Determine the required thickness of the supporting skirt. Allowable tensile stress is 16 ksi. Use Fig. 2.4 for axial compression calculations. The temperature of the skirt is 200°F at the base and 800°F at the top.
Eb, 0.8I1-i~~-+---t___'f--~+-+--+--t--+--+-+---1-f---+-+--+--t--+--t--l
,~
0.6 ~~ ~
~:I ~
~~
'~€
.... iI~ ~
Q4r-+~~;~~~~~~.~~I~~~r-r-~n~-r-+~--r-~+-~-r-r-+~
&~,g
400
800 1200 1600 2000 2400 2800 3200 3600 4000.

R/t
FIG. 2.5
C FACTOR AS A FUNCTION OF RfT (Jawad, 1994)
40
Chapter 2
32 psf
I
--
ii
2~ psf
I r-LCo
N
20
psf
~
FIG. E2.8
Solution Assume t
= 3/8 in.
Axial force = 60
251 = 311 kips
Axial compressive
stress = force/area
=
of material in skirt
311,000/'11 X 96 X 0.375
= 2750 psi The wind moment at
tht<
bottom of skirt using a vessel projected area of 8 ft is

X 8 X 36
M = 32
(36/2
34
26)
24 X 8 X 34 X (34/2
26)
20 X 8 X 26 X (26/2)
Cylindrical Shells
41
= 718,848
280,704
54,080
"" 1,053,632 ft-lb
Notice that in many applications, the projected area must be increased beyond 8 ft to take into consideration such items as insulation, ladders, and platforms. Also, the moment may have to be modified for shape and drag factors. The bending stress is obtained from the classical equation for the bending of beams:
Stress = Mel!
where c I M
=
=
maximum depth of the cross section from the neutral axis moment of inertia applied moment
and for thin circular cross sections, this equation reduces to
= 1,053,632 X 12/(11
= 4660 psi
X 482 X 0.375)
Total compressive
stress = 2750
4660
= 7410 psi
The allowable compressive stress is calculated from Eq. (2.21).

A = 0.125/(48/0.375)
= 0.00098 From Fig. 2.4 with A = 0.00098 and temperature of 200°F, we get B = 12,000 psi, which is the allowable compressive stress. Thus, the selected thickness is adequate at the bottom of the skirt. Note that the thickness would have been inadequate if the temperature at the bottom was 800°F. Now let us check the thickness at the top of the skirt The axial stress due to dead load stays the same. The bending moment becomes
M = 32 X 8 X 36 X (36/2
34
26 -
16)
24 X 8 X 34 X (34/2
26 -
16)
20 X 8 X 10 X (1012)
= 571,392
176,256
8000
= 755,648 ft-lb
42 Cha ter 2
Bending stress = 755,648 X 12/('iT
= 3340 psi
X 482 X 0.375)
Total compressive
stress = 2750
3340
= 6090 psi From Fig. 2.4 with A = 0.00098 and temperature of 800°F, we get B thickness is adequate at the top of the skirt. 7,000 psi. Thus, the selected
Maximum tensile stress at bottom of skirt = 4660 - 2750 = 1910 psi
Maximum tensile stress at top of skirt = 3340 - 2750

= 590 psi
Both of these values are less than 16,000 psi, which is the allowable tensile stress for the skirt These calculations show that t = 3/8 in. for the skirt is satisfactory. This thickness may need to be increased in actual construction to take into account such items as opening reinforcements, corrosion, outof-roundness considerations, and handling factors. Example 2.9 Problem What is the allowable compressive stress in an internal cylinder with Do temperature = 900°F? Use Fig. 2.4 for the External Pressure Chart. Solution From Eq. (2.21),
A = 0.125/(120/0.1875)
= 24
ft, t = 3/16 in., and design
= 0.0002 From Fig. 2.4, this A value falls to the left of the curve for 900°F. Therefore, Eq. (2.22) must be used. The value of E is obtained from Fig. 2.4 as 20.8 X 106 psi for 900°F. Hence, allowable compressive stress B is B
= 0.0002
20,800,000/2
= 2080 psi
2.4 EXTERNAL PRESSURE

External pressure on cylindrical shells causes compressive forces that could lead to buckling. The equations for the buckling of cylindrical shells under external pressure are extremely cumbersome to use directly in
Cylindrical Shells 43 design (Jaw ad, 1994). However, these equations can be simplified for design purposes by plotting them so that the minimum buckling strain is expressed in terms of length, diameter, and thickness of the cylinder. These plots are utilized by the ASME as discussed next. The rules and factors of safety in VllI-l and VID-2 are identical for external pressure. Accordingly, references in this section are made to paragraphs in VID-l only.
2.4.1 External Pressure for Cylinders with Dolt
2:
10
The ASME uses plots to express the lowest critical strain, A, in terms of the ratios LIDo and Dolt of the cylinder, as shown in Fig. 2.6. The designer calculates the known quantities LIDo and Doft and then uses the figure .to. detennine buckling Strlllfl,,A,.Toc()rrelatebUckling Strllln to allowable extteI1lal press we, . the designer uses the stress-strain diagram of Fig. 2.4 to obtain a B value. The allowable external pressure can then be determined from this B value, as explained below. Accordingly, the procedure in ASME Vill-l Paragraph UO-28 for determining the allowable external pressure for cylinders with Dol t ratios equal to or greater than ten consists of the following steps: 1. Assume a value of t for the cylinder. 2. Calculate the quantities LIDo and Dolt. 3. Use Fig. 2.6 with the calculated values of LIDo and Dol t and establish an A value.
).
s, .... "
) .,_
,~.
3".
i-
~~
m·
t.
A FIG. 2.6
GEOMETRIC CHART FOR CYLINDRiCAL VESSELS UNDER EXTERNAL PRESSURE (Jawad and Fan, 1989)
44
eha ter 2
4. Use an External Pressure Chart such as Fig. 2.4 with an A value and determine a B value from the appropriate temperature chart. 5. Calculate the allowable external pressure from the equation
P = (4/3)(B)/(Dolt) (2.26)
6. When A falls to the left of the curves, the value of Pis determined from
P = 2AEI3(Dolt) (2.27)
where ...E=modulusofelasticity·. Note that the curves in Fig. 2.6 are based on a thin cylinder simply supported at the ends with external pressure acting laterally and on the ends. These curves can also be used, conservatively, for cases where the pressure is on the sides only, as is the case with jacketed vessels. These curves can also be used, conservatively, for cylinders with fixed rather than simply supported ends. The effective length of a cylinder, L, needed to use Fig. 2.6, can sometimes be difficult to establish when the cylinder is attached to other components, such as heads and transition sections. Figure 2.7 is provided by VIII-l to define the effective length of some commonly encountered cylinders. The effective length of cylinders with spiral stiffeners or variable thicknesses between supports is not addressed by VIII-I. The factor of safety for the allowable external pressure obtained by using Eq, (2.26) or (2.27) is three against buckling and also against yield.
Example 2.10 Problem What is the required thickness of a cylindrical shell with length equal to 20 ft and outside diameter equal to 5 ft? The cylinder is subjected to an external pressure of 15 psi at 500°F? Use Fig. 2.4 for the External Pressure Chart. Solution Try t = 0.50 in. Then LIDo = 4.0 and Dol t = 120.00. From Fig. 2.6, we obtain a value of A = 0.00022. Then from Fig. 2.4, B = 3000 psi. The allowable pressure is obtained from Eq. (2.26) as
P "" (4/3) (3000)/120.00
= 33.3 psi
Since this value is higher than 15 psi, try a new thickness of 0.375 in. Then LIDo = 4.0 and Dol t = 160.00. From Fig. 2.6, we get A = 0.00017, and from Fig. 2.4, we see that A falls to the left of the curves. Thus, from Eq. (2.27) P
=2
x 0.00017 x 27.0 x 10"((3 x 160.00)
= 19.1 psi
A new trial of t = 5/16 in. results in an unacceptable allowable pressure of 12.2 psi. Thus, the required thickness to be used is 3 I 8 in.
CylindricalShells 45
(a - 1)
[a -
2)
(bJ
[Notes I 1} end (2)]
(Note (3))
T r lr 11 1
L
t
h
T L
L
(e) (ff
[e -
1J
(e - 2)
(d)
[Notes 11 J and (2J]
(Note (3))
(Note (3ll
NOTES: (1) When the cone-to cylinder or the knuckle-to-cylinder
junction is not a line of support, the nominal thickness of the cone, knuckle, or toriconical section shell not be less then the minimum required thickness of the adjacent cylindrioal shell. (2) Caloulations shall be made using the diameter and oorresponding thickness of each section with dimension L as shown. (3) When the cone-to-cylinder or the knuckle-to-cylinder junction is II line of support, tha moment of inertia shall be provided in accordance with 1-8.
FIG. 2.7
SOME LINES OF SUPPORT OF CYLINDRICAL SHELLS UNDER EXTERNAL (ASME Vm~1) PRESSURE
46 Cha ter 2 The allowable compressive hoop stress is then
s = P Rolt
19.1
30/0.375
1530 psi
2.4.2 External Pressure for Cylinders with Dolt < 10

When Do It is less than 10, theall()wableexterl1al from the following two equations:
Pal
pressure
is taken as the smallerofthevaluesdetermined
[2.167 I (Do I t) ~ 0.0833] B 2S
(2.28)
Pa2
= Dolt [1 -
11 (Dol t»)
(2.29)
where B is obtained as discussed above. For values of (Dolt) of less than or equal to 4, the A value is calculated from
A = Ll/(Doft? (2.30)
For values of A greater than 0.10, use a value of 0.10. The value of S is taken as the smaller of two times the allowable tensile stress, or 0.9 times the yield stress of the material at the design temperature. The yield stress is obtained from the External Pressure Chart of the material by using twice the B value obtained from the extreme right-hand side of the termination point of the appropriate temperature curve. The factor of safety in Eqs. (2.28) and (2.29) varies from 3.0 for Dol t = 10 to about 1.67 for Dol t = 2. This gradual reduction in the factor of safety as the cylinder gets thicker is justified since buckling ceases to be a consideration and the factor of safety for external pressure is kept the same as that for internal pressure, which is 2 I 3
s;
Example 2.11 Problem The inside cylinder of a jacketed vessel has an outside diameter of 20 in., a length of 72 in., and a thickness of 5 in. What is the maximum allowable jacket pressure? Use Fig. 2.4 for an External Pressure Chart. Let the design temperature be 300°F. The allowable stress from tension is 17,500 psi. Solution Calculations give LIDo
3.60 and Dolt
= 4.0. And since Dolt

A
4.0, Eq. (2.30) must be used. Hence,
1.l/(4.0? 0.0688
=
From Fig. 2.4, B 17,800 psi.
CylindricalShells 47 From Eq. (2.28),

Pol
= [(2.167/4.0) = 8160 psi
0.0833]17,800
The yield stress of the material is (0.9)(2B) or 32,040 psi. Twice the allowable stress is 35,000 psi. Hence S = 32,040 psi is to be used. From Eq. (2.29),
P d1 = (2 x 32,040/4.0)(1 ..= 12;020psi 1/4.0)
Therefore, the allowable jacket pressure in accordance with VIII-l is 8160 psi. Notice, however, that this pressure is greater than 0.385S, an indication that thick-shell equations may have to be used. Such equations for external pressure are not in VIII -1 yet.
2.4.3 Empirical Equations

It is of interest to note that Fig. 2.6 can only be used for (Dolt) of up to 1000. Larger values are not permitted presently by the ASME. One approximate equation (Jawad, 1994) that is frequently used by designers for large (Dol t) ratios was developed by the U.S. Navy and is given by
P = 0.866E/(LlDo)(Dolt?5
(2.31)
where
E
P = allowable external pressure, psi

This equation incorporates a factor of safety of 3 and a Poisson's ratio of 0.30. Many pressure vessels are subjected routinely to vacuum as well as .axial loads from wind and dead load. Section VIII does not give any method for calculating the allowable compressive stress due to combined effect of vacuum and axial loads. One such method is given by Bergman (Bergman, 1955). It uses an equivalent external pressure to account for the axial compression effect on external pressure. Another method that is used to combine axial and external pressure is that of Gilbert (Gilbert and Pol ani, 1979). This method uses an interactive equation similar to the one used for calculating the buckling of beam columns. Example 2.12 Problem Solve Example 2.10 using Eq. (2.31). Solution From Example 2.10, t
modulus of elasticity, psi
= 0.375 in., LIDo
4.0, Dolt
160.00, andE
= 27,000 ksi. Then from Eq. (2.31),
P = 0.866 X 27,000,000/(4.0)(160)2-5
= 18.1 psi
This approximate value differs from the answer in Example 2.10 by about 6%.
48
Chapter 2
2.4.4 Stiffening Rings

The required thickness of a shell with a given diameter under a specified external pressure can normally be reduced by shortening the shell's effective length. The length can be reduced by providing stiffening rings at various intervals, as shown in Fig. 2.7. The required moment of inertia of such rings is determined from
1, = [D~L,(t
+
or
A, 1L,.)A] !l4
(2.32)
1;
[D~lt
+ A,IL,)A]110.9
(2.33)
where I, = required moment of inertia, Fig. 2.8(a), of the cross section of the ring about its neutral axis, in." l' = required moment of inertia, Fig. 2.8(b), of the cross section of the ring and effective shell about s their combined neutral axis, in.' The effective length of the shell is taken as 1.10(Dot,)1I2. L, = half the distance from the center line of the stiffening ring to the next line of support on one side, plus half the distance from the center line of the ring to the next line of support on the other side. A line of support is (1) a stiffening ring, (2) jacket bar, (3) circumferential line on a head at onethird the depth of the head, (4) cone-to-cylinder junction. As = area of the stiffening ring, in.? t = minimum thickness of the shell, in. t, = nominal thickness of the shell, in. To design the stiffening ring 1. Assume first an area, A" of the stiffening ring and calculate the available moment of inertia. I, or Is. 2. Calculate B from the equation
B = 0.75 [PDo! (I
A) L,)]
(2.34)
3. Use the appropriate External Pressure Chart and determine an A value. 4. If B falls below the left end of the temperature line, calculate A from
A = 2BIE (235)
5. Solve Eq. (2.32) or (2.33) for the required moment of inertia. 6. The furnished moment of inertia must be greater than the required one.
Example 2.13 Problem Calculate the required thickness of the shell and the required moment of inertia of the stiffening ring shown in Fig. E2.13(a). The shell and ring material are SA 285-C. External pressure is 12 psi, and the design temperature is lOO°F.
I
1 :l
Cylindrical Shells
49
Solution Try t = 1/4 in. Then Dolt 0.00016, and from Eq. (2.27), 240.0 and LIDo = 12(10
2.5/3)/60
2.17. From Fig. 2.6, A
P=2
0.00016 X 29.000,000/3(240)
= 12.9 psi
Thus, a shell thickness of 114 in. is adequate. For the stiffening ring, try a 3 Xli 4 in. hard way bar, as shown in Fig. E2.13. For ease of calculations, assume that the stiffening ring is not integral with the shell, Hence, Eq. (2.32) can be used. The moment .......ofineItia6fthebatisbd3j 12. Thus,
1= 0.25 X 3.(}l112
= 0.56 in."
5' 0.0.
.. -<..
-.
"
FIG. E2.13
r---
.671'
50
Chapter 2
From Eq, (234),

B = 0.75[12 = 2110
X 60.0/(0.25
0.25
3.01140,0)]
psi
Since this value falls below the left end of the material line in Fig. 2.4, we use Eq. (2.35):
A =2
2110/29,000,000
"" 0.000146
From Eq. (2.32),

Is = [60.01 X 140.0(0.25
0.25
3.0/140.0)
0.000146]/14
= 1.34 in." Since this number is larger than the actual moment of inertia of the ring (0.56 in.'), the assumed ring is inadequate and a larger ring is required. However, before such a new ring is chosen, let us use the effective moment of inertia of the existing ring and shell and compare that to Eq. (2.33). From Eq. (2.33),
I:
= [60.02 X 140.0(0.25
= 1.72 in."
0.25 X 3.01140.0) 0.000146] 110.9
The effective centroid of the shell-ring section, Fig. E2.13(b), is

h = [4.26 X 0.25 X 0.125
= 0.796 in.
0.25 X 3.0 (1.5
0.25)]/(4.26
X 0.25
0.25 X 3.0)
The actual moment of inertia is

1 = 4.26 X 0.253/12 "" 0.006
= 1.73
+ +
4.26 X 0.25 X 0.67F
0.25 X 3.0'/12
0.25 X 3.0 X 0.9542
+
in."
0.480
0.563
0.683
Thus, using the composite section results in a 1/4 in. X 3 in. stiffener that is adequate.
2.4.5 Attachment of Stiffening Rings

Details of the attachment of stiffening rings to the shell are given in Fig. UO-30 of VIII-l, which is reproduced in Fig. 2.8. The welds must be able to support a radial pressure load from the shell of PL,. This is based on the code assumption that the stiffening rings must support the total lateral load if the shell segments between the rings collapse. Also, the code requires that the welds support a shear load of 0.01 PL)Jo. This shear load is arbitrary and is based on the assumption that if the rings buckle, bending moments
CylindricalShells 51
·11··············
~ l---tw
In-Line Intermittent Weld
Staggered Intermittent Weld
Continuous
Fillet Weld
One Side, Intermittent Other Side
5 .;; St external stiffeners 5 .;; 12 t intarnal stiffeners
(a]
(bl
Ie)
Continuous full
w
(dl
leI
FIG. 2.8 SOME DETAILS FOR ATTACHING STIFFENER RINGS (ASME VIII-1)
52
Cha ter 2
occur and generate shear forces. VllI-l also has other requirements pertaining to stitch welding and gaps between the rings and the shell. These requirements are given in Paragraphs UG-29 and UG-30 of VIII-I.
Example 2.14 Problem Calculate the required size of the double fillet welds attaching the stiffening ring shown in Fig. E2.J3(b) of Example 2.13 to the shell. Let the allowable tensile stress of SA 285-C at lOOoP be 15,700 psi. Solution The radialloadrFr; on the-rings: is equal to PL;,·
PL, = 12.0 X 140
= 1680 lb f in. of
circumference 8635 psi. The total load carried by
Allowable tensile stress in the fillet weld from Table 1.4 is 0.555 weld is
Total load "'" number of welds attaching ring X size of weld X allowable stress
2 X W X 8635
Hence, the required weld size, W, is

W = 16801 (2 X 8635)
= 0.10 in.
Use 2 1 (4 in. continuous fillet welds, in accordance with the minimum requirements of UG-30(f). Shearing force, V, on the weld is
= 0,01 PL/Jo = 0.01 X 12 X 140
60
"" 10081b
Allowable shearing stress in fillet weld from Table 1,4 is 0.555 the equation for shear stress is given by
1" =
= 8635 psi. From strength of materials,
VQIlt
where Q is at the location of the weld, as shown in Fig, E2.13(b), and is given by Q
= 4.26 X 0.25(0.796
- 0.125) = 0.71 in?
Cylindrical Shells
S3
Hence,
l'
= 1008
0.71 11.73(2
0.25)
= 827 psi
<
8635 psi
2.5
CYLINDRICAL SHELL EQUATIONS, VIII-2
The equation for the design of thin cylindrical shells is given in Section AD·201 ofVIU·2 and is expressed as
t = PRI(S
- 0.5P),
for
P < 0045
(2.36)
where
P = internal pressure R = inside radius S = allowable stress at design temperature t = required thickness
This equation is derived from the basic equation for thin cylindrical shells, t = PRIS, by substituting for the inside radius, R, the quantity R + t/2. Equation (2.36) is applicable as long as the axial tensile force, F, is not larger than 0.5PR. When F exceeds 0.5PR, then the required thickness of the cylinder is governed by the equation for longitudinal tensile stress as follows:
t = (0.5PR
F)/(S - 0.5P)
(2.37)
For values of P > OAS, the ratio of tlR increases beyond the scope of Eq, (2.36) and the design equation in VIIl-2 is based on the plastic theory of thick cylindrical shells (Prager and Hodge, 1965). This equation is given by t
= R(ePls 1)
for
>
OA5
(2.38)
Example 2.15 Problem Calculate the required thickness of a cylindrical shell with an inside diameter of 60 in. and an allowable stress of 20 ksi. Let the pressure be (a) 1000 psi, (b) 8500 psi. Solution
(a) Since P
< OAS, Eq. (2.36) applies. Then,

t = 1000
30/(20,000
- 0.5 X 1000)
1.54 in.
S4 Chapter 2 (b) Since P
> O.4S, Eq. (2.38) applies. Then,

t = 30( e8SOO120000
~
1)
= 15.89 in.
2.6 MISCELLANEOUS SHELLS

2.6.1 Mitered Cylinders
........Miteredcylinders, Fig; 2.9, are used in nozzle connections; transitio nsections-andreducers, · NeitherVm-l·· nor VllI-2 give design rules. The piping code (AS ME B31.3, 1993) gives design equations for various miters. The basic equation for the allowable pressure in a shell with a single miter is
P "" (SEtIR){ 11 [1
0.643(R/ t)Jl2 tan 1.2S(Rlt)"2
en
for for
6 6
<
22.so
(2.39) (2.40)
P = (SEtIRHl/[l
tan 6])
> 22S
where E "" Joint Efficiency Factor P = allowable pressure, psi R = radius of the shell in accordance with Fig. 2.9, in. S = allowable stress t = shell thickness, in. a = angle of change in the direction of the miter joint, Fig. 2.9 6= a. 12 ASME B31.3 gives further information, such as equations for multiple miters, curved miters, and length of tapers. Example 2.16 Problem A mitered cylinder has an inside radius of 24 in., a of 40°, a design pressure of 500 psi, an allowable stress of 15,000 psi, and a joint efficiency of 0.85. Determine the required thickness. Solution From Eq. (2.1) for a straight cylinder,
t = 500 X 24/(15,000
= 0.96 in. X 0.85 ~ 0.6 X 500)
Try t
:=
1 in.
= 40/2
= 20 Thus, Eq. (2.39) governs.

0 •
P = (15,000 = 248 psi
0.85
1.0/24){ 1 I [1
0.643(24/1.0)'12
tan
20n
Cylindrical Shells
55
FIG. 2.9 MITERED BEND
This is inadequate. Try t = 1.8125 in.

P
<=
(15,000 X 0.85 X 1.8125/24)(1/[1
0.643(24/1.8125)ln
tan 20])
= 520 psi
>
500 psi
Use t
113/16 in.
2.6.2 Elliptical Shells

Elliptical shells, Fig. 2.10, are encountered occasionally by the pressure vessel designer. The stresses, away from discontinuities, in the shell due to internal pressure can be approximated by using the membrane theory of elliptical cylinders (Plugge, 1967). The basic equation for hoop stress is expressed as
(2.41)
where
a = major radius of the ellipse, in. b = minor radius of the ellipse, in. E = Joint Efficiency Factor P = design pressure, psi S = allowable stress, psi t = thickness, in. <p = angle as defined in Fig. 2.10
56
Cha ter 2
-r--~ ---
FIG. 2.10 ELLIPTICAL CYLINDER
Example 2.17 Problem Solve Example 2.1 if the cylinder is elliptical in cross section with the major diameter equal to 100 in. and the minor diameter equal to 92 in. Solution a
= 50.125 in., b = 46.125 in., P = q,

= 0°, Eq. (2.41) gives
100 psi, S
= 20,000 psi, and E = 0.85
For
t = 100 X 50.1252 X 46.1252/20,000

= 0.320
X 0.85(50.1252
sin20
+ 46.1252 cos- 0)31" + 0.125
0.125
= 0.45 in.
For ¢
= 90°, Eq. (2.41) gives

t = 100 X 50.12SZ X 46.1252/20,000
= 0.250
X 0.85(50.1252
sin2 90
46.1252 cos" 90?12
0.125
0.125
= 0.375 in.
Use t = 0.45 in. This thickness is about 10% higher than that for a cylinder with a circular cross section having an average diameter of 96 in.
CHAPTER
3
SPHERICAL SHELLS, HEADS, AND TRANSITION SECTIONS
3.1 INTRODUCTION
Sections VIII-I and VIII-2 contain rules for the design of spherical shells, heads and transition sections. Head configurations include spherical, hemispherical, torispherical, and ellipsoidal shapes. Transition sections include conical and toriconical shapes. The design rules for most of these shapes differ significantly in VIII-l and VIII-2. This difference is due to the design approach used in developing the equations for VIII-l and VIII-2. In this chapter a brief description of the various kinds of heads is given.
3.2 SPHERICAL SHELLS AND HEMISPHERICAL HEADS, VIn~1

3.2.1 Internal Pressure in Spherical Shells and Pressure on Concave Side of Hemispherical Heads
The required thickness of a thin spherical shell due to internal pressure is listed in Paragraph UG-27 and is given by
t = PRI(2SE
- O.2P),
when
< O.356R
or
P < 0.665SE
(3.1)
where E
= = =
Joint Efficiency Factor internal radius stress in the material thickness of the head
P = internal pressure
R
S t
This equation can be rewritten to calculate the maximum pressure when the thickness is known. It then takes the form
P=
2SEt I (R + O.2t)
(3.2)
57
58
Chapter 3
Notice the similarity between Eq. (3.1) and the classical equation for the membrane stress in a spherical shell (Beer and Johnston, 1992), given by
t
=>
PRI2SE
(3.3)
The difference is in the additional term of 0.2P in the denominator. This term was added by the ASME to take into consideration the nonlinearity in stress that develops in thick spherical shells. The designer should be aware that Eq. (3.1) determines the thickness based on pressure only. Large spherical shells for liquid storage usually have low internal pressure. Thus, the governing thickness is controlled by the liquid weight rather than Eq. (3.1). One method for determining the thickness in such spheres is given in (API 620, 1990). In some instances, the outside radius ofa shell is known rather than the inside radius. In this case the governing equation is obtained from Eq. (3.1) by substituting (Ra - t) for R. The resulting equation is given in VIII-I, Appendix 1, Article 1-1, as
t = PRo/(2SE
O.SP),
with
< 0.356Ro
or
< 0.665SE
(3.4)
or
P = 2SEtl(Ro O.St) (3.5)
As the ratio of t/R increases beyond 0.356, the thickness given by Eq. (3.1) becomes nonconservative, This is similar to the case for cylindrical shells discussed in 2.2. The ASME VIII-l equation for thick spherical shells is given by
t = R(fi/l
1)
(3.6)
where
Y = 2(SE
P)/(2SE
- P)
Equation (3.6) used in Appendix 1-3 of VIII-l to determine the required thickness in thick spherical shells for the conditions where t > 0.356R or P > 0.665SE. This equation can also be written in tenus of pressure as
P = 2SE[(Y l)/(Y
is
2)]
(3.7)
where
Y = [(R
t) 1R]l
The thick shell expressions given by Eqs. (3.6) and (3.7) can be stated in terms of outside radii as
with t
>
O.356R
or
P>
0.665SE
(3.8)
where
Y = 2(SE
P)/(2SE
- P)
Spherical Sheils, Heads, and Transition Sections or in terms of pressure,

P = 2SE[(Y l)/(Y
59
+ 2)J
0.9)
where
Y = (RoIR)' '" [Ro/(Ro -
tW
Equations (3.1) through (3.9) are also applicable to hemispherical heads with pressure on the concave side. This is illustrated in Fig. 3.1. For an applied internal pressure in compartment A, the hemispherical heads abc and def are subjected to concave pressure and Eqs. (3.1) through (3.9) may be used. Paragraph UG-32(f) of VIH- J gives the rules forthe design of hemispherical heads due to pressure on the concave side. Example 3.1 Problem A pressure vessel is constructed of SA 516-70 material and has an inside diameter of 8 ft. The internal design pressure is 100 psi at 450oP. The corrosion allowance is 0.125 in. and the joint efficiency is 0.85. What is the required thickness of the hemispherical heads if the allowable stress is 20,000 psi? Solution The quantity 0.665SE = 11,300 psi is greater than the design pressure of 100 psi. Thus, Eq. (3.1) applies. The inside radius in the corroded condition is equal to
A
d
11-------iI
SHELL
INTERMEDIATE HEAO
e
SHELL B
SKIRT
FIG. 3.1
60 Chapter 3
R = 48
0.125
= 48.125 in.
The total head thickness is

t = PR 1(2SE - 0.2P)
corrosion
100 X (48.125)/(2
X 20,000 X 0.85 - 0.2 X 100)
0.125
= 0.142
0.125
= 0.27 in.
The calculated thickness is less than 0.356R. Thus, Eq. (3.1) is applicable. Example 3.2 Problem A pressure vessel with an internal diameter of 120 in. has a head thickness of 1.0 in. Determine maximum pressure if the allowable stress is 20 ksi. Assume E = 0.85. Solution The maximum pressure is obtained from Eq. (3.2) as
P=
the
X 20,000
X 0.85 X 1.0/(60
0.2 X 1.0)
= 565 psi
Example 3.3 Problem A vertical unfired boiler is constructed of SA 516-70 material and built in accordance with the requirements of VIII-l. It has an outside diameter of 8 ft and an internal design pressure of 450 psi at 5S0 OF. The corrosion allowance is 0.125 in. and the joint efficiency is 1.0. Calculate the required thickness of the hemispherical head if the allowable stress is 19,700 psi. Solution From Eq. (3.4), the required head thickness is
t = 450 X 48/(2
= 0.543 X 19,700 X 1.0
0.8 X 450)
0.125
0.125
= 0.67 in.
Spherical Shells, Heads, and Transition Sections
61
Example 3.4 Problem Calculate the required hemispherical head thickness of an accumulator with P S = 15,000 psi, and E = 1.0. Assume a corrosion allowance of 0.25 in. Solution The quantity 0.6655£
10,000 psi, R
18 in.,
= 9975 psi is less than the design pressure of 10,000 psi. Thus, Eq. (3.6) applies.
Y = 2(SE
P)/(2SE
~ P) X15,OOOxl.O"" 10,000)
=
=
2(1S;OOO
LO+1O,OOO)/(2
2.5
t = R(Y!!]
1)
~
= (18.25)(2.5li3
1.0)
= 6.52 in. Total head thickness = 6.52 + 0.25 = 6.77 in. The required thickness of the shell for this vessel is calculated in Example 2.5. Attaching the head to the shell requires a transition with a 3:1 taper, as shown in Fig. UW-13.1 of VIII-l. This taper, however, is impractical to make in this case since the thickness of the head is about two-thirds the radius. One method of attaching the head to the shell is shown in Fig. E3.4.
3.2.2 External Pressure in Spherical Shells and Pressure on Convex Side of Hemispherical Heads
The procedure for calculating the external pressure on spherical shells is given in Paragraph UG-28(d) of VIII-l and consists of calculating the quantity
A "" 0.12S/(Roft)
(3.10)
where
A = strain
Ro t
outside radius of the spherical shell thickness
and then using a stress-strain diagram similar to Fig. 2.4 to determine a B value. The allowable external pressure is calculated from
P,
= B/(Ro/f)
(3.11)
62
Cha ter 3
6.77"
..........-t
= 13.61"
FIG. E3.4
If the calculated value of A falls to the left of the stress-strain line in a given External Pressure Chart, then P; must be calculated from the equation
P, = O.0625EI(Rolt)2 (3.12)
where E
= modulus of elasticity of the material at design temperature
The modulus of elasticity, E, in Eq. (3.12) is obtained from the actual stress-strain diagrams furnished by the ASME, such as those shown in Fig. 2.4. Equations (3.10) and (3.11) are also applicable to hemispherical heads with pressure on the convex side, as mentioned in Paragraph UG-33(c) of VIII-I. This is illustrated in Fig. 3.1. For an applied internal pressure in compartment B, the hemispherical head def is subjected to convex pressure and Eqs. (3.10) and (3.11) may be used.
63
Example 3.5 Problem Determine the required thickness of the head in Example 3.1 due to an external pressure of 10 psi. Solution From Example 3.1, the required thickness for internal pressure is 0.14 in. We will use this thickness as our assumed t. Then from Eq. (3.10),
A = 0.125/[(4&
0.125
0.14)/0.14]
From Fig. 2.4, B = 4,700 psi. And fromEq.
(3.11),
P; = 4,700/(4&.265/0.14) = 13.6 psi
Since this pressure is larger than the design pressure of 10 psi, the minimum calculated thickness of 0.14 in. is adequate. Example 3.6 Problem What is the required thickness of a hemispherical head subjected to external pressure of 15 psi? Let Ro = 150 in. and design temperature r= 900"P. The material is SA 516-70. Solution Assume t
0.25 in. Then from Eq. (3.10),

A = 0.125/(150/0.25)
= 0.00021
Since the A value is to the left of the 900°F material line in Pig. 2.4, we have to use Eq. (3.12).
Pa
:=
0.0625
20,800,000/(150/0.25)2
= 3.6 psi Since this value is less than 15 psi, a larger thickness is needed. Try t
A = 0.125/(15010.50) = 0.00042
=
0.50 in.
64
Chapter 3
From Fig. 2.4, B
4500 psi. And from Eq. (3.11),
r,
= 4500/(15010.50) =
15 psi
The selected thickness of 0.50 in. is adequate for the 15 psi external pressure. The thickness may have to be increased due to handling and fabrication requirements.
3.3 .SPHERICAL. SIIELLS AN])IIEMlSraERICALHEAPS,

The required thickness of a spherical shell due to internal pressure VIII-2 as
t = O.SPR/(S 0.2SP),
Y:[]I-2
when
< OAS
(3.13)
where
P = internal pressure R = internal radius S = stress in the material t = thickness of the hemisphere
As the ratio of PIS increases beyond 0.4, the thickness given by Eq. (3.13) becomes nonconservative. This is similar to the case for cylindrical shells discussed in 2.2. The VIII-2 equation for thick hemispherical heads is given by In [(R This can be written also as
t=
R(eO.5PIS -
t)/ R] = O.SP/S,
when
> 0.4S
(3.14)
1),
when
> O.4S
(3.1S)
When meridional forces, F (for instance, wind and earthquake loads) are present on the head, then Eq. (3.13) is modified as follows:
t = (O.5PR
F)/(S
O.25P),
when
<
O.4S
(3.16)
where
F = Meridional force, lb/ in, of circumference. F is taken as positive when it is in tension and negative
when it is in compression. When F is larger than O.5PR, then buckling could occur and the rules for external pressure must be considered. The rules for calculating the required thickness of hemispherical heads subjected to pressure on the concave side are given in Paragraph AD-204.1 of VIII-2. The rules are identical to those for spherical shells given by Eqs. (3.13) through (3.16). The procedure and the factors of safety for calculating the allowable external pressure on spherical shells in Vill-2 are given in Paragraph AD-320. The rules and factors of safety are identical to those given in VIII -1. Similarly, the rules in VIII- 2 for calculating the allowable pressure on the convex side of hemispherical heads are given in Paragraph AD-350.1. They are identical to those given in VIII-l.
65
Example 3.7 Problem

Determine the required thickness for a hemispherical head subjected to an internal pressure of 10,000 psi. Let S = 20 ksi, R = 20 in.
Solution
p /S
=
0.5. Since this ratio is larger than OA, Eq. (3.15) must be used.
t = 20 (eo.s
= 5.68
x 10.000120.000 -
1)
in.
3.4 ELLIPSOIDAL HEADS, VIII-1

3.4.1 Pressure on the Concave Side
A commonly used ellipsoidal head has a ratio of base radius to depth of 2:1 (Fig. 3.2a). The shape can be approximated by a spherical radius of 0.9D and a knuckle radius of 0.17D, as shown in Fig. 3.2(b). The required thickness of 2:1 heads due to pressure on the concave side is given in Paragraph UG-32(d) of VIU-l. The thickness is obtained from the following equation:
t = PD / (2SE - 0.2P) (3.17)
or in terms of required pressure,

P
= 2SEt/(D
0.2t)
(3.18)
where D E
inside base diameter
= Joint Efficiency Factor
P = pressure on the concave side of the head
S = allowable stress for the material t = thickness of the head Ellipsoidal heads with a radius-to-depth ratio other than 2:1 may also be designed to the requirements of VIII-l. The governing equations are given in Appendix 1-4 of VIII-l as
t = PDKI(2SE - D.2P) (3.19)
where
K = (1/6)[2
(D 12h)2J
and D 12h varies between 1.0 and 3.0. The 1.0 factor corresponds to a hemispherical head. The K equation is given in Article 1·4(c) of Appendix 1 of VIII-I. Equation (3.19) can be expressed in terms of the required pressure as P
= 2SEt/(KD
D.2t)
(3.20)
66
Chapter 3
O/2h
2:1
.. I
o
(b)
FIG. 3.2
These equations can also be written in terms of the outside diameter, Do. Thus,
t = PDQKI[2SE
2P(K - 0.1)]
(3.21)
or in terms of required pressure

P = 2SEtl[KDo - 2t(K - 0.1)] (3.22)
It is of interest to note that VIII-l does not give any PIS limitations for the above equations. Nor does it have any rules for ellipsoidal heads when the ratio of PIS is large.
67
3.4.2 Pressure on the Convex Side

The thickness needed to resist pressure on the convex side of an ellipsoidal head is given in Paragraph UG-33 of VIII -1. The required thicknes s is the greater of the two thicknesses determined from the steps below. 1. Multiply the design pressure on the convex side by the factor 1.67. Then use this new pressure and a joint efficiency of E = 1.0 in the appropriate equations listed in Eqs. (3.17) through (3.22) to determine the required thickness. 2. Determine first the crown radius of the ellipsoidal head. Then use this value as an equivalent spherical radius to calculate a permissible external pressure in a manner similar to the procedure given for spherical shells in Section 3.2.2. The procedure consists of calculating the quantity
A = O.125/(KoDo/t) (3.23)
where
A = strain
Ko = function of the ratio Dol2ho and is obtained from Table 3.1 Do = outside base diameter of the ellipsoidal head t "" thickness
Then, using a stress-strain diagram similar to Fig. 2.4, determine the B value. The allowable pressure is calculated from
r,
= BI(KoDolt)
(3.24)
If the calculated value of A falls to the left of the stress-strain line in a given External Pressure Chart, then P, must be calculated from the equation P; = 0.0625EI (KoDo/
tl
(3.25)
where E = modulus of elasticity of material at design temperature' The modulus of elasticity, E, in Eq. (3.25) is obtained from the actual stress-strain diagrams, such as those shown in Fig. 2.4, furnished by the ASME.
Example 3.8 Problem Calculate the required thickness of a 2.2:1 head with an inside base diameter of 18 ft, design temperature of lOO°F, concave pressure of 200 psi, convex pressure of 15 psi, allowable stress is 17,500 psi, and joint efficiency of 0.85. The head is made of low-carbon steel.
TABLE 3.1 FACTOR

Do/2ho Ko Dol2ho
Ko FOR AN ELLIPSOIDAL HEAD WITH PRESSURE ON THE CONVEX SIDE

3.0 1.36 1.8 0.81 2.8 1.27 1.6 0.73 2.6 1.18 1.4 0.65 2.4 1.08 1.2 0.57 2.2 0.99 1.0 0.50 2.0 0.90
«;
68
Cha ter 3
Solution For Concave Pressure From Eq. (3.19), with K
(116)[2
(2.2)2J
1.14,
X 17,500 X 0.85 - 0.2 X 200)
t = (200
216,0 X 1.14)/(2
= 1.66 in.
For Convex Pressure 1. First calculate the pressure and thickness.

P = 1.67 x 15 = 25.1 psi
t = (25.1
216.0
1.14)/(2
17,500
1.0 - 0.2
25.1)
= 0.18 in.
2. For external pressure, we determine Ko from Table 3.1 as 0.99. Let minimum Do From Eq. (3.23),
A = 0.125/(0.99
= 0.00096 = 216
= 1.66 in.
(2 X 1.66) = 219.32
X 219.321L66)
From Fig. 2.4, B
12,000 psi.
r, =
12,000/(0.99 X 2]9.32/1.66)
= 91.7 psi
Thus, minimum t
1.66 in.
3.5 TORISPHERICAL HEADS, vnr-r

3.5.1 Pressure on the Concave Side
Shallow heads, which are commonly referred to as Flanged and Dished heads, or F&D, can also be built to VIII-l rules, in accordance with Paragraph UG-32(e). The most commonly used F&D heads can be approximated by a spherical radius, L, of 1.0D and a knuckle radius, r, of O.06D, as shown in Fig. 3.3. The required thickness of such heads due to pressure on the concave side is obtained from
t = 0.885PLI(SE - O.lP) (3.26)
69
1-
o
L=D
~I
FIG. 3.3
P = SEt I (O.885L
O.lt)
(3.27)
where
E = Joint Efficiency Factor
L= P= S= t=
inside spherical radius pressure on the concave side of the head allowable stress for the material thickness of the head
Torispherical heads with various spherical and knuckle radii may also be designed to the requirements of VIII-I. The governing equations are given in Appendix 1-4 as
t = PLMI(2SE O.2P) (3.28)
70
Cha ter 3
where
M = (114)[3 + (Llr)lI1J
and Llr varies between 1.0 and 16.67. The 1.0 ratio corresponds to a hemispherical shell. The M equation is given in Article 1-4(d) of Appendix 1 of VIII-I. Equation (3.28) can be expressed in terms of the required pressure as P
=
2SEtl(IM
+ D.2t)
(3.29)
These equations can also be written in terms of the outside radius, Lo, as
t = PLaM/[2SE
P(M -
0.2)]
(3.30)

P = 2SEt/[MLo - t(M - 0.2)] (3.31)
The theoretical membrane stress distribution in the circumferential, N6, and meridional, Nq" directions in shallow heads due to internal pressure are shown in Fig. 3.4. Both the circumferential and meridional stresses at the crown of the head are tensile with a magnitude of S ee Pa2l2bt. However, at the base of the head, the meridional stress is tensile with magnitude S = Pa12, while the circumferential stress is compressive with a value of S = (PaI2t)[2 - (alb)Z]. This compressive stress, which is not considered by Eq. (3.28), could cause buckling of the shallow head as the ratio of Dlt increases. One way to avoid such failure is to calculate the thickness based on an equation (Shield and Drucker, 1961) that takes buckling into consideration and is expressed as
nPiSy = (0.33
5.5r/D)(tIL)
28(1 - 2.2rID)(tIL?
- 0.0006
a FIG. 3.4
Spherical Shells, Heads, and Transition Sections where
71
D
L n
P
r
S,
t
base diameter of head, in. cap radius, in. safety = design pressure, psi = knuckle radius, in. = yield stress of the material, psi = thickness, in.
= spherical = factor of
This equation normally results in a thickness that is greater than that calculated from Eqs. (3.26), (3.28), .. or (3,30)forshallowheadswithlargeDltratios . ....... ··paragfaph UG-32(e) (){VIII-l states that the maxirnllIIl allowable stress used to calculate thereq1lired· thickness of tori spherical heads cannot exceed 20 ksi, regardless of the strength of the material. This requirement was added in the code to prevent the possibility of buckling of the heads as the thickness is reduced due to the use of materials with higher strength.
3.5.2 Pressure on the Convex Side

For pressure on the convex design, the buckling rules for calculating F&D head thicknesses are the same as those for ellipsoidal heads, with the exception that the outside crown radius of the F&D head is used in lieu of the quantity KoDo. Example 3.9 Problem Calculate the required thickness of an F&D head with an inside base diameter of 18 ft, design temperature of lOO°F, internal (concave) pressure of 200 psi, external (convex) pressure of 15 psi, allowable stress is 17,500 psi, and joint efficiency of 0.85. The head is made of low-carbon steel. Solution For Concave Pressure Using L = 216.0 in., r from Eq. (3.28)
= 0.06
216
13.0 in., and M = (1/4)[3
(216/ 13)1I2]
1.77, we get
t = (200
x 216.0 x
1.77)/(2
x 17,500 x 0.85 - 0.2 x 200)
"" 2.58 in. For Convex Pressure 1. Find the pressure and the thickness. P
=
1.67 x 15
25.1 psi
= =
(25.1 x 216.0 0.27 in.
X 1.77)/(2
17,500
1.0 - 0.2
25.1)
72 eha ter 3 2. Let t
2.58 in. Then Outside radius

=
216 + (2 X 2.58) "" 221.16 in.
From Eq. (3.23),

A = 0.125/(221.16/2.58)
"" 0.0015 .... rom Fig. 2.4; B···=. ·14,000··psi:' F

P, = 14,000/(221.16/2.58)
=
163 psi> 15 psi
Thus, t
2.58 in.
3.6 ELLIPSOIDAL AND TORISPHERICAL HEADS, VID-2

The required thickness for ellipsoidal as well as tori spherical heads is obtained from Paragraph AD-204 and Article 4-4 of VIII-2. The procedure utilizes a chart, Fig. 3.5, which takes into consideration the possibility of buckling of thin shallow heads, as discussed in the previous section. The design consists of calculating the quantities PIS and riD first and then using Fig. 3.5 to obtain the quantity t/L, and thus t. The thickness for 2:1 ellipsoidal heads is obtained by using the riD = 0.17 curve, while the thickness for a standard F&D head is obtained by using the r/D = 0.06 curve. Figure 3.5 is plotted from the following equation: (3.32) where A =Al + Az + A3 A[ = -1.26176643 - 4.5524592 (riD) + 28.933179 (rlDp Az = [0.66298796 - 2.2470836 (riD) + 15.682985 (rIDf] [In(P J 5')] AJ = [0.26878909 X 10-4 - 0.42262179 (riD) + 1.8878333 (rID)2][ln(PIS)F D = base diameter, in. L = crown radius, in. P = design pressure, psi r = crown radius, in. S = allowable stress, psi t = thickness Example 3.10 Problem An F&D head with a 6% knuckle is subjected to 40 psi of pressure. What is the required thickness if D 168 in.? Use VIII-2 and then VIII-1 rules. S = 20,000 psi for VIII-2, and 15,000 psi for VIII-l.
Spherical Shells, Heads, and Transition Sections 0.10 .-----r--..,.--,----r--,-.,..--r-r-~r___._-._,__r_r......,._, 0.09 i----I--+--+--+-+--t-t-t---t----t-t-t-t--t--+-l
73
0.08 1----I--+--+--+--+--t-t-t---t----t-t-t-t-~-j~?7;'i2~'l' 0.07 1----I--~--+---+--l--+-t-+--i---+-;--;--1f;~"'W7Wt--l 0.065 i-----+----t--+--+-+--+-+-+--+----t-+-+~/V~ h+-l 0.06 1----+--+--+--+-+--t-t-+--t----t-+-i7~';j'jf7,?/WL~i-t 0.055 1----I--+---+--l--+--+-t-t----t----+-+-7"H/.~~'iJri7'V,-+-i 0.05 1----I--+---+--+--+--+-t-t---i---t-i7.M~:lJl//,ff:l(/t-i---t--f
:::51---+--+--!--+-l-+-H---t--+'d"-hA~~~~W:~~:-~--I
0.035
1----+---+--+--+-+--+-+-+--+------,;t;~~rff1"-+(_t_-t-+-+
..............•......................•........
."'0.031
.~~W ~~
0.025 r----+-----li---t--+-----+-+-+-+-A~~~-t--+---i---t--t-l
0.02
~-+-rlo-+--f--+--+-+-+ld-+-_~~~r?L--+---t--+---+-+-+-t
(2:1 Elll
0.Q15 :---0.17
\
010
0.20 ~ 0.15
\ O.O~
h~~~~
~~~r
PIS
r~~ L / .2J =:
-F-
I Of2_:j-~
NOTE: For 2: 1 Ellipsoi·
dal heads use L ~ 0.90 to calculate t/t:
-~
-f-
0.00 1
t'I 0 0
M
ci
q
0
s q
0
II)
0 0
ci
8 0
0
0 0
r-,
00 0 0
(; ci ci
0 0
Ol
U'l
s 0
q
0
('I
In N
II)
<'l
<'l
s 0
<t 0
In
ci
q
0
In
tiL
FIG. 3.5
74
Chapter 3
Solution Using VIIl-2 rules,

PIS
= 0.002
and
riD "'- 0.06
From Fig. 3.4, tiL"" 0.003 Using VIII-l rules and Eq. (3.26), t
>
and
:=
0.003 x 168 = 0.50 in.
0.885>< 40X 0040 in.
J68/(J5,OOQ ><1:°:::-°.1>(40)
Note that normally the thickness obtained for a given component is 33% higher in VIIl-l than that obtained from VIII-2, since the allowable stress in VIII-l is 4.0 while that in VIII-2 is 3.0. However, as is illustrated above, in the case of F&D heads, this may not be so. The reason is in the safety factors imbedded in the equations of VIII-l and VIII-2.
3.7 CONICAL SECTIONS,
vm.r
Conical shells and transition sections have a variety of configurations, as shown in Fig. 3.6. The required thicknesses of the conical and knuckle regions are calculated in a different manner. In addition, conical sections without a knuckle that are attached to shells result in an unbalanced force at the junction that must be considered by the designer. VIII-l provides rules for the design of the junctions. These rules differ for internal and external pressure.
3.7.1
Internal Pressure
For internal pressure, the design equation for a conical section is given by
t = PD![2 cos a (5E - 0.6P)], where (3.33)
where
t
P = internal pressure, psi
= required thickness, in.
D = inside diameter of conical section under consideration, in. S = allowable tensile stress, psi E = Joint Efficiency Factor Equation (3.33) can be expressed in terms of internal pressure as
P = 2SEt cos a I (D
1.2t cos 0:)
(3.34)
Equations (3.33) and (3.34) can also be expressed in terms of outside diameter as
t "'" PDal [25E cos a
P(2 -
1.2 cos
a)]
(3.35)
75
t
t
L
Portion
of a cone
t t
L
(al
(b)
DL$
.Lc
DL
DL$
T(X
Lc
DL Df Dss
Dff
(e)
(d'
(e)
FIG. 3.6
P = 2SEt cos o:/[Do
t(2 -
1.2 cos rx)]
(3.36)
Equations (3.33) to (3.36), which are applicable at any angle 0::, are limited by VIII-l to a :5 30°. When the angle 0:: exceeds 30°, then VITI-! requires a knuckle at the large end, as shown in Fig. 3.6(c) and (e). This type of construction will be discussed later in this section. After determining the thickness of the cone for internal pressure, the designer must evaluate the coneto-shell junction. The cone-to-shell junction at the large end of the cone is in compression due to internal pressure, in most cases. The designer must check the junction for required reinforcement needed to contain the unbalanced forces in accordance with Paragraph 1-5 of Appendix 1 of VIII -1. The required area is obtained from
(3.37)
where,
ArL
E)
= Joint
required area at the large end of the cone, in. Efficiency Factor of the longitudinal joint in the cylinder
76
Chapter 3 modulus of elasticity of the cone, psi modulus of elasticity of the reinforcing ring, psi modulus of elasticity of the cylinder, psi 1 when additional area of reinforcement is not required y/ SeE,. but not less than 1.0 when a stiffening ring is required QL = axial load at the large end, lb/in., including pressure end-load RL = large radius of the cone, in. S, = allowable stress in the cone, psi S, = allowable stress in the reinforcing ring, psi Ss = allowable stress in the cylinder, psi y = S,£, for the reinforcing ring on the shell ...ScEcJor the reinforcing ring Oil the COne ,i = angle obtained from Table 3.2 E; = E, = E, = k=
The area calculated from Eq. (3.37) must be furnished at the junction. Part of this area may be available at the junction as excess area. This excess area can be calculated from the equation
(3.38)
where A,I. t
t, = nominal cone thickness, in. t, = minimum required thickness of the cone, in. t, = nominal shell thickness, in.
= available area at the junction, in.2 = minimum required thickness of the shell, in.
If this excess area is less than that calculated from Eq. (3.37), then additional area in the form of stiffening rings must be added. The cone-to-shell junction at the small end of the cone is in tension due to internal pressure, in most cases. The designer must check the junction for required reinforcement in accordance with Paragraph 1-5 of Appendix 1 of VIII-l. The required area at the small end of the cone is obtained from A" = (k Q, RJS, El)(l - Ala) tan a where A" = required area at the small end of the cone, in.' Q, "" axial load (including pressure end load) at small end, lb lin. R, = small radius of the cone, in. Ll. "" angle obtained from Table 3.3
(3.39)
TABLE 3.2 VALUES OF Ll. FOR JUNCTIONS AT THE LARGE CYLINDER DUE TO INTERNAL PRESSURE
PISsEj
L'>., deg.
0.001
11
0.006 25
0.002 15 0.007 27
0.003 18
0.004 21 0.009' 30
0.005 23
PISsEj
L'>., deg.
O.OOS 2S.5
NOTE: (1) L'>. = 30° for greater values of PI SsE,.
77
The area calculated from Eq. (3.39) must be furnished at the junction. Part of this area may be available at the junction as excess area. This excess area can be calculated from the equation
A" = 0.78(R,t,)1I2
ret, -
t)
(t< -
tr)/cos
«]
(3.40)
If this excess area is less than that calculated from Eq. (3.39), then additional area in the form of stiffening rings must be added. When the angle IX exceeds 30°, VIII-1 requires a knuckle at the large end, as shown in Fig. 3.6(c) and (e). The required thickness for the knuckle (called a flange) at the large end of the cone is obtained from the equation
t "" PLMI(2SE - D.2P)
(3,41)"
where M L
= D,!2 cos IX D, = inside diameter

=
= (1/4)[3 +
(L/r)ll2]
at the knuckle-to-cone junction D - 2r (1 - cos IX) r = inside knuckle radius, in.
Equation (3.41) can be expressed in terms of the required pressure

P = 2SEt I (LM
D.2t)
(3.42)
Equations (3.41) and (3.42) can also be written in terms of the outside diameter, Do, as
t = PL,N I [2SE
P(M ._ 0.2»)
(3.43)
or in terms of required pressure, P

= 2SEt/[MLo
- t(M - 0.2)]
(3.44)
When a knuckle is used at the cone-to-shell junction, the diameter at the large end of the cone is slightly less than the diameter of the cone without a knuckle, as shown in Fig. 3.6. Thus, the design of the cone as given by Eq. 3.33 is based on diameter D, rather than on the shell diameter. ASME VIII-] does not give rules for the design of knuckles (flues) at the small end of cones. One design method uses the pressure-area procedure (Zick and Germain, 1963) to obtain the required thickness. Referring
TABLE 3.3 VALUES OF

PIS.E\ fl, deg. PIS.Ej fl, deg. NOTE: (1) il = 30° 0.002 4
Jl FOR JUNCTIONS AT THE SMALL CYLINDER DUE TO

INTERNAL PRESSURE
0.005 0.010
6
0.10 27
9
0.125' 30
0.02 12.5
0.04 17.5
0.08
24
for greater
values
of PI
S,E1
78
Chapter 3
ex
FIG. 3.7
to Fig. 3.7 for terminology, we can determine the required thickness based on membrane forces in the flue and adjacent cone and shell areas from
(3.45)
where E = Joint Efficiency Factor KI = 0.125 (2r + DY tan ex - ct'lTr2/360 K2 = 0.28DlD1t,Y12

K3
K4 = O.78tc(K6tc)112 K, = 0.55t/D1t,)lJ2 K6 = [Dl + 2r(1 - cos a)]/2 cos ex P = internal pressure, psi
O.78K6(K6fc)112
Spherical Shells, Heads, and Transition Sections S "" allowable stress, psi tc = thickness of the cone, in. tf "" thickness of the flue, in. t, = thickness of the shell, in. a = flue angle, deg. The flue angle is normally the same as the cone angle.
79
Example 3.11 Problem Determine the required thickness of the cone, the two cylinders, and the area at the cone-to-cy linder junctions shown in Fig.·E3 .11 .Let-axial-compressive load at cone vicinity frcm.mounted.equipmentr= ..50 .. ips, k
Small Cylinder
Allowable stress, psi Joint Efficiency Factor Modulus of elasticity, ksi Pressure, psi 15,000 0.85 27,000 100
Cone
16,000
Large Cylinder
17,500 0.85 25,000 100
Reinforcing Ring
13,000 30,000
LO
29,000 100
Rs
5'-0"
FIG. E3.11
80
Chapter 3
Solution Small Shell The required thickness from Eq. (2.1) is

t = 100 X 60/(15,000
= 0.47 in.
X 0.85 - 0.6 X 100)
Use t Cone
= 1/2 in.
From Eq. (3.33), the cone thickness is calculated as

t = 100 X 2 X 7 X 12/[2 cos 28(16,000 X LO - 0.6 X 100)]
= 0.60 in.
Use t = 5/8 in. Large Shell Again, using Eq. (2.1), we get
t = 100 X 7 X 12/(17,500 X 0.85 - 0.6 = 0.57 in.
100)
Use t
5/8 in.
Large Cone-to-Shell Junction Assume that a reinforcing ring, if needed, is to be added to the shell. Then from Eq. (3.37), we calculate the stiffness ratio, k, as
k
17,500 X 25,000,000/(13,000
X 30,000,000)
"" 1.12
The axial loads are given by

QL = PRLI2
- axial equipment load

[50,000/(2'1l'84)J
= 100 X 84/2 = 4105 lb/in.

Guidebook For The Design of Asme Section Viii Pressure Vessel

Hochgeladen von

Dokumentinformationen

Originalbeschreibung:

Originaltitel

Copyright

Verfügbare Formate

Dieses Dokument teilen

Dokument teilen oder einbetten

Freigabeoptionen

Stufen Sie dieses Dokument als nützlich ein?

Sind diese Inhalte unangemessen?

Copyright:

Verfügbare Formate

Guidebook For The Design of Asme Section Viii Pressure Vessel

Hochgeladen von

Copyright:

Verfügbare Formate

FOR THE DESIGN OF ASME SECTION VIII

Library of Congress Cataloging-in-Publication Data

Cover Photo Courtesy of Nooter Corp.

'"Chapter 4" Flat Plates, Covers, and Flanges

xi 157 157 164

5.7 Chapter 6.1 6.2

Special Components, Vlfl-I

233 233 233 239 240 245 249

. , "., ,", ,." , ,................................................... ,.................... , ,',........... ,.,.,.................................

251 253 255 277 283 285

410 Stainless Steels

E3.4 3.2 3.3 3.4

139 141 143 146 147

7.12 ES.! 8.1 ESA 8.2

171 171 174

190 191 192 193 198

209 214 221

C.14 C.I5 C.16 C.l7 C.18 C.19 C.20.E

266 267 268 269 270 271

278 279 280 281

1.2 ALLOWABLE STRESSES

CRITERIA FOR ESTABLISHING

C,...p Rate 1.0S~

S,.Ry Or O.9S,.Ry £NQt~ un

Welded pipe or lube, ferrous and ncnterrcus

~!)x G.SSSy 3.5

'/~ )( D.SS.)!"R!" or 0.9 x O,S5SrRy

CRITERIA FOR ESTABLISHING

1.3 JOINT E}"'FICIENCY FACTORS

Class/ Cond.! Temper Size/ Thick, in. P·No Group No. 2 2

Min. Tensile Line No. 28 29 Strength ksl 70 70

Min. Yield Stress ksi 38 38 1000 850

External Pressure Chart

No. CS-2 CS-2

Notes G10, Gl0,

Note: Gl0, 81, T2 are described

in 11-0 and pertain to metallurgical

VIII-2 Type of Stress

'Sm = stress intensity values for VIII-2 construction

FIG. 1.1 WELDED JOINT CATEGORIES

TABLE 1.5 JOINT EFFICIENCIES1,5 FOR ARC~ AND GAS-WELDED

Degree of Radiographic Examination Type No.

0.55 0.55 0.50

Single ful! fillet lap joints without plug welds

Channel-to-tubesheet Nozzle-to-channel Flange-to-nozzle

weld neck weld

Pass partition-to-tubesheet Tube-to-tubesheet Shell-to-tubesheet weld weld

Jacket to bar-to-outer-shell Nozzle-to-jacket Longitudinal Head-to-shell Longitudinal Skirt-to-head weld

shell seam seam jacket seam seam

1.4 BRITTLE FRACTURE CONSIDERATIONS

TABLE 1.6 OF MATERIALS TO CURVES (ASME VIII~1)

L7, L7A, L7M, L43 1, 2 BC

+20 -20 +10

Spec. No. SA-194 SA-540

Grade 2, 2H, 2HM, 3, 4, 7, 7M, and 16 823IB24

Impact Test Exemption Temperature,

f(f2 = fA (seamless) or f8 (welded)

---<t,lal Butt Welded Components

tg1 = the thinner

t(f2 = the thinner

fg3 = the thinner

(bl Welded Connection with Reinforcement Plate Added