Mandatory Appendices

Transcription

1 Mandatory Appendices

2

3 MANDATORY APPENDICES A99 Appendix 1 Supplementary Design Formulas Appendix 2 Rules for Bolted Flange Connections With Ring Type Gaskets Appendix 3 Definitions Appendix 4 Rounded Indications Charts Acceptance Standard for Radiographically Determined Rounded Indications in Welds Appendix 6 Methods for Magnetic Particle Examination (MT) Appendix 7 Examination of Steel Castings Appendix 8 Methods for Liquid Penetrant Examination (PT) Appendix 9 Jacketed Vessels Appendix 10 Quality Control System Appendix 11 Capacity Conversions for Safety Valves Appendix 12 Ultrasonic Examination of Welds (UT) Appendix 13 Vessels of Noncircular Cross Section Appendix 14 Integral Flat Heads With a Large, Single, Circular, Centrally-Located Opening Appendix 16 Submittal of Technical Inquiries to the Boiler and Pressure Vessel Committee Appendix 17 Dimpled or Embossed Assemblies Appendix 18 Adhesive Attachment of Nameplates Appendix 19 Electrically Heated or Gas Fired Jacketed Steam Kettles Appendix 20 Hubs of Tubesheets and Flat Heads Machined From Plate Appendix 21 Jacketed Vessels Constructed of Work-Hardened Nickel Appendix 22 Integrally Forged Vessels Appendix 23 External Pressure Design of Copper, Copper Alloy, and Titanium Alloy Seamless Condenser and Heat Exchanger Tubes with Integral Fins Appendix 24 Design Rules for Clamp Connections Appendix 25 Acceptance of Testing Laboratories and Authorized Observers for Capacity Certification of Pressure Relief Valves Appendix 26 Pressure Vessel and Heat Exchanger Expansion Joints Appendix 27 Alternative Requirements for Glass-Lined Vessels Appendix 28 Alternative Corner Weld Joint Detail for Box Headers for Air-Cooled Heat Exchangers Appendix 29 Requirements for Steel Bars of Special Section for Helically Wound Interlocking Strip Layered Pressure Vessels Appendix 30 Rules for Drilled Holes Not Penetrating Through Vessel Wall Appendix 31 Rules for Cr Mo Steels With Additional Requirements for Welding and Heat Treatment

4

5 APPENDIX 1 SUPPLEMENTARY DESIGN FORMULAS 1-1 THICKNESS OF CYLINDRICAL AND SPHERICAL SHELLS (a) The following formulas, in terms of the outside radius, are equivalent to and may be used instead of those given in UG-27(c) and (d). (1) For cylindrical shells (circumferential stress), Where t is known and P is desired, P p SE Z 1 Z +1 (2) t p PR o SE + 0.4P or P p SEt R o 0.4t (1) where where R o p outside radius of the shell course under consideration, in. (2) For spherical shells, t p PR o 2SE + 0.8P or P p 2SEt R o 0.8t Other symbols are asdefined in UG-27. (2) Z p R + t R 2 p R o R 2 p R o R o t 2 (2) Longitudinal Stress (Circumferential Joints). When the thickness of the cylindrical shell under internal design pressure exceeds one-half of the inside radius, or when P exceeds 1.25SE, the following formulas shall apply: When P is known and t is desired, 1-2 THICK CYLINDRICAL SHELLS (a)(1) Circumferential Stress (Longitudinal Joints). When the thickness of the cylindrical shell under internal design pressure exceeds one-half of the inside radius, or when P exceeds 0.385SE, the following formulas shall apply: When P is known and t is desired, (Z1 1) t p R(Z 1)p R o (1) Z 1 2 where t p R (Z 1)p R o Z 2 1 Z 1 2 (3) Z p P SE +1 where When t is known and P is desired, Z p SE + P SE P P p SE(Z 1) (4) 315

6 SECTION VIII DIVISION where Z p R + t R 2 p R o R 2 p R o R o t 2 Symbols are as defined in UG-27 and THICK SPHERICAL SHELLS When the thickness of the shell of a wholly spherical vessel or of a hemispherical head under internal design pressure exceeds 0.356R, or when P exceeds 0.665SE, the following formulas shall apply: When P is known and t is desired, where t p R(Y 1)p R o Y 3 1 Y 1 3 (1) Y p 2(SE + P) 2SE P When t is known and P is desired, where P p 2SE Y 1 Y +2 (2) R + t Y p R 3 R p o R o t 3 Symbols are as defined in UG-27 and 1-1. (b) The symbols defined below are used in the formulas of this paragraph (see Fig. 1-4): tp minimum required thickness of head after forming, in. Pp internal design pressure (see UG-21), psi Dp inside diameter of the head skirt; or inside length of the major axis of an ellipsoidal head; or inside diameter of a cone head at the point under consideration measured perpendicular to the longitudinal axis, in. D o p outside diameter of the head skirt; or outside length of the major axis of an ellipsoidal head; or outside diameter of a cone head at the point under consideration measured perpendicular to the longitudinal axis, in. Sp maximum allowable working stress, as given in Subsection C, psi, except as limited by footnote 1 to 1-4(c) and (d), UG-24, UG-32(e), and UW-12. Ep lowest efficiency of any Category A joint in the head (for hemispherical heads this includes head-to-shell joint). For welded vessels, use the efficiency specified in UW-12. rp inside knuckle radius, in. Lp inside spherical or crown radius for torispherical and hemispherical heads, in. Lp K 1 D for ellipsoidal heads in which K 1 is obtained from Table UG-37, in. L o p outside spherical or crown radius, in. L/rp ratio of the inside crown radius to the inside knuckle radius, used in Table Mp a factor in the formulas for torispherical heads depending on the head proportion L/r hp one-half of the length of the minor axis of the ellipsoidal head, or the inside depth of the ellipsoidal head measured from the tangent line (head-bend line), in. Kp a factor in the formulas for ellipsoidal heads depending on the head proportion D /2h D /2hp ratio of the major to the minor axis of ellipsoidal heads, which equals the inside diameter of the skirt of the head divided by twice the inside height of the head, and is used in Table p one-half of the included (apex) angle of the cone at the center line of the head (c) Ellipsoidal Heads FORMULAS FOR THE DESIGN OF FORMED HEADS UNDER INTERNAL PRESSURE (a) The formulas of this paragraph provide for the design of formed heads of proportions other than those given in UG-32, in terms of inside and outside diameter. t p PDK 2SE 0.2P or P p 2SEt KD + 0.2t 1 Ellipsoidal heads designed under K > 1.0 and all torispherical heads made of materials having a specified minimum tensile strength exceeding 80,000 psi shall be designed using a value of S equal to (1) 316

7 1-4 APPENDIX 1 MANDATORY 1-4 FIG. 1-4 PRINCIPAL DIMENSIONS OF TYPICAL HEADS TABLE VALUES OF FACTOR K (Use Nearest Value of D/2h; Interpolation Unnecessary) D/2h K D/2h K or t p P p PD o K 2SE +2P(K 0.1) 2SEt KD o 2t(K 0.1) (2) Numerical values of the factor K are given in Table Example 1. 2 Determine the required thickness t of a seamless ellipsoidal head, exclusive of provision for corrosion for the following conditions: D p 40 in; h p 9 in; P p 200 psi; S p 13,750 psi; E p where K p D 2h 2 D 2h p p ,000 psi at room temperature and reduced in proportion to the reduction in maximum allowable stress values at temperature for the material as shown in the appropriate table (see UG-23). 2 This calculation is intended only to illustrate the use of the formula herein. Other paragraphs in this Division may have to be satisfied to permit use of the full tabular stress value. 317

8 SECTION VIII DIVISION From Table 1-4.1, K p Substituting in Eq. (1), t p p 0.33 in. [2 13,750 (1.00) ( )] Example 2. 2 Determine the maximum allowable working pressure P of a seamless ellipsoidal head for the following conditions: D p 30 in.; h p 7.5 in.; total thickness p 1 2 in. with no allowance for corrosion; maximum operating temperature p 800 F; E p From the appropriate table given in Subpart 1 of Section II, Part D, S p 10,200 psi. D 2h p p 2.0 From Table 1-4.1, K p 1.0. Substituting in Eq. (1), P p 2 10, [ ( )] (d) Torispherical Heads 1 p 339 psi Numerical values of the factor M are given in Table Example 1. 2 Determine the required thickness t, exclusive of allowance for corrosion, of a torispherical head for the following conditions: D p 40 in.; L p 40 in.; r p 4 in.; P p 200 psi; S p 13,750 psi; E p 1.00 (seamless head). L r p 40 4 p 10 and from Table 1-4.2, M p Substituting in Eq. (3), t p p 0.45 in. [2 13,750 (1.00) ( )] Example 2. 2 Determine the maximum allowable working pressure P of a torispherical head for the following conditions: D p 30 in.; L p 24 in.; r p 2.00 in.; E p 1.00 (seamless head); total thickness p 0.5 in. with no allowance for corrosion; material conforms to SA-515 Grade 70; maximum operating temperature p 900 F. From the appropriate table given in Subpart 1 of Section II, Part D, S p 6500 psi. t p PLM 2SE 0.2P or P p 2SEt LM + 0.2t (3) L r p p 12.0 From Table 1-4.2, M p Substituting in Eq. (3), t p PL o M 2SE + P(M 0.2) P p (e) Conical Heads p 167 psi or t p PD 2 cos (SE 0.6P) P p 2SEt ML o t(m 0.2) (4) or where P p 2SEt cos D + 1.2t cos (5) M p L r t p PD o 2 cos (SE + 0.4P) 318

9 1-4 APPENDIX 1 MANDATORY 1-5 TABLE VALUES OF FACTOR M (Use Nearest Value of L/r; Interpolation Unnecessary) L/r M L/r M L/r M NOTE: (1) Maximum ratio allowed by UG-32(j) when L equals the outside diameter of the skirt of the head. or P p 2SEt cos D o 0.8t cos 1-5 RULES FOR CONICAL REDUCER SECTIONS AND CONICAL HEADS UNDER INTERNAL PRESSURE (a) The formulas of (d) and (e) below provide for the design of reinforcement, if needed, at the cone-tocylinder junctions for conical reducer sections and conical heads where all the elements have a common axis and the half-apex angle 30 deg. Subparagraph (g) below provides for special analysis in the design of cone-to-cylinder intersections with or without reinforcing rings where is greater than 30 deg. In the design of reinforcement for a cone-to-cylinder juncture, the requirements of UG-41 shall be met. (b) Nomenclature A rl p required area of reinforcement at large end of cone, in. 2 A rs p required area of reinforcement at small end of cone, in. 2 A el p effective area of reinforcement at large end intersection, in. 2 A es p effective area of reinforcement at small end intersection, in. 2 E s p modulus of elasticity of cylinder material, psi E c p modulus of elasticity of cone material, psi E r p modulus of elasticity of reinforcing ring material, psi E 1 p efficiency of longitudinal joint in cylinder. For compression (such as at large end of cone), E 1 p 1.0 for butt welds. E 2 p efficiency of longitudinal joint in cone. For compression, E 2 p 1.0 for butt welds. (6) f 1 p axial load at large end due to wind, dead load, etc., excluding pressure, lb/in. f 2 p axial load at small end due to wind, dead load, etc., excluding pressure, lb/in. Pp internal design pressure (see UG-21), psi Q L p algebraical sum of PR L /2 and f 1, lb/in. Q s p algebraical sum of PR s /2 and f 2, lb/in. R s p inside radius of small cylinder at small end of cone, in. R L p inside radius of large cylinder at large end of cone, in. S s p allowable stress of cylinder material at design temperature, psi S c p allowable stress of cone material at design temperature, psi S r p allowable stress of reinforcing ring material at design temperature, psi tp minimum required thickness of cylinder at cone-to-cylinder junction, in. t c p nominal thickness of cone at cone-to-cylinder junction, in. t r p minimum required thickness of cone at coneto-cylinder junction, in. t s p nominal thickness of cylinder at cone-to-cylinder junction, in. p half-apex angle of cone or conical section, deg. p angle indicating need for reinforcement at coneto-cylinder junction having a half-apex angle 30 deg. When, no reinforcement is required at the junction (see Tables and 1-5.2), deg. yp cone-to-cylinder factor p S s E s for reinforcing ring on shell p S c E c for reinforcing ring on cone (c) For a cone-to-cylinder junction, the following values shall be determined at large end and again at the small end in order that both the large end and the small end can be examined: 319

10 SECTION VIII DIVISION TABLE VALUES OF FOR JUNCTIONS AT THE LARGE CYLINDER FOR 30 deg. P/S s E , deg P/S s E , deg NOTE: (1) p 30 deg. for greater values of P/S s E 1. TABLE VALUES OF FOR JUNCTIONS AT THE SMALL CYLINDER FOR 30 deg. P/S s E , deg P/S s E , deg NOTE: (1) p 30 deg. for greater values of P/S s E 1. Determine P/S s E 1 and then determine at the large end and at the small end, as appropriate, from Tables and Determine k: kp 1 when additional area of reinforcement is not required p y/s r E r when a stiffening ring is required, but k is not less than 1.0 (d) Reinforcement shall be provided at the junction of the cone with the large cylinder for conical heads and reducers without knuckles when the value of obtained from Table 1-5.1, using the appropriate ratio P/S s E 1, is less than. Interpolation may be made in the Table. The required area of reinforcement shall be at least equal to that indicated by the following formula when Q L is in tension: A rl p kq LR L S s E 1 1 tan (1) At the large end of the cone-to-cylinder juncture, the PR L /2 term is in tension. When f 1 is in compression and the quantity is larger than the PR L /2 term, the design shall be in accordance with U-2(g). The calculated localized stresses at the discontinuity shall not exceed the stress values specified in 1-5(g)(1) and (2). The effective area of reinforcement can be determined in accordance with the following formula: A el p (t s t) R L t s + (t c t r ) R L t c /cos (2) Any additional area of reinforcement which is required shall be situated within a distance of R L t s from the junction of the reducer and the cylinder. The centroid of the added area shall be within a distance of 0.25 R L t s from the junction. (e) Reinforcement shall be provided at the junction of the conical shell of a reducer without a flare and the small cylinder when the value of obtained from Table 1-5.2, using the appropriate ratio P/S s E 1, is less than. The required area of reinforcement shall be at least equal to that indicated by the following formula when Q s is in tension: A rs p kq sr s S s E 1 1 tan (3) At the small end of the cone-to-cylinder juncture, the PR s /2 term is in tension. When f 2 is in compression and the quantity is larger than the PR s /2 term, the design shall be in accordance with U-2(g). The calculated localized stresses at the discontinuity shall not exceed the stress values specified in 1-5(g)(1) and (2). The effective area of reinforcement can be determined in accordance with the following formula: A es p 0.78 R s t s [(t s t) +(t c t r )/cos ] (4) Any additional area ofreinforcement which is required shall be situated within a distance of R s t s from the junction, and the centroid of the added area shall be within a distance of 0.25 R s t s from the junction. (f) Reducers not described in UG-36(e)(5), such as those made up of two or more conical frustums having different slopes, may be designed in accordance with (g). (g) When the half-apex angle is greater than 30 deg., cone-to-cylinder junctions without a knuckle may be used, with or without reinforcing rings, if the design is based on special analysis, such as the beam-onelastic-foundation analysis of Timoshenko, Hetenyi, or Watts and Lang. See U-2(g). When such an analysis is made, the calculated localized stresses at the discontinuity shall not exceed the following values. (1) (Membrane hoop stress) + (average discontinuity hoop stress) shall not be greater than SE, where the average discontinuity hoop stress is the average 320

11 1-5 APPENDIX 1 MANDATORY 1-6 FIG. 1-6 SPHERICALLY DISHED COVERS WITH BOLTING FLANGES hoop stress across the wall thickness due to the discontinuity at the junction, disregarding the effect of Poisson s ratio times the longitudinal stress at the surfaces. (2) (Membrane longitudinal stress) + (discontinuity longitudinal stress due to bending) shall not be greater than 4SE. The angle joint (see 3-2) between the cone and cylinder shall be designed equivalent to a double buttwelded joint, and because of the high bending stress, there shall be no weak zones around the angle joint. The thickness of the cylinder may have to be increased to limit the difference in thickness so that the angle joint has a smooth contour. The joint efficiencies E shall be in accordance with UW SPHERICALLY DISHED COVERS (BOLTED HEADS) (a) Circular spherical dished heads with bolting flanges, both concave and convex to the pressure and conforming to the several types illustrated in Fig. 1-6, shall be designed in accordance with the formulas which follow. (b) The symbols used in the formulas of this paragraph are defined as follows: tp minimum required thickness of head plate after forming, in. Lp inside spherical or crown radius, in. rp inside knuckle radius, in. Pp internal pressure (see UG-21) for the pressure on concave side, and external pressure for the pressure on convex side [see UG-28(f)], psi Sp maximum allowable stress value, psi (see UG-23) Tp flange thickness, in. M o p the total moment, in.-lb, determined as in 2-6 for heads concave to pressure and 2-11 for heads convex to pressure; except that for heads of the type shown in Fig. 1-6 sketch (d), H D and h D shall be as defined below, and an additional moment H r h r (which may add or subtract) shall be included where H r pradial component of the membrane load in the spherical segment, lb, acting at the intersection of the inside of the flange ring with the center line of the dished cover thickness 321

12 SECTION VIII DIVISION ph D cot 1 h r plever arm of force H r about centroid of flange ring, in. H D paxial component of the membrane load in the spherical segment, lb, acting at the inside of the flange ring p0.785 B 2 P h D pradial distance from the bolt circle to the inside of the flange ring, in. 1 pangle formed by the tangent to the center line of the dished cover thickness at its point of intersection with the flange ring, and a line perpendicular to the axis of the dished cover parc sin B 2L + t NOTE: Since H r h r in some cases will subtract from the total moment, the moment in the flange ring when the internal pressure is zero may be the determining loading for flange design. Ap outside diameter of flange, in. Bp inside diameter of flange, in. Cp bolt circle, diameter, in. (c) It is important to note that the actual value of the total moment M o may calculate to be either plus or minus for both the heads concave to pressure and the heads convex to pressure. However, for use in all of the formulas which follow, the absolute values for both P and M o are used. (d) Heads of the type shown in Fig. 1-6 sketch (a): (1) the thickness of the head t shall be determined by the appropriate formula in UG-32 for pressure on concave side, and UG-33(a)(1) for pressure on convex side; (2) the head radius L or the knuckle radius r shall comply with the limitations given in UG-32; (3) the flange shall comply at least with the requirements of Fig. 2-4 and shall be designed in accordance with the provisions of 2-1 through 2-7 for pressure on concave side, and 2-11 for pressure on convex side. (Within the range of flange standards listed in Table U-3, the flange and drillings may conform to the standards, and the thickness specified therein shall be considered as a minimum requirement.) (e) Heads of the type shown in Fig. 1-6 sketch (b) (no joint efficiency factor is required): (1) head thickness (a) for pressure on concave side, (b) for pressure on convex side, the head thickness shall be determined based on UG-33(c) using the outside radius of the spherical head segment; (2) flange thickness for ring gasket T p M o SB A + B A B (2) (3) flange thickness for full face gasket T p 0.6 P B(A + B)(C B) S A B (3) NOTE: The radial components of the membrane load in the spherical segment are assumed to be resisted by its flange. (Within the range of flange standards listed in Table U-3, the flange and drillings may conform to the standards, and the thickness specified therein shall be considered as a minimum requirement.) (f) Heads of the type shown in Fig. 1-6 sketch (c) (no joint efficiency factor is required): (1) head thickness (a) for pressure on concave side, t p 5PL (4) 6S (b) for pressure on convex side, the head thickness shall be determined based on UG-33(c) using the outside radius of the spherical head segment; (2) flange thickness for ring gasket for heads with round bolting holes where T p Q M o (C + B) SB(7C 5B) Q p PL 4S C + B 7C 5B (3) flange thickness for ring gasket for heads with bolting holes slotted through the edge of the head (5) t p 5PL 6S (1) T p Q M o(c + B) SB(3C B) (6) 322

13 1-6 APPENDIX 1 MANDATORY 1-7 where Q p PL 4S C + B 3C B and M J p o SB A + B A B (4) flange thickness for full-face gasket for heads with round bolting holes T p Q + Q2 + 3BQ(C B) L (7) (h) These formulas are approximate in that they do not take into account continuity between the flange ring and the dished head. A more exact method of analysis which takes this into account may be used if it meets the requirements of U-2. where Q p PL 4S C + B 7C 5B (5) flange thickness for full-face gasket for heads with bolting holes slotted through the edge of the head where T p Q + Q2 + 3BQ(C B) L Q p PL 4S C + B 3C B (6) the required flange thickness shall be T as calculated in (2), (3), (4), or (5) above, but in no case less than the value of t calculated in (1) above. (g) Heads of the type shown in Fig. 1-6 sketch (d) (no joint efficiency factor is required): (1) head thickness (a) for pressure on concave side, t p 5PL (9) 6S (b) for pressure on convex side, the head thickness shall be determined based on UG-33(c) using the outside radius of the spherical head segment; (2) flange thickness where (8) T p F + F 2 + J (10) F p PB 4L 2 B 2 8S(A B) 1-7 LARGE OPENINGS IN CYLINDRICAL SHELLS 1-7(a) Openings exceeding the dimensional limits given in UG-36(b)(1) shall be provided with reinforcement that complies with the following rules. Twothirds of the required reinforcement shall be within the following limits: 1-7(a)(1) parallel to vessel wall: the larger of three-fourths times the limit in UG-40(b)(1), or equal to the limit in UG-40(b)(2); 1-7(a)(2) normal to vessel wall: the smaller of the limit in UG-40(c)(1), or in UG-40(c)(2). 1-7(b) Openings for radial nozzles which exceed the limits in UG-36(b)(1) 1-7(b)(1) and which also are within the range defined by the following limits (a) vessel diameters greater than 60 in. I.D.; (b) nozzle diameters which exceed 40 in. I.D. and also exceed 3.4 Rt; the terms R and t are defined in Figs and 1-7-2; (c) the ratio R n /R does not exceed 0.7; for nozzle openings with R n /R exceeding 0.7, refer to (c) below and/or U-2(g) shall meet the requirements in 1-7(b)(2), (3), and (4) that follow. The rules are limited to radial nozzles in cylindrical shells that do not have internal projections, and do not include any analysis for stresses resulting from externally applied mechanical loads. For such cases U-2(g) shall apply. 1-7(b)(2) The membrane stress S m as calculated by Eq. (1) or (2) below shall not exceed S, as defined in UG-37 for the applicable materials at design conditions. The maximum combined membrane stress S m and bending stress S b shall not exceed 1.5S at design conditions. S b shall be calculated by Eq. (5) below. 1-7(b)(3) Evaluation of combined stresses from internal pressure and external loads shall be made in accordance with U-2(g)

14 SECTION VIII DIVISION FIG (b)(4) For membrane stress calculations, use the limits defined in Fig , and comply with the strength of reinforcement requirements of UG-41. For bending stress calculation, the greater of the limits defined in Fig or Fig may be used. The strength reduction ratio requirements of UG-41 need not be applied, provided that the allowable stress ratio of the material in the nozzle neck, nozzle forging, reinforcing plate, and/or nozzle flange divided by the shell material allowable stress is at least Case B (See Fig ) S m p P R(R n + t n + R m t) + R n (t + R nm t n ) A s Cases A and B (See Fig or Fig ) (2) M p R 3 n 6 + RR ne P (3) NOTE: The bending stress S b calculated by Eq. (5) is valid and applicable only at the nozzle neck-shell junction. It is a primary bending stress because it is a measure of the stiffness required to maintain equilibrium at the longitudinal axis junction of the nozzleshell intersection due to the bending moment calculated by Eq. (3). Case A (See Fig ) S m p P R(R n + t n + R m t) + R n (t + t e + R nm t n ) A s (1) a p e +t/2 (4) S b p Ma I (5) 1-7(b)(5) Nomenclature. Symbols used in Figs and are as defined in UG-37(a) and as follows: 324

15 1-7 APPENDIX 1 MANDATORY 1-7 FIG A s p shaded (cross-hatched) area in Fig , Case A or Case B, in. 2 Ip moment of inertia of the larger of the shaded areas in Fig or Fig about neutral axis, in. 4 ap distance between neutral axis of the shaded area in Fig or Fig and the inside of vessel wall, in. R m p mean radius of shell, in. R nm p mean radius of nozzle neck, in. ep distance between neutral axis of the shaded area and midwall of the shell, in. S m p membrane stress calculated by Eq. (1) or (2), psi S b p bending stress at the intersection of inside of the nozzle neck and inside of the vessel shell along the vessel shell longitudinal axis, psi S y p yield strength of the material at test temperature, see Table Y-1 in Subpart 1 of Section II, Part D, psi 1-7(c) It is recommended that special consideration be given to the fabrication details used and inspection employed on large openings; reinforcement often may be advantageously obtained by use of heavier shell plate for a vessel course or inserted locally around the opening; welds may be ground to concave contour and the inside corners of the opening rounded to a generous radius to reduce stress concentrations. When radiographic examination of welds is not practicable, liquid penetrant examination may be used with nonmagnetic materials and either liquid penetrant or magnetic particle inspection with ferromagnetic materials. If magnetic particle inspection is employed, the prod method is preferred. The degree to which such measures should be used depends on the particular application and the severity of the intended service. Appropriate proof testing may be advisable in extreme cases of large openings approaching full vessel diameter, openings of unusual shape, etc. 325

16 SECTION VIII DIVISION RULES FOR REINFORCEMENT OF CONE-TO-CYLINDER JUNCTION UNDER EXTERNAL PRESSURE (a) The formulas of (b) and (c) below provide for the design of reinforcement, if needed, at the cone-tocylinder junctions for reducer sections and conical heads where all the elements have a common axis and the half-apex angle 60 deg. Subparagraph (e) below provides for special analysis in the design of cone-tocylinder intersections with or without reinforcing rings where is greater than 60 deg. In the design of reinforcement for a cone-to-cylinder juncture, the requirements of UG-41 shall be met. The nomenclature given below is used in the formulas of the following subparagraphs: Ap factor determined from Fig. G and used to enter the applicable material chart in Subpart 3 of Section II, Part D A el p effective area of reinforcement at large end intersection, in. 2 A es p effective area of reinforcement at small end intersection, in. 2 A rl p required area of reinforcement at large end of cone, in. 2 A rs p required area of reinforcement at small end of cone, in. 2 A s p cross-sectional area of the stiffening ring, sq in. A T p equivalent area of cylinder, cone, and stiffening ring, sq in., where A TL p L Lt s 2 + L ct c 2 + A s for large end A TS p L st s 2 + L ct c 2 + A s for small end Bp factor determined from the applicable material chart in Subpart 3 of Section II, Part D for maximum design metal temperature, psi [see UG-20(c)] D L p outside diameter of large end of conical section under consideration, in. D o p outside diameter of cylindrical shell, in. (In conical shell calculations, the value of D s and D L should be used in calculations in place of D o depending on whether the small end D s, or large end D L, is being examined.) D s p outside diameter at small end of conical section under consideration, in. E 1 p efficiency of longitudinal joint in cylinder. For compression (such as at small end of cone), E 1 p 1.0 for butt welds. E 2 p efficiency of longitudinal joint in cone. For compression, E 2 p 1.0 for butt welds. E c p modulus of elasticity of cone material, psi E r p modulus of elasticity of stiffening ring material, psi E s p modulus of elasticity of shell material, psi E x p E c,e r, or E s f 1 p axial load at large end due to wind, dead load, etc., excluding pressure, lb / in. f 2 p axial load at small end due to wind, dead load, etc., excluding pressure, lb / in. Ip available moment of inertia of the stiffening ring cross section about its neutral axis parallel to the axis of the shell, in. 4 I p available moment of inertia of combined shellcone or ring-shell-cone cross section about its neutral axis parallel to the axis of the shell, in. 4 The nominal shell thickness t s shall be used, and the width of the shell which is taken as contributing to the moment of inertia of the combined section shall not be greater than 1.10 Dt s and shall be taken as lying one-half on each side of the cone-to-cylinder junction or of the centroid of the ring. Portions of the shell plate shall not be considered as contributing area to more than one stiffening ring. CAUTIONARY NOTE: Stiffening rings may be subject to lateral buckling. This should be considered in addition to the requirements for I s and I s [see U-2(g)]. I s p required moment of inertia of the stiffening ring cross section about its neutral axis parallel to the axis of the shell, in. 4 I s p required moment of inertia of the combined shell-cone or ring-shell-cone cross section about its neutral axis parallel to the axis of the shell, in. 4 If the stiffeners should be so located that the maximum permissible effective shell sections overlap on either or both sides of a stiffener, the effective shell section for that stiffener shall be shortened by one-half of each overlap. kp 1 when additional area of reinforcement is not required p y/s r E r when a stiffening ring is required, but k is not less than 1.0 Lp axial length of cone, in. 326

17 1-8 APPENDIX 1 MANDATORY 1-8 L c p length of cone between stiffening rings measured along surface of cone, in. For cones without intermediate stiffeners, L c p L 2 +(R L R s ) 2 L L p design length of a vessel section, in., taken as the largest of the following: (1) the center-to-center distance between the cone-to-large-shell junction and an adjacent stiffening ring on the large shell; (2) the distance between the cone-to-largeshell junction and one-third the depth of head on the other end of the large shell if no other stiffening rings are used. L s p design length of a vessel section, in., taken as the largest of the following: (1) the center-to-center distance between the cone-to-small-shell junction and adjacent stiffening ring on the small shell; (2) the distance between the cone-to-smallshell junction and one-third the depth of head on the other end of the small shell if no other stiffening rings are used. Pp external design pressure, psi Q L p algebraical sum of PR L /2 and f 1, lb/in. Q s p algebraical sum of PR s /2 and f 2, lb/in. R L p outside radius of large cylinder, in. R s p outside radius of small cylinder, in. S c p allowable stress of cone material at design temperature, psi S r p allowable stress of stiffening ring material at design temperature, psi S s p allowable stress of cylinder material at design temperature, psi tp minimum required thickness of cylinder at cone-to-cylinder junction [see UG-28(c)], in. t c p nominal thickness of cone at cone-to-cylinder junction, in. t r p minimum required thickness of cone at coneto-cylinder junction, in. t s p nominal thickness of cylinder at cone-to-cylinder junction, in. y p cone-to-cylinder factor p S s E s for stiffening ring on shell p S c E c for stiffening ring on cone p one-half the included (apex) angle of the cone at the center line of the head p value to indicate need for reinforcement at coneto-cylinder intersection having a half-apex angle 60 deg. When, no reinforcement is required at the junction (see Table 1-8.1). TABLE VALUES OF FOR JUNCTIONS AT THE LARGE CYLINDER FOR 60 deg. P/S s E , deg P/S s E , deg P/S s E Note (1), deg NOTE: (1) p 60 deg. for greater values of P/SE. (b) Reinforcement shall be provided at the junction of the cone with the large cylinder for conical heads and reducers without knuckles when the value of obtained from Table using the appropriate ratio P/S s E 1 is less than. Interpolation may be made in the Table. The required area of reinforcement shall be at least equal to that indicated by the following formula when Q L is in compression: A rl p kq LR L tan S s E PR L Q L Q L (1) At thelarge end of the cone-to-cylinder juncture, the PR L /2 term is in compression. When f 1 is in tension and the quantity is larger than the PR L /2 term, the design shall be in accordance with U-2(g). The calculated localized stresses at the discontinuity shall not exceed the stress values specified in 1-5(g)(1) and (2). The effective area of reinforcement can be determined in accordance with the following formula: A el p 0.55 D L t s (t s + t c /cos ) (2) Any additional area of stiffening which is required shall be situated within a distance of R L t s from the junction of the reducer and the cylinder. The centroid of the added area shall be within a distance of 0.25 R L t s from the junction. When the cone-to-cylinder or knuckle-to-cylinder juncture is a line of support, the moment of inertia for a stiffening ring at the large end shall be determined by the following procedure. Step 1. Assuming that the shell has been designed and D L,L L, and t are known, select a member to be used for the stiffening ring and determine cross-sectional 327

18 SECTION VIII DIVISION area A TL. Then calculate factor B using the following formula: where B p 3 4 F LD L A TL F L p PM + f 1 tan M p R L tan + L L R L 2 R 2 s 3R L tan Step 2. Enter the right-hand side of the applicable material chart in Subpart 3 of Section II, Part D for the material under consideration at the value of B determined by Step 1. If different materials are used for the shell and stiffening ring, use the material chart resulting in the larger value of A in Step 4 below. Step 3. Move horizontally to the left to the material / temperature line for the design metal temperature. For values of B falling below the left end of the material / temperature line, see Step 5 below. Step 4. Move vertically to the bottom of the chart and read the value of A. Step 5. For value of B falling below the left end of the material / temperature line for the design temperature, the value of A can be calculated using the formula A p 2B / E x. For value of B above the material/ temperature line for the design temperature, the design shall be either per U-2(g) or by changing the cone or cylinder configuration, stiffening ring location on the shell, and/or reducing the axial compressive force to reduce the B value to below or at the material/temperature line for the design temperature. Step 6. Compute the value of the required moment of inertia from the formulas for I s or I s. For the circumferential stiffening ring only, I s p AD L 2 A TL 14.0 For the shell-cone or ring-shell-cone section, I s p AD L 2 A TL 10.9 Step 7. Determine the available moment of inertia of the ring only I or the shell-cone or ring-shell-cone I. Step 8. When the ring only is used, I I s and when the shell-cone or ring-shell-cone is used, I I s If the equation is not satisfied, a new section with a larger moment of inertia must be selected, and the calculation shall be done again until the equation is met. The requirements of UG-29(b), (c), (d), (e), and (f) and UG-30 are to be met in attaching stiffening rings to the shell. (c) Reinforcement shall be provided at the junction of the conical shell of a reducer without a flare and the small cylinder. The required area of reinforcement shall be at least equal to that indicated by the following formula when Q s is in compression: A rs p kq sr s tan S s E 1 (3) At the small end of the cone-to-cylinder juncture, the PR s /2 term is in compression. When f 2 is in tension and the quantity is larger than the PR s /2 term, the design shall be in accordance with U-2(g). The calculated localized stresses at the discontinuity shall not exceed the stress values specified in 1-5(g)(1) and (2). The effective area of reinforcement can determined in accordance with the following formula: A es p 0.55 D s t s [(t s t) +(t c t r )/cos ] (4) Any additional area of stiffenerwhich is required shall be situated within a distance of R s t s from the junction, and the centroid of the added area shall be within a distance of 0.25 R s t s from the junction. When the cone-to-cylinder or knuckle-to-cylinder juncture is a line of support, the moment of inertia for a stiffening ring at the small end shall be determined by the following procedure. Step 1. Assuming that the shell has been designed and D s,l s, and t are known, select a member to be used for the stiffening ring and determine cross-sectional area A TS. Then calculate factor B using the following formula: B p 3 4 F sd s A TS 328

19 1-8 APPENDIX 1 MANDATORY 1-8 where F s p PN + f 2 tan Step 7. Determine the available moment of inertia of the ring only I or the shell-cone or ring-shell-cone I. Step 8. When the ring only is used, N p R s tan + L s R L 2 R 2 s 6R s tan Step 2. Enter the right-hand side of theapplicable material chart in Subpart 3 of Section II, Part D for the material under consideration at the value of B determined by Step 1. If different materials are used for the shell and stiffening ring, use the material chart resulting in the larger value of A in Step 4 below. Step 3. Move horizontally to the left to the material / temperature line for the design metal temperature. For values of B falling below the left end of the material / temperature line, see Step 5 below. Step 4. Move vertically to the bottom of the chart and read the value of A. Step 5. For values of B falling below the left end of the material / temperature line for the design temperature, the value of A can be calculated using the formula A p 2B / E x. For value of B above the material/ temperature line for the design temperature, the design shall be either per U-2(g) or by changing the cone or cylinder configuration, stiffening ring location on the shell, and/or reducing the axial compressive force to reduce the B value to below or at the material/ temperature line for the design temperature. Step 6. Compute the value of the required moment of inertia from the formulas for I s or I s. For the circumferential stiffening ring only, I s p AD s 2 A TS 14.0 For the shell-cone or ring-shell-cone section, I s p AD s 2 A TS 10.9 I I s and when the shell-cone or ring-shell-cone is used: I I s If the equation is not satisfied, a new section with a larger moment of inertia must be selected, and the calculation shall be done again until the equation is met. The requirements of UG-29(b), (c), (d), (e), and (f) and UG-30 are to be met in attaching stiffening rings to the shell. (d) Reducers not described in UG-36(e)(5), such as those made up of two or more conical frustums having different slopes, may be designed in accordance with (e). (e) When the half-apex angle is greater than 60 deg., cone-to-cylinder junctions without a knuckle may be used, with or without reinforcing rings, if the design is based on special analysis, such as the beam-onelastic-foundation analysis of Timoshenko, Hetenyi, or Watts and Lang. See U-2(g). The effect of shell and cone buckling on the required area and moment of inertia at the joint is to be taken into consideration in the analysis. When such an analysis is made, the calculated localized stresses at the discontinuity shall not exceed the following values. (1) (Membrane hoop stress) + (average discontinuity hoop stress) shall not be greater than SE. (2) (Membrane longitudinal stress) + (discontinuity longitudinal stress due to bending) shall not be greater than 4SE, where the average discontinuity hoop stress is the average hoop stress across the wall thickness due to the discontinuity at the junction, disregarding the effect of Poisson s ratio times the longitudinal stress at the surfaces. 329

20

21 APPENDIX 2 RULES FOR BOLTED FLANGE CONNECTIONS WITH RING TYPE GASKETS GENERAL 2-1 SCOPE (a) The rules in Appendix 2 apply specifically to the design of bolted flange connections with gaskets that are entirely within the circle enclosed by the bolt holes and with no contact outside this circle, and are to be used in conjunction with the applicable requirements in Subsections A, B, and C of this Division. These rules are not to be used for the determination of the thickness of supported or unsupported tubesheets integral with a bolting flange as illustrated in Fig. UW sketches (h) through (l) or Fig. UW-13.3 sketch (c). Appendix S provides discussion on Design Considerations for Bolted Flanged Connections. These rules provide only for hydrostatic end loads and gasket seating. The flange design methods outlined in 2-4 through 2-8 are applicable to circular flanges under internal pressure. Modifications of these methods are outlined in 2-9 and 2-10 for the design of split and noncircular flanges. See 2-11 for flanges with ring type gaskets subject to external pressure, 2-12 for flanges with nut-stops, and 2-13 for reverse flanges. Proper allowance shall be made if connections are subject to external loads other than external pressure. (b) The design of a flange involves the selection of the gasket (material, type, and dimensions), flange facing, bolting, hub proportions, flange width, and flange thickness. See Note 1, 2-5(c)(1). Flange dimensions shall be such that the stresses in the flange, calculated in accordance with 2-7, do not exceed the allowable flange stresses specified in 2-8. All calculations shall be made on dimensions in the corroded condition. (c) It is recommended that bolted flange connections conforming to the standards listed in UG-44 be used for connections to external piping. These standards may be used for other bolted flange connections within the limits of size in the standards and the pressure temperature ratings permitted in UG-44. The ratings in these standards are based on the hub dimensions given or on the minimum specified thickness of flanged fittings of integral construction. Flanges fabricated from rings may be used in place of the hub flanges in these standards provided that their strength, calculated by the rules in this Appendix, is not less than that calculated for the corresponding size of hub flange. (d) Except as otherwise provided in (c) above, bolted flange connections for unfired pressure vessels shall satisfy the requirements in this Appendix. (e) The rules of this Appendix should not be construed to prohibit the use of other types of flanged connections provided they are designed in accordance with good engineering practice and method of design is acceptable to the Inspector. Some examples of flanged connections which might fall in this category are as follows: (1) flanged covers as shown in Fig. 1-6; (2) bolted flanges using full-face gaskets; (3) flanges using means other than bolting to restrain the flange assembly against pressure and other applied loads. 2-2 MATERIALS (a) Materials used in the construction of bolted flange connections shall comply with the requirements given in UG-4 through UG-14. (b) Flanges made from ferritic steel and designed in accordance with this Appendix shall be given a normalizing or full-annealing heat treatment when the thickness of the flange section exceeds 3 in. (c) Material on which welding is to be performed shall be proved of good weldable quality. Satisfactory qualification of the welding procedure under Section IX is considered as proof. Welding shall not be performed on steel that has a carbon content greater than 0.35%. All welding on flange connections shall comply 331

22 SECTION VIII DIVISION with the requirements for postweld heat treatment given in this Division. (d) Fabricated hubbed flanges shall be in accordance with the following. (1) Hubbed flanges may be machined from a hot rolled or forged billet or forged bar. The axis of the finished flange shall be parallel to the long axis of the original billet or bar. (This is not intended to imply that the axis of the finished flange and the original billet must be concentric.) (2) Hubbed flanges [except as permitted in (1) above] shall not be machined from plate or bar stock material unless the material has been formed into a ring, and further provided that: (a) in a ring formed from plate, the original plate surfaces are parallel to the axis of the finished flange. (This is not intended to imply that the original plate surface be present in the finished flange.) (b) the joints in the ring are welded butt joints that conform to the requirements of this Division. Thickness to be used to determine postweld heat treatment and radiography requirements shall be the lesser of t or (A B) 2 where these symbols are as defined in 2-3. (3) The back of the flange and the outer surface of the hub are examined by either the magnetic particle method as per Appendix 6 or the liquid penetrant method as per Appendix 8. (e) Bolts, studs, nuts, and washers shall comply with the requirements in this Division. It is recommended that bolts and studs have a nominal diameter of not less than 1 2 in. If bolts or studs smaller than 1 2 in. are used, ferrous bolting material shall be of alloy steel. Precautions shall be taken to avoid over-stressing small-diameter bolts. 2-3 NOTATION The symbols described below are used in the formulas for the design of flanges (see also Fig. 2-4): Ap outside diameter of flange or, where slotted holes extend to the outside of the flange, the diameter to the bottom of the slots, in. A b p cross-sectional area of the bolts using the root diameter of the thread or least diameter of unthreaded position, if less, sq in. A m p total required cross-sectional area of bolts, taken as the greater of A m1 and A m2,sqin. A m1 p total cross-sectional area of bolts at root of thread or section of least diameter under stress, required for the operating conditions, sq in. p W m1 / S b A m2 p total cross-sectional area of bolts at root of thread or section of least diameter under stress, required for gasket seating, sq in. p W m2 / S a Bp inside diameter of flange, in. When B is less than 20g 1, it will be optional for the designer to substitute B 1 for B in the formula for longitudinal stress S H. B 1 p B + g 1, in., for loose type flanges and for integral type flanges that have calculated values h / h o and g 1 / g o which would indicate an f value of less than 1.0, although the minimum value of f permitted is 1.0. B 1 p B + g o, in., for integral type flanges when f is equal to or greater than one bp effective gasket or joint-contact-surface seating width, in. [see Note 1, 2-5(c)(1)] b o p basic gasket seating width, in. (from Table 2-5.2) Cp bolt-circle diameter, in. cp basic dimension used for the minimum sizing of welds, in., equal to t n or t x, whichever is less dp factor, in. 3 d p U V h og o 2 for integral type flanges d p U V L h o g o 2 for loose type flanges ep factor, in. 1 e p F h o for integral type flanges e p F L h o for loose type flanges Fp factor for integral type flanges (from Fig ) F L p factor for loose type flanges (from Fig ) fp hub stress correction factor for integral flanges from Fig (When greater than one, this is the ratio of the stress in the small end of hub to the stress in the large end.) (For values below limit of figure, use f p 1.) Gp diameter, in., at location of gasket load reaction. Except as noted in sketch (1) of Fig. 2-4, G is defined as follows (see Table 2-5.2): When b o 1 4 in., G p mean diameter of gasket contact face, in. 332

23 2-3 APPENDIX 2 MANDATORY 2-3 When b o > 1 4 in., G p outside diameter of gasket contact face less 2b, in. g o p thickness of hub at small end, in. g 1 p thickness of hub at back of flange, in. Hp total hydrostatic end force, lb p 0.785G 2 P H D p hydrostatic end force on area inside of flange, lb p 0.785B 2 P H G p gasket load (difference between flange design bolt load and total hydrostatic end force), lb p W H H p p total joint-contact surface compression load, lb p 2b 3.14 GmP H T p difference between total hydrostatic end force and the hydrostatic end force on area inside of flange, lb p H H D hp hub length, in. h D p radial distance from the bolt circle, to the circle on which H D acts, as prescribed in Table 2-6, in. h G p radial distance from gasket load reaction to the bolt circle, in. p (C G)/2 h o p factor, in. p Bg o h T p radial distance from the bolt circle to the circle on which H T acts as prescribed in Table 2-6, in. Kp ratio of outside diameter of flange to inside diameter of flange p A/B Lp factor p te +1 T + t 3 d M D p component of moment due to H D, in.-lb p H D h D M G p component of moment due to H G, in.-lb p H G h G M 0 p total moment acting upon the flange, for the operating conditions or gasket seating as may apply, in.-lb (see 2-6) M T p component of moment due to H T, in.-lb p H T h T mp gasket factor, obtain from Table [see Note 1, 2-5(c)(1)] Np width, in., used to determine the basic gasket seating with b o, based upon the possible contact width of the gasket (see Table 2-5.2) Pp internal design pressure (see UG-21), psi. For flanges subject to external design pressure, see Rp radial distance from bolt circle to point of intersection of hub and back of flange, in. For integral and hub flanges, R p C B g 1 2 S a p allowable bolt stress at atmospheric temperature, psi (see UG-23) S b p allowable bolt stress at design temperature, psi (see UG-23) S f p allowable design stress for material of flange at design temperature (operating condition) or atmospheric temperature (gasket seating), as may apply, psi (see UG-23) S n p allowable design stress for material of nozzle neck, vessel or pipe wall, at design temperature (operating condition) or atmospheric temperature (gasket seating), as may apply, psi (see UG- 23) S H p calculated longitudinal stress in hub, psi S R p calculated radial stress in flange, psi S T p calculated tangential stress in flange, psi Tp factor involving K (from Fig ) tp flange thickness, in. t n p nominal thickness of shell or nozzle wall to which flange or lap is attached, in. t x p two times the thickness g 0, when the design is calculated as an integral flange, in., or two times the thickness, in., of shell nozzle wall required for internal pressure, when the design is calculated as a loose flange, but not less than 1 4 in. 333

24 SECTION VIII DIVISION Up factor involving K (from Fig ) Vp factor for integral type flanges (from Fig ) V L p factor for loose type flanges (from Fig ) Wp flange design bolt load, for the operating conditions or gasket seating, as may apply, lb [see 2-5(e)] W m1 p minimum required bolt load for the operating conditions, lb [see 2-5(c)]. For flange pairs used to contain a tubesheet for a floating head for a U-tube type of heat exchangers, or for any other similar design, W m1 shall be the larger of the values as individually calculated for each flange, and that value shall be used for both flanges. W m2 p minimum required bolt load for gasket seating, lb [see 2-5(c)] wp width, in., used to determine the basic gasket seating width b 0, based upon the contact width between the flange facing and the gasket (see Table 2-5.2) Yp factor involving K (from Fig ) yp gasket or joint-contact-surface unit seating load, psi [see Note 1, 2-5(c)] Zp factor involving K (from Fig ) 2-4 CIRCULAR FLANGE TYPES (a) For purposes of computation, there are three types: (1) Loose Type Flanges. This type covers those designs in which the flange has no direct connection to the nozzle neck, vessel, or pipe wall, and designs where the method of attachment is not considered to give the mechanical strength equivalent of integral attachment. See Fig. 2-4 sketches (1), (1a), (2), (2a), (3), (3a), (4), and (4a) for typical loose type flanges and the location of the loads and moments. Welds and other details of construction shall satisfy the dimensional requirements given in Fig. 2-4 sketches (1), (1a), (2), (2a), (3), (3a), (4), and (4a). (2) Integral Type Flanges. This type covers designs where the flange is cast or forged integrally with the nozzle neck, vessel or pipe wall, butt welded thereto, or attached by other forms of arc or gas welding of such a nature that the flange and nozzle neck, vessel or pipe wall is considered to be the equivalent of an integral structure. In welded construction, the nozzle neck, vessel, or pipe wall is considered to act as a hub. See Fig. 2-4 sketches (5), (6), (6a), (6b), and (7) for typical integral type flanges and the location of the loads and moments. Welds and other details of construction shall satisfy the dimensional requirements given in Fig. 2-4 sketches (5), (6), (6a), (6b), and (7). (3) Optional Type Flanges. This type covers designs where the attachment of the flange to the nozzle neck, vessel or pipe wall is such that the assembly is considered to act as a unit, which shall be calculated as an integral flange, except that for simplicity the designer may calculate the construction as a loose type flange provided none of the following values is exceeded: g 0 p 5 8 in. B/g 0 p 300 P p 300 psi operating temperature p 700 F See Fig. 2-4 sketches (8), (8a), (9), (9a), (10), (10a), and (11) for typical optional type flanges. Welds and other details of construction shall satisfy the dimensional requirements given in Fig. 2-4 sketches (8), (8a), (9), (9a), (10), (10a), and (11). 2-5 BOLT LOADS (a) General Requirements (1) In the design of a bolted flange connection, calculations shall be made for each of the two design conditions of operating and gasket seating, and the more severe shall control. (2) In the design of flange pairs used to contain a tubesheet of a heat exchanger or any similar design where the flanges and / or gaskets may not be the same, loads must be determined for the most severe condition of operating and / or gasket seating loads applied to each side at the same time. This most severe condition may be gasket seating on one flange with operating on the other, gasket seating on each flange at the same time, or operating on each flange at the same time. Although no specific rules are given for the design of the flange pairs, after the loads for the most severe conditions are determined, calculations shall be made for each flange following the rules of Appendix 2. (b) Design Conditions (1) Operating Conditions. The conditions required to resist the hydrostatic end force of the design pressure tending to part the joint, and to maintain on the gasket or joint-contact surface sufficient compression to assure a tight joint, all at the design temperature. The minimum load is a function of the design pressure, the gasket A99 334

25 2-5 APPENDIX 2 MANDATORY 2-5 FIG. 2-4 TYPES OF FLANGES material, and the effective gasket or contact area to be kept tight under pressure, per Formula (1) in (c)(1) below, and determines one of the two requirements for the amount of the bolting A m1. This load is also used for the design of the flange, per Formula (3) in (d) below. (2) Gasket Seating. The conditions existing when the gasket or joint-contact surface is seated by applying an initial load with the bolts when assembling the joint, at atmospheric temperature and pressure. The minimum initial load considered to be adequate for proper seating is a function of the gasket material, and the effective gasket or contact area to be seated, per Formula (2) in (c)(2) below, and determines the other of the two requirements for the amount of bolting A m2. For the design of the flange, this load is modified per Formula (4) in (d) below to take account of the operating conditions, when these govern the amount of bolting required A m, as well as the amount of bolting actually provided A b. (c) Required Bolt Loads. The flange bolt loads used in calculating the required cross-sectional area of bolts shall be determined as follows. (1) The required bolt load for the operating conditions W m1 shall be sufficient to resist the hydrostatic end force H exerted by the maximum allowable working pressure on the area bounded by the diameter of gasket reaction, and, in addition, to maintain on the gasket or joint-contact surface a compression load H p, which experience has shown to be sufficient to assure a tight joint. (This compression load is expressed as a multiple m of the internal pressure. Its value is a function of the gasket material and construction. See Note 1.) NOTE 1: Tables and give a list of many commonly used gasket materials and contact facings, with suggested values of m, b, and y that have proved satisfactory in actual service. These 335

26 SECTION VIII DIVISION FIG. 2-4 TYPES OF FLANGES (CONT D) values are suggested only and are not mandatory. Values that are too low may result in leakage at the joint without affecting the safety of the design. The primary proof that the values are adequate is the hydrostatic test. The required bolt load for the operating conditions W m1 is determined in accordance with Formula (1). W m1 p H + H p p 0.785G 2 P +(2b 3.14GmP) (1) (2) Before a tight joint can be obtained, it is necessary to seat the gasket or joint-contact surface properly by applying a minimum initial load (under atmospheric temperature conditions without the presence of internal pressure), which is a function of the gasket material and the effective gasket area to be seated. The minimum initial bolt load required for this purpose W m2 shall be determined in accordance with Formula (2). W m2 p 3.14bGy (2) For flange pairs used to contain a tubesheet for a floating head for a U-tube type of heat exchanger, or for any other similar design, and where the flanges and/or gaskets are not the same, W m2 shall be the larger of the values obtained from Formula (2) as individually calculated for each flange and gasket, and that value shall be used for both flanges. The need for providing sufficient bolt load to seat the gasket or joint-contact surfaces in accordance with Formula (2) will prevail on many low-pressure designs and with facings and materials that require a high seating load, and where the bolt load computed by Formula (1) for the operating conditions is insufficient to seat the joint. Accordingly, it is necessary to furnish bolting and to pretighten the bolts to provide a bolt load sufficient to satisfy both of these requirements, each one being individually investigated. When Formula (2) governs, flange proportions will be a function of the bolting instead of internal pressure. (3) Bolt loads for flanges using gaskets of the self-energizing type differ from those shown above. (a) The required bolt load for the operating conditions W m shall be sufficient to resist the hydrostatic end force H exerted by the maximum allowable working pressure on the area bounded by the outside diameter of the gasket. H p is to be considered as 0 for all self- 336

27 APPENDIX 2 MANDATORY Fig. 2-4 FIG. 2-4 TYPES OF FLANGES (CONT D) 337

28 Table SECTION VIII DIVISION 1 TABLE GASKET MATERIALS AND CONTACT FACINGS 1 Gasket Factors m for Operating Conditions and Minimum Design Seating Stress y Min. Design Gasket Seating Facing Sketch Factor Stress y, and Column Gasket Material m psi Sketches in Table Self-energizing types (0 rings, metallic, elastomer, other gasket types considered as self-sealing) Elastomers without fabric or high percent of asbestos fiber: Below 75A Shore Durometer (1a),(1b),(1c),(1d), 75A or higher Shore Durometer (4),(5); Column II Asbestos with suitable binder for operating conditions: 8 in. thick in. thick in. thick (1a),(1b),(1c),(1d), (4),(5); Column II Elastomers with cotton fabric insertion (1a),(1b),(1c),(1d), (4),(5); Column II Elastomers with asbestos fabric insertion (with or without wire reinforcement): 3-ply ply (1a),(1b),(1c),(1d), (4),(5); Column II 1-ply Vegetable fiber (1a),(1b),(1c),(1d), (4)(5); Column II Spiral-wound metal, asbestos filled: Carbon ,000 (1a),(1b); Column II Stainless, Monel, and nickel-base ,000 alloys Corrugated metal, asbestos inserted, or corrugated metal, jacketed asbestos filled: Soft aluminum Soft copper or brass Iron or soft steel (1a),(1b); Column II Monel or 4% 6% chrome Stainless steels and nickel-base alloys

29 APPENDIX 2 MANDATORY Table TABLE (CONT D) GASKET MATERIALS AND CONTACT FACINGS 1 Gasket Factors m for Operating Conditions and Minimum Design Seating Stress y Min. Design Gasket Seating Facing Sketch Factor Stress y, and Column Gasket Material m psi Sketches in Table Corrugated metal: Soft aluminum Soft copper or brass Iron or soft steel Monel or 4% 6% chrome Stainless steels and nickel-base alloys (1a),(1b),(1c),(1d); Column II Flat metal, jacketed asbestos filled: Soft aluminum Soft copper or brass Iron or soft steel Monel % 6% chrome Stainless steels and nickel-base alloys (1a),(1b),(1c), 2 (1d) 2 ;(2) 2 ; Column II Grooved metal: Soft aluminum Soft copper or brass Iron or soft metal Monel or 4% 6% chrome Stainless steels and nickel-base alloys ,100 (1a),(1b),(1c),(1d), (2),(3); Column II Solid flat metal: Soft aluminum Soft copper or brass ,000 (1a),(1b),(1c),(1d), Iron or soft steel ,000 (2),(3),(4),(5); Monel or 4% 6% chrome ,800 Column I Stainless steels and nickel-base alloys ,000 Ring joint: Iron or soft steel ,000 Monel or 4% 6% chrome ,800 Stainless steels and nickel-base alloys ,000 (6); Column I NOTES: (1) This Table gives a list of many commonly used gasket materials and contact facings with suggested design values of m and y that have generally proved satisfactory in actual service when using effective gasket seating width b given in Table The design values and other details given in this Table are suggested only and are not mandatory. (2) The surface of a gasket having a lap should not be against the nubbin. 339

30 Table SECTION VIII DIVISION 1 TABLE EFFECTIVE GASKET WIDTH 2 (1a) Basic Gasket Seating Width b o Facing Sketch (Exaggerated) Column I Column II (1b) N 2 N 2 See Note (1) (1c) (1d) w N w + T 2 ; w + N max 4 w + T 2 ; w + N max 4 See Note (1) w N (2) 1 64 in. nubbin w N/2 w + N 4 w +3N 8 (3) 1 64 in. nubbin w N/2 N 4 3N 8 (4) See Note (1) 3N 8 7N 16 (5) See Note (1) N 4 3N 8 (6) w 8... Effective Gasket Seating Width, b b p b o, when b o 1 4 in.; b p 0.5 b o, when b o > 1 4 in. NOTE: (1) Where serrations do not exceed 1 64 in. depth and 1 32 in. width spacing, sketches (1b) and (1d) shall be used. (2) The gasket factors listed only apply to flanged joints in which the gasket is contained entirely within the inner edges of the bolt holes. 340

31 2-5 APPENDIX 2 MANDATORY 2-7 TABLE 2-6 MOMENT ARMS FOR FLANGE LOADS UNDER OPERATING CONDITIONS h D h T h G For gasket seating, W p (A m + A b )S a 2 (4) Integral type flanges [see Fig. 2 4 sketches (5), R + 0.5g 1 R + g 1 + h G 2 (6), (6a), (6b), and (7)]; and optional type flanges calculated as integral type [see Fig. 2-4 sketches (8), (8a), (9), (9a), (10), (10a), and (11)] Loose type, except lapjoint flanges [see Fig. 2-4 sketches (2), (2a), (3), (3a), (4), and (4a)]; and optional type flanges calculated as loose type [see Fig. 2-4 sketches (8), (8a), (9), (9a), (10), (10a), and (11)] Lap-type flanges [see Fig. 2-4 sketches (1) and (1a)] C B 2 C B 2 h D + h G 2 C G 2 C G 2 C G 2 C G 2 energizing gaskets except certain seal configurations which generate axial loads which must be considered. (b) W m2 p 0. Self-energizing gaskets may be considered to require an inconsequential amount of bolting force to produce a seal. Bolting, however, must be pretightened to provide a bolt load sufficient to withstand the hydrostatic end force H. (d) Total Required and Actual Bolt Areas, A m and A b. The total cross-sectional area of bolts A m required for both the operating conditions and gasket seating is the greater of the values for A m1 and A m2 where A m1 p W m1 / S b and A m2 p W m2 / S a. A selection of bolts to be used shall be made such that the actual total crosssectional area of bolts A b will not be less than A m. (e) Flange Design Bolt Load W. The bolt loads used in the design of the flange shall be the values obtained from Formulas (3) and (4). For operating conditions, W p W m1 (3) S a used in Formula (4) shall be not less than that tabulated in the stress tables (see UG-23). In addition to the minimum requirements for safety, Formula (4) provides a margin against abuse of the flange from overbolting. Since the margin against such abuse is needed primarily for the initial, bolting-up operation which is done at atmospheric temperature and before application of internal pressure, the flange design is required to satisfy this loading only under such conditions (see Note 2). NOTE 2: Where additional safety against abuse is desired, or where it is necessary that the flange be suitable to withstand the full available bolt load A b S a, the flange may be designed on the basis of this latter quantity. 2-6 FLANGE MOMENTS In the calculation of flange stress, the moment of a load acting on the flange is the product of the load and its moment arm. The moment arm is determined by the relative position of the bolt circle with respect to that of the load producing the moment (see Fig. 2-4). No consideration shall be given to any possible reduction in moment arm due to cupping of the flanges or due to inward shifting of the line of action of the bolts as a result thereof. For the operating conditions, the total flange moment M o is the sum of the three individual moments M D, M T, and M G, as defined in 2-3 and based on the flange design load of Formula (3) with moment arms as given in Table 2-6. For gasket seating, the total flange moment M o is based on the flange design bolt load of Formula (4), which is opposed only by the gasket load, in which case M o p W (C G) CALCULATION OF FLANGE STRESSES The stresses in the flange shall be determined for both the operating conditions and gasket seating condition, whichever controls, in accordance with the following formulas: (5) 341

32 SECTION VIII DIVISION FIG VALUES OF T, U, Y, AND Z (Terms Involving K) (a) for integral type flanges [Fig. 2-4 sketches (5), (6), (6a), (6b), and (7)]; for optional type flanges calculated as integral type [Fig. 2-4 sketches (8), (8a), (9), (9a), (10), (10a), and (11)]; and for loose type flanges with a hub which is considered [Fig. 2-4 sketches (1), (1a), (2), (2a), (3), (3a), (4), and (4a)]: Longitudinal hub stress S H p fm o Lg 1 2 B (6) Tangential flange stress S T p YM o t 2 B ZS R (8) (b) for loose type flanges without hubs and loose type flanges with hubs which the designer chooses to calculate without considering the hub [Fig. 2-4 sketches (1), (1a), (2), (2a), (3), (3a), (4), and (4a)] and optional type flanges calculated as loose type [Fig. 2-4 sketches (8), (8a), (9), (9a), (10), (10a), and (11)]: Radial flange stress S T p YM o t 2 B (9) S R p (1.33te +1)M o Lt 2 B (7) S R p 0 S H p 0 342

33 2-8 APPENDIX 2 MANDATORY 2-8 FIG VALUES OF F (Integral Flange Factors) 2-8 ALLOWABLE FLANGE DESIGN STRESSES (a) The flange stresses calculated by the formulas in 2-7 shall not exceed the following values: (1) longitudinal hub stress S H not greater than S f for cast iron 1 and, except as otherwise limited by (1)(a) and (1)(b) below, not greater than 1.5 S f for materials other than cast iron: (a) longitudinal hub stress S H not greater than the smaller of 1.5S f or 1.5S n for optional type flanges designed as integral [Fig. 2-4 sketches (8), (8a), (9), (9a), (10), (10a), and (11)], also integral type [Fig. 2-4 sketch (7)] where the neck material constitutes the hub of the flange; (b) longitudinal hub stress S H not greater than the smaller of 1.5S f or 2.5S n for integral type flanges 1 When the flange material is cast iron, particular care should be taken when tightening the bolts to avoid excessive stress that may break the flange. The longitudinal hub stress has been limited to S f in order to minimize any cracking of flanges. An attempt should be made to apply no greater torque than is needed to assure tightness during the hydrostatic test. with hub welded to the neck, pipe or vessel wall [Fig. 2-4 sketches (6), (6a), and (6b)]. (2) radial flange stress S R not greater than S f ; (3) tangential flange stress S T not greater than S f ; (4) also (S H + S R ) / 2 not greater than S f and (S H + S T ) / 2 not greater than S f. (b) For hub flanges attached as shown in Fig. 2-4 sketches (2), (2a), (3), (3a), (4), and (4a), the nozzle neck, vessel or pipe wall shall not be considered to have any value as a hub. (c) In the case of loose type flanges with laps, as shown in Fig. 2-4 sketches (1) and (1a), where the gasket is so located that the lap is subjected to shear, the shearing stress shall not exceed 0.8 S n for the material of the lap, as defined in 2-3. In the case of welded flanges, shown in Fig. 2-4 sketches (3), (3a), (4), (4a), (7), (8), (8a), (9), (9a), (10), and (10a) where the nozzle neck, vessel, or pipe wall extends near to the flange face and may form the gasket contact face, the shearing stress carried by the welds shall not exceed 0.8 S n. The shearing stress shall be calculated on the basis of W m1 or W m2 as defined in 2-3, whichever is 343

34 SECTION VIII DIVISION FIG VALUES OF V (Integral Flange Factors) greater. Similar cases where flange parts are subjected to shearing stress shall be governed by the same requirements. 2-9 SPLIT LOOSE FLANGES 2 Loose flanges split across a diameter and designed under the rules given in this Appendix may be used under the following provisions. (a) When the flange consists of a single split flange or flange ring, it shall be designed as if it were a solid flange (without splits), using 200% of the total moment M o as defined in 2-6. (b) When the flange consists of two split rings each ring shall be designed as if it were a solid flange (without splits), using 75% of the total moment M o as 2 Loose flanges of the type shown in Fig. 2-4 sketch (1) are of the split design when it is necessary to install them after heat treatment of a stainless steel vessel, or when for any reason it is desired to have them completely removable from the nozzle neck or vessel. defined in 2-6. The pair of rings shall be assembled so that the splits in one ring shall be 90 deg. from the splits in the other ring. (c) The splits should preferably be midway between bolt holes NONCIRCULAR SHAPED FLANGES WITH CIRCULAR BORE The outside diameter A for a noncircular flange with a circular bore shall be taken as the diameter of the largest circle, concentric with the bore, inscribed entirely within the outside edges of the flange. Bolt loads and moments, as well as stresses, are then calculated as for circular flanges, using a bolt circle drawn through the centers of the outermost bolt holes. 344

35 2-11 APPENDIX 2 MANDATORY 2-11 FIG VALUES OF F L (Loose Hub Flange Factors) FIG VALUES OF V L (Loose Hub Flange Factors) A FLANGES SUBJECT TO EXTERNAL PRESSURES (a) The design of flanges for external pressure only [see UG-99(f)] 3 shall be based on the formulas given in 2-7 for internal pressure except that for operating conditions: For gasket seating, M o p H D (h D h G )+H T (h T h G ) (10) where W p A m2 + A b 2 S a H D p 0.785B 2 P e H T p H H D H p 0.785G 2 P e (11a) (11b) (11c) (11d) M o p Wh G (11) 3 When internal pressure occurs only during the required pressure test, the design may be based on external pressure, and auxiliary devices such as clamps may be used during the application of the required test pressure. P e pexternal design pressure, psi See 2-3 for definitions of other symbols. S a used in Formula (11a) shall be not less than that tabulated in the stress tables (see UG-23). (b) When flanges are subject at different times during operation to external or internal pressure, the design shall satisfy the external pressure design requirements 345

36 SECTION VIII DIVISION FIG VALUES OF f (Hub Stress Correction Factor) given in (a) above and the internal pressure design requirements given elsewhere in this Appendix. NOTE: The combined force of external pressure and bolt loading may plastically deform certain gaskets to result in loss of gasket contact pressure when the connection is depressurized. To maintain a tight joint when the unit is repressurized, consideration should be given to gasket and facing details so that excessive deformation of the gasket will not occur. Joints subject to pressure reversals, such as in heat exchanger floating heads, are in this type of service FLANGES WITH NUT-STOPS (a) When flanges are designed per this Appendix, or are fabricated to the dimensions of ASME/ ANSI B16.5 or other acceptable standards [see UG-44(a)], except that the dimension R is decreased to provide a nut-stop, the fillet radius relief shall be as shown in Fig. 2-4 sketches (12) and (12a) except that: (1) for flanges designed to this Appendix, the dimension g 1 must be the lesser of 2t (t from UG-27) or 4r, but in no case less than 1 2 in., where rp the radius of the undercut (2) for ASME/ANSI B16.5 or other standard flanges, the dimension of the hub g o shall be increased as necessary to provide a nut-stop REVERSE FLANGES (a) Flanges with the configuration as indicated in Fig shall be designed as integral reverse flanges and those in Fig shall be designed as loose A99 346