Read ASME_Ch21_p027-111.qxd text version

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 27

CHAPTER

21

SECTION VIII­DIVISION 1: RULES FOR CONSTRUCTION OF PRESSURE VESSELS

Urey R. Miller and Thomas P. Pastor

21.1 INTRODUCTION 21.2

The history of the industrial revolution is not favorable from the perspective of safety and the protection of the general public from the failure of boilers and other pressure equipment. Because of the frequent occurrences of boiler and pressure vessel failures, the American Society of Mechanical Engineers appointed a committee to develop a boiler code in 1911. The objective of the ASME Boiler & Pressure Vessel Code is the same today as it was in 1911. The objective is to provide rules defining the minimum requirements to assure that boilers and pressure vessels are constructed in a manner that they may be safely operated. Section VIII, Division 1 of the ASME Boiler & Pressure Vessel Code provides rules for the construction of pressure vessels. Since the ASME Code is a "safety" code , it is not a purchase specification that defines all the criteria that are required for specific applications. This becomes clear when it is recognized that Section VIII, Division 1, is written to cover a wide range of industrial and commercial pressure vessel applications. For example, Section VIII, Division 1, is applicable to small compressed air receivers that are sold commercially to the general public as well as to very large pressure vessels needed by the petrochemical and refining industry. The user of the equipment must define the requirements that are needed for a specific application. For example, some details that may be appropriate for a small, low-pressure piece of equipment may not be appropriate for a large, high-pressure piece of equipment subjected to severe service conditions. Thus, it is necessary that the user of the Codes be knowledgeable and experienced in the principles of pressure vessel engineering in order to assure that the selected requirements and details are appropriate for the specific service conditions that the pressure vessel is expected to experience. This chapter provides an overview to each of the parts of Section VIII, Division 1, of the ASME Boiler & Pressure Vessel Code. The intent is to provide a broad perspective for the reader to have better understanding of the Code's intent, and to point out, by example, some of the subtleties that may not be evident. It is not the objective of this chapter to provide the reader with a detailed "how to" handbook. There are many handbooks on pressure vessel that provide a clear illustration of how to design pressure vessels [10].

SECTION VIII, DIVISION 1, FOREWORD

Section VIII, Division 1, is intended for the construction of new pressure vessels. The ASME Code, in and of itself, does not have a mandate for its use for the construction of pressure vessels. Its use is implemented by the enactment of laws in those jurisdictions that require the use of code rules. There are numerous states and cities that require pressure vessels installed within their jurisdiction to satisfy all the rules of the ASME Boiler & Pressure Vessel Code. A summary of the jurisdictional requirements for boilers and pressure vessels may be found in Synopsis of Boiler and Pressure Vessel Laws [1]. Likewise, there are numerous states that have not enacted laws mandating the use of the Code for pressure vessels. In those jurisdictions, the use of the ASME Code becomes a contractual requirement between the purchaser and the manufacturer. However, it should be pointed out that for facilities that fall under the scope of Occupational Safety and Health Administration (OSHA) 1910.119, Process Safety Management for Hazardous Materials [2], use of the ASME Code should be considered the de facto standard. Owners of facilities covered by OSHA 1910.119 [2] are obligated to demonstrate that process equipment complies with "Recognized and Generally Accepted Good Engineering Practice" (RAGAGEP). If process vessels in OSHA 1910.119 covered facilities do not meet all the provisions of the ASME Code (including the application of the code stamp), the owner has accepted an onerous task if he is ever required to justify that these pressure vessels meet the RAGAGEP criteria. The ASME Code is generally accepted in the United States (and in many foreign countries) as the recognized minimum safety standard for the construction of pressure vessels. Once the code stamp is applied, the provisions of the ASME Code are satisfied. Anything done to the pressure vessel after the application of the code stamp is outside the scope of the ASME Code. It is a commonly held belief that any modifications done to a pressure vessel that has been in service must satisfy the provisions of the ASME Code in order to maintain "Code status." This is not an entirely accurate statement. There are other non-ASME practices and guidelines that are applicable for the repair, alteration, and rerating of pressure vessels. These guidelines, in fact, may reference the Section VIII, Division 1 requirements; however,

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 28

28 · Chapter 21

they may have requirements or offer alternatives different from Section VIII that are deemed appropriate for the modification of pressure vessels already in service. Once the ASME Code stamp is applied, the rules of Section VIII are satisfied and the vessel no longer falls under its scope for any additional work done on it. The available guidelines for the repair, alteration, or rerating of existing pressure vessels are API 510 [4] and National Board Inspection Code [5]. Again, the required application of NBIC or API-510 will be as defined by a jurisdiction. Otherwise, the use of these guidelines should be as defined by their respective scopes. One of the most important statements in the Foreword is as under: The Code is not a handbook and cannot replace education, experience, and the use of engineering judgment. The phrase engineering judgment refers to technical judgments made by knowledgeable designers experienced in the application of the Code. Engineering judgments must be consistent with Code philosophy and such judgments must never be used to override mandatory requirements of specific prohibitions of the Code. This statement is important because it cautions that people who do not understand the engineering principles related to pressure vessel design and fabrication should not apply the principles contained in the document. The code sets the minimum requirements that the Committee has deemed necessary for a pressure vessel to perform in a safe and reliable manner. However, it must be recognized that rules of Section VIII, Division 1 are applied for pressure vessels in a vast diversity of services. The chemical, petrochemical, refinery, and pharmaceutical industries have unique conditions that may make certain details allowed by Section VIII, Division 1, inappropriate for long-term, reliable operation of pressure vessels in such services.

pressure vessel with a design pressure less than 15 psi may still be code stamped if all the applicable code rules are satisfied. (2) It is not the responsibility of ASME to decide when a pressure vessel should or should not be constructed in accordance with VIII-1 and stamped. Instead, it is the responsibility of the legal jurisdiction or regulatory body in force at the location where the equipment will be operated to decide whether or not the vessel should be code stamped. In fact, the enforcement or regulatory body has the right to establish the mandatory applicability of the code in whole or in part, as well as require code compliance even for equipment that would otherwise be outside its scope. 21.3.1.1 Scope: Vessel Classification Some of the more important vessel scope definitions are provided below: In accordance with paragraph U-1(c)(2), Division 1 of Section VIII is not intended to apply for vessels that are (a) included within the scope of other sections of the ASME Boiler & Pressure Vessel Code; (b) fired process tubular heaters; and (c) pressure containers that are an integral part of machinery such as compressors, pumps, hydraulic or pneumatic cylinders where "the primary design considerations and/or stresses are derived from the functional requirements of the device." Further discussion of the above equipment items is in order. Clearly, vessels that are within the scope of other sections of the ASME Code should be constructed according to that appropriate section. It is common practice in the petrochemical processing industry to construct fired process tubular heaters to the provisions of API 530 [6]. This recommended practice addresses the specific requirements that are needed for the successful long-term operation of such units. The class of equipment defined by (c) above is more inexact. It is clear that the design basis for many pressure containing parts of rotating equipment have their dimensions set by deformation limits required for the proper functioning of the equipment. This would apply to such components as the barrel of a centrifugal compressor, a pump housing, and so on. However, other components of machinery systems may fall under the scope of Section VIII. Examples are the pulsation dampeners of reciprocating compressors, oil tanks and coolers, and so on. These components have their basic design thickness determined only by the basis of the required pressure capability. As such, such components fall under the scope of Section VIII. Paragraph U-1(c)(2)(d) exempts piping systems from the scope of Section VIII. The distinction between piping system and a pressure vessel can become blurred to some extent. There are piping components that may have the characteristics of a pressure vessel; likewise, some pressure vessels may have characteristics of piping components. If the primary function of the pressure container is to transfer fluid from one point in the system to another, then the component should be considered as piping and constructed according to the applicable piping code. Components such as strainers, orifices, valves, headers for distributing or controlling flow that are part of a piping system may be considered outside the scope of Section VIII, provided they are made of " . . . generally recognized piping components or accessories." Clearly, components that have distillation trays, level control connections, demisting pads, and so on do not have a primary function of

21.3

21.3.1

SECTION VIII, DIVISION 1, INTRODUCTION

Scope

Paragraph U-1 defines the scope of coverage for Section VIII, Division 1, of the ASME Code. The term scope refers to both the types of pressure equipment considered in the development of these rules, as well as the geometric scope of a vessel that is stamped as meeting Section VIII, Division 1 (VIII-1). Paragraph U-1(a)(2) VIII-1 scope addresses pressure vessels that are defined as containers for the containment of pressure, internal or external. The definition of a pressure vessel as given in U-1(a)(2) can be interpreted to cover a broad range of pressure equipment some of which was not considered in the development of the rules. To minimize confusion among code users concerning what types of pressure vessels are covered by VIII-1, the committee elected to list the types of equipment "not considered in the development of the rules" instead of trying to list all types of equipment that were considered in the development of the rules. Paragraph U-1(c) describes the types of pressure equipment considered not within the scope of VIII-1. In addition, it makes two other very important points: (1) Any pressure vessel meeting all the applicable requirements of VIII-1, even if outside the scope of the Code, may have the "Code" stamp applied. As an example, a

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 29

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 29

transporting fluid. They are performing processing functions, and, as such, would fall under the scope of Section VIII. Paragraph U-1(c)(2)(f) provides an exemption for vessels containing water under pressure (including those vessels containing air as a cushion), provided the design pressure does not exceed 300 psi and the design temperature does not exceed 210°F. Paragraph U-1(c)(2)(g) exempts hot water supply tanks heated by steam or other indirect means, provided (1) the heat input does not exceed 200,000 BTU/h, (2) the water temperature does not exceed 210°F, and (3) the nominal capacity does not exceed 120 gal. Paragraph U-1(c)(2)(h) provides scope definition regarding the design pressure intended for Section VIII, Division 1, applications. If a pressure container has a design pressure at the top of the vessel not greater than 15 psi (internal or external pressure), then the vessel may be considered outside the scope of Section VIII, Division 1. Thus, a vessel with an internal design pressure of 14.5 psi is considered to be exempted from the scope of Section VIII, Division 1, and normally would not require code stamping. Likewise, a vessel that may experience full vacuum (14.7 psi external) need not be considered within the scope of Section VIII and would not require code stamping for external pressure. However, it is important to reemphasize that any vessel meeting all the requirements of the code may be code stamped. Although a vessel that is expected to experience vacuum (a maximum external pressure of 14.7 psi) may not be within the scope of the code, it may be code stamped for external pressure if all the code requirements have been satisfied. This exemption rule needs to be carefully reviewed when applying it to multichambered vessels such as jacketed vessels and shell and tube heat exchangers in which a component may be subject to a differential pressure greater than 15 psi, thereby placing it within the scope of VIII-1, even though the maximum pressure in either chamber is less than 15 psi. In accordance with paragraph U-1(c)(2)(i), pressure vessels that have an internal diameter, width, height, or diagonal dimension that does not exceed 6 in. are not considered to be within the scope of Section VIII, Division 1, regardless of the vessel length or the design pressure. Thus, according to code requirements, any vessel with a cross-sectional dimension not exceeding 6 in. is generally not required to be constructed even if the pressure exceeds 15 psi. Paragraph U-1(c)(2)(j) states that pressure vessels for human occupancy are exempted from Section VIII. These vessels are those used for oceanic service that may experience significant external pressure from diving operations. Such vessels must balance the weight and thickness needed for proper function against the external pressure design experienced at the ocean depths. ANSI/ASME PVHO-1 [7] provides the construction requirements of such vessels. U-1(d) provides information related to vessels that are in high pressure service. This paragraph states that the design principles and construction practices used in Division 1 are deemed appropriate for pressures not exceeding 3000 psi. However, a pressure vessel can be constructed and stamped in accordance with Section VIII, Division 1, when the design pressure exceeds 3000 psi. As stated, additional provisions may be needed for the design and construction of vessels in high pressure service, and if all the provisions of Section VIII have been satisfied, a vessel with a design pressure in excess of 3000 psi may still have the "U" stamp applied. In such instances, the designer of the pressure vessel

must decide what other provisions are applicable when the design pressure exceeds 3000 psi. One of the scope provisions of Section VIII, Division 1, that is often misunderstood is given in paragraph U-1(g). This paragraph states, "The following pressure vessels in which steam is generated shall be constructed with the rules of this Division: (1) vessels known as evaporators or heat exchangers; (2) vessels in which steam is generated by the use of heat resulting from the operation of a processing system containing a number of pressure vessels such as used in the manufacture of chemical and petroleum products." Many petrochemical processing units have steam generation as an integral part of their processing scheme. Steam may be generated in steam-generating heat exchangers by heat transfer from the process fluid. Steam- generating heat exchangers may be in the convection section of a fired process heater, or in the processing area where steam is generated by the transfer of heat from the process stream. Vessels generating steam in such units fall under the scope of Section VIII, Division 1. For example, the steam-generating components of a pyrolysis unit used for the manufacture of ethylene or the steam-generating components of a hydrocarbon reforming unit used in the manufacture of ammonia and hydrogen should be constructed to the rules of Section VIII, Division 1. The rules of this division may be applied even though the heat to generate the steam may be from a fired process heater. The heat generated in the process heater is required to cause a chemical reaction, and the residual heat of combustion is recovered to generate steam. There are a number of code interpretations that support that such components should be constructed according to Section VIII, Division 1. Two interpretations presented below show the code intent. Interpretation: VIII-1-83-104 Subject: Section VIII-1, U-1(g) and Section I, Preamble Date Issued: March 4, 1983 File: BC79-780 Question: U-1(g) of Section VIII, Division 1, states: Unfired steam boilers as defined in Section I shall be constructed in accordance with the rules of Section I or this Division [see UG-125(b) and UW-2(c)]. The following pressure vessels in which steam is generated shall be constructed in accordance with the rules of this Division: (1) vessels known as evaporators or heat exchangers; (2) vessels in which steam is generated by the use of heat resulting from operation of a processing system containing a number of pressure vessels, such as used in the manufacture of chemical and petroleum products. Are these provisions in conflict with the third from the last paragraph of the Preamble of Section I? Reply: No, as illustrated by reasons as follows: (a) The definitions of pressure vessels in which steam is generated, but which are not "unfired steam boilers," are identical in both Sections I and VIII. Such vessels are not within the Scope of Section I. They are within the Scope of Section VIII, and the special rules applicable to "unfired steam boilers," such as UW-2(c), are not required. (b) The second from the last paragraph of the Preamble of Section I requires "unfired steam boilers" to be constructed in accordance with its rules or the applicable rules of Section VIII.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 30

30 · Chapter 21

(c) As quoted in the Question, Section VIII requires "unfired steam boilers" to be constructed in accordance with its special, applicable rules, such as UW-2(c). We caution you that the laws or regulations at the point of installation may dictate the construction. As indicated by footnote 1 to U-1 Scope of Section VIII, such laws or regulations should be reviewed to determine requirements that may be different or more restrictive than the Code rules. Some applicable laws or regulations may require such vessels to be constructed either with the provisions of UW-2(c) applied or under the rules of Section I. Interpretation: VIII-1-86-201 Subject: Section VIII, Division 1, U-1(g) Date Issued: January 25, 1988 File: BC87-486 Question (1): Are vessels in which steam is generated as described in U-1(g), except those known as an unfired steam boilers, required to meet the special requirements of UG116(d), UG-125(b), and UW-2(c)? Reply (1): No. Question (2): If the reply to Question (1) is no, who determines if a vessel as described in U-1(g) is defined as an unfired steam boiler or one of the other "vessels in which steam is generated?" Reply (2): See footnote 1 of the Introduction. Another important scope definition is found in U-1(h). This paragraph states, "Pressure vessels or parts subjected to direct firing from the combustion of fuel, which are not within the scope of Sections I, III, or IV, may be constructed to the rules of this Division." This means that fired equipment can be constructed by Section VIII, Division 1 rules. A fired vessel that generates steam in a petrochemical processing unit can be constructed in accordance with Section VIII as demonstrated by the following interpretation. Interpretation: VIII-1-95-48 Subject: Section VIII, Division 1 (1992 Edition, 1993 Addenda), U-1(h) Date Issued: March 13, 1995 File: BC94-670 Question: Steam is generated in a processing system containing a number of pressure vessels used in the manufacture of chemical and petroleum products. When the steam generated does not provide all the steam needed for the operation of the chemical or petroleum process plant, fired auxiliary steamgenerating equipment, which is an integral part of the plant, is necessary for the operation of the plant. May fired auxiliary steam-generating equipment that generates steam for the operation of a chemical or petroleum processing system be constructed to the rules of Section VIII, Division 1, satisfying the requirements of U-1(h)? Reply: Yes, see footnote 1 of U-1. Paragraph U-1(j) provides guidance for a particular class of code-stamped vessel. The requirements of this paragraph are not a code exemption per se, but provide criteria for a waiver of third

party inspection by the Authorized Inspector. This class of pressure vessel is designated as "UM.-" The "UM" Manufacturer shall designate one individual from the company as the Certified Individual, whose responsibility is to provide oversight to ensure that the use of the "UM" symbol is in accordance with the requirements of this Division. In essence, the Certified Individual takes on the role of the Authorized Inspector. The Certified Individual signs the Certificate of Shop Compliance on Form U-3. For vessels that are not required to be fully radiographed, which are not provided with quick actuating closures, and do not exceed the following pressure/volume limits may be exempted from inspection by Inspectors, provided that they comply with all other provisions of Section VIII, Division 1: (1) 5 cu ft. in volume and 250 psi design pressure, or (2) 3 cu ft. in volume and 350 psi design pressure, or (3) 11/2 cu ft. in volume and 600 psi design pressure. (Interpolation is allowed for intermediate volumes and pressures.) Vessels fabricated by the rules of this paragraph are marked with the UM symbol. As such, they are "Code- Stamped" pressure vessels. The only exemption is the waiver of third party inspection. No discussion of the scope of Section VIII, Division 1, is complete without referring to paragraph U-1(c)(1). This paragraph states that the scope of Division 1 has been established to identify the components and parameters in formulating the rules of this division. However, the jurisdictional authorities at the location where the vessel is to be installed define the mandatory applicability of the code by the laws that have been enacted within that jurisdiction. It is possible that the laws for a jurisdiction may be different and more restrictive than those defined by the scope exemptions of paragraph U-1. The user and Manufacturer of code-stamped equipment must understand the jurisdictional requirements at the location where pressure vessel equipment is to be installed. The jurisdictional requirements are summarized in the "Synopsis of Boiler and Pressure Vessel Laws" [1]. When jurisdictional scope requirements differ from those of Section VIII, Division 1, the jurisdictional requirements will prevail. If the jurisdictional requirements are not satisfied, it is possible that the necessary approvals or operating permits may be withheld. Caution is advised regarding steam-generating equipment because some jurisdictional requirements may not be aligned with the scope provisions of Section VIII, Division 1, and may require the application of other sections of the ASME Code. 21.3.1.2 Geometric Scope Paragraph U-1(e) defines the geometric scope of VIII-1. When constructing a pressure vessel to be code stamped, it is important to know where the Section VIII, Division 1, rules apply, and where beyond a certain boundary some other standard applies. The most common geometric scope definitions are as follows: U-1(e)(1)--Connections to piping, other pressure vessels, and mechanical equipment. The vessel boundary ends at (a) the welding and connection for the first circumferential joint for welded connections. The most common example would be welded connections to piping. However, note that the vessel boundary can never end at a welded connection directly to the vessel wall, but instead a nozzle must be welded to the vessel, and then the vessel boundary can be terminated at the weld prep of the nozzle for a connection to piping.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 31

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 31

(b) the first threaded joint for screwed connections; (c) the face of the first flange for bolted, flange connections; note that although the flange attached to the nozzle must meet all applicable code requirements, as per Interpretation VIII-1-98-57 the bolting used for that flange is outside the geometric scope of the code; (d) the first sealing surface for proprietary connections. U-1(e)(2)--Attachment of nonpressure parts to pressure boundary material. Any material welded directly to the internal or external pressure retaining surface of the vessel must satisfy all applicable requirements specified in code. "For example, per UW-5(b) all material used for nonpressure parts that are welded directly to the pressure vessel shall be of proven weldable quality." This is accomplished by using a welding procedure qualified in accordance with Section IX. Another example would be where a nonpressure part is classified as essential to the structural integrity of the vessel. If so, UCS-66(a) requires that this nonpressure material satisfy the toughness requirements given in UCS-66. U-1(e)(3)--Pressure-retaining covers for openings. In all openings in a vessel in which a pressure-retaining cover will be installed during operation of the vessel, the pressure-retaining cover is automatically within the scope of the vessel. Common examples are manway covers and handhold covers. U-1(e)(4)--The first sealing surface for proprietary fittings or components. A common example would be a sight class installed within a framed opening of the vessel.

The user must determine if the fluids of the process meet the code definition for "lethal service." Specifically, lethal service is defined at footnote 1 of paragraph UW-2. The definition is given as follows: By "lethal substances" are meant poisonous gases or liquids of such a nature that a very small amount of the gas or of the vapor of the liquid mixed or unmixed with air is dangerous to life when inhaled. For the purposes of this Division, this class includes substances of this nature which are stored under pressure or may generate a pressure if stored in a closed container. This is the only definition given by Division 1 for lethal service. As may be seen, it is not precise. The term "lethal" as used in the context of this Division only determines specific construction details and fabrication requirements. There is no "iron clad" definition or "cook book" that is used by industry for defining when a service should be classified as lethal. In the petrochemical processing industry, there are many toxic or dangerous materials that are routinely handled; however, they may not meet the definition of lethal service when evaluated by the user. Typically, substances that are designated as lethal, in the context of the code, are such highly toxic that a very small amount inhaled will cause death or severe health effects. Almost any substance can be injurious to health. Certainly, there are many deaths each year associated with drowning; however, that does not mean that water is lethal in the context of the code. The same is true with other fluids where the substance may be dangerous, but it takes more than a "whiff" to cause the undesired effects. The user must evaluate the characteristics of the fluid being handled in the process and determine if the lethal service designation and the resulting construction details are appropriate for his installation. There are services where environmentally assisted stress corrosion cracking will occur for certain materials. As with the basic corrosion allowance, the user must define if postweld heat treatment is required to avoid environmental cracking resulting from residual stresses. Typically, carbon steel vessels in caustic or amine services are specified to be postweld heat treated to minimize stress corrosion cracking. Because of the special nature of stress corrosion and other potential residual-stress-related phenomenon, a corrosion specialist should be consulted when making the determination for the need for postweld heat treatment for service conditions. As described in paragraphsU-1(g) and U-1(h), vessels that generate steam may fall within the scope of Section VIII. Section VIII, Division 1, does not contain specific steam system (related to piping and instrumentation) requirements. It is considered good engineering practice to review the Section I requirements for steam systems that are to operate in a chemical or petrochemical facility in order to assure the system functional requirements are appropriate for steam service. In accordance with paragraph U-2(a)(4), the user is responsible for determining the need for piping, valves, instruments, and so on to perform the functions covered by the system design requirements of Section I, paragraphs PG-59 through PG-61. Although not specifically listed in paragraph U-2, the user (or his designated agent) is reminded that the pertinent loading conditions must also be specified to the Manufacturer. Such loading conditions include design pressure, design maximum and minimum temperature, and any other loads that may have an influence on the design (such as those listed in paragraph UG-22), including

21.3.2

Responsibilities

Paragraph U-2 provides general information regarding the use of the code. Included in this paragraph are the responsibilities of the user, the Manufacturer, and the Authorized Inspector. Paragraph U-2 (a) clearly states that the purchaser (the user or his specified agent) of the pressure vessel must define to the Manufacturer, in sufficient detail, how the vessel is intended to be operated in order that the vessel can be properly designed. The following specific considerations are listed: (1) the need for corrosion allowances; (2) if the vessel is in "lethal" service; (3) the need for postweld heat treatment (PWHT) based on service conditions; and (4) if the vessel is to generate steam, definition of how the system requirements of ASME Section I are to be implemented. The user of the vessel is responsible to understand its intended operation in order to define the expected corrosion or erosion rate. Generally, this is an internal corrosion/erosion allowance; however, it could also be an external corrosion allowance. The corrosion allowance specified is to be added to all the design calculations in order to assure the thickness of the vessel includes provisions for corrosion during operation. This does not mean that every vessel has to have a corrosion allowance added to the design thickness. If the user determines that a corrosion allowance is not required, then it is acceptable to specify zero corrosion allowance. Also note that different corrosion allowances can be specified for different zones of the vessel (see Interpretation VIII-1-01-81). It is common for stainless steel and nonferrous materials to have zero corrosion allowance specified because of the excellent corrosion resistance of such material. It is strongly recommended that a corrosion specialist be consulted for the determination of required corrosion allowances.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 32

32 · Chapter 21

any expected abnormal events such as deflagration. These loadings may include static, cyclic, or dynamic loads from an attached piece of equipment or piping not in the Manufacturer's scope of supply. Likewise, if the vessel will experience significant cyclic thermal gradients during normal operation of the vessel, these operating conditions need to be specified by the user so that they can be considered in the design of the vessel. Paragraph U-2(b) provides specific defined responsibilities for the Manufacturer. The Manufacturer is responsible for complying with all the applicable provisions of Division 1. In this context, the Manufacturer is defined as the certificate holder whose stamp will be placed on the vessel. The Manufacturer is responsible for the design, welding, nondestructive examination, postweld heat treatment, and so on. If the Manufacturer subcontracts part of the work to others, he is still responsible, by proper certification, to assure that this work complies with all the code requirements. The Manufacturer is responsible to design the vessel, using the information provided by the user, such that all applicable design requirements are satisfied. When the Manufacturer subcontracts parts of a vessel to others, he still is responsible for the design of the vessel. It should be noted that the code documentation requires the Manufacturer to certify that the design complies with the code. Paragraph U-2(e) defines the responsibility of the Inspector. The Authorized Inspector is responsible for conducting all the inspections that the code defines that he is supposed to make. This does not mean that the Inspector is required to act as the fabricator's quality assurance function. The Inspector is charged with monitoring the quality control and the examinations made by the Manufacturer. The Inspector has the leeway to monitor and conduct any inspections, needed by his judgment, which will confirm the Manufacturer has satisfied all applicable code requirements. Likewise, the Inspector has a specific duty to verify that all applicable calculations have been made on file at the Manufacturer's facility. The Inspector is not responsible to verify the accuracy of the calculations; the Manufacturer is responsible for the accuracy of the design calculations. As an example, if a vessel has a custom-designed flange, the Inspector is expected to verify that calculations in accordance with Appendix 2 of Division 1 have been made; however, he is not required to verify that they have been done correctly. This same logic applies when calculations are performed to satisfy paragraph U-2(g) (see Interpretation VIII-79-45). Of course, if an Inspector was to notice an obvious mistake, it is appropriate and likely that he would point it out to the Manufacturer, but that is not his obligation under code requirements.

construction is prohibited, or need not be considered. For example, VIII-1 does not contains fatigue design rules; however, if a vessel to be code stamped is subjected to cyclic loading, then a fatigue analysis must be carried out. For such cases, the Manufacturer may use principles of pressure vessel design/engineering and fabrication that are consistent with the philosophy and safety margins of Division 1, and the Inspector has to accept these design and fabrication details. There are instances when the Code Committee receives requests for interpretations for design details that do not have specific requirements in Division 1. Since the Committee cannot endorse specific designs, the inquirer is advised that the code does not have rules related to the subject of the inquiry and the provisions of paragraph U-2 (g) should be applied.

21.3.4

Referenced Standards

Another important point related to the Introduction of Division 1 is contained in paragraph U-3. Table U-3 presents references to various standards such as ANSI and other ASME standards that cover pressure­temperature ratings, dimensions, and procedures. These standards are allowed and may be used to establish the pressure­temperature ratings for components that fall within their scope, sometimes with supplemental requirements as defined in paragraph UG-44. For example, flanges meeting the dimensions, material, and pressure­temperature ratings of the ASME B16.5 [22] standard may be used without performing the detailed calculations of Appendix 2 of Division 1. However, Table U-3 may not list the latest published edition of those standards. The editions that are listed are those editions that have been accepted by the Code Committee for use in Division 1. It is intended that the specific edition of each standard listed in Table U-3 be used for Division 1 applications. Earlier or later editions than those listed in Table U-3 for the applicable standards may not be used.

21.3.5

U-4 Units of Measurement

21.3.3

Design and Construction Details Not Covered

The 2004 Edition of the ASME Boiler & Pressure Vessel Code marked the first time that units other than U.S. customary may be used for the construction of pressure equipment, including stamping and certification. What this means for Manufacturers is that their drawings, engineering work, and fabrication documents may now be prepared in SI or other local units. The only caveat is that the units used to prepare the fabrication drawings must be the same as used on the Manufacturer's Data Report (MDR) and nameplate stamping. Another consequence of the "metrication" of the code was the removal of all units from the variable definitions found in the nomenclature within the codebooks. Also, allowable stresses are now published in both U.S. customary and SI units in Section II, Part D.

Paragraph U-2 (g) provides very important and often overlooked criteria for the design and construction of pressure vessels. This paragraph states: This Division of Section VIII does not contain rules to cover all details of design and construction. Where complete details are not given, it is intended that the Manufacturer, subject to the acceptance of the Inspector, shall provide details of design and construction which will be as safe as those provided by the rules of this Division. This paragraph makes it clear that if a particular detail or type of construction is not given in Division 1, it does not mean that such

21.4

SUBSECTION A: GENERAL REQUIREMENTS FOR ALL METHODS OF CONSTRUCTION AND ALL MATERIALS

Part UG of Division 1 contains rules that are applicable to all pressure vessels regardless of the type of construction and type of materials used. The rules of Part UG are to be supplemented, as applicable, by the rules relating to specific types of construction (welding, forged, brazing, etc.), for specific materials (carbon and low alloy, austenitic stainless steel, and nonferrous), and applicable mandatory Appendices.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 33

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 33

21.4.1

Materials

21.4.1.1 General Requirements Paragraphs UG-4 through UG-15 define general material requirements. There are several points related to the general material requirements that will be highlighted here. As a basic rule, all pressure containing material shall conform to a material that is listed in Section II of the ASME Boiler & Pressure Vessel Code; however, the material must also be listed in the tables of permitted specifications in the applicable part of Subsection C. See Tables UCS-23, UHA-23, UNF-23, UCI-23, UCD-23, UHT-23, and ULT-23. Not all materials listed in Section II are automatically accepted for use in Division 1 applications. Only those materials listed in the above-referenced tables have allowable stress values given for Section VIII, Division 1, application in Section II, Part D. Nonpressure parts such as supports, lugs, brackets, and internal components do not have to conform to the same requirements as the material for which they are attached or to a permitted code specification. However, if unidentified material is welded to a pressure part, then the "non-Code" material has to be of weldable quality. Paragraph UW-5(b) defines weldable quality as the ability of a butt-welded test coupon of the material to pass the guided bend test specified in paragraph QW-451 of Section IX of the ASME Boiler & Pressure Vessel Code. Likewise, paragraph UG-4(b) specifies that allowable stress used for material that is not identified with a permitted specification shall not exceed 80% of the maximum allowable stress value for a similar material that is permitted by Division 1. For example, if a pressure vessel support skirt material is not a code identifiable material, the design allowable stress is limited to 80% of that for a similar material that is listed in the appropriate table in Subsection C. 21.4.1.2 UG-8 Pipe and Tube Permitted material specifications for pipes and tubes include seamless and welded construction. Footnote 3 to paragraph UG-8 clarifies that welded pipe and tubing using a fusion-welded process with added filler material may not be used unless it is fabricated in accordance with the rules of Division 1. This means that pipe fabricated to a specification (such as SA-658) that is fusion welded using filler metal must be done by an organization having a Code Certificate of Authorization, qualified welding procedures, qualified welders, and so on. Additionally, such material must be furnished with a Code Partial Data Report for Welded Construction (Form U-2 or U-2A). These requirements do not apply to electric resistance welded (ERW) pipe and tube material since they do not use filler metal in the welding process. Paragraph UG-8(b) provides requirements for finned tubing that is commonly used for heat transfer equipment. The most significant requirement regarding the design of finned tubes is that the maximum allowable working pressure (MAWP) (internal or external) must be based on the minimum thickness at the root diameter of the finned section. The minimum thickness at the finned section after thinning can be considerably less than the tube thickness before the finning operation. This can have a pronounced effect on the design of such equipment. For external pressure design for some alloys, Appendix 23 allows the use of proof testing of finned tubes to establish the allowable external working pressure. Use of Appendix 23 may result in considerable increase of the allowed external pressure for finned tubes. Likewise, paragraph UG-8(b)(5) specifies additional pressure test requirements that are above and beyond the testing requirements of the material specification for finned tubes. Finned tubes for

Section VIII, Division 1, application must be exposed to either an internal pneumatic test pressure of no less than 250 psi for at least 5 s or the hydrostatic test required by paragraph UG-99. Each test requires the tube to be examined for leakage. The pneumatic test requires easy visual detection such as air under water or pressure differential methods. 21.4.1.3 Material Recertification Paragraph UG-10 provides rules that allow material to be used in pressure vessels when the material is not identified with or produced to a specification recognized by Division 1. This paragraph addresses the situation where material that is produced to a specification that has not been adopted by the code may still be used if it can be shown that this material satisfies all of the requirements of a material specification that has been adopted by the code. This "recertification" process may be carried out by either the material manufacturer, or the vessel Manufacturer. In addition, provisions are given for the situation where material for which traceability has been lost may still be used for code construction when its properties can be established by additional testing. It is commonly asked if material produced according to an American Society for Testing and Materials (ASTM) specification must be recertified under SA/SB specification as per UG-10 to be used for code construction. The answer in most cases is no, so long as the ASTM specification is listed as equivalent to its SA/SB counterpart in Section II, Parts A and/or B. See Interpretation VIII-1-92-149 given below. Interpretation: VIII-1-92-149 Subject: Section VIII, Division 1 (1992 Edition, 1992 Addenda), UG-9 and UG-10 Date Issued: May 20, 1993 File: BC93-111 Question (1): Do materials that are manufactured according to an ASTM specification and are stated in the material specification in Section II, Parts A and B to be identical, and which have been adopted by Section VIII, Division 1, require recertification under the rules of UG-10? Reply (1): No. Question (2): Do materials that are manufactured to an ASTM specification and for which a companion ASME specification has not been adopted by Section VIII, Division 1, require recertification under the rules of UG-10? Reply (2): Yes. Question (3): May welding materials conforming to an AWS specification, but not an SFA specification, be used in Section VIII, Division 1, construction, as given in UG-9, without recertification under UG-10? Reply (3): Yes, provided the provisions of the welding procedure specification are met. 21.4.1.4 Prefabricated or Preformed Pressure Parts Paragraph UG-11 addresses prefabricated and preformed pressure parts, commonly referred to as "miscellaneous standard parts." These are parts that are produced either according to an ASME/ ANSI reference standard or to a Manufacturer's Standard that establishes the pressure/temperature rating for the part, and may be used within a Section VIII, Division 1, pressure vessel without any further design calculations. Common examples of miscellaneous

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 34

34 · Chapter 21

standard parts include flanges produced as per B16.5 or B16.47, fittings produced according to B16.9 or B16.11, or quick opening closures or manway closures produced as per Manufacturers Standard. Miscellaneous standard parts do not require Partial Data Reports even when they contain welding. However, welded miscellaneous standard parts cannot be used for the shell or head of a pressure vessel unless accompanied by a Partial Data Report (this means the welded shell or head must be fabricated by a Certificate Holder). One caution concerning the use of parts produced according to a Manufacturers Standard is that per footnote 6 to UG-11, the vessel Manufacturer is ultimately responsible for assuring that the part is suitable for the intended design and that the part meets all applicable requirements in the Code, including material and welding requirements. Paragraph UG-12 defines requirements for bolts and studs for use in pressure vessels. The bolting specifications allowed for Section VIII, Division 1, applications are given in the tables referenced in UG-23(a). (Note that only material listed in Table 3 of Section II, Part D, may be used for bolting; see Interpretation VIII-1-95-131.) The requirements for threaded connections apply to bolts and studs used to attach removable covers or attachments such as blind flanges or full opening closures. This paragraph requires that all studs and bolts be either threaded for their full length or have the shank machined down to the thread root diameter. This requirement represents good engineering practice for the design of threaded connections. If the shank is not machined to the thread root diameter, the transition between the thread and shank results in highly localized stressed region that may initiate a fatigue failure or brittle fracture. However, this paragraph does allow an unthreaded portion with the same diameter as the major diameter of the threads if there is a transition area of at least 0.5 diameters adjacent to the threaded portion. The length of the threaded portion of the bolt must be at least 11/2 times the bolt diameter. The requirements of UG-12 are only applicable for bolting that connects pressure-retaining components that are within the scope of the code. See the interpretation, below, which provides clarification. Interpretation: VIII-1-98-57 Subject: Section VIII, Division 1 (1998 Edition, 1998 Addenda), U-1(e)(1)(c) Date Issued: January 26, 1999 File: BC98-368 Question: In meeting the requirements of U-1(e)(1)(c) in Section VIII, Division 1, are proprietary bolts supplied with a flange that attaches a non-code item to a code nozzle within the scope of the code? Reply: No. Paragraph UG-14 addresses the use of rod and bar material in pressure vessels. Rod and bar stock is commonly used for pressure parts such as flange rings, stiffening rings, stays and stay bolts, and in some cases fabricated nozzles. Occasionally, it is difficult for a Manufacturer to purchase pipe or tube in a specific alloy for use as a nozzle, but they are able to purchase solid bar material that is then machined to produce the hollow cylindrical part. However, such parts are limited in size to NPS 4 when machined from rolled or forged bar, unless Code Case 2156-1 is used. In addition, it is prohibited to fabricate flanges of all types, elbows, return bends, tees, and header tees directly from bar stock.

Paragraph UG-15 addresses the situation where VIII-1 has adopted the use of a particular alloy in one or more product forms (e.g., plate, forging) but not for other product forms (e.g., pipe or tube) that is needed for a vessel to be constructed according to the code. In this situation it is acceptable to use this alloy in these other product forms that are not listed in Section II, Part D, so long as this alloy does have published stresses in at least one product form, and the desired product forms are accepted by VIII-1 for other alloys. It is best to illustrate this with an example. Let us assume that a manufacturer desires to use stainless steel bar, SA-479, in Type 316H (UNS S31009). A review of Table UHA-23 shows that this particular alloy is not listed as available for SA-479; however, it is available for use in plate product form SA-240. UG-15 would allow the manufacturer to use SA-479 Type 316H for code construction, and would also use the allowable stress published under SA-240 Type 316H [also, Interpretation VIII-1-89-28].

21.4.2

Design

21.4.2.1 General Regardless of the required thickness that results from calculations, Section VIII, Division 1, specifies minimum thickness for shells and heads for pressure vessels. These minimum thickness requirements are somewhat arbitrary and based on practicable considerations for pressure vessel fabrication, handling, and general robustness. The minimum shell and head thickness requirements are specified in paragraph UG-16. The specified minimum thickness of pressure vessel shells and heads, after forming and not including any allowance for corrosion, is 1/16 in. However, there are a number of exceptions to this rule that are delineated in the subsequent paragraphs of UG-16. The minimum thickness does not apply to heat exchanger tubes (or tubes or pipe that are fully protected by duct work or other enclosures), the heat transfer plates of plate and frame heat exchangers, and the inner pipe of double pipe heat exchangers if the inner pipe is NPS 6 or less in diameter. These exemptions generally apply to components that are protected by other components from external mechanical damage. Likewise, paragraph UG-16 provides more conservative minimum thickness requirements for other types of vessels; carbon steel pressure vessels in air, water, or steam service shall not be less than 3/32 in. in the corroded condition; the minimum thickness of shells and heads in unfired steam boilers shall be 1/4 in. exclusive of corrosion allowance. Paragraphs UG-16(c) and UG-16(d) provide requirements related to how thickness tolerances of plates and piping are to be considered for pressure boundary components. Plate material may not be ordered to a thickness that is less than that required by the design formulae. However, plate material received by the Manufacturer with an actual supplied thickness that is thinner than the required thickness by 0.01 in. or 6% of the ordered thickness may be used at the full design pressure without any penalty. This effectively means that the plate thickness under tolerance, as allowed by the material specification, does not need to be considered if it is the lesser of 0.01 in. or 6% of the plate thickness. If the material specification provides for thickness under tolerance that is greater, then the ordered thickness must consider this greater tolerance. Different requirements apply for pipe and tubes that are used for pressure boundary components. If a pipe or tube is ordered as nominal thickness (which is common industry practice), the mill under tolerance must be considered in determining the ordered thickness. The most commonly used pipe specifications allow a thickness under tolerance of 121/2% of the nominal wall thickness. Thus, the specified nominal thickness of piping

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 35

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 35

components, multiplied by 0.875, must be equal to or greater than the required minimum thickness in the corroded condition. In accordance with paragraph UG-16(d), consideration of piping under tolerance is not required for opening reinforcement area calculations. The nominal thickness of pipe material should be used for the opening or nozzle reinforcement calculations of UG-37 and UG-40. Paragraph UG-16(e) states that all the dimensional symbols in the design formulae represent the geometry in the corroded condition. This also applies to the various figures in the code that provide dimensional requirements, such as sizing fillets welds for nozzle attachments as per Fig. UW-16.1. If there is a corrosion allowance specified, then that corrosion allowance must be added to the minimum thickness determined by the design rules, including the figures, of Section VIII, Division 1. Paragraph UG-19 addresses vessels of special construction, such as vessels that consist of more than one independent pressure chamber, or are of shapes other than cylindrical or spherical. In the case of vessels consisting of more than one independent pressure chamber, such as a shell and tube heat exchanger or jacketed vessel, UG-19(a)(1) and UG-19(a)(2) provide rules for designing the common elements for a differential pressure and mean metal temperature. Because the differential pressure acting on a common element such as a tubesheet in many cases is lower than the maximum pressure in one or the other chamber, it is necessary to mark the vessel and the Manufacturers Data Report to clearly identify the limiting pressure for the common element. Furthermore, UHX-19.2.2 requires the following caution statement to be marked on heat exchangers utilizing fixed tubesheets: The Code required pressures and temperatures marked on the heat exchanger relate to the basic design conditions. The heat exchanger design has been evaluated for specific operating conditions and shall be reevaluated before it is operated at different operating conditions. 21.4.2.2 Design Conditions and Allowable Stresses In order that the Manufacturer may properly design a pressure vessel, the design conditions must be established. As already noted, it is the responsibility of the user or his designated agent to define the design conditions for which the pressure vessel will be operated. The key design parameters are given in paragraphs UG-20, UG-21, and UG-22. Paragraph UG-20 provides requirements for establishing the maximum and minimum design temperatures. The maximum design temperature is given as . . . the maximum temperature used in design shall not be less than the mean metal temperature (through the thickness) expected under normal operating conditions for the part considered. It is common practice to set the maximum design temperature equal to the maximum anticipated fluid temperature with an appropriate allowance; however, the above definition does allow a much less conservative approach to be used. If a pressure vessel is insulated, then the mean metal temperature and process fluid temperature are essentially the same at steady state conditions. However, for noninsulated vessels, the mean metal temperature through the thickness can be considerably different than the process fluid temperature. This can be used to advantage for those applications where the mean metal temperature is significantly cooler than the process fluid. For example, if the maximum anticipated process fluid temperature is 800°F and

heat transfer calculations show that the temperature at the midwall of the shell was only 550°F, it is acceptable to use the lower temperature as the maximum design temperature. The same rationale may be used to establish the maximum design temperature for heat exchanger tubesheets that are exposed to significantly different fluid temperatures on either side. It is acceptable to establish the maximum temperature by use of heat transfer calculations to determine the mean metal temperature across the tubesheet. This method is not allowed for components that are directly fired; requirements are given in UW-2 (d)(3) for determining maximum design temperatures for directly fired equipment. Equally important is the correct determination of the minimum design metal temperature (MDMT). Carbon and low alloy ferritic materials undergo an abrupt transition in notch toughness at a temperature known as the toughness transition temperature (nil-ductility transition temperature). At temperatures colder than the toughness transition temperature, the material behaves in a brittle manner and may fail by catastrophic brittle fracture. Above the transition temperature, the material behaves in a ductile manner and brittle fracture is not a failure mechanism of concern. It is very important to correctly define the minimum design temperature in order that the correct material can be specified to avoid the possibility of brittle fracture during operation conditions. As with the maximum design temperature, the MDMT may be established as the coldest mean metal temperature through the thickness of the component. The MDMT that is marked on the nameplate shall be the coldest temperature that is coincident with the maximum allowable working pressure that is shown on the nameplate. As will be discussed later in this chapter, there are instances where a vessel may be safely operated at a temperature colder than the MDMT if the coincidental pressure is less than the MAWP. Consideration shall be given to the lowest temperature expected during operation, the effects of the atmospheric temperature when the vessel is under pressure, autorefrigeration effects, and any other effects that may cool the vessel. If a vessel's metal temperature is affected by the ambient temperature, then the coldest anticipated seasonable temperature should be considered in establishing the MDMT. Some liquid hydrocarbons that are stored at pressure/temperature conditions well above their boiling point will cool to their atmospheric pressure boiling temperature when the pressure is removed. These temperatures can be quite cold. For example, the atmospheric pressure boiling temperature of ethylene is approximately ­155°F, and ethylene liquid under pressure will cool to this temperature when the pressure is removed. This is the "autorefrigeration" effect that is to be considered as noted in paragraph UG-20(b). If parts of pressure vessel have different operating temperatures, then multiple temperatures may be specified for different sections of the vessel. Paragraph UG-20(f) provides general impact of test exemption guidelines for carbon steel material. Experience has shown that within certain parameters, carbon steel material has operated quite satisfactorily at temperatures as cold as­20°F without a need to validate the material toughness by impact testing. This paragraph allows vessels made of carbon steel (P-No. 1, Gr. Nos. 1 and 2 only) to be exempted from the impact of test requirements of UG84 for temperatures as cold as ­20°F, provided (1) materials that require the use of curve A of Fig. UCS-66 are not thicker than 1/2 in., or (2) materials that require the use of curves B, C, or D of Fig. UCS-66 are not thicker than 1 in.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 36

36 · Chapter 21

Additional provisos apply for the use of this general impact test exemption. The vessel must be hydrostatically tested in accordance with the rules of UG-99(b) or UG-99(c). This requirement is based on the fact that the hydrostatic test results in localized yielding, or blunting of notches, or imperfections in a pressure vessel. This blunting of imperfections improves the apparent notch toughness of the material and makes it much less likely to fail in a brittle manner. Also, the maximum design temperature of vessel cannot exceed 650°F. A limit is placed on the maximum design temperature because it has been shown that beneficial effects of the hydrostatic test are significantly diminished when the vessel operates at elevated temperature. The limit of 650°F assures that the beneficial effects of the hydrostatic test related to brittle fracture are not lost. This impact test exemption may not be used if thermal or mechanical shock and/or cyclic loads are part of the design basis. Shock loads are a concern because they are much more likely to result in brittle fracture than static loads. Cyclic loading may initiate and propagate fatigue cracks in the vessel to such an extent that they may become significant from a fracture mechanics consideration that may make the vessel more vulnerable to brittle fracture. Paragraph UG-21 defines the considerations for establishing the design pressure of a vessel. This paragraph requires the vessel to be designed for at the most severe pressure and temperature that is coincidentally expected in normal operation. This means those conditions such as start-up, shutdown, and any identified upset conditions must be considered when establishing the maximum operating pressure. For heat exchangers and other multicompartment vessels, the most severe combination of pressure must be considered. For example, if the maximum operating pressure on the shell side of a heat exchanger is 100 psi and the tube side is under full vacuum, then the common elements between the shell side and tube side (tubes and tubesheets) must be designed for a differential pressure of 115 psi. Similar consideration must be given for jacketed pressure vessels. The design pressure is significant not only for the design of the pressure vessel, but also for the minimum set pressure of the pressure relief device (PRD) that protects the vessel from overpressurization. In an operating system, the set pressure of the pressure relief device must be above the operating pressure by a sufficient amount so that the device does not actuate unintentionally. The PRD shall not be set to actuate at a pressure that is greater than the design pressure of the vessel or vessels it is protecting. It is normal practice to set the design pressure (set pressure of the PRD) at least 10% above the maximum operating pressure when spring-actuated relief valves are used. This margin may be relaxed if pilot-operated relief valves are used. A pressure vessel may be satisfactorily operated with pilot-actuated relief device having a set pressure of about 5% above the vessel maximum operating pressure. When establishing the proper margin of the PRD set pressure, it is recommended that the PRD supplier be consulted for direction. Paragraph UG-22 provides information that is part of the userspecified design conditions to be used by the Manufacturer for designing the vessel. In addition to internal and external pressure, there are other loads that must be considered in the design. Often overlooked are those loads from other equipment or components that must be supported by the vessel. This would include ladders and platforms, vessel internals, and other pressure vessels that rest on the vessel. The hydrostatic pressure due to the head of liquid in the vessel during normal operation must be considered in the design. It is noted that the design pressure (or MAWP) is referenced to the top of the pressure vessel. The coincident pressure

resulting from the head of liquid must be added to the design pressure when designing those sections below the top. This effect can be quite pronounced and cannot be ignored. For example, a 100-ft. tall vessel that operates full of water will have 43.3 psi of hydrostatic head pressure at the bottom. If the design pressure at the top is 20 psi, the bottom section must be designed to 63.8 psi (the design pressure at the top plus 43.3 psi of liquid head). Strictly speaking, the same consideration should be given to horizontal vessels. The design pressure is referenced at the top and the head of liquid acts at lower elevations. Normally this is an insignificant effect; however, there are instances where the effect may have a minor influence on the design. Note that when a vessel is filled with a liquid only during the pressure test, then the liquid static head present during the test does not need to be considered when calculating the required wall thickness of the vessel. However, the vessel designer does need to consider the weight of the test fluid and the manner in which the vessel will be supported during the pressure test to assure that the vessel is not damaged during the test. The effect of any static or dynamic reactions on the vessel from any attached equipment must also be considered in the design. It is the user's responsibility to define loading conditions that are expected in normal operation. Failure by the user to define these loading conditions may result in excessive deformation, cracking, or possible failure of the vessel. Consideration is also required for wind, snow, and seismic reactions. The location of the vessel will dictate whether these loads are significant to the vessel design. Criteria for determining wind and seismic loadings at a specific location may be found in ASCE 7-95 [8] or The Uniform Building Code [9]. If a vessel is to be subjected to abnormal conditions during operation, such as possible internal deflagration, then this should be included in the load considerations. (See Nonmandatory Appendix H.) Paragraph UG-22 provides a rather extensive list of loads to be considered; however, no specific guidance is provided regarding how these loads are to be analyzed. It is the responsibility of the owner or user to define the loading conditions, and it is the responsibility of the Manufacturer to assure that the design considers the load conditions. In the absence of defined code rules, the specific design methods to consider these loads are at the Manufacturer's discretion if not covered by the contractual requirements between the Manufacturer and the user. The Manufacturer must satisfy the guidelines given in U-2(g). It is an accepted practice to use the analysis provisions of Appendices 4 and 5 of Section VIII, Division 2, of the ASME Code for considering the localized effects of the loads listed in UG-22. These appendices provide direction regarding classification of local, secondary, and peak stresses that are generally required to evaluate most of the loads that are listed in UG-22. Certainly, it is not appropriate to limit secondary and highly localized stresses to the basic Division 1 allowable stress. Paragraph UG-23 defines the maximum allowable stress for internal and external pressure to be used in the design of Division 1 pressure vessels. The allowable tensile stresses are tabulated in Section II, Part D, of the Boiler & Pressure Vessel Code. There are several points regarding the allowable stress to be used in design that merit discussion. UG-23(a) states that for material identified as meeting more than one material specification or grade, the allowable stress for either specification or grade may be used, provided that all the limitations of the specification is satisfied. For example, it is a normal practice for material suppliers to "dual certify" stainless steel products as TP304/TP304L. As such,

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 37

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 37

this material is certified to satisfy all the specification requirements for both grades, including limits on chemistry and minimum required yield and tensile strength. The allowable stress for this material may be taken from either the TP304 or TP304L stress lines in the applicable table of Section II, Part D. This may provide an advantage to the designer because, in this case, the low carbon grade, TP304L, of stainless steel can be selected for its corrosion properties, while the larger allowable stress for the straight grade, TP304, may be used. For further information on the use of dual certified material, see Section II, Part D Appendix 7. UG-23 also provides criteria for the maximum allowable longitudinal compressive stress to be used for cylindrical shells or tubes that are subjected to longitudinal compressive loads. The first criterion is that the maximum allowable longitudinal compressive stress cannot be greater than the maximum allowable tensile stress. The second criterion is based on buckling of the component. Paragraph UG-22(b) provides a step-by-step procedure for the determination of the allowable compressive stress considering buckling or instability. This procedure uses the external pressure charts of Section II, Part D. The rules of this paragraph are also used to determine the maximum allowable compressive stress resulting from a moment applied across the section of the vessel such as that resulting from wind and seismic moments. Paragraph UG-23(c) states, The wall thickness of a vessel computed by these rules shall be determined such that, for any combination of loadings listed in UG-22 that induce primary stress and are expected to occur simultaneously during normal operation of the vessel, the induced maximum general primary membrane stress does not exceed the maximum allowable stress value in tension (See UG-23), except as provided in (d) below. There are some important concepts that must be understood to correctly apply these rules. The first is that the wall thickness, as a minimum, is established by general primary membrane stress. A general primary membrane stress is defined in Appendix 3 of Section VIII, Division 1, as a stress in the structure such that no redistribution of load occurs as a result of yielding. A more extensive definition and examples of general primary membrane stress are found in Appendix 4 of ASME Section VIII, Division 2. Basically, a general primary membrane stress is the stress remote from any discontinuities such as nozzles and transitions. It does not include any thermal stresses nor does it include any localized or peak stress effects. The "design by rule" methods given in the applicable parts of Division 1 and its appendices indirectly consider the secondary stresses due to pressure-induced loads in areas of discontinuities by the requirements for nozzle reinforcement, cone-to-cylinder transitions, and so on. UG-23(c) states, It is recognized that high localized discontinuity stresses may exist in vessels designed and fabricated in accordance with these rules. Insofar as practical, design rules for details have been written to limit such stresses to a safe level consistent with experience. The reader is urged to refer to Chapter 22 of this book for further explanation of detailed stress analysis subjects contained in Section VIII Division 2. Another key consideration of paragraph UG-23 is given in UG23(d). This provision allows the general primary membrane stress resulting from either seismic or wind loading, when combined

with other applicable loadings of UG-22, to be as great as 1.2 times the code allowable stress. The 1.2 increase permitted is equivalent to a load reduction factor of 0.833. Some standards that define applicable load combinations do not permit an increase in allowable stress, however a load reduction factor (typically 0.75) is applied to multiple transient loads (e.g., wind plus live load, seismic plus live load, and others). The increased allowable stress basis may be used for tension, external pressure, and longitudinal compression under wind or seismic loading. It is not required to consider that the design wind and seismic loads occur simultaneously because it is a remote possibility that a design basis earthquake will occur during a design basis wind storm. Although this combination of loadings is remotely possible, it is not considered a credible operating case. (See paragraph L-2.1 of Section VIII, Division 1, for an example problem illustrating the consideration of wind loads.) Paragraph UG-25 provides a general discussion related to corrosion and other vessel wall thinning mechanisms such as abrasion or erosion. The user shall define when such considerations are required; likewise, the user shall define the amount of allowance to be included in the design of a pressure vessel. The user-specified allowance for thinning expected during the vessel's life is to be added to the thickness derived from the design formulae. All dimensions given in the design formulae and the associated figures do not include any thickness needed for corrosion or other thinning mechanisms. 21.4.2.3 Internal Pressure Design Definition of the minimum required thickness (or maximum allowable pressure) for a shell under internal pressure is provided in paragraph UG-27. For cylindrical shells, the applicable equations for circumferential stress (the stress acting across the longitudinal seam) are as follows: t PR SE 0.6P or P SEt R 0.6P (21.1)

For cylindrical shells for longitudinal stress (the stress acting across the circumferential joints), the applicable equations are t t P R S E PR 0.4P or P 2SEt 0.4t (21.2)

2SE

R

minimum required thickness of shell, in. (in the corroded condition) internal design pressure, psi inside radius of shell under consideration, in. (in the corroded condition) maximum allowable stress from the applicable allowable stress table in Section II, Part D the lesser of the joint efficiency for welded joints (as defined in Table UW­12), or the ligament efficiency between openings (when applicable, determined using UG-53).

For spherical shells, t PR 0.2P or P 2SEt 0.2t (21.3)

2SE

R

These equations are very straightforward and do not require much discussion. However, there are some pertinent issues that are

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 38

38 · Chapter 21

discussed. These equations are based on thin wall theory, but do include provisions to account for the variation of stress through the wall of the vessel (called the Lame' effect) that becomes significant for very thick cylinders. For example, the "20.6 P" term included in the cylinder equations is added in a simplistic way to account for variation in stress through the thickness. The limits that are provided for using these equations assure that the simplification of the Lame' effect is not used for very thick shells where the effect is more accurately defined by the equations given in Appendix 1. Use of Equation (21.1) is limited to a pressure that does not exceed 0.385 SE or where the thickness does not exceed 1 2 of the inside radius. For commonly used carbon steel, such as SA-516 Gr. 70, the limiting pressure is about 7700 psi. If the limits of these equations are exceeded, the equations provided in Appendix 1 must be used. Equation (21.1) defines the required thickness or the maximum allowable working pressure of a cylinder due to hoop (circumferential) stress. The hoop stress acts normal to the longitudinal joint of the cylinder. As such, the welded joint efficiency, E, to be used in this equation is based on the degree of radiography of the longitudinal weld seam (see Table UW-12). Equation (21.2) applies to the longitudinal stress that acts normal to the circumferential weld seam; thus, the welded joint efficiency used in this equation is based on the degree of radiography of the circumferential seam. Since the longitudinal stress in a cylinder is about 1 2 of the hoop stress resulting from pressure, it can be seen that Equation (21.2) will not govern the design of a cylinder unless the joint efficiency of the circumferential joint is less than 1 2 of the longitudinal joint. However, if longitudinal stresses are imposed from other loadings, such as the overturning moment from wind or seismic events, the longitudinal stress can govern the design. The weld joint efficiency used for spherical shells will be the lesser of the E value for the weld seams of multiple pieces required to make the head or the E value required for the head to shell connection. 21.4.2.4 External Pressure Design The method used to design shells and tubes when external pressure is specified as a design load is given in paragraph UG-28. If vacuum is specified for a vessel, the scope of Section VIII, Division 1, allows exclusion from the mandatory requirements for external pressure since the pressure is not greater than 15 psi [see paragraph U-1(h)]. However, if the external pressure resulting from vacuum is to be shown on the data report and the nameplate, then all applicable requirements pertaining to external pressure must be satisfied. ASME Code Case 2286 provides alternative rules for the design of vessels and vessel components under external pressure, and these rules may be used in lieu of the rules in UG-28. Generally, use of Code Case 2286 will result in a less conservative design, that is, less thickness or fewer stiffening rings than required by paragraph UG-28. Code Case 2286 has the same theoretical basis as API publication 2U [31] that has been used for the design of submerged components of offshore structures. The methods used for applying the rules for external pressure are well illustrated in the Appendix L examples (see Examples L.3 and L.4) and various pressure vessel design handbooks [10] and are not repeated here. However, there are some important concepts that must be understood to correctly apply these rules. Probably the most important issue is the correct application of the length between lines of support (between stiffening rings). The definitions for various geometries are pictorially shown in Fig. 21.1 (Fig. UG-28.1). Figure 21.1 (a-1) shows a cylindrical vessel with heads having no stiffening rings. The heads provide stiffening for the

cylinder for external pressure, and the effective stiffening location is the depth of the head divided by 3. Thus, the unstiffened length for this example is the length of straight shell plus 1/3 the head depth at each end. This establishes the value of L to be used in the external pressure calculations. When a vessel has cone-to-cylinder and/or cone knuckle-tocylinder junctions, determination of the unstiffened length L to be used in the calculations is more complicated. Notes 1 and 3 of Fig. 21.1 provide important requirements for these instances. If the cone-to-cylinder (with or without a knuckle) does not have sufficient moment of inertia, then it may not be considered as a line of support. In order to consider the cone-to-shell junction as a line of support, the moment of inertia at the junction must be equal to or larger than the required moment of inertia as determined in paragraph 1-8 of Appendix 1. If the cone-to-cylinder junction does not have sufficient moment of inertia to be considered as a line of support, then the unstiffened length (L) must be taken between lines of support on either side of the joint as shown in Fig. 21.1 (a-2 and c-2). (See Example 3.13 of Farr and Jawad's book [10] for an example of how to determine the moment of inertia of the cone-to-cylinder junction.) The designer is cautioned to carefully consider whether such junctions qualify as a line of support (adequate moment of inertia) when determining the stiffening ring location for these types of vessels. It is unusual that the actual moment of inertia at a cone to shell junction will satisfy the required moment of inertia unless there is a stiffener present. In practice, the junction between a cone and shell (with or without a knuckle) will normally require a stiffening ring if it is to be considered a line of support. Once the lines of support and vessel thickness are established, the allowable external pressure is determined using the external pressure charts of Section II, Part D. This is a "trial and error" process. If the allowable external pressure is less than the specified pressure, then either the vessel thickness or the length between lines of support must be changed. The most practicable approach for external pressure design is to add stiffening rings to reduce the length between lines of support rather than increasing the thickness of the vessel. When stiffening rings are added, the required moment of inertia is defined in paragraph UG-29. Two methods are given to evaluating the required moment of inertia. The value of Is considers only the moment of inertia of the stiffening ring with no consideration given to the combined ring-shell contribution to the total moment of inertia. Is' includes the structural effects of the shell by including that portion of the shell that is considered effective as contributing to the total moment of inertia. When Is' is used, smaller stiffening rings may be used. See the example problem L.5 in Appendix L for application of the stiffening ring design. When using the detailed design procedures of UG-29 for the design of stiffening rings, there are instances where the external pressure curves are horizontal. In these cases, there are multiple values of A possible for the calculated value of B. When this occurs, the smallest value of A on the horizontal part of the external pressure chart should be used. Stiffening rings may be located either on the inside of the vessel or on the outside. When stiffening rings are located on the inside, consideration should be given to corrosion effects and the ability to inspect the rings for in-service deterioration. Another more subtle consideration is that stiffening rings located on the outside of the vessel are inherently more stable than those located on the inside. For example, if the stiffening becomes "cocked" for some reason, the loading is such that rings on the outside of a

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 39

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 39

FIG. 21.1 DIAGRAMMATIC REPRESENTATION OF LINES OF SUPPORT FOR DESIGN OF CYLINDRICAL VESSELS SUBJECTED TO EXTERNAL PRESSURE (Source: Fig. UG-28.1 of Section VIII Div. 1 of the ASME 2007 Code)

vessel will tend to pull the ring back to its normal or more stable position. However, for rings located on the inside surface, an initially eccentric ring may become more eccentric under the loads resulting from external pressure. Because of these reasons (and the ease of fabrication), stiffening rings are often located on the outside of the vessel; however, there are instances where they are put on the inside of the vessel. The code has no restrictions on

either. Stiffening rings, if not proportioned correctly, may be subject to lateral buckling. Lateral instability is possible when the width of the ring is large compared to its thickness. The code does not provide specific requirements to avoid ring instability, and it is recommended that the AISC Manual for Steel Construction [32] be consulted for guidelines to avoid lateral instability of structural members.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 40

40 · Chapter 21

FIG. 21.2 MAXIMUM ARC OF SHELL LEFT UNSUPPORTED BECAUSE OF GAP IN STIFFENING RING OF CYLINDRICALSHELL UNDER EXTERNAL PRESSURE (Source: Fig.UG-29.2 of Section VIII Div.1 of the ASME Code)

When stiffening rings are used to resist external pressure, the provided stiffness has to be continuous around the circumference of the vessel. Gaps are allowed between the ring and the shell; however, the ring has to be continuous and the arc of the gap is limited by Fig. 21.2. If the arc of the gap between the ring and shell does not meet the Fig. 21.2 requirements, then the additional requirements of UG-29(c)(1) through UG-29(c)(4) must be satisfied. In addition to limits on the number and relative orientation of gaps in adjacent rings, a stiffening ring with a gap that exceeds the value allowed by Fig. 21.2 may not be considered to be a line of support for determining the unstiffened length of the shell. The stiffening ring may be interrupted if the required stiffness is provided by another component such as a support saddle, tray ring, or internal baffles, provided they are located at the ring and supply the necessary moment of inertia. Stiffening rings may be attached by stitch welds (staggered or in-line), or continuous welds on both sides of the ring, or continuously welded on one side and stitch welded on the opposite side of the ring. Details for sizing the attachment welds are given in paragraph UG-29 and are illustrated by Example L-5 in Appendix L. 21.4.2.5 Formed Heads Rules for the design of formed heads and sections with the pressure on the concave side of the component are given in paragraph UG-32.

The required thickness of ellipsoidal (with the major diameter twice the minor diameter) heads is given by t = PD 2SE - 0.2P or P = 2SEt D + 0.2t (21.4)

D diameter of the ellipse major axis Other terms are as given for the shell design formulae of UG-27. Ellipsoidal heads that do not have a major to minor diameter ratio of 2:1 shall be designed in accordance with Appendix 1-4. The required thickness for a torispherical head with the knuckle radius equal to 6% of the inside crown radius and the inside crown radius equal to the outside diameter of the skirt (straight section of the head attached to the adjacent shell) is given by t = where: L 0.885PL SE - 0.1P or P = SEt 0.885L + 0.1t (21.5)

inside crown radius of the formed head

The above equations for a torispherical head represent only one of many possible combinations of knuckle radius and crown radius. As it turns out, the head described by the above equation represents the "flatest" formed head with smallest knuckle radius

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 41

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 41

allowed by Division 1 [see paragraph UG-32(j)] that is not stayed. A head with these dimensions is commonly referred to as a "Code flanged and dished" head. This is not a precise description and the use of this terminology is discouraged. Other combinations of knuckle radii and crown radii are acceptable, provided that the knuckle radius is greater than 6% (but not less than three times the head thickness) of the crown radius and the crown radius is smaller than the outside diameter of the head skirt. Thus, a formed head with a knuckle radius of 10% of the crown radius and a crown radius of 80% of the skirt diameter is also a "Code F&D" head. Formed torispherical heads with dimensions other than those to which the above formulae are applicable shall be designed in accordance with Appendix 1-4. It should also be recognized that the requirements of paragraph UG-32 do not account for the possibility of elastic instability at the knuckle for thin heads. For thin formed heads with internal pressure, the membrane stress in the knuckle region can be compressive and can result in local buckling or wrinkling of the knuckle. For thin heads (with t/L 0.002), the requirements of paragraph 1-4(f) must also be satisfied. The rules of 1-4(f) assure that the compressive stress in the knuckle of formed heads does not result in elastic instability (wrinkling) of the head. The rules of paragraph UG-33 provide rules or formed heads with external pressure. The rules for ellipsoidal and torispherical heads under external pressure require the thickness to be checked using the internal pressure rules (UG-32) with P 1.67 times the external design pressure and E 1.0. The required thickness of the head is larger of that determined by this calculation or that determined by the external pressure rules of UG-33(d) or UG-33(e) as applicable. 21.4.2.6 Conical Heads and Sections Conical heads and sections, which do not have a transition knuckle, must meet the following thickness requirements: t = PD 2cos a(SE - 0.6P) D or P = 2SEt cos a D + 1.2t cos a (21.6)

a localized region. Unless a detailed stress analysis is performed, see paragraph 1-5(g), conical shells must have a half apex angle of 30° or less; otherwise, conical sections with a half apex angles greater than 30° must have transition knuckles. In order to reduce the discontinuity stresses at cone-to-shell connections, a toriconical head or section may be used. The minimum knuckle radius at the transition must be at least 6% of the outside skirt diameter and not less than three times its thickness. It should be noted that toriconical heads are required when the cone half apex angle exceeds 30° unless a detailed stress analysis is performed to assure that the discontinuity stresses are not excessive. Also, if the service requirements of Part UW require the Category B weld seam to be radiographically examined, then a transition knuckle must be provided if the half apex angle exceeds 30°. This is because a half apex angle of more than 30° results in the weld joint being classified as an angle joint that cannot be effectively RT examined. 21.4.2.7 Flat Heads and Covers Rules for unstayed flat heads and covers are given in paragraph UG-34. In summary, blind flanges conforming to the flange standards listed in U-3 may be used in accordance with the pressure­temperature limits of the standard; however, the final configuration of the blind flange or cover must comply with the geometry allowed by the standard. For example, ASME B16.5 allows single central openings in a blind flange of a limited size. If a ASME B16.5 blind flange is drilled with larger openings than provided for by the standard or if multiple openings are drilled into the blind flange, the pressure­temperature rating of the flange is no longer valid, and the thickness and reinforcement requirements of code must be satisfied. The required thickness for circular flat heads/covers that do not have an edge bolting moment; for example, they are attached by welding, is t = d2CP>SE where: d C E (21.7)

where:

one-half of the cone's apex angle the inside diameter of the cone at the point of consideration

A number of interesting points arise in the design of cones. These equations are the same as given for a straight shell; however, the radius of curvature normal to the cone's surface is used. It is apparent from the definition of D in the above formulae that the required thickness of a cone continually varies along its length since D is continually changing. Thus, a cone could be continually tapered; however, this is not feasible or practicable. When a conical section is made of more than one shell course, the thickness of each section may be based on the largest diameter for that section. In order to assure that the discontinuity stresses at the cone-tocylinder connection are not excessive, the cone-to-shell reinforcement rules of Appendix 1-5 shall be satisfied. The rules of Appendix 1-5 assure that the bending stress and the localized membrane stress in the area of the cone-to-shell connection are not excessive. The rules of Appendix 1-5 effectively limit the bending stress at the large end of the cone to SPS. This bending stress is a secondary stress and the greater stress limit is appropriate. Likewise, the rules of Appendix 1-5 effectively limit the membrane stress at the small end of the cone to 1.5 S. This limit assures that there is not an excessive amount of plastic deformation at this juncture. Again, this is a discontinuity stress acting in

S P

diameter over which pressure acts (see Fig. 21.3 for specific geometry) factor that is dependent on the edge restraint of the component weld joint efficiency for any welded seams made within the head or cover allowable stress from Section II, Part D design pressure

The values of C are dependent on the amount of edge restraint that the attached shell provides to stiffen the flat head. When classical flat plate theory is reviewed, it is found that the stress in the center of a flat head is dependent on the edge or boundary conditions. If the edge is fixed, then the stress at the center of the flat head is less than that when the edge is simply supported. The C value provided in Fig. 21.3 accounts for this edge effect. The geometries that provide more effective edge constraint receive a smaller value of C that results in a smaller required thickness for those flat heads. It is also noted that the values of C for covers welded to an adjacent cylinder includes a factor that allows the stress in the head to be as large as 11/2 S. The stress at the center of a flat cover is considered a primary bending stress. The allowable stress of 11/2 S is consistent with the allowable primary bending stress found in Section VIII, Division 2. The specific provisions of paragraph UG-34(d) must be applied to determine the value of C for a particular geometry. The value of E (weld joint efficiency) only applies when the flat cover

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 42

42 · Chapter 21

FIG. 21.3 SOME ACCEPTABLE TYPES OF UNSTAYED FLAT HEADS AND COVERS (THE ABOVE ILLUSTRATIONS ARE DIAGRAMMATIC ONLY; OTHER DESIGNS THAT MEET THE REQUIREMENTS OF UG-34 ARE ACCEPTABLE) (Source: Fig. UG-34 of Section VIII Div.1 of the ASME Code)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 43

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 43

contains a butt weld, such as when two plates are butt welded together to form a single plate. The value of E depends on the extent of radiographic examination done on that weld. If there are no butt welds in the plate, then E 1. Note that joint efficiency is not a design consideration for nonbutt-welded corner joints that attach flat heads to a cylinder. The required thickness for circular flat heads and covers that are bolted to the attached shell is t = d2CP>SE + 1.9WhG>SEd3 where: W hG (21.8)

the design bolt load for the flange calculated by Appendix 2 gasket moment arm as calculated in Appendix 2 Other terms have been previously defined.

cross section through the opening. Likewise, the reinforcing elements must be attached to the shell and nozzle by welds that have sufficient strength (strength path) to transmit and distribute the loads resulting from pressure in the vicinity of the opening. Use of the nozzle reinforcement rules and strength path determination are shown by the examples given in paragraph L-7 of nonmandatory Appendix L and will not be repeated here. There are several peripheral issues regarding opening reinforcement that warrant some discussion. Paragraph UG-36(c)(3) provides exemption from reinforcement calculations for welded or brazed connections in vessels not subject to rapid fluctuations if the finished opening is not larger than 31/2 in. diameter­in vessel shells or heads of 3/8 in. or less thickness, or 23/8 in. diameter­in vessel shells or heads over 3/8 in. thickness. The "finished opening" is the opening of the completed assembly. (See Fig. UG-40 for definition of finished opening, d.) For example, if a 4-in. diameter hole is cut into a 3/8-in. thick vessel to accept a nozzle with a thickness of 1/4 in., the finished opening is 31/2 in. (4 in. minus two times 1/4 in.) and the opening is exempt from the reinforcement rules. As noted, this exemption does not apply for vessels in cyclic pressure service. If the "area replacement" requirement is not satisfied, large localized stresses may exist at the opening, and this may have an adverse effect if cyclic pressure loadings occur. If the vessel is not in cyclic pressure service, these localized stresses are not deemed to be significant for the size of openings allowed by the exemption. Paragraph UG36(c)(3) also provides limitations regarding the spacing of openings that are exempt from the reinforcement rules. When an exemption from reinforcement is not provided by paragraph UG-36(c)(3), then the opening must be reinforced as per the rules given in paragraphs UG-37 through UG-43. As stated earlier, these rules are based on the area replacement method, and are only applicable where the wall thickness of the shell or head in which the opening is placed is controlled by internal or external pressure. These rules do not address the requirements for openings under the action of externally applied loadings, such as type reactions, or a vessel subject to wind or seismic loads. When externally applied loadings are to be considered, see U-2(g). Figure 21.4 is often misinterpreted or is not fully understood, and an explanation of the requirement is in order. The area of reinforcement required at any cross section through the opening must be equal to the shell area required for the design pressure that is removed by the opening (i.e., minimum required shell thickness times the finished diameter of the opening). In a cylindrical shell, the stress varies from the longitudinal plane to the circumferential plane; the circumferential pressure stress is twice as large as the longitudinal pressure stress. Thus, required pressure thickness in the circumferential plane of an opening (i.e., 90° from the longitudinal axis) is 50% of that required for the longitudinal plane. Accordingly, the amount of reinforcement area required for the circumferential plane is only one half of that required for the longitudinal plane. For integrally reinforced openings in cylinders and cones, the value of F, as shown in Fig. 21.4, may be applied to the required replacement area at any given plane in order to account for this effect. (An integrally reinforced opening does not use a separate reinforcing element such as a reinforcement plate.) However, the correction factor F is set equal to 1.0 for all openings, except for integrally reinforced openings in cylinders and cones. Paragraph UG-39 provides the reinforcement rules for openings in flat heads. Since the wall thickness of a flat head is predominantly

This equation accounts for the bending stress in the flat cover that is introduced by the moment produced by the reaction between the bolts and gasket. It should be noted that the increased allowable stress (11/2 S for welded covers) is not allowed for bolted covers because the resulting additional deformation may cause gasket sealing problems. Paragraph UG-34 also provides the design basis for noncircular flat covers with and without edge bolting moments. 21.4.2.8 Openings and Reinforcements When an opening is made in a pressure vessel, there is a stress intensification resulting from the hole that is formed in the shell. This is analogous to the classical stress concentration effect of a hole in a plate that is loaded in traction. The code reinforcement rules do not consider loads other than pressure. For example, external loadings due to piping or supported components are not evaluated by these rules and separate consideration is required when applicable. Openings in shells should preferably be round, elliptical, or obround. When a circular connection is made normal to the surface of the vessel, a round opening in the vessel results. If the connection is oblique to the surface of the shell, the opening in the shell is elliptical. Obround openings are made by connections formed by parallel sides with semicircular ends. Openings of other shapes shall be provided with a suitable radius to minimize stress concentration effects in the shell. When the strength of vessels with such openings cannot be determined with accuracy or when doubt exists regarding its strength, the proof test provisions of paragraph UG-101 shall be applied. There is no limit to the size of an opening that may be installed in a pressure vessel. The opening reinforcement rules given in UG-36 through UG-43 apply to openings not exceeding the following: for vessels of 60 in. inside diameter and less, the opening may be as large as one half the vessel diameter, but not to exceed 20 in.; for vessels over 60 in. inside diameter, the opening may be as large as one third the vessel diameter, but not to exceed 40 in. Openings that exceed these limits shall also satisfy the supplemental rules of Appendix 1, 1-7, or alternatively the rules of 1-10. The method given in Appendix 1,1-10, is based on pressure­area calculations, and is essentially identical to the opening reinforcement rules published in the 2007 Section VIII, Division 2. These alternative rules provide a more accurate solution for determining the reinforcement requirements for large openings, and in many cases results in less reinforcement required as determined by the rules given in Appendix 1,1-7. The basic philosophy of the code reinforcement rules assures that at any cross section through an opening, the area of the shell that is removed by the opening is replaced by material of adequate strength and equivalent area adjacent to the opening. The reinforcement must be balanced on either side of the opening at any

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 44

44 · Chapter 21

FIG. 21.3A SOME REPRESENTATIVE CONFIGURATIONS DESCRIBING THE REINFORCEMENT DIMENSION te AND THE OPENING DIMENSION d (Source: Fig. UG-40 of Section VIII, Division 1 2007 Edition of the ASME Code).

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 45

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 45

(2) If the vessel is designed for external pressure, set the pressure equal to the external design pressure and recalculate the required thickness as in (1) above. (3) If the vessel is designed for both internal and external pressure, use the larger of (1) or (2) above. (4) The minimum thickness of standard wall pipe plus any required corrosion allowance for the connection. The minimum thickness of standard wall pipe is the nominal wall thickness less 121/2%. As an example, assume that UG-22 loadings require a nozzle neck minimum thickness to be 0.2 in. including corrosion allowance. If the calculation in (1) shows the required vessel thickness (including corrosion allowance) to be 0.45 in. at the location of the nozzle attachment and there is no external pressure design, the larger of (1) and (2) is 0.45 in. The minimum thickness for standard weight pipe is 0.375 in. times 0.875 or 0.328 in. If the nozzle has 0.063 in. corrosion allowance, then (4) requires the minimum neck thickness to be 0.391 in.(0.328 plus 0.063). Thus, the smallest of (1) through (4) above is 0.391 in. Since this is larger than the thickness required by the UG-22 loadings, the required minimum nozzle neck thickness is 0.391 in. Since 0.391 in. is greater than the minimum thickness of standard weight pipe (0.328 in.), the next heavier schedule pipe must be used to satisfy the UG-45 requirements. 21.4.2.9 Attachments to Pressure Vessels General rules for the attachment of nonpressure parts to pressure components are provided by paragraphs UG-54 and UG-55. The rules presented are not prescriptive. Any connections made to a pressure vessel shall be done with consideration of the loading conditions defined in paragraph UG-22. Using all applicable loading conditions, the design shall be such that neither the attachment nor the vessel wall is overstressed. Appendix G provides recommended good practice for the design of supports and attachments to pressure vessels. 21.4.2.10 Inspection Openings All pressure vessels for use with compressed air and those subject to internal corrosion shall be provided with suitable manhole, handhole, or other inspection openings for examination and cleaning. Several exemptions for inspection openings are given, including a declaration on the Manufacturer's Data Report that the vessel is intended for "noncorrosive service." This raises an interesting question: if the user specifies a corrosion allowance for a vessel that is in noncorrosive service, should most of this vessel be supplied with inspection openings? Based on Interpretation VIII-1-95-46, the answer is no. Interpretation: VIII-1-95-46 Subject: Section VIII, Division 1 (1992 Edition, 1993 Addenda), UG-46(a) Date Issued: February 6, 1995 File: BC94-679 Question (1): If a user specification states that a vessel is to be designed with a corrosion allowance, may this fact be sufficient verification that the vessel is intended for corrosive service? Reply (1): Yes. Question (2): May a vessel that has been designed with a corrosion allowance be specified under remarks on the Manufacturer's Data Report Form "for noncorrosive service?"

FIG. 21.4 CHART FOR DETERMINING VALUE OF F, AS REQUIRED IN UG-37 (Source: Fig. UG-37 of Section VIII Div. 1 of the ASME Code)

based on bending stresses, the required area of reinforcement for an opening in a flat head is typically half of that required for a similar opening in a cylindrical shell or formed head. The amount of available reinforcement area adjacent to an opening is based on the definition of the limits of reinforcement as given in paragraph UG-40. Limits are specified parallel and perpendicular to the surface of the shell or head in which the opening is placed. Another critical variable is the finished opening dimension d. As shown in Fig. 21.3A, the finished opening d varies depending on the nozzle attachment configuration. Paragraph UG-45 provides rules for minimum nozzle neck thickness. A nozzle neck or any other connection shall not be thinner than that required to satisfy the thickness requirements for the loads defined in paragraph UG-22. Except for manways and other openings that are provided only for access, additional requirements of paragraph UG-45 may require a thicker nozzle neck. Paragraph UG-45(b) requires the neck to be no thinner than the smallest of the following: (1) The required thickness for internal pressure (plus any required corrosion allowance) for the shell or head at the location where the nozzle is attached. This calculation is to be done with E 1.0. This thickness cannot be less than the minimum required thickness given in UG-16(b).

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 46

46 · Chapter 21

Reply (2): Yes, provided noncorrosive service has been specified by the user. 21.4.2.11 General Fabrication Requirements Paragraphs UG-75 through UG-85 provides general fabrication requirements applicable for all types of materials and methods of construction. Paragraph UG-77 requires the Manufacturer to maintain the traceability of material used in a vessel to the original material identification. There are several accepted ways of providing the material identification as follows: (1) Transfer the original material marking to the vessel where they are visible on the completed vessel. (2) Identification by a coded marking that is traceable to the original material marking. (3) Recording the required markings on material tabulations or sketches that assure that each piece may be identified during fabrication and subsequently in the completed vessel. Paragraph UG-78 allows the Manufacturer to repair defects in materials, provided that the Inspector accepts the repair method and extent of repair. Material that is found to be defective and cannot be repaired, or the repair is not acceptable to the Inspector, shall be rejected. Repair of material pertains to material that is manufactured to one of the allowed material specifications, and a defect in the material is found subsequent to the delivery of the material in the Manufacturer's facility. In accordance with the rules of UG-79, any process may be used to form shells and heads that will not unduly impair the physical properties of the material. The Manufacturer may form the shells and heads or he may contract to have the forming done by others. The Manufacturer should know the method to be used in the forming operation because other requirements of Section VIII, Division 1, may apply depending on the forming methods used. For example, restrictions apply to cold formed carbon and low alloy material [see UCS-79(d) and UHT-79(a)]. When carbon and low alloy steel material is hot formed (heated above the lower transformation temperature of the material), the provisions of UCS-85 may require simulated heat treatment of the mill test specimens. The Manufacturer must consider all the applicable requirements for the materials and types of construction when the method of forming is selected. Paragraph UG-80 provides rules for the determination of the permissible out-of-roundness of shells. For internal pressure, the vessel out-of-roundness measured by the difference between the maximum and minimum diameter at a cross section cannot be greater than 1% of nominal diameter. This tolerance applies to all sections, other than at nozzles, normal to the axis of the vessel including circumferential joints and joints between a shell and head. Because some additional allowance is needed for the local deformation at nozzles, the permissible out-of-roundness tolerance is 2% of the nominal diameter at cross sections near an opening. The out-of-roundness tolerance provides a limitation on the amount of bending stress imposed in a shell under the application of internal pressure. When the vessel will experience external pressure, additional requirements are specified in order to limit the "as fabricated" shape of the vessel. The requirements that are applicable for internal pressure also apply to external pressure; however, the deviation from true circular form cannot exceed the value of the term e that is derived from Fig. 21.5. The value of e is dependent on the vessel outside diameter to thickness ratio and the design length (for cylinders, this is the unstiffened length as defined and used in

paragraph UG-28) to outside diameter ratio. Special rules are presented in paragraph UG-80 for determining the value for L and Do for cones and conical sections. Deviation from true circular form is measured by using a segmental circular template with a chord length equal to twice the arc length as determined by Fig. 21.2. If the measurements are to be taken from the outside, then the radius of the template shall be the design outside radius. If the measurements are made from the inside, the template radius shall be the design inside radius. Sample problem L-4 in nonmandatory Appendix L demonstrates the use of the tolerance rules for vessels under external pressure. When the nominal thickness used in the construction of the vessel exceeds the minimum thickness required for external pressure, the allowed deviation from circular form may be increased. Paragraph UG-80(b)(9) allows the value of e to be increased by the ratio of the B factor (from the external pressure chart of Section II, Part D) using the actual nominal thickness divided by the B factor using the minimum required thickness. Some pipe and tube material specifications (See Section II) allow out-of-roundness greater than that allowed by Division 1. Paragraph UG-80(b)(10) allows the use of these materials for external pressure without any additional consideration. The designer may wish to consider imposing additional dimensional requirements for those components made according to these material specifications especially when they are subjected to large external pressure, such as heat exchanger tubing with high pressure on the external surface. The tolerances of formed heads are given in paragraph UG-81. For vessels under internal pressure, the inner surface of a torispherical, toriconical, hemispherical, or ellipsoidal head shall not deviate outside its specified shape by more than 11/4% of the nominal inside diameter of the vessel at the point where the head is attached. Likewise, the formed head shall not deviate inside its specified shape by more than 5/8% of the shell diameter. The knuckle radius of formed heads shall not be less than that specified. These requirements assure that a formed head does not have a flat spot that will experience significant bending under the action of internal pressure. For vessels under external pressure, spherical heads and the spherical portions of formed heads shall also meet the requirements of UG-80(b) using a value of 0.5 for L/Do. The tolerance on deviation from true circular form for cylinders and heads is required to validate the design methods used for external pressure. External pressure can cause the vessel to buckle by elastic or elastic­plastic instability. As with any stability problem, the initial imperfections will have a dramatic effect on the value of the failure load. The tolerance on deviation from true circular form will assure that margins included in the design for external pressure are not compromised. Caution should be used when the external pressure design rules of Section VIII are used to evaluate an existing vessel or for a vessel that will not be stamped for external pressure [see paragraph U-1(c)(2)(h) and UG-28(f)]. In these instances, consideration should be given to shape of the vessel and if it cannot be validated that the out-of-roundness and deviation from true circular form meet the requirements of UG-80, engineering judgment should be exercised in establishing the maximum allowable external pressure of the vessel. 21.4.2.12 Impact Test Requirements When impact testing is required, the provisions of UG-84 shall be satisfied. Impact testing is required as determined elsewhere in Division 1 [see paragraphs UG-20(f), UCS-66, and UHA-51]. Paragraph UG-84(b) defines the procedures to be used for conducting Charpy V-Notch (CVN)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 47

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 47

FIG. 21.5 MAXIMUM PERMISSIBLE DEVIATION FROM A CIRCULAR FORM e FOR VESSELS UNDER EXTERNAL PRESSURE (Source: Fig. UG-80.1 of Section VIII Div. 1 of the ASME Code)

impact tests of base material, welds, and heat-affected zones. Each set of impact tests requires three specimens that must conform to the size and shape requirements of Fig. 21.6. Each specimen must be tested at a temperature not warmer than the minimum design metal temperature (unless allowed otherwise by Table UG-84.4). If the material thickness is such that full-sized impact specimens cannot be obtained, it is allowed to impact test of smaller specimens at a temperature lower than the minimum design metal temperature as given in paragraph UG-84(c)(5)(b). The impact test acceptance criteria are given in Fig. 21.7. The required absorbed energy for each specimen is a function of the nominal thickness of the material or weld and the specified minimum yield strength (SMYS) of the material. For the thickness and SMYS of the material or weld being impact tested, the average of the three specimens must be at least the value given in Fig. 21.7, and the minimum absorbed energy of any one specimen may not be less than 2/3 of the required average. For example, consider SA-516 Grade 70 material that is 2 in. thick. This material has a SMYS of 38 ksi and the required average value of the absorbed energy of the three specimens from the Charpy V-Notch test is 15 ft-lbs and no single specimen can have an absorbed energy less than 2/3 of 15 ft-lbs or 10 ft-lbs. When impact testing is required for steel vessels of welded construction, in addition to the base material impact tests, impact

FIG. 21.6 SIMPLE BEAM IMPACT TEST SPECIMENS (CHARPY TYPE TEST) (Source: Fig. UG-84 of Section VIII Div. 1)

tests shall be done as part of the Welding Procedure Qualification (WPQ) and impact specimens shall be taken from welds made during fabrication (production impact tests). These impact tests assure that the welding procedures and the welding operators will produce actual welds and heat-affected zones of sufficient toughness for the specified MDMT. One note about production impact tests. As per UG-84(i)(1), when production impact tests are required, only the welds of Categories A and B joints need to be tested. For Category A joints, the test plate shall, where practicable, be welded as an extension to the end of the production joint

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 48

48 · Chapter 21

FIG. 21.7 CHARPY V-NOTCH IMPACT TEST REQUIREMENTS FOR FULL SIZE SPECIMENS FOR CARBON AND LOW-ALLOY STEELS, HAVING A SPECIFIED MINIMUM TENSILE STRENGTH OF LESS THAN 95 KSI, LISTED IN TABLE UCS-23 (Source: Fig. UG-84.1 of Section VIII Div. 1 of the ASME 2007 Code)

so that the test plate weld will represent as nearly as practicable the quality and type of welding in the vessel joint. For Category B joints that are welded using a different welding procedure than used on Category A joints, a test plate shall be welded under the production welding conditions used for the vessel using the same

type of equipment, the same procedures as used for the joint, and it shall be welded concurrently with the production welds. Production test of Category C (e.g., corner joints and flange attachment welds) and Category D (nozzle attachment welds) is not required.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 49

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 49

21.4.3

Inspection and Tests

21.4.3.1 General Paragraphs UG-90 through UG-103 provide general requirements for inspections and tests. These requirements are in addition to the inspection and testing requirements found elsewhere in Division 1. The general requirements of UG-90 summarize in some detail the Manufacturer'sresponsibilities. They include the following: (1) Obtain the Certificate of Authorization from ASME that allows the Manufacturer to apply the code symbol. (2) Prepare drawings and calculations. (3) Identification of all materials used in the vessel. (4) Obtain partial data reports for fabrication work involving welding done by other organizations. (5) Provide access to the Inspector. (6) Examination of all materials before fabrication. (7) Documentation of impact test results when impact testing is required. (8) Obtain concurrence of the Inspector before any repairs are made. (9) Examination of all shell and heads to assure they comply with the required tolerances. (10) Qualification of welding and/or brazing procedures and welders/brazers. (11) Examination of fit-up to assure proper cleaning and alignment. (12) Assure provisions are in place to meet all heat-treatment requirements when required. (13) Maintain records of nondestructive examination. (14) Conduct the required pressure test of the vessel. (15) Apply the proper marking to the vessel nameplate. (16) Prepare the Manufacturer's Data Report and have it certified by the Inspector. Likewise, the duties of the Inspector are summarized in UG90(c)(1). The Inspector shall make all inspections that are specifically required by the code and any other inspections he believes necessary for him to certify that the vessel has been constructed in accordance with all applicable rules. Some of the required inspections are provided below: (1) Verify the Manufacturer has a valid Certificate of Authorization. (2) Verify that design calculations are available. (3) Verify that the materials of construction are properly identified. (4) Verify the welding/brazing procedures have been qualified. (5) Verify that welders/brazers have been qualified. (6) Verify that all required heat treatments have been performed. (7) Verify that material defects have been repaired in an acceptable manner. (8) Verify that all required NDE has been done and that the results are acceptable. (9) Verify that all material identification markings have been properly transferred. (10) Conduct a visual inspection to confirm that no material or dimensional defects are present. (11) Witness the pressure test. (12) Verify that the required marking is applied to the vessel nameplate. (13) Sign the Certificate of Authorization on the Manufacturer's Data Report.

Paragraph UG-90(c)(2) provides for the special circumstance where a Manufacturer makes duplicate pressure vessels on high production line basis, such as may be the case for small air receiver vessels that are manufactured and sold on a mass scale. For these situations, it is not practicable for the Inspector to perform all of his required duties on each pressure vessel. For these situations, the Manufacturer, in collaboration with the Inspector, shall prepare detailed procedures that define how all the requirements of the code will be satisfied. This procedure has to be included in the Manufacturers Quality Control System and be accepted by the jurisdictional authority, the inspection agency, and the representative (designee) of ASME. The term "Inspector" is widely used throughout Division 1, and is defined in paragraph UG-91. The Inspector must be regularly employed by an ASME-accredited Authorized Inspection Agency. An accredited agency can be the inspection organization of a state or municipality of the United States, or Canadian Province, or an insurance company authorized to write boiler and pressure vessel insurance. However, when a user has a pressure vessel manufacturing facility to fabricate pressure vessels for its own use and not for resale, the Inspector may be an inspector employed by the user. This arrangement allows a user to fabricate, possibly on an emergency basis, pressure vessels for use in their own facilities. Otherwise, the Inspector shall not be in the employ of the Manufacturer. All Inspectors shall have been qualified by a written examination under the rules of any state of the United States or province of Canada that has adopted the code. Normally, the states and provinces recognize the inspector certification process conducted by the National Board of Boiler & Pressure Vessel Inspectors. The Manufacturer is required to inspect materials to be used in the fabrication of pressure vessels. Materials accepted for use must be in compliance with the requirements of paragraph UG93. These are summarized as follows. (1) Proper documentation must be provided. For plates, a material test report or certificate of compliance is required (as provided for in the material specification) and the Inspector shall examine the documentation and assure it complies with the material specification requirements. For materials other than plate (e.g., pipe, tube, forgings, castings, and others), documentation may include marking of each piece or the marking of each bundle, lift, or container with the proper material designation as provided for by the material specification. (2) When tests are required by Division 1 on material and those tests are not conducted by the material supplier, the Manufacturer must obtain supplementary material test reports or certificates of compliance that demonstrate that the materials meet the requirements and represent the material supplied. For example, if a heat treatment (e.g., quench and temper) required by the material specification is not done by the material supplier, the vessel Manufacturer must obtain the necessary documentation, including the results of additional mechanical tests, to confirm that material complies with the material specification. Likewise, when Division 1 requirements exceed or supplement those of the material specification, the Manufacturer must obtain any supplementary documentation that shows compliance with the Division 1 requirements. (3) Before fabrication begins, all material shall be examined to assure that there are no imperfections that could affect the

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 50

50 · Chapter 21

safety of the vessel. This includes visual inspection of the cut edges of the material for laminations, cracks, and other defects. For material that is required to be impact tested, the surfaces shall be inspected for surface cracks and repaired, if necessary. (4) During fabrication, paragraph UG-93 requires nondestructive examinations for certain corner joint weld details. When a pressure part is attached to an unsupported flat plate thicker than 1/2 in. to form a corner joint, the weld edge preparation and the outside edge of the flat plate require examination by magnetic particle or liquid penetrant examination. The flat plate is considered to be supported if at least 80% of the pressure load is carried by tubes, stays, or braces. The code does not provide guidance of how to determine the amount of load that is carried by the tubes, stays, or braces. 21.4.3.2 Maximum Allowable Working Pressure Paragraph UG-98 provides a discussion about maximum allowable working pressure for a pressure vessel. Although this term is used widely, there is a chance of misunderstanding of what it really means and how it relates to the design pressure. The following is taken from Appendix 3 for the definition of maximum allowable working pressure and design pressure. (a) Design pressure. The pressure used in the design of a vessel component together with the coincident design metal temperature, for the purpose of determining the minimum permissible thickness or physical characteristics of the different zones of a vessel. When applicable, static head shall be added to the design pressure to determine the thickness of any specific zone of the vessel (see UG-21). (b) Maximum Allowable Working Pressure. The maximum gauge pressure permissible at the top of a completed vessel in its operating position at the designated coincident temperature for that pressure. This pressure is the least of the values for the internal or external pressure to be determined by the rules of this division for any of the pressure boundary parts, including the static head thereon, using nominal thicknesses exclusive of allowances for corrosion and considering the effects of any combinations of loadings listed in UG-22 that are likely to occur (see UG-98) at the designated coincident temperature [see UG-2(a)]. It is the basis for the pressure setting of the pressure-relieving devices protecting the vessel. The design pressure may be used in all cases in which calculations are not made to determine the value of the maximum allowable working pressure. As may be seen, the MAWP of a vessel may be determined by using the thickness provided for fabrication, not including any allowance for corrosion, to determine the allowable pressure. This would include confirming nozzle reinforcement, cone-to-shell reinforcement, and so on to establish the MAWP. It is often more convenient to arbitrarily let the MAWP of the vessel be equal to the design pressure, and not conduct the calculations to determine the actual MAWP. The definition of the MAWP from Appendix 3 expressly allows this simplification. When the MAWP is set to the design pressure, the MAWP is "limited by design" and may not be a true representation of the actual pressure containing ability of the vessel. If the user wishes that calculations be conducted to determine the MAWP of a vessel in lieu of using the design pressure, then that should be a contractual requirement between the user and the Manufacturer.

21.4.3.3 Pressure Tests All completed pressure vessels must be subjected to a pressure test. The requirements for pressure tests are provided in paragraphs UG-99 and UG-100. The pressure test achieves a number of beneficial effects. The most obvious is that a pressure test will expose leaks at flanged connections and at welded and/or brazed joints. Additionally, the pressure test will expose a gross error in the design of the pressure containment. The pressure test will also result in a mechanical "stress relief" of the vessel. Local regions of high stress, such as occurs at stress concentrations and crack-like imperfections, will undergo local yielding at the test pressure. After the release of the pressure, a more favorable stress pattern is achieved because these local areas will have residual stresses that will reduce the local stress in subsequent applications of pressure. This latter benefit is especially important when considering the potential of brittle fracture [see paragraph UG20(f)]. The most desirable pressure test is a hydrostatic test where the vessel is filled with a liquid and pressurized in excess of its maximum allowable working pressure. The hydrostatic test must be conducted after the vessel has been completed. It is noted that no welding shall be done on a vessel after the hydrostatic test. It is permissible to machine nozzle ends and conduct cosmetic grinding that does not affect minimum wall thickness after the test. The following interpretation illustrates this point. Interpretation: VIII-80-10 Subject: Section VIII, Division 1, Welding of Attach ments after the Final Hydrostatic Test, UG99(a) Date Issued: February 7, 1980 File: BC80-74 Question: May nonpressure attachments be welded to a Section VIII, Division 1, pressure vessel subsequent to the final hydrostatic test required by UG-99(a) without subsequent hydrostatic testing and prior to the stamping of the completed vessel? Reply: No, Section VIII, Division 1, requires that the final hydrostatic test as required by UG-99(a) be the last operation prior to the stamping of the completed vessel. The minimum required hydrostatic test pressure is determined as follows, Ptest Where: Ptest (Sa/S)min 1.3 MAWP(Sa/S)min (21.9)

minimum required hydrostatic test pressure smallest ratio of the allowable stress at test temperature to the allowable stress at design temperature for all the materials used in the vessel.

It is noted that (Sa/S)min is the smallest value for the materials used for pressure containment in the vessel. For example, a vessel is constructed predominantly of material that gives (Sa/S)min 2.3; however, if there is one component in the vessel where 1.0, then the minimum hydrostatic test pressure is (Sa/S)min 1. Heat exchangers and other vessels with based on (Sa/S)min multiple chambers that can act independently should have each chamber pressure tested separately without pressure in the adjacent chambers.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 51

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 51

Paragraph UG-99(c) provides rules for an alternative hydrostatic test that will generally result in a larger hydrostatic test pressure that may be used when agreed to by the Manufacturer and user. This paragraph allows a "calculated test pressure" to be used to establish the test pressure. The calculated test pressure is the calculated MAWP of the vessel in "new and cold" condition. This means that the actual thickness (including any thickness included for corrosion) of the vessel is used with the allowable stress at the test temperature to determine the maximum allowed pressure. Once the calculated test pressure is determined, the minimum hydrostatic test pressure is 1.3 times the calculated test pressure (corrected for any liquid head at any location). This is an alternative approach to the minimum test pressure of UG99(b) and is used when contractually agreed between the Manufacturer and user. When this approach is used, the inspector has the right to assure that the basis for test pressure calculations have been conducted. Paragraph UG-99(b) provides the minimum test pressure that may be used, and UG-99(c) provides an optional basis; however, the code does not specify a maximum pressure to which a vessel can be hydrostatically tested. If the vessel experiences visible permanent deformation as a result of the pressure test, the Inspector has the right to reject the vessel. This, in effect, defines the upper limit on hydrostatic test pressure. The test pressure should not be so large as to result in visible permanent deformation anywhere in the vessel. Following the application of the required hydrostatic test pressure, the pressure shall be reduced to a value not less than the test pressure divided by 1.3. The vessel shall then be inspected for leakage at this lower pressure. Except for temporary covers that may be required for welded connections to be made in the field, no observable leakage is allowed at permanent bolted connections or welds. Any nonhazardous liquid may be used for the hydrostatic test medium if the test temperature is below its boiling point. The metal temperature of the vessel should be sufficiently warm to preclude the possibility of brittle fracture. Paragraph UG-99 recommends that the metal temperature be maintained at 30°F above the minimum design metal temperature (see UG-20). For personnel safety, it is recommended that the temperature of the test fluid must not exceed 120°F to avoid the possibility of burns from potential leaks. Unless it is in lethal service (see paragraph UW-2), a pressure vessel may be painted, coated, or lined, either inside or outside, prior to the hydrostatic test. Because of the nature of lethal fluids, it is important that no pinholes or other small leak paths be covered or plugged prior to the pressure test, and coating and/or lining a vessel that is in lethal service prior the pressure test is not allowed. It should be understood that painting, coating, or lining can provide an effective barrier to leaks that otherwise would be discovered during the hydrostatic test. If the user does not wish to allow painting/coating/lining of the vessel prior to the pressure test, then that must be defined in the contractual documents. There are pressure vessels where a hydrostatic test is not desired because of the nature of the service or because of the nature of the design. This may include vessels in a service where the test liquid can contaminate the process or where it is impracticable to design the vessel such that it can be supported with the weight of the test liquid. For such instances, the Manufacturer may conduct a pneumatic test. Any decision to conduct a pneumatic test requires very careful consideration because the stored

energy of a compressible fluid is considerably greater than when a liquid is used. In the unlikely event when a component fails during the pressure test, extensive damage may occur to the vessel and its surroundings. Before a pneumatic test is conducted, the examination requirements of paragraph UW-50 shall be satisfied. This paragraph requires all welds around openings and all attachment welds be subjected to magnetic particle or liquid penetrant examination prior to the pressure test. The minimum required test pressure for a pneumatic test is Ptest 1.1 MAWP(Sa/S)min (21.10)

The pneumatic test shall not exceed 1.1 times the basis for calculated test pressure. During a pneumatic test, it is mandatory that the metal temperature be at least 30°F warmer than the minimum design metal temperature as defined in UG-20 in order to minimize the risk of brittle fracture. As with the hydrostatic test, the examination for leakage is done after the pressure is reduced to the test pressure divided by 1.1. Leakage is not allowed, except at those temporary test closures for openings to welded in the field. In addition to a hydrostatic or pneumatic pressure test, paragraph UG-101 allows a pressure proof test to be used for any geometry where the maximum allowable working pressure cannot be assured analytically. The proof test may be based on yielding or bursting of the vessel, with a suitable safety margin applied to the proof test pressure to determine the MAWP of the vessel. If a vessel that has been proof tested is to be put into service, it does not require an additional pressure test. Vessels that are duplicates of a vessel where the design has been proof tested must undergo either a hydrostatic test or a pneumatic test. The Inspector must witness any proof test of a vessel or a component. A proof test report describing the test, the instrumentation and the methods of calibration used, and results obtained must be prepared by the Manufacturer and certified by the Inspector. Provisions are given in UG-101(a)(4) for transfer of proof test results when a Manufacturer is acquired by a new owner. 21.4.3.4 Marking and Reports Paragraphs UG-115 through UG-120 define the required marking to be placed on the vessel and the reports that must be prepared. The information required on the nameplate and on all the reports shall be in dimensions that are either U.S. customary units, SI units, or local customary units (see U-4). Each pressure vessel built in accordance with the code requirements shall be marked with the following information: (1) (2) (3) (4) (5) (6) the official Code U or UM [see U-1(j)] Symbol the name of the manufacturer the MAWP at the coincident maximum design temperature the MDMT at the MAWP the Manufacturer's serial number the year built

The nameplate stamping shall include a letter designation indicating the type of construction. This letter is to be directly under the code symbol and shall be one of the following: W­ for arc or gas welded P­ for pressure welded (except for resistance welding) B­ for brazed RES­ for resistance welded

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 52

52 · Chapter 21

If the vessel satisfies special service provisions, the following designation (to be located after the letter designating the type of construction) shall be marked on the nameplate: L­ for lethal service UB­ for unfired steam boiler DF­ for direct firing In addition, when the vessel has been radiographed (or ultrasonically examined as per Code Case 2235), the following designation shall be provided under the code symbol: RT 1­ when all butt welds [except those of Categories B and C welds in nozzles exempted by UW-11(a)(4)] in the vessel have been fully radiographed RT 2 ­ when the complete vessel satisfies the requirements of UW-11(a)(5) and spot radiography is done in accordance with UW-11(a)(5)(b) RT-3­ when the complete vessel satisfies the spot radiography requirements of UW-11(b) RT-4 ­ when only a part of the vessel meets the requirements for "full radiography" or none of the above radiography categories are applicable. Additionally, designation "HT" is required on the nameplate when the complete vessel has been postweld heat treated. If only part of the vessel is heat treated, then the letters "PHT" are required. The Manufacturer's Data Report will contain much more details regarding the vessel. Detailed instructions for completing the MDR may be found in nonmandatory Appendix W. Paragraph UG-117 addresses Certificates of Authorization and Code Symbol Stamps. A company who desires to fabricate pressure vessels in accordance with ASME Code with stamping must apply to the Boiler and Pressure Vessel Committee of the Society for a Certificate of Authorization. In Section VIII, Division 1, authorization to use the "U" and "UM" stamps are granted at a single location only. This means that if a company operates at more than one location, then they will need to apply for a Certificate of Authorization for each of their locations where they intend to fabricate and stamp pressure vessels. The Certificate of Authorization is granted for a three-year period, after which another audit by the society is required. This audit of the Manufacturers Quality Control System is carried out by an ASME designee as well as a representative of the inspection agency. 21.4.3.5 Pressure Relief Devices Paragraphs UG-125 through UG-137 provide requirements for pressure relief devices. Pressure relief devices, as described by these requirements, must protect all pressure vessels that receive the code symbol. An exception to this requirement is possible for some vessels by applying the Code Case 2311 "Overpressure Protection by System Design." (The user must specify when this option is to be applied.) Since most pressure vessels are installed in a processing system that are beyond the control of the vessel Manufacturer, the user is responsible to assure that all required pressure relief devices are properly installed prior to initial operation. It is not a requirement that the vessel Manufacturer supply the needed pressure relief devices. Pressure relief devices must provide protection for each vessel as defined by the following: (1) When a single relief device is used, the pressure in the vessel cannot exceed the MAWP by the larger of 10% or 3 psi. (2) When multiple relief devices are used [see paragraph UG134(a)], the pressure in the vessel cannot exceed the MAWP by the larger of 16% or 4 psi.

These limitations refer to the allowed pressure accumulation in the vessel after the relief device actuates. The set pressure (or the actuation pressure) of the relief device cannot be greater than the MAWP of the vessel that it is protecting. When multiple devices are used to provide the required relief capacity, at least one of the devices must have a set pressure not greater than the MAWP. The other relief devices may have a larger set pressure; however, none of the multiple relief devices shall have a set pressure greater than 105% of the MAWP. If there is a possibility that the vessel can be exposed to fire or other unexpected sources of heat from abnormal events such as an accidental release and ignition of a fluid, supplemental fire protection relief devices shall be provided. The allowed accumulated pressure for fire relief considerations is 21% above the MAWP of the vessel. It is not mandatory that every vessel has to be provided with a fire protection relief device, and the code does not provide information to determine if a fire case is credible. The need for fire relief devices is defined by the user and requires consideration of the service, location of the vessel, and other factors that may affect the likelihood of the vessel being exposed to fire. A specialist with experience in fire protection and loss protection should be consulted regarding the need for overpressure protection from potential fires. In accordance with paragraph UG-125(g), the pressure relief device shall be installed directly on the vessel. However, if the source of pressure is external to the vessel and is under positive control such that it prevents the pressure from exceeding the MAWP, then the relief device may be installed remote from the vessel. The note in paragraph 125(g) cautions that mechanical and/or electrical devices are not considered to be acceptable for providing positive pressure control when the relief device is remotely located. The means that control valves, check valves, and other instrumented systems cannot be considered effective at controlling the pressure to a vessel in lieu of a pressure relief device. However, if the system is such that it is not possible to expose a vessel to a pressure that is greater than the MAWP, then the relief device may be remotely located in the piping system or on other vessels in the same system. As an example, if a vessel is connected to a pump and the maximum pressure that can be generated by the pump is less than the MAWP of the vessel, the relief device does not have to be located on the vessel. It may be located elsewhere in the system, typically on the pipe at the pump discharge. Note that the pressure relief device is still required for this scenario, only its location is affected by the system properties. This is supported by the following interpretation (see Appendix M for further guidance): Interpretation: VIII-1-83-29 Subject: Section VIII, Division 1, UG-125, Relief Devices Date Issued: October 4, 1982 File: BC81-268 Question (1): UG-125(h) states, "The protective devices required in (a) need not be installed directly on a pressure vessel when the source of pressure is external to the vessel and is under such positive control that the pressure in the vessel cannot exceed the maximum allowable working pressure at operating temperature, except as permitted in (c) (See UC98)." A note states that control instruments, except for pilot operated valves, cannot be considered for such positive control. What constitutes "positive control" as used in UG125(h)?

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 53

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 53

Reply (1): Without excluding other possible methods of positive control, it is often taken to mean the maximum pressure that could be developed in a vessel based upon engineering calculations applicable to the system. One example might be the maximum shutoff head plus maximum suction pressure of a pump that pressurized the vessel. Another example might be the pressure drop at maximum flow conditions when a vessel discharges through a system to a known pressure, such as the atmosphere. Question (2): If a vessel is pressurized by a centrifugal pump and the maximum allowable working pressure of the vessel exceeds the shutoff head of the pump plus its maximum suction pressure, must the vessel still be protected by a relief device(s)? Reply (2): Yes, see UG-125(a). However, this source of pressure need not be considered in determining the required capacity of the relief device(s). Question (3): If the reply to Question (2) is yes, then under what formula or guidelines does one determine the required capacity of the relief device? Reply (3): Except for unfired steam boilers as covered by UG125(b) and for the references contained in M-1I concerning fire conditions, the code does not provide requirements or guidance on the determination of required relief capacities. That is the responsibility of the user or his designated agent. Specific requirements are given in paragraphs UG-126 and UG-127 for pressure relief valves (PRV) and nonreclosing pressure relief devices, respectively. Nonreclosing pressure relief devices include rupture disk devices, breaking pin or buckling pin devices, and spring-loaded nonreclosing devices. The interested reader should refer to these paragraphs for additional information. The main components that make up a pressure relief device must satisfy all applicable code requirements, such as materials, toughness consideration, including a final hydrostatic pressure test. The rules for the hydrostatic pressure test given in UG136(d)(2) and UG-137(d)(2) were completely revised in the 2007 Edition of VIII-1. The exemption from pressure testing based on size and pressure were removed, such that all sizes of pressure relief valves for any pressure rating must now be pressure tested. A Manufacturer that has received the appropriate Certificate of Authorization shall supply all pressure relief devices, and they must be marked with the appropriate code symbol (UV or UD).

specific services. Paragraph UW-2(a) provides requirements for welded joints for vessels that the user has defined to be in lethal service. Lethal service is defined by footnote 1 to paragraph UW-2 as under: By "lethal substances" are meant poisonous gases or liquids of such a nature that a very small amount of the gas or liquid mixed or unmixed with air is dangerous to life when inhaled. For purposes of this Division, this class includes substances of this nature which are stored under pressure or may generate a pressure if stored in a closed vessel. It is the responsibility of the user to determine if a fluid se service should be categorized as lethal, and the user shall inform the Manufacturer for vessels that are intended to be in lethal service. The Manufacturer is responsible to comply with the construction provisions applicable for lethal service. Vessels in lethal service that are fabricated by welding must have all butt welds fully radiographed with the following exceptions: (1) Electric resistance welded pipe or tube is not permitted to be used as a shell or nozzle in lethal service applications, whether or not radiography is performed on the weld seam. (2) Electric resistance welded tubes used in a heat exchanger may be used without radiography, provided that they are totally enclosed in a vessel that meets the requirement of full radiography of all butt welds. For example, ERW tubes may be used in a heat exchanger in lethal service if both the shell side and tube side meet all the lethal service requirements. (3) If one side of a heat exchanger contains a lethal substance, the other side does not need to be built to the lethal service rules if the heat exchanger tubes are seamless; or ERW tubes are used that meet all the following tests and examinations: (a) hydrostatic test in accordance with the material specification; (b) pneumatic test under water in accordance with SA-688; (c) ultrasonic test or eddy current test. For vessels in lethal service, all Category A welds (see Fig. 21.8 for joint category definition) shall be butt welds obtained by double welding or other means that will result in the same quality (Type 1 butt weld). Welds using backing strips that remain in place shall not be used. All Categories B and C joints shall be full penetration butt welds; however, single-welded joints and joints using backing strips are allowed (Type 1 or Type 2 butt weld). Special provisions are made for fabricated lap joint stub ends. Vessels in lethal service are often constructed of high alloy materials and lap joint type flanges are used at the bolted connections. The weld at the lap joint stub end is a Category C weld; however, it is sometimes not practicable to make this a butt weld. Accordingly, Fig. UW-13.5 provides for the construction of a "special" corner joint that is acceptable for lap joint stub ends for vessels in lethal service. This method consists of depositing a weld buildup, conducting full radiography of the weld, and then ultrasonic examination of the fusion line between the weld buildup and nozzle neck. Then, the lap ring is attached with a butt weld that is fully radiographed. Alternatively, the lap ring may be machined from the plate. If the hub of the lap ring is in the through thickness direction of the plate, then the provisions of Appendix 20 must be satisfied. All welds that attach nozzles to a vessel in lethal service shall be full penetration welds through the vessel wall or the nozzle wall.

21.5

SUBSECTION B ­ REQUIREMENTS PERTAINING TO METHODS OF FABRICATION OF PRESSURE VESSELS

Subsection B contains rules that apply to specific methods used for the fabrication of pressure vessels. The specific methods covered by Subsection B include welded construction, forging, and brazing. All rules of Subsection B must be used in conjunction with the requirements of Subsections A and C.

21.5.1

Part UW : Requirements for Pressure Vessels Fabricated by Welding

21.5.1.1 General Part UW applies to pressure vessels that are fabricated by welding. Paragraph UW-2 defines requirements for

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 54

54 · Chapter 21

FIG. 21.8

ILLUSTRATION OF WELDED JOINT LOCATIONS TYPICAL OF CATEGORIES A, B, C, AND D (Source: Fig. UW-3 of Section VIII Div. 1 of the ASME Code)

Vessels that are fabricated from UCS material with a minimum design metal temperature colder than­55°F and vessels made from UHA material where impact testing is required shall have welded joints that comply with paragraph UW-2(b). The basic requirements for vessels in low temperature service are that Category A welds shall be double-welded butt welds (Type 1) and Category B welds shall be double- or single-welded butt joints (Type 1 or Type 2). Categories C and D joints shall be full penetration welds extending through the full thickness of the joint. Note that some exceptions to these rules are offered for UHA material and the specific provisions of UW-2(b) should be reviewed when impact-tested stainless steel material is used. UW-2(c) requires that unfired steam boilers, having a design pressure greater than 50 psi, have double butt welds (Type 1) for Category A joints and either double butt or single butt welds (Type 1 or Type 2) for all Category B joints. All butt welds require full radiography, except for butt welds in nozzles or other communicating chambers that exceed neither NPS 10 nor 11/8 in. Note that there are no specific requirements for Categories C and D joints for vessels in this service. UW-2(d) addresses the welded joint requirements for vessels that are subject to direct firing. Direct firing means that the vessel wall is directly exposed to the radiant heat resulting combustion of a fuel. Such vessels shall have double butt welds for all Category A joints. When the joint thickness exceeds 5/8 in., Category B joints shall be either double or single butt welds (Type 1 or Type 2). It should be noted that paragraph UW-2(d)(3) provides requirements for determination of the design temperature of vessels that are directly fired. Calculations that determine the design temperature of the vessel shall be made available to the Inspector. The mean metal temperature may be used to establish the vessel design temperature if all the joints are Type 1 or Type 2 butt welds. If the vessel has welded joints other than Type 1 or Type 2, then the vessel design temperature shall not be less than the maximum metal surface temperature (not the maximum mean metal temperature). Paragraph UW-3 defines weld categories. Weld categories refer to the location of a welded joint in a vessel. The types of welds are defined in Table UW-12 and elsewhere. Weld joint categories are used to conveniently designate special requirements (joint type) and degree of examination that are used elsewhere in Division 1. Weld joint categories do not apply to every welded joint, and special requirements for any welded joint apply only

when specifically stated. Division 1 designates four weld joint categories (see Fig. 21.8). (1) Category A. Longitudinal joints in the main shell, cones, nozzles, tubes, and pipes; any weld connecting segments of spherical vessels, sections of formed or flat heads, and within the side plates of a flat-sided vessel (away from the corner joint); and the circumferential weld attaching a hemispherical head to a shell. (2) Category B. Circumferential welded joints in shells other than the circumferential joint attaching a hemispherical heads and circumferential welded joints in nozzles. (3) Category C. Joints attaching flanges, laps, tubesheets, and flat heads to the main shell and joints attaching the side plates to one another in flat-sided vessels. (4) Category D. Joints attaching nozzles or communicating chambers (such as sumps or boots) to the main shell. It is noted that the nozzle at the small end of a cone is a Category B joint and not Category D. Another important point is made in paragraph UW-3(b). When Category B joints are required to be butt joints (see UW-2), the angle between the connected parts cannot exceed 30° . For example, if a cone with a 40° half-apex angle is to be attached to a cylinder, and a butt weld is required by a service condition of paragraph UW-2, the cone must have transition knuckles at the ends in order to comply with this requirement. 21.5.1.2 Design of Welded Joints Paragraph UW-9 provides general requirements for the design of welded joints. The permissible types of welded joints are listed in Table 21.1. (Source: Table UW-12 of Section VIII, Division 1 2007 Edition of the ASME Code) The welding grooves of butt joints must be such that complete fusion and penetration is achieved during welding. The Welding Procedure Qualification is an acceptable confirmation that the groove is satisfactory. If there is an offset between the surfaces of abutting welded sections in a vessel, a tapered transition shall be provided if the difference in thickness of the parts joined is more than 1/8 in. or one-fourth the thickness of the thinner part, whichever is less. The tapered transition must have length of three times the offset between the adjacent surfaces. The transition may be made by removing material of the thicker section or by adding weld material to the thinner section. If material is removed from the thicker section, the minimum thickness in the reduced section must satisfy the minimum thickness requirements of UG-23. When weld

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 55

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 55

TABLE 21.1

MAXIMUM ALLOWABLE JOINT EFFICIENCIES FOR ARC- AND GAS-WELDED JOINTS (Source: Table UW-12 of Section VIII, Div. 1 2007 Edition of the ASME Code)

Degree of Radiographic Examination Type No. (1) Joint Description Butt joints as attained by double-welding or by other means which will obtain the same quality of deposited weld metal on the inside and outside weld surfaces to agree with the requirements of UW-35. Welds using metal backing strips which remain in place are excluded. Single-welded butt joint with backing strip other than those included under (1) Single-welded butt joint without use of backing strip None Limitations Joint Category A, B, C & D (a) Full [Note (1)] 1.00 (b) Spot [Note (2)] 0.85 (c) None 0.70

(2)

(a) None except as in (b) below (b) Circumferential butt joints with one plate offset; see UW-13(b)(4) and Fig. UW-13.1, sketch (i) Circumferential butt joints only, not over 5/8 in. (16 mm) thick and not over 24 in. (600 mm) outside diameter (a) Longitudinal joints not over 3 in. 5 (10 mm) thick (b) Circumferential joints not over 5/8 in. (16 mm) thick (a) Circumferential joints [Note (4)] for attachment of heads not over 24 in. (600 mm) outside diameter to shells not over 1/2 in. (13 mm) thick (b) Circumferential joints for the attachment to shells of jackets not over 5/8 in. (16 mm) in nominal thickness where the distance from the center of the plug weld to the edge of the plate is not less than 11/2 times the diameter of the hole for the plug. (a) For the attachment of heads convex to pressure to shells not over 5/8 in. (16 mm) required thickness, only with use of fillet weld on inside of shell; or (b) for attachment of heads having pressure on either side, to shells not over 24 in. (600 mm), inside diameter and not over 1/4 in. (6 mm) required thickness with fillet weld on outside of head flange only

A, B, C & D A, B & C

0.90 0.90

0.80 0.80

0.65 0.65

(3)

A, B & C

NA

NA

0.60

(4)

Double full fillet lap joint

A B&C [Note (3)] 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

C

NA

NA

0.50

(6)

Single full fillet lap joints without plug welds

A&B

NA

NA

0.45

A&B

NA

NA

0.45

(Continued)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 56

56 · Chapter 21

TABLE 21.1

MAXIMUM ALLOWABLE JOINT EFFICIENCIES FOR ARC- AND GAS-WELDED JOINTS (CONTINUED) (Source: Table UW-12 of Section VIII, Div. 1 2007 Edition of the ASME Code)

Degree of Radiographic Examination Type No. (7) Joint Description Corner joints, full penetration, partial penetration, and/or fillet welded Angle joints Limitations As limited by Fig. UW-13.2 and Fig UW-16.1 Design per U-2(g) for Category B and C joints Joint Category C&D [Note (5)] (a) Full [Note (1)] NA (b) Spot [Note (2)] NA (c) None NA

(8)

B, C & D

NA

NA

NA

General Notes: (a) The single factor shown for each combination of joint category and degree of radiographic examination replaces both the stress reduction factor and the joint efficiency factor considerations previously used in this Division. (b) E = 1.0 for butt joints in compression. Notes: (1) See UW-12 (a) and UW-51. (2) See UW-12 (b) and UW-52. (3) For Type No. 4 category C joint, limitation not applicable for bolted flange connections. (4) Joints attaching hemispherical heads to shells are excluded. (5) There is not joint efficiency E in the design formulas of this Division for Category C and D corner joints. When needed, a value of E not greater than 1.00 may be used.

metal buildup is used, the buildup region must comply with UW-42, which requires qualification of the weld procedure, magnetic particle, or liquid penetrant examination, and that portion containing the weld shall be included in any required radiography. When shell courses with longitudinal welds seams are joined, paragraph UW-9(d) requires that the longitudinal weld seams between them be offset by at least five times the thickness of the thinner plate. This requirement may be waived if the longitudinal weld is radiographed for 4 in. each side of the girth weld. Paragraph UW-9(f) provides general guidance regarding welded joints subjected to bending. Fillet welds shall be added, as necessary, to reduce the stress concentration; however, corner joints with fillet welds only shall not be used unless the plates are stayed [also, see paragraph UW-18(b)]. As defined in paragraph UW-9(g), the designer shall consider the loading conditions given in UG-22 for sizing fillet welds and partial penetration welds. However, these welds can be no smaller than the minimum size given in the other parts of Section VIII regardless of the expected load. Table 21.1 lists three types of butt joints: Types 1, 2, and 3. A Type 1 butt joint is attained by double welding or by other means (typically a single-sided joint), which will obtain the same quality of deposited weld metal on the inside and outside weld surfaces to agree with the requirements of UW-35. A Type 2 butt joint is a single-welded butt joint with a backing strip left in place. A Type 3 butt joint is a single-sided butt joint for which visual examination of the backside of the weld joint is not possible. It is assumed that since access to the backside of a Type 3 weld joint is not possible, the volumetric examination is not possible and a joint efficiency equal to 0.60 is assigned to that joint. It is often asked that if it is possible to radiograph a single-sided butt joint, does that automatically qualify it as a Type 1 joint. The answer lies in the definition given for a Type 1 joint, where it requires confirmation that the quality of the deposited weld metal on the inside weld surface must satisfy the requirements of UW-35 (undercut, weld reinforce-

ment, etc.). This is normally achieved by visual inspection. If a Manufacturer can demonstrate that the backside of a single-sided butt weld satisfies the criteria of UW-35, he can classify the joint as a Type 1 joint. Paragraph UW-11 provides requirements for three radiographic and ultrasonic examination categories used for butt welds. The categories of examination are Full, Spot, and No Radiography. Full radiography requires the butt joint to be radiographed to its full length, as described in paragraph UW-51. Spot radiography requires examination of a local portion at a determined frequency and location of the weld, as described in paragraph UW-52. No radiography category applies when no part of the butt weld is radiographed. The provisions of full radiography must be applied for the following: (1) All butt welds in shells and heads of vessels are designated to be in lethal service. Categories B and C butt welds in nozzles and communicating chambers that do not exceed NPS 10 or 11/8-in. wall thickness do not require any radiographic examination [except as may be required by UHT-57(a)]. (2) All butt welds in vessels where the nominal thickness at the joint is greater than 11/2 in. or exceeds the thickness requiring full examination found elsewhere in Division 1 (e.g., see UCS-57, UNF-57, UHA-33, UCL-35, and UCL-36). Categories B and C butt welds in nozzles and communicating chambers that do not exceed NPS 10 or 11/8-in. wall thickness do not require any radiographic examination [except as may be required by UHT-57(a)]. (3) All butt welds in the shells and heads of unfired steam boilers with a design pressure greater than 50 psi. Categories B and C butt welds in nozzles and communicating chambers that do not exceed NPS 10 or 11/8-in. wall thickness do not require any radiographic examination [except as may be required by UHT-57(a)].

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 57

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 57

(4) All Categories A and D butt welds in vessels and heads when the design is based on a joint efficiency given by Table 21.1 Column (a). (E of 1.0 for Type 1 and 0.9 for Type 2 butt welds.) It should be noted that for this case, Categories A and B welds must be Type 1 or Type 2. This may dictate the type of construction required at transitions within the vessel [see UW-3(b)]. Likewise, it should be noted that Category B or Category C butt welds that intersect a Category A weld designed with the full radiography joint efficiencies must, at least, meet the requirements for spot radiography. Category B or Category C butt joints that connect seamless sections designed on the basis of E 1.0 must meet the requirements for spot radiography given in UW-52. These spot radiographs cannot be used to satisfy the weld increment required by UW-52 for any other joint. (5) All butt welds made with electrogas welding when any single pass is greater than 11/2 in. (6) All butt welds made by the electroslag process. It is noted in UW-11(a)(7) that ultrasonic examination, in accordance with UW-53, may be used as a substitute for radiography only in the final closure seam if the construction of the vessel does not permit interpretable radiographs. This is frequently the case for fixed tubesheet heat exchangers where the tube bundle precludes the possibility of obtaining a meaningful radiograph. However, Code Case 2235 provides alternative requirements that allow the use of ultrasonic examination in lieu of radiography for a much broader scope. It is cautioned that code cases are permissive and their use may not be acceptable to all jurisdictional authorities. Type 1 or Type 2 butt welds that are not required to be fully radiographed may be examined by spot radiography in accordance with UW-52. If spot radiography is applied to the entire vessel, Categories B and C butt welds in nozzles and communicating chambers that do not exceed NPS 10 or 11/8-in. wall thickness do not require any radiographic examination. Note in paragraph UW-11(b) provides clarification that radiography requirements only apply to Types 1 and 2 butt welds but does not preclude the use of other types, such as fillet or corner welds, for attaching nozzles, manways, flat covers, tubesheets, and so on. When these other types of nonbutt welds are not precluded by the service restrictions of paragraph UW-2, they may be used and no radiography is required. Butt welds in vessels do not require any radiography if the vessel is designed for external pressure only and is not in lethal service, or if the design is based on the joint efficiencies given in Table 21.1, Column (c) (E 0.7 for Type 1 butt joints and E 0.65 for Type 2 butt joints). The joint efficiencies given in Table 21.1 are to be used for the design of arc- or gas-welded pressure vessels. Except for Categories B and C butt joints that intersect a fully radiographed Category A joint, or that attach seamless sections, the joint efficiency of a weld depends on the type and the degree of examination for that joint. This is an important change that was introduced into the code in the late 1980s. Prior to that time, the design of the entire vessel was based on the degree of examination of the welded joints that resulted in the smallest joint efficiency. The present rules allow the design of a joint to use the joint efficiency applicable to that joint. It is noted in Table 21.1 that the user shall establish the type of joint and extent of required examination.

A review of Table 21.1 shows that joint efficiencies are provided for butt joints, double full-fillet lap joints, single full-fillet lap joints with plug welds, and single full-fillet lap joints without plug welds, corner joints, and angle joints. The table defines the joint category, joint efficiency, and limitations of which each joint may be used. The use of the lap joint welds is severely limited to joint type and thickness of the joined parts. Since these are not buttwelded joints, there are no provisions for radiography and the joint efficiency of Column (c) for no radiography applies. Likewise, these joint types (Types 3, 4, 5, and 6) may not be used when a butt joint is a service requirement of UW-2. Joint Types 6 (corner joints) and 7 (angle joints) were introduced in the 2002 Addenda. Radiographic examination is not applied to these joint types, and when designing a component utilizing one of these joint types, E not greater than 1.00 may be used in design formulas that call for a joint efficiency. For butt joints that satisfy the full radiography requirements, the joint efficiency given by Column (a) "Full" may be used in the design equations. However, it is also a requirement of "full radiography" that Category B or Category C butt welds that intersect a Category A weld must at least meet the requirements for spot radiography. If this requirement is not satisfied, the joint efficiency of Column (b) "Spot" applies. Column (b) applies to buttwelded joints that satisfy the spot radiography requirements of UW-52. Column (c) applies to all welded joints that are neither fully radiographed nor spot radiographed. Special consideration is required for the joint efficiency of seamless components [see UW-12(d)]. The joint efficiency E used in the design of seamless components may be equal to 1.0, provided that the attaching weld is spot radiographed as per UW11(a)(5)(b). If the attachment weld is not spot radiographed or is a Type 3, Type 4, Type 5, or Type 6, then the seamless component must be designed with E 0.85. Welded pipe or tubing (not in addition of filler metal) shall be treated as seamless when applying the rules of UW-12 for determination of joint efficiency. The allowable tensile stress shall be taken from the welded product values (a 0.85 joint efficiency is automatically built in to the allowable stress values for welded pipe and tube), and the requirements of UW-12(d) applied. 21.5.1.3 Attachment Details: Heads and Flat Plates Paragraph UW-13 provides rules for weld details, welds attaching heads and intermediate heads to shells, and flat plates to pressure parts that form corner joints. Figure 21.9 provides the requirements for attaching formed heads. The use of the lap joint details (Fig. 21.9a­d) and the "butt weld with one edge offset" (Fig. 21.9i) are limited, as defined in Table 21.1 and elsewhere. When heads are attached to a shell and the thickness difference between them is the lesser of 1/8 in. or one-fourth the thickness of the thinner part, a 3:1 transition is required. Likewise, the required minimum thickness of the cylindrical shell must be maintained according to the tangent line where the head meets the shell. This means that the transition in thickness cannot result in the shell as being thinner than its minimum required thickness on the shell side of the tangent line. There are instances where the centerlines of the head and the shell are different. For example, this may occur when a smooth surface is required on the inside of the vessel and the shell and head are of different thickness. This is allowed, as shown in Fig. 21.9 j­m, as long as the centerline difference does not exceed one-half the difference in their thickness and a 3:1 taper is

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 58

58 · Chapter 21

provided. This offset allowed by design does not affect the manufacturing tolerances allowed by paragraph UW-33. Figure 21.9 f-1, f-2, g, and h shows details that are not acceptable to attach an unstayed head to the shell. Use of these details would result in significant bending across the weld. It is noted

that such details may be used if the head is stayed or otherwise supported [see UW-9(f) and UW-18(b)]. Paragraph UW-13(d) provides rules for the attachment of unstayed flat head plates that form corner joints. The required weld size for corner joints is given by Fig. 21.10. The weld size is

FIG. 21.9 HEADS ATTACHED TO SHELLS (Source: Fig. UW-13.1 of Section VIII, Div. 1 2007 Edition of the ASME Code)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 59

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 59

FIG. 21.9 (CONTINUED) (Source: Fig. UW-13.2 of Section VIII, Div. 1 2007 Edition of the ASME Code)

a function of the terms "a" and "b" as shown in the figure. These dimensions are established by projecting the line of fusion of the weld parallel and perpendicular to the part being welded to establish "a", the weld width, and "b", the weld depth into the plate. For attaching a flange ring of a bolted connection, the sum of "a" and "b" shall not be less than three times the nominal thickness of the abutting part. For supported tubesheets that are extended as

a flange and all the details given in Fig. 21.10 a­l, the sum of "a" and "b" shall not be less than two times the thickness of the abutting part. In previous editions of the code, unsupported tubesheets with a flange extension were required to have a weld size of "a" plus "b" equal to or greater than three times the nominal thickness of the abutting part. However, an analysis done by the committee showed that the stresses at the tubesheet-to-shell joint were not

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 60

60 · Chapter 21

FIG. 21.10 ATTACHMENT OF PRESSURE PARTS TO FLAT PLATES TO FORM A CORNER JOINT (Source: Fig. UW-13.2 of Section VIII Div. 1 of the 2007 Edition of the ASME Code)

significantly affected whether the tubesheet was supported or unsupported. Accordingly, the Standards Committee approved action to delete the increased weld size for unsupported tubesheet-to-shell weld with a flange. This has a significant effect

by reducing the required shell-to-tubesheet weld size for many heat exchangers. When the tubesheet or flat head is to be attached using a butt weld (as may be required by the UW-2 service restrictions), the

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 61

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 61

FIG. 21.11 TYPICAL PRESSURE PARTS WITH BUTT-WELDED HUBS (Source: Fig. UW-13.3 of Section VIII Div. 1 of the 2007 Edition of the ASME Code)

1/ in. (6 mm) min. radius 4

1/ in. min. (6 mm) 4

30 deg max.

18.5 deg max.; 14 deg min.

radius 30 deg max. 18.5 deg max.; 14 deg min.

tn [Note (1)] t rn

tn [See Note (1)]

t rn

See Note (2)

See Note (2)

30 deg max.

t1 [See Note (3)]

18.5 deg max.; 14 deg min.

1/ in. (6 mm) min. radius 4

t1 [See Note (3)]

(a) (b) NOTES: (1) As defined in UG-40. (2) Weld bevel is shown for illustration only. (3) t1 is not less than the greater of: (a) 0.8tr n where tr n = required thickness of seamless nozzle wall (b) Minimum wall thickness of connecting pipe

FIG. 21.12

NOZZLE NECKS ATTACHED TO PIPING OF LESSER WALL THICKNESS (Source: Fig. UW-13.4 of Section VIII Div. 1 of the ASME Code)

hub dimensions must satisfy the requirements of Fig. 21.11When the tubesheet or plate is made from a forging, the mechanical test specimens must be taken at an orientation to demonstrate that the hub has sufficient strength and ductility in the direction normal to the plate surface. Hubbed tubesheets and flat heads may be machined from plate material if the provisions of Appendix 20 are satisfied. 21.5.1.4 Attachment Details: Connections and Openings The design requirements of ASME B31.3 may result in a smaller

thickness for piping to be attached to a pressure vessel nozzle. When vessel nozzles are to be welded to thinner piping, the nozzle neck may be tapered to meet the thickness of the attached pipe even though this thickness may be less than that required by Section VIII. The tapered transition must, however, meet the requirements of Fig. 21.12. The requirements of paragraph UW-14 allow any type of opening that meets the reinforcement rules of UG-37 or UG-39 to be located in any welded seam without any additional examinations of the weld. However, if the opening was exempt from reinforcement

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 62

62 · Chapter 21

calculations by UG-36(c)(3), additional requirements apply, including radiography of the seam on each side of the opening. In such instances, the designer has the option of showing that such openings meet the reinforcement rules of UG-37 or providing the additional radiography. Paragraph UW-15 provides general requirements for welded connections. It is required that nozzles, other connections, and their reinforcing elements be welded with sufficient strength to develop the strength of the reinforcing parts through shear or tension in the weld. This requirement is consistent with the reinforcement rules given in UG-41. When strength calculations are required for nozzle attachment welds, the allowable stress values for groove and fillet welds (as a percentage of the allowable stress for the vessel material) are given below: (1) groove weld in tension (2) groove weld in shear (3) fillet weld in shear 74% 60% 49%

an ASME/ANSI Reference Standard or to a Manufacturers Standard, as defined in UG-11(a). This then precludes the use of shop-fabricated fittings unless they are produced according to a Manufacturers Standard, complete with pressure/temperature ratings defined. 21.5.1.5 Attachment Details: Fillet Welds Paragraph UW-18 provides general guidance for the use of fillet welds. Fillet welds may be used as strength welds within the provisions and limitations as given elsewhere in Division 1. As stated in UW-18(d), the weld joint efficiency to be used for fillet welds is E 0.55 unless otherwise defined elsewhere. For example, the joint efficiencies of paragraph UW-18(d) do not supersede the weld efficiencies given in Table 21.1. Paragraph UW-20 provides mandatory requirements for tubeto-tubesheet welds. It is intended that the provisions of UW-20 be used for heat exchangers where the tubes act as stays for the tubesheet, such as the fixed tube type heat exchanger. The provisions of this paragraph apply for tube-to-tubesheet joints that rely only on the strength of the weld to resist the tube loads, that is, the strength of the joint does not consider tube expansion or rolling. (See Appendix A for tube-to-tubesheet joints where the strength of the joint considers the combination of welding and tube expansion.) Paragraph UW-20 considers three types of tubeto-tubesheet welds. Full strength welds are sized such that the strength of the weld is equal to or greater than the tube strength in the axial direction. For full strength welds, it is expected that the tube would fail before the joint; thus, it would develop the full strength of the tube. There are situations where it may not be required that the joint must develop the full tube strength. Such situations may include exchangers made of very expensive material where the size of weld needs to be minimized. Partial strength welds are intended to apply for these situations. In such cases, it is acceptable to size the tube-to-tubesheet weld to resist the loads that are expected during normal operation; however, such welds may not develop the full strength of the tube. When the partial strength weld option is used, the weld size may be determined in accordance with UW-18 or Appendix A. It is noted that UW-20 does not require mock-up testing for tube-to-tubesheet joints. If the user wants mock-up testing to be done on full strength or partial strength welds (designed using UW-18), then that must be defined in the contractual requirements. Seal welds are not considered to contribute any strength to the tube-to-tubesheet joint; thus, they are not considered to provide any contribution to the staying of the tubesheet. The design of the seal-welded tube-totubesheet joint considers that all tube axial loads are resisted by the expansion of the tube into the tubesheet (see Appendix A for determining the maximum allowable load for such joints). It is the responsibility of the user to define the type of tube-to-tubesheet joint that is required for a heat exchanger. Paragraph UW-21 provides attachment details for ASME B16.5 socket-welded and slip-on flanges. Slip- on flange attachment details based on B31.3 rules will be introduced in a future revision to VIII-1. 21.5.1.6 Welding Fabrication Requirements The Manufacturer is responsible for the quality of welding. No production welding may be done on a vessel unless the welding procedures and welders have been qualified as required by Section IX of the ASME Code. Code welding is only permitted at the shop location listed on the Manufacturer's Certificate of Authorization and at the field sites.

Examples L-7.2­L7.6 of Appendix L illustrate the proper use of these "efficiencies" for nozzle strength calculations. It should be noted that integrally reinforced nozzles that are attached to the shell with full penetration welds are exempt from strength path calculations. Reinforcement plates and saddles of nozzles attached to the outside of a pressure vessel must have at least one "telltale" hole no larger than NPS 1/4 tap [See UG-37(g)]. The reinforcement plate or saddle should be pressurized with compressed air while the welds that seal off the connection to the inside of the vessel are coated with a bubble-forming solution. Compressed air at a pressure of 25 psi is typically considered adequate for determining the leak tightness of these welds. It is not intended that these telltale holes be fitted with a plug that is capable of retaining pressure during operation. Plugging of the telltale holes can have very serious consequences if there is a leak from the vessel into the space between the reinforcement pad and the vessel. The strength of the welds and reinforcing element does not consider the forces that would have to be resisted if pressure is acting between the reinforcing element and the vessel. If it is desired to plug the telltale holes, then a suitable material that will not sustain pressure must be used. Paragraph UW-16 provides the minimum requirements for attachment welds at openings. Nozzles, connections, necks, tubes, fittings, pads, and so on are terms that describe a construction that forms a Category D weld joint. Figure 21.13 provides a partial listing of some acceptable types of welded nozzles and other connections to shells. The weld dimensions given in Fig. 21.13 represent the minimum size weld allowed for those types of nozzle connections described by the figure. However, if necessary, the welds must be increased in order to satisfy the strength calculation requirements defined in paragraph UW-15 and described above. Likewise, the weld sizes shall be increased by an appropriate amount if a corrosion allowance is specified. Under special circumstances, small nozzle necks and fittings may be attached with a single fillet weld. As per UW-16(e), necks and tubes up to NPS 6 may be attached with a single fillet weld, as shown in Fig. 21.13 w-1 and w-2. Note that although Fig. 21.13 w-1 and w-2 shows the welds on the inside, they may also be placed on the outside. A significant revision concerning the attachment of small fittings as per UW-16(f)(3)(a) occurred in the 2007 Edition. This revision clarified that the attachment of fittings with a single fillet weld is only permitted for standard fittings produced according to

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 63

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 63

FIG. 21.13

SOME ACCEPTABLE TYPES OF WELDED NOZZLES AND OTHER CONNECTIONS TO SHELLS, HEADS, ETC. (Source: Fig. UW-16.1 of Section VIII Div. 1 of the ASME 2007 Code Edition)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 64

64 · Chapter 21

FIG. 21.13

(CONTINUED)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 65

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 65

TABLE 21.2 ALLOWABLE OFFSET AT BUTT WELDS (Source: Table UW-33 of Section VIII Div. 1)

Joint Categories Section Thickness, in. Up to 1/2, incl. Over to 1/2 to 1/4 incl. Over to 1/4 to 11/2 incl. Over to 1/2 to 2 incl. Over 2

1 1

A /4 t /8 in. 1 /8 in. 1 /8 in. Lesser of 1 /16 t or 3/8 in.

1 3

B, C, & D /4 t /4 t 3 /16 in. 1 /8 t Lesser of 1 /8 t or 3/4 in.

Paragraph UW-27 provides a listing of the welding processes and restrictions, which may be used in the construction of a pressure vessel. These include arc welding, electron beam, electroslag, inertia or friction welding, laser beam, oxyfuel, resistance, and explosive welding.Paragraph UW-33 provides the allowable alignment tolerance applicable for butt welds. Sections to be butt welded shall be fitted, positioned, and held before and during welding such that the alignment tolerances given in Table 21.2 are not violated. Any offset within the allowable tolerance must be contoured to a 3:1 taper. These tolerances assure that any bending at a butt weld resulting from an offset is not significant. The allowed tolerance for Category A butt joints is generally more restrictive than required for other weld categories because the membrane stress due to pressure across these joints is greater. Also, note that these alignment tolerances are in addition to the centerline offset as given in UW-9(c). Paragraph UW-35 provides requirements for the finished butt welds in longitudinal and circumferential joints in the pressure shell. All butt-welded joints shall be full penetration joints with complete fusion. The surface of the weld must be such that proper interpretation of the radiograph and any other required examinations can be made. Weld undercut is allowed, provided that it does not infringe on the required minimum wall thickness of the part and the depth of undercut is not more than 1/32 in. or 10% of the part, whichever is less. In order to assure that the butt weld groove is completely filled, weld reinforcement may be added. The amount of weld reinforcement at each face of the weld is limited, as defined in UW-35(d). Excessive weld reinforcement can have a deleterious effect for vessels that experience cyclic loading. The designer should consider stress concentration effects of the weld surface and reinforcement at butt welds for vessels that will experience cyclic loading. It is cautioned that the stress concentration effects are greatly affected by the reentrant angle of the weld reinforcement with the welded part. It may be that vessels in severe cyclic service will require grinding of the welds to reduce the stress concentration effects of the weld reinforcement. Paragraph UW-37(f) requires each welder and welding operator to apply an identifying mark on or near each weld that he makes. Alternatively, the Manufacturer may keep a record of the welder or weld operator employed on each welded joint. This record shall be made available to the Inspector. 21.5.1.7 Postweld Heat Treatment Requirements Paragraph UW-40 provides criteria and information related to procedures to be used for PWHT for welded components. The requirements for establishing when PWHT is required are covered in Subsection C for the specific materials used. When PWHT is required, it shall be done in accordance with the provisions of UW-40. PWHT may be

accomplished in a number of ways. The entire vessel may be put into a furnace and heated to the required temperature and hold time. If a vessel is too large to fit into a furnace, it is allowed to heat treat only a portion of the vessel at a time in the furnace, provided there is an overlap of the heated sections of at least five ft. The vessel may be heated from the inside by firing gas burners or any other appropriate heating method. Vessel sections, before final assembly, can be PWHT in a furnace, and the subsequent circumferential joints subjected to a local heat treatment. When local heat treatment is conducted of any circumferential joint, the heated band must extend around the full circumference of the part. It should be noted that UW-40 (a)(3) allows local PWHT of the longitudinal joints of cylinders prior to joining to the completed vessel. This allows local PWHT for long seams in individual shell cans prior to assembly with other shells. Also, paragraph UW40(a)(7) allows local PWHT of welds in formed heads or spherical sections by heating only the area around the weld joint. PWHT shall be conducted prior to the required pressure test of UG-99 or UG-100. Generally, the code does not define the relative timing of PWHT and the required nondestructive examinations. Paragraph UW-40(a) should be consulted for determining the specific requirements related to PWHT. UW-40(f) provides the definition of nominal thickness to be used in application of the PWHT requirements given in Tables UCS-56, UCS-56.1, UHA-32, and UHT-56. In the context of the requirements for postweld heat treatment, the nominal thickness is the thickness of the welded joint and not necessarily the thickness of the parts being joined. For example, the nominal thickness of a groove weld is the depth of the groove, and the nominal thickness of a fillet weld is the throat dimension of the fillet regardless of the thickness of the components being joined. Also, the nominal thickness of a weld repair is the depth of the repair weld. This means that a 4-in.thick vessel made of P No. 1 material may have a weld repair of 11/2-in.deep (provided preheat is applied during welding) without a subsequent PWHT (refer to Table 21.3). To correctly apply the need for PWHT and the correct hold time, it is essential to understand the definitions in this paragraph for nominal thickness. Refer to Interpretation VIII-1-83-100 that is provided in the discussion for UCS-56 for more examples. 21.5.1.8 Surface Weld Metal Buildup Paragraph UW-42 allows the buildup of base material and welds by the addition of weld metal. This may be required to restore the base material to the minimum required thickness if local areas have been damaged and to provide the required transition between joints between parts that are of differing thickness and/or misaligned. When such weld buildup is used, the weld procedure must be properly qualified and the entire surface of the weld metal buildup area must be examined by the liquid penetrant or magnetic particle methods. Any weld buildup over a weld that requires radiography must be done before the radiography is conducted. 21.5.1.9 Inspection and Tests Paragraphs UW-47, UW-48, and UW-49 define specific responsibilities of the Authorized Inspector related to welded pressure vessels. The Inspector has the responsibility to assure himself that the welding procedures used to construct a pressure vessel have been properly qualified in accordance with Section IX of the code. The Manufacturer shall submit documentation to the Inspector to this effect. Likewise, the Manufacturer shall certify that the welders and weld operators are properly qualified in accordance with Section IX and shall make certified records available to the Inspector. The Inspector may, at

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 66

66 · Chapter 21

TABLE 21.3

POSTWELD HEAT TREATMENT REQUIREMENTS FOR CARBON AND LOW-ALLOY STEELS (Source: Table UCS-56 of Section VIII, Div. 1 2007 Edition of the ASME Code)

Material P-No. 1 Gr. Nos. 1, 2, 3 Gr. No. 4

Normal Holding Temperature, °F (°C), Minimum 1,100 (595)

Minimum Holding Time at Normal Temperature for Nominal Thickness [See UW-40(f)] Up to 2 in (50 mm) 1 hr/in. (25 mm), 15 min minimum None Up to 2 in. to 5 in. (50 mm to 125 mm) 2 hr plus 15 min for each additional inch (25 mm) over 2 in. (50 mm) None Over 5 in. (125 mm) 2 hr plus 15 min for each additional inch (25 mm) over 2 in. (50 mm) None

NA

Notes: (1) When it is impractical to postweld heat treat at the temperature specified in this Table, it is permissible to carry out the postweld heat treatment at lower temperatures for longer periods of time in accordance with Table UCS-56.1. (2) Postweld heat treatment is mandatory under the following conditions: (a) for welded joints over 11/2 in. (38 mm) nominal thickness; (b) for welded joints over 11/4 in. (32 mm) nominal thickness through 11/2 in. (38 mm) nominal thickness unless preheat is applied at a minimum temperature of 200°F (95°C) during welding; (c) for welded joints of all thickness if required by UW-2, except postweld heat treatment is not mandatory under the conditions specified below: (1) for groove weld not over 1/2 in. (13 mm) size and fillet welds with a throat not over 1/2 in. (13 mm) that attach nozzle connections that have a finished inside diameter not greater than 2 in. (50 mm), provided the connections do not form ligaments that require an increase in shell or head thickness, and preheat to a minimum temperature of 200°F (95°C) is applied; (2) for groove welds not over 1/2 in (13 mm) in size or fillet welds with a throat thickness of 1/2 in. (13 mm) or less that attach tubes to a tubesheet when the tube diameter does not exceed 2 in. (50 mm). A preheat of 200°F (95°C) minimum must be applied when the carbon content of the tubesheet exceeds 0.22%. (3) for groove welds not over 1/2 in. (13 mm) in size or fillet welds with a throat thickness of 1/2 in. (13 mm) or less used for attaching nonpressure parts to pressure parts provided preheat to a minimum temperature of 200°F (95°C) is applied when the thickness of the pressure part exceeds 11/4 in. (32 mm); (4) for studs welded to pressure parts provided preheat to a minimum temperature of 200°F (95°C) is applied when the thickness of the pressure part exceeds 11/4 in. (32 mm); (5) for corrosion resistant weld metal overlay cladding or for welds attaching corrosion resistant applied lining (see UCL-34) provided preheat to a minimum temperature of 200°F (95°C) is maintained during application of the first layer when the thickness of the pressure part exceeds 11/4 in. (32 mm). NA not applicable

any time, require retest of a welding procedure or welder. The Inspector shall also assure himself that all PWHT has been correctly done and temperature and hold time readings conform to the requirements. UW-51 defines requirements for radiographic and radioscopic examination of welded joints. Radiographic examination is conducted in accordance with Article 2 of Section V of the ASME Boiler & Pressure Vessel Code, except as specified in this paragraph. Paragraph UW-51 is applicable for butt welds that require full radiography. Radiography provides a permanent record of the weld examination; however, the Manufacturer is required to maintain the radiographs and records only until the Manufacturer's Data Report has been signed by the Inspector. If the user wishes to have access to the radiographs for future reference, this must be by contractual agreement between the Manufacturer and the user. Personnel who perform or interpret radiographic examinations must be qualified and certified in accordance with their employers written practice, using SNT-TC-1A [33] as a guideline. Alternatively, the ASNT Central Certification Program (ACCP) or CP-189 may be used to fulfill the examination and demonstration requirements of SNT-TC-1A and the employers written practice. Imperfections that are shown on the radiograph are unacceptable and require repair when the following characterizations are made to the imperfection: (1) crack or zone of incomplete fusion or penetration; (2) any other elongated indication which has a length greater than

­ 1/4 in. (6 mm) for weld thicknesses up to 3/4 in. (19 mm); ­ 1/3 t for t ranging from 3 in. (19 mm) to 21/4 in. (57 mm); 4 ­ 3/4 in. (19 mm) for t greater than 21/4 in. (57 mm). Also, UW-51 provides acceptance criteria for aligned indications that are in the same proximity. UW-51 does not provide acceptance criteria for weld porosity that is detected by radiography. Acceptance criteria for weld porosity are provided by mandatory Appendix 4. Paragraph UW-52 provides radiographic examination requirements for welds that require spot radiography, including those welds specified to be radiographed by paragraph UW-11(a)(5)(b). It should be understood that spot radiography is used as an aid to quality control. Note to UW-52 provides very useful information. Spot radiographing of a welded joint is recognized as an effective inspection tool. The spot radiography rules are also considered to be an aid to quality control. Spot radiographs made directly after a welder or an operator has completed a unit of weld proves that the work is or is not being done in accordance with a satisfactory procedure. If the work is unsatisfactory, corrective steps can then be taken to improve the welding in the subsequent units, which unquestionably will improve the weld quality. Spot radiography in accordance with these rules will not ensure a fabrication product of predetermined quality level throughout. It must be realized that an accepted vessel under these spot radiography rules may still contain defects which might be disclosed on further examination. If all radiographically disclosed defects must be

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 67

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 67

eliminated from a vessel, then 100% radiography must be employed. Thus, spot radiography will not assure that a vessel is free from defects if full radiography were applied. The user must determine whether spot radiography is appropriate for the intended service given the operating conditions. Spot radiography requires the examination of one spot [6.0 in. (150 mm)]of a butt weld for each 50-ft. increment of weld or fraction thereof when the joint efficiency of the butt weld is based on spot radiography. For each increment of weld to be examined, spot radiographs must be taken to examine each welder or weld operator. The spot must be taken at a location determined by the Inspector as soon as possible after the weld increment is completed. It should be noted that any spot radiographs required by paragraph UW-11(a)(5)(b) cannot be used to satisfy the spot radiographic requirements. This means that if a section of a vessel is specified to have full radiography [UW-11(a)] but the Category B welds are radiographed in accordance with UW-11(a)(5)(b), and another section of the vessel is specified to be spot radiographed, then the spot radiographs taken to satisfy UW-11(a)(5)(b) cannot be used to satisfy the UW11(b) requirements. The acceptance criteria for spot radiography are less restrictive than that for full radiography. Any indications that are characterized as cracks or lack of fusion or penetration are unacceptable. However, slag inclusions and cavities within certain sized limits and spacing may be deemed acceptable. Porosity is not evaluated for spot radiography. If a spot radiograph reveals an unacceptable indication, two additional radiographs must be taken away from the first radiograph, but within the same weld increment. If these additional radiographs do not reveal additional unacceptable indications, the weld increment is acceptable, provided the defects revealed by the first radiograph are repaired (and reexamined). If either of the additional radiographs reveals unacceptable indications, then the entire weld increment is rejected and must be repaired (and reexamined).

required after heat treatment and the actual tensile strength shall not exceed the specified minimum value by more than 20 ksi. Likewise, paragraph UF-12 specifies that stress risers must be minimized for certain grades of SA-372. The yield strength to tensile strength ratio is relatively large for these grades of forging material, and it is important to minimize localized stresses that may not be able to redistribute, by yielding, without cracking the material. Because of this concern, the provisions of UG-36(c)(3) that exempts reinforcement calculations for small openings do not apply when these materials are used. 21.5.2.2 Design and Fabrication Part UF supplements the fabrication requirements of Part UG for forged construction. The out-ofroundness of the body of a forged vessel is limited to 1% of mean diameter of that section. However, provisions are made to allow greater out-of-roundness (up to 3%) if the vessel MAWP is reduced in accordance with Paragraph UF-27. The reduced pressure, P , is defined in UF-27 as 1.25 Sb P + 1Q S

P¿ = P

(21.11)

where:

P S Sb

the MAWP of the forging if the out-ofroundness did not exceed 1% allowable stress of the material at the design temperature the additional bending stress caused by the outof-round condition as determined by Sb = 1.5PR1t (D1 - D2) P t3 + 3 R1R2 a E (21.12)

D1 and D2 R1 Ra E

21.5.2

Part UF: Requirements for Pressure Vessels Fabricated by Forging

Part UF is applicable to forged pressure vessels, without longitudinal joints, fabricated from carbon or low alloy steel or of high alloy steels that are within the limitations given in Part UHA. The rules of Part UF are used in conjunction with the applicable parts of Subsections A and C. The rules of Subsections A and C must be satisfied unless they are superseded by rules of Part UF. 21.5.2.1 Materials Forging material must comply with the requirements of Part UG, except that the carbon content of the forging heat analysis shall not exceed 0.35% when welding is done in the vessel construction. However, if the welding only involves minor welding, such as seal welds and minor nonpressure attachment welds, the carbon content may be as great as 0.50%. If the carbon content of the forging exceeds 0.50%, then no welding is allowed on the forging. Special provisions are defined for SA-372 materials that have been subjected to liquid quenching and tempering heat treatments to enhance the material strength. These special provisions do not apply for austenitic material or to materials that have a specified minimum tensile strength not exceeding 95 ksi. Forgings made from SA-372 materials may be subjected to a quenching and tempering heat treatment in order to obtain the specified mechanical properties; however, surface examination is

(measured maximum and minimum inside diameter, respectively (average inside radius at section as determined from D1 and D2 (average radius to the midwall using D1 and D2 and the thickness at the section (modulus of elasticity of the material taken from Table UF-27.

When a vessel is out of round, bending stresses result in the shell when pressure is applied. The above procedure accounts for the added bending stress and an out-of-round forging may be used if the reduced pressure is acceptable for the intended service. This procedure is identical to the provisions of Appendix 27 that allow glass-lined vessels to exceed the roundness tolerance presented in paragraph UG-80. 21.5.2.3 Heat Treatment Vessels that are fabricated from liquid quenched and tempered SA-372 material shall be subjected to heat treatments, as defined in SA-372, after the forging and welding is completed (seal welds of threaded connections may be done after the heat treatment). After heat treatment, the outside surface and those portions of the inside surface that are accessible shall be examined by the magnetic particle method for ferromagnetic material and by the liquid penetrant method for nonmagnetic materials. Likewise, liquid quenched and tempered material (except for austenitic material) must be subjected to Brinell hardness testing.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 68

68 · Chapter 21

The average of the Brinell hardness readings at a section shall not be more than 10% below or more than 25% above the required Brinell hardness value. The highest average hardness number at one section shall not exceed the lowest average hardness reading taken at another section of the vessel by more than 40 points. The required Brinell hardness value is determined by dividing the specified minimum tensile strength by 500. The hardness testing requirements are to assure that the heat treatment has resulted in the proper strength and to assure that the strength is uniform throughout the entire vessel. 21.5.2.4 Welding For forged vessels with material having a carbon content greater than 0.35%, no welding is allowed except for seal welding of threaded connections, welds attaching minor, nonpressure parts, or repair welds. When welding is done, the welding procedure and welders must be qualified in accordance with the provisions of UF-32(b). These rules require that the qualification tests be based on material that has seen a similar heat treatment as that required for the finished vessel. Paragraph UF-37 provides the requirements for the repair of surface defects on forged vessels. Defects shall be removed by grinding. A local thin area may have a thickness less than the required thickness if it can be justified by using the rules for reinforcement contained in Part UG. If the local thin area does not satisfy the reinforcement provisions, repair by welding is allowed if the welding meets the requirements of UF-37(b) and is acceptable to the Inspector.

MAXIMUM DESIGN TEMPERATURES FOR BRAZING FILLER METAL (Source: Table UB-2 of Section VIII, Div. 1 2007 Edition of the ASME Code)

TABLE 21.4

Column 1 Temperature, °F (°C), Filler Metal Below Which Section IX Classification Tests Only Are Required BCup BAg BcuZn Bcu BalSi BNl BAu BMg 300 (150) 400 (200) 400 (200) 400 (200) 300 (150) 1200 (650) 800 (430) 250 (120)

Column 2 Temperature, °F (°C), Requiring Section IX and Additional Tests 300­350 (150­180) 400­500 (200­260) 400­500 (200­260) 400­650 (200­340) 300­350 (150­180) 1200­1500 (650­815) 800­900 (430­480) 250­275 (120­135)

General Note: Temperatures based on AWS recommendations

21.5.3

Part UB: Requirements for Pressure Vessels Fabricated by Brazing

21.5.3.1 General Part UB applies to pressure vessels and parts that are fabricated by brazing. Brazing is defined as a group of welding processes that produce a coalescence of the materials by heating them to the brazing temperature in the presence of a filler metal having liquidus above 840°F (450°C) and below the solidus of the base material. The filler material is drawn into the gap between closely fitting surfaces of the joint by capillary attraction. The brazing processes are categorized by the method of heating and include the following: (1) (2) (3) (4) (5) torch brazing furnace brazing induction brazing electrical resistance brazing dip brazing

21.5.3.2 Design Considerations The allowable operating temperature of a brazed joint is dependent on the filler material and the base materials of the joined parts. Temperature limits for brazing filler material is given in Table 21.4. Table 21.4, Column 1, defines the maximum design temperature for filler material when only the qualification tests of Section IX of the ASME Boiler & Pressure Vessel Code are required. Higher temperatures are allowed by Column 2 if the additional tests of paragraph UB-12 are satisfied. Brazed joints are not allowed in vessels that are in lethal service, used for unfired steam boilers, or subjected to direct firing. When brazed joints are used, the Manufacturer is responsible to assure that the brazed joint will have adequate strength at the design temperature. The strength of the brazed joint shall not be less than the strength of the weaker of the base materials to be

joined. When a brazed joint will be used at a design temperature in accordance with Column 2 of Table 21.4, mechanical testing of production-type joints is required at the design temperature and at a temperature that is 5% warmer than the design temperature. In either test, the joint shall not fail in the braze metal. The joint efficiency to be used for brazed joints depends on the ability to visually examine the joint to assure that the filler material has fully and completely penetrated the entire joint. If such visual examination can confirm that the joint is completely filled, a joint efficiency of 1.0 may be used in the design. However, if it cannot be confirmed that the filler has penetrated the entire joint, a joint efficiency of 0.5 shall be used. Figure 21.14 provides guidance for the application of filler material and the appropriate joint efficiency to be used. The application of the brazing filler material should be considered as part of the joint design in order to assure that the needed visual observation of penetration of the material is possible. Some acceptable types of brazed joints are shown in Fig. 21.15. The strength of brazed joints is dependent on the clearance between the two parts being joined. The strength of the joint decreases as the clearance increases. In order to assure that the fabricated brazed joints have the required strength, the clearances between the parts to be brazed shall be within the tolerances set by the design and as used in the qualification testing of the brazing process. Table 21.5 provides recommended joint clearances at the brazing temperature for various types of filler material. The maximum strength of the brazed joint is obtained when the smallest possible clearances are maintained. 21.5.3.3 Fabrication Requirements Paragraphs UB-31 and UB-32 provide specific requirements for the qualification of brazing procedures and brazers and brazing operators. The requirements for brazing parallel those for welding. Each procedure must be qualified in accordance with Section IX. Each brazer and operator must be issued an identifying mark that can be used to identify their work, and the Manufacturer must maintain records of the qualification tests and the identification of each of the brazers of operators. Likewise, paragraph UB-43 requires the Manufacturer to certify that the brazing has been done by brazers or operators who have been qualified by the requirements of Section IX, and the Inspector shall assure himself that only qualified brazers have been used. The Inspector shall visually inspect both sides of a

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 69

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 69

FIG. 21.14

EXAMPLES OF FILLER METAL APPLICATION (Source: Fig. UB-14 of Section VIII of the ASME Code)

FIG. 21.15 SOME ACCEPTABLE TYPES OF BRAZED JOINTS (OTHER EQUIVALENT GEOMETRIES YIELDING SUBSTANTIALLY EQUAL RESULTS ALSO ARE ACCEPTABLE) (Source: Fig. UB-16 of Section VIII Div. 1 of the ASME Code)

brazed joint to assure that the filler metal has completely penetrated the joint. If one side of the joint is not visible, then the Inspector shall review design in order to assure that the correct joint efficiency has been used. The following conditions are grounds for rejection of a brazed joint: (1) evidence that the filler metal did not completely fill the joint (2) concavity in a butt joint

(3) cracks in the brazing filler material (liquid penetrant examination may be used but is not mandatory) (4) cracks in the base material adjacent to the brazed joint (5) pinholes or open defects in the brazed joint (6) rough fillets, especially those with a convex appearance Except for cracks in the base material, repair or rebrazing of the joint is allowed as defined in paragraph UB-44.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 70

70 · Chapter 21

RECOMMENDED JOINT CLEARANCES AT BRAZING TEMPERATURE (Source: Table UB-17 of Section VIII, Div. 1 2007 Edition of the ASME Code)

TABLE 21.5

Brazing Filler Metal BAlSi

Clearance, in. (mm) [Note (1)] 0.006­0.010 (0.15­0.25) for laps less than or equal to 1/4 in. (6 mm) 0.010­0.025 (0.25­0.64) for laps greater than 1/4 in. (6 mm) 0.001­0.005 (0.02­0.13) 0.002­0.005 (0.05­0.13) 0.002­0.005 (0.05­0.13) 0.000­0.002 (0.05­0.13) [Note (2)] 0.001­0.005 (0.02­0.13)

ness exemption curves also requires that the backing strip material specification minimum tensile strength shall not exceed that of the pressure part material specification, and the backing strip material specification minimum per cent elongation shall be at least equal to that for the pressure part material specification. 21.6.1.2 Requirements for Postweld Heat Treatment Welds and HAZs of welds in carbon and low alloy steel material may have a hardened, martensitic grain structure if the weld is cooled too rapidly. In these instances, there is an increased likelihood that the lack of notch toughness would cause cracking and possibly failure of the weld seam. In order to preclude the undesirable effects of hardened welds and heated-affected zones, PWHT may be required. Paragraph UCS-56 requires postweld heat treatment of all welds in carbon and low alloy material unless specifically exempted in the notes to Table UCS-56 and Table UCS-56.1. It is noted that, as the starting point, all welds joining Part UCS material require a PWHT. However, certain welds may be exempted as defined in the notes of Table UCS-56. (Table 21.3 provides a portion of Table UCS-56 for reference.) One of the key parameters causing the undesirable characteristics of welds and heat-affected zones for these materials is the cooling rate of the weld. The faster the cooling rate of the weld, it is more likely that undesirable properties will develop. It is a simple matter of heat transfer to show that the larger the heat sink of the parts being welded, the faster the weld and HAZ will be cooled. Thus, the cooling rate of the weld is a function of the thickness of the material being welded, and the PWHT exemptions are thickness dependent. Experience has shown that carbon steel and some low alloy material have acceptable properties in the weld and HAZ, within certain thickness limitations, without postweld heat treatment. This experience provides the basis of the PWHT exemptions in the notes of Table UCS-56. For example, experience has shown that carbon steel (P-No. 1 material) has satisfactory weld and HAZ properties without PWHT if the thickness of the parts being welded do not exceed 11/4 in. If the material is preheated prior to welding, the weld pool will cool at a slower rate. Thus, Table 21.3 allows the PWHT thickness exemption to increase to 11/2 in. for carbon steel if a 200°F preheat is applied during welding. Before the PWHT exemptions can be correctly applied, a clear understanding of the definition of nominal thickness is required. The nominal thickness to be used in Table UCS-56 is not always the thickness of the parts joined by welding. The definitions of nominal thickness to be used in applying PWHT holding times and exemptions are found in paragraph UW-40(f). For example, paragraph UW-40(f) defines the nominal thickness of a butt weld between two parts as the total depth of the weld, not including any weld reinforcement. For groove welds, the nominal thickness to be used in Table UCS-56 is the depth of the groove. For fillet welds, the nominal thickness is the throat of the fillet. If a groove and fillet weld is used, the nominal thickness is the greater of the depth of the groove or the throat dimension. At a shell-totubesheet or shell-to-flat head weld, the nominal thickness is the thickness of the shell. For a nozzle attachment, the nominal thickness is the greater of the weld thickness across the shell, the thickness of the weld across the reinforcement pad, or the thickness of the attaching fillet weld. A careful review of the nominal thickness definitions of paragraph UW-40(f) shows that the nominal thickness to be used in Table UCS-56 is basically the thickness of the weld that has been deposited. The requirements for PWHT for a given material are organized by the material's P-No. from Section IX of the ASME

BCuP BAg BcuZn BCu BNi

Notes: (1) In the case of round or tubular members, clearance on the radius is intended. (2) For maximum strength, use the smallest possible clearance.

21.6

SUBSECTION C: REQUIREMENTS PERTAINING TO CLASSES OF MATERIALS

Part UCS: Requirements for Pressure Vessels Constructed of Carbon and Low Alloy Steels

21.6.1

21.6.1.1 Materials Part UCS provides rules that are applicable to pressure vessels and vessel parts that are constructed from carbon or low alloy steel. These rules are used in conjunction with the general rules in Part A and the specific requirements pertaining to the methods of fabrication that are defined in Part B of Section VIII, Division 1. Except as provided in UG-10 and UG-11, all carbon and low alloy material used must be a material listed in Table UCS-23. Additionally, steel with a carbon content exceeding 0.35% shall not be used for welded construction or for material exposed to oxygen cutting. This requirement is based on the tendency for steels with high carbon content to exhibit high hardness and poor toughness when welded or thermally cut. Limits are also placed on the use of structural grades of plate material (SA-36 and SA-283). These materials are allowed only when (1) the vessel is not in lethal service (2) the material is not part of an unfired steam boiler (3) if strength welding is applied (other than flanges, bolted covers, and stiffening rings), the thickness does not exceed 5/8 in. In the 2006 Addenda to VIII-1, a much needed clarification regarding backing strip material was added. Prior to this revision, backing strip material was required to be made of code material to satisfy the toughness requirements in UCS-66. Now backing strip material may be from any specification so long as (1) The specification maximum composition limits for the backing strip material do not exceed those of the material specification for the pressure part to which it is attached. (2) The backing strip material and its heat-affected zones (HAZs) and weld metal shall be impact tested as per UG 84 or the material shall be assigned to Curve A and of the MDMT shall be determined as per Fig. UCS-66. This last option to establish the MDMT using the tough-

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 71

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 71

Code. P-Numbers are assigned to base metals dependent on the characteristics such as composition, weldability, and mechanical properties. The P-No. for each base material is listed in the allowable stress tables of Section II, Part D. P-Numbers may be further categorized by assigning a Group Number. When joining materials of differing P-Numbers, the PWHT requirements of the materials requiring the higher PWHT temperature and holding time shall control. It is useful to demonstrate some of the PWHT exemption rules of Table UCS-56. For example, consider a vessel constructed from carbon steel of a P-No. 1, Gr. No. 1 material. If the vessel thickness is 13/8 in.(35 mm), the butt welds in this vessel do not require PWHT, provided that a 200°F (95°C) preheat be applied. These welds are exempt from PWHT by note (2)(b) of Table 21.3. For a vessel of the same material that is 2 in. (50 mm) thick, the butt welds require PWHT at 1100°F (595°C) minimum temperature for 2 h. However, if a nozzle neck of 1 in. (25 mm) thickness is welded to the shell using a "set-on" detail as given in Fig. 21.13a PWHT of this weld is not required because the thickness of the weld through the nozzle neck [defined by UW-40(f) as the nominal thickness] is less than 11/4 in. (32 mm). Another example that demonstrates the use of the nominal thickness [as defined in paragraph UW-40(f)] and how it is applied for PWHT exemptions is given in the ASME Code Interpretation given below. Interpretation: VIII-1-83-100 Subject: Section VIII-1, UCS-56 and Table UCS-56 Date Issued: March 3, 1983 File: BC82-825 Question: Is PWHT required by Table UCS-56, Section VIII, Division 1 for the weld joints in the following conditions? In each case all material is P-No. 1, and the service requirements of UCS-67 and UW-2(a) are not applicable. (a) A 5/8-in. thick shell is attached to a 3-in. thick tubesheet using the attachment weld geometry of Fig. UW-13.2f of Section VIII, Division 1. (b) A 5/8-in. thick nozzle that is 8 in. in diameter to a 3-in. thick shell using the attachment weld geometry of Fig. UW-16.1 a of Section VIII, Division 1. (c) A 5/8-in. thick clip (nonpressure part) is attached to a 3-in. thick shell using a full penetration weld through the thickness of the clip. Reply: (a) No. (b) No. (c) No. Table UCS-56 makes provisions for a lower PWHT temperature at a longer hold time than the minimum specified time. When this option is selected, the corresponding hold time for a decrease in minimum specified PWHT temperature is given in Table 21.6. This table may be used only if it is referenced by the requirements for a specific P-No. in Table UCS-56. The rationale behind Table 21.6 is that the tempering (softening) of the heat-affected zone of a weld will occur at a temperature less than the minimum specified temperature; however, it will take a longer period of time to achieve adequate tempering. The hold time values for the temperature decrease in PWHT presented by Table 21.6 are experienced based. It should be noted that none of the PWHT exemptions provided in UCS-56 apply when PWHT is a requirement of the service

TABLE 21.6 ALTERNATIVE POSTWELD HEAT TREATMENT REQUIREMENTS FOR CARBON AND LOW-ALLOY STEELS (Source: Table UCS-56.1 of Section VIII Div. 1)

Decrease in Temperature Below Minimum Specified Temperature.°F 50 100 150 200

Minimum Holding Time [Note (1)] at Decreased Temperature, hr 2 4 10 20

Notes ... ... (2) (2)

Notes: (1) Minimum holding time for 1 in. thickness or less. Add 15 minutes per inch of thickness greater than 1 in. (2) These lower postweld heat treatment temperatures permitted only for P-No. 1 Gr. Nos. 1 and 2 materials.

conditions of paragraph UCS-68. Paragraph UCS-68 makes provisions for reducing the impact test exemption temperature obtained from Fig. 21.16 by 30°F (17°C) when PWHT is performed when it is not otherwise required. Likewise, PWHT exemptions are more restrictive for certain fluid services (see paragraph UW-2) because of the improved level of quality provided by the PWHT for welded joints of UCS material. For P-No. 1 material, the PWHT exemptions for the service conditions given in UW-2 are defined in note (2)(c) of Table 21.3. Low alloy materials are generally more susceptible to a hardened heat-affected zone and, consequently, the PWHT exemptions are more restrictive for these materials. For example, PWHT exemptions given for P-Nos. 4 and 5 materials are very restricted. These are the 11/4 Cr­1/2 Mo and 21/4 Cr­1 Mo materials that are susceptible to delayed hydrogen embrittlement after welding if the hardness of the weld and HAZ is not reduced by PWHT. Note that separate PWHT exemption tables are provided for the P-No. 5 material; one table is provided for P-Nos. 5A, 5B, Gr. No. 1, and 5C, Gr. No. 1 and a separate table for P-No. 5B, Gr. 2. Additional research has identified two material degradation modes for P-No. 5B, Gr. 2 material (Grade 91, etc.) when the lower critical temperature (LCT) is exceeded: ferrite and martensite formation. In addition to the minimum holding temperature, a maximum holding temperature is also specified. Thus, in order to determine if a vessel or weld is exempt from PWHT, the nominal thickness of the weld should be established in accordance with UW-40(f) and the P-No. of the material should be determined from the allowable stress tables in Section II, Part D, for the materials being welded. If there is an exemption to PWHT given in the appropriate part of Table UCS-56, then the weld may not require PWHT. When a vessel or vessel part is to be PWHT in a furnace, the temperature and hold time must be based on the nominal thickness and material that provides the highest temperature and hold time. For nonpressure parts that are welded to pressure parts, the PWHT requirements of the pressure part govern. When PWHT is required, paragraph UW-40 provides the general procedures to be followed; however, additional requirements for Part UCS material are defined by paragraph UCS-56(d). These additional requirements include provisions for heating/cooling rates and allowable gradients to control thermal stress induced by

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 72

72 · Chapter 21

140 (60)

120 (49)

100 (38) Minimum Design Metal Temperature, F ( C) A 80 (27)

B

60 (16) C 40 (4) D 20 (­7)

0 (­18)

­20 (­29)

­40 (­40) ­55 (­48) ­60 (­50) Impact testing required ­80 (­62) 0.394 (10) 1 (25) 2 (51) 3 (76) Nominal Thickness, in. (mm) [Limited to 4 in. (102 mm) for Welded Construction] 4 (102) 5 (127)

FIG. 21.16

IMPACT TEST EXEMPTION CURVES (Source: FIG. UCS-66 of Section VIII DIV. 1 of the ASME Code)

the PWHT process. Also, the furnace atmosphere shall be controlled to avoid excessive oxidation of the material during the PWHT operation. After a vessel has been subjected to a PWHT, there are instances where weld repair is required for either the base material or the weld. Paragraph UCS-56(f) provides rules for weld repairs after PWHT without the need for a subsequent PWHT. It should be noted that that all weld repairs must be completed prior to the pressure test. Material and welds may be repaired without subsequent PWHT if the total repair depth does not exceed 11/2 in. (38 mm) for P-No.1 and 5/8 in.(16 mm) for P-No.3 material. The total depth of repair shall be taken as the sum of the depth from both sides at a given location. This PWHT exemption for weld repairs is not applicable when the service requirements of UCS68 apply, if the vessel is in Lethal Service [UW-2(a)], or the repair

does not meet the exemptions for nominal thickness given in Table UCS-56. Additional requirements are placed on the welding procedures to be used for the repair welding. Preheat of 200°F (95°C) is required for P-No. 1 materials, and a temper bead weld technique is to be applied for P-No. 3 materials. This technique allows each deposited layer of weld to be tempered by the subsequent layers of weld. This method effectively provides the beneficial effects obtained by PWHT. When PWHT exemption is applied for weld repairs, the manufacturer must obtain the user's acceptance and the repairs shall be noted on the Data Report. 21.6.1.3 Radiographic Examination Table 21.7 defines the thickness of UCS material, organized by P-Number, for which full radiography is required of butt-welded joints. Carbon steel of PNo.1 requires full radiographic examination for butt welds that

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 73

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 73

have a thickness that exceeds 11/4 in. (32 mm). Butt welds in P-No. 4 (11/4 Cr­ 1/2 Mo) material require full radiographic examination when the thickness exceeds 5/8 in. (16 mm). Butt welds in P-No. 5 (21/4 Cr­ 1 Mo) material require all butt welds to be fully radiographed. These requirements, as defined in paragraph UCS-57, are in addition to the requirements of paragraph UW-11. 21.6.1.4 Low Temperature Operation Figure 21.16 is used as a guide for the selection of materials and provides the basis for impact test exemption of the pressure-retaining metal at the minimum design metal temperature coincident with full design or operating pressure. Unless otherwise exempted by rules given elsewhere in Division 1, impact testing is required for a combination of MDMT and thickness that is below the applicable curve for the material in Fig. 21.16. If the MDMT and thickness combination is on or above the applicable curve, then impact testing of the base material is not required. Figure 21.16 applies to all pressure-retaining material and any nonpressure parts that are essential to the vessel integrity (e.g., support skirts and support lugs) that are welded to the vessel. Figure 21.16 was developed from the work done by members of the Special Working Group Toughness (SWGT), a Committee reporting to Subcommittee VIII of the Boiler & Pressure Vessel Code. The basis for the curves is given in papers by Corten [11] and Selz [12]. A subsequent paper by Yukawa [13] describes the inherent fracture safety margins built into these curves. Some of the essential elements included in the development and basis of the impact exemption curves are discussed below. Notch Blunting. The mandatory overload hydrostatic pressure test enhances performance at low temperatures. This warm prestressing causes mechanical stress relief of built-in residual stresses, which is called notch blunting or notch nullification. Strain Rate. Another conservatism built into the curves is the effect that the strain rate has on the apparent toughness of the material. The curves of Fig. 21.16 are based on dynamic fracture toughness, using Charpy V-Notch tests. The actual loading rate (strain rate) or rate of pressurization in a pressure vessel is much less. The apparent notch toughness of a material is dependent on the rate of loading. For a slower strain rate, the apparent toughness of material increases. As pointed out by Selz [12], loading rates during CVN testing are about 30,000,000 psi/s, and during severe transient loading, the loading rate is about 30,000 psi/s. The slower actual loading rate would permit a temperature shift (improvement of apparent notch toughness) of 65°F of the toughness/temperature curve from the toughness curve derived from dynamic loads. Therefore, CVN toughness, for example, determined by testing at 20°F would be equivalent to the toughness during a nondynamic loading at 85°F. To allow for unknown factors and to avoid the need for controls to assure that the temperature shift is valid, as implemented in the code, Table UG-84.4 allows only a 10°F temperature shift for materials with specified minimum yield strength not greater than 40 ksi. Barsom and Rolfe [14] provide an in-depth explanation of the fracture toughness shifts that occur at different strain rates. Section VIII, Division 1, contains provisions in paragraph UCS66(d) that permit the use of nonimpact-tested thin wall carbon steel pipe and tubing at a temperature much colder than allowed by Fig. 21.16. The impact test exemption temperature is 155°F for

TABLE 21.7 THICKNESS ABOVE WHICH FULL RADIOGRAPHIC EXAMINATION OF BUTT WELDED JOINTS IS MANDATORY (Source: Table UCS-57 of Section VIII, Div. 1 2007 Edition of the ASME Code)

P-No. & Group No. Classification of Material 1 Gr. 1, 2, 3 3 Gr. 1, 2, 3 4 Gr. 1, 2 5A Gr. 1,2 5B Gr. 1, 2 5C Gr. 1 9A Gr. 1 9B Gr. 1 10A Gr. 1 10B Gr. 1 10C Gr. 1 10F Gr. 1

Nominal Thickness Above Which Butt Welded Joints Shall Be Fully Radiographed, in. (mm) 11/4 (32) 3 /4 (19) 3 /4 (16) 0 (0) 0 (0) 0 (0) 5 /8 (16) 5 /8 (16) 3 /4 (19) 5 /8 (16) 5 /8 (16) 3 /4 (19)

pipe or tubing of P-No. 1 materials, NPS 4 or smaller, and within the following yield strength and thickness limits: SMYS, ksi (MPa) 20 to 35 (140 to 240) 36 to 45 (250 to 310) 46 (320) and higher Thickness, in. (mm) 0.237 (6.0) 0.125 (3.2) 0.10 (2.5)

This has significant implications for materials used in equipment exposed to cold temperatures. The main thrust of the rules of UCS-66(d) is to permit the use of nonimpact-tested carbon steel tubing in exchanger services that operate at temperatures as cold as liquid ethylene at atmospheric pressure; thus, nonimpact-tested carbon steel tubing may be used at service temperatures as cold as 155°F. The fracture toughness of thin carbon steel sections that are not highly restrained is well known and its acceptability at these colder temperatures was demonstrated by validation studies. In order to use Fig. 21.16 correctly, an understanding of the nominal thickness is very important. Paragraph UCS-66(a) provides definitions of governing thickness. A detailed example of the procedure to determine an MDMT, including the use of governing thickness, is given in Appendix L. The example illustrates the complete procedure in detail. As pointed out in the example, the governing thickness is normally the thickness at the welded joint at the point of interest. Figure 21.17 clarifies the definition of governing thickness, including a definition for flat heads and tubesheets. The definition of governing thickness for flat heads and tubesheets is the greater of the flat head or tubesheet thickness divided by four or the thickness at its welded connection. The t/4 thickness definition is widely used in other international codes and it recognizes that the design of flat covers and tubesheets is primarily controlled by bending stresses. Paragraph UCS-66(b) can be used to establish the acceptability of operating a vessel at temperatures colder than the MDMT, coincident with a reduced pressure. In order to understand the basis for the low stress exemptions as provided in paragraph UCS-66(b), reference is made to the fracture analysis diagram (FAD) shown in Fig. 21.18 , originally developed by Pellini at the Naval Research

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 74

74 · Chapter 21

FIG. 21.17

SOME TYPICAL VESSEL DETAILS SHOWING THE GOVERNING THICKNESSES AS DEFINED IN UCS-66 (Source: Fig. UCS-66.3 of Section VIII Div. 1 of the ASME Code)

Labs [15]. The bottom curve on the FAD represents the crack arrest temperature (CAT). The CAT line indicates the stress relative to the material's nil-ductility temperature (NDT), below which brittle fracture will not occur regardless of what size flaw may be present. The CAT stress limit is generally considered as 10% of the material's minimum tensile strength for materials listed in Fig. 21.18. For Division 1 designs, Fu/3.5 0.35 0.10 Fu or 10% of the minimum ultimate tensile stress. This is how the 0.35 design pressure impact exemption limit in UCS-66(b) is derived. The curved portion of the CAT line also illustrates that as the temperature is increased above the NDT, the allowable CAT stress increases. The application of these concepts is included in the code, as shown by Fig. 21.19. This figure allows operation of a vessel in a temperature colder than the MDMT if the applied stress is lower than the allowable design stress. This is supported by the following code interpretation:

Interpretation: VIII-1-89-268 Subject: UG-20(b) and Fig. UCS-66.1 File: BC 90-225 Question: Is it permissible to design a vessel for operations below the minimum design metal temperature stamped on the nameplate, provided the reduction in MDMT for the coincident design stress in tension results in a temperature that is no colder than that permitted by Fig. UCS-66.1. Reply: Yes. This interpretation indicates that a vessel may be safely operated below its MDMT if the primary tensile membrane stress ratio at coincident temperature remains within the limits governed by Fig. 21.19. Nonpressure loads that result in primary tensile membrane stresses must also be considered; however, there are

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 75

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 75

FIG. 21.18 FRACTURE ANALYSIS DIAGRAM FOR THE ENGINEERING SELECTION OF FRACTURE-SAFE STEELS BASEDON THE NDT TEMPERATURE

no specific rules presented for loads listed in UG-22 and the provisions of paragraph U-2(g) must be applied. Since the issuance of the above interpretation, provisions have been incorporated into the code in paragraph UCS-160 that implements the intent stated by this interpretation. It should be noted that unless the stress ratio is less than 0.35, impact testing is required if the operating temperature is colder than 55°F ( 48°C). For example, if the MDMT of a vessel is 30°F and the vessel is subjected to a decrease in temperature coincident with a pressure decrease. For a stress ratio of 0.38, the allowed reduction in operating temperature is 100°F. Thus, by using Fig. 21.19, the allowable operating temperature would be ­70°F ( 57°C); however, since this temperature is colder than ­55°F ( 48°C), impact testing is required. For such instances, the code does not define what the impact test temperature has to be, and if the materials for this example were impact tested at 30°F ( 1°C), then operation at ­70°F ( 57°C) would be acceptable. Also, if the minimum operating temperature is below­55°F ( 48°C) and the stress ratio is not less than 0.35, then all the requirements of UCS-68 apply. As given in paragraph UCS-67, impact testing in accordance with paragraph UG-84 of specimens taken from weld metal and heat-affected zones must be included in the Welding Procedure Qualification for welds made with filler material when any of the following apply: (1) When either of the base materials require impact testing. (2) When the base materials are exempt from impact testing by Fig. 21.16, Curves C or D material and the MDMT is colder than ­20°F ( 29°C), but not by more than ­55°F ( 48°C), unless the welding filler metal has been classified by impact tests at a temperature not warmer than the MDMT by the applicable SFA specification from Section II, Part C;

(3) When joining base metals are exempted from impact testing by UCS-66(g) when the MDMT is colder than ­55°F ( 48°C). Procedures for welds made without the use of filler material shall be impact tested when the thickness at the weld exceeds 1/2 in. (13 mm) regardless of the MDMT. This does not include autogenously welded pipe or ERW pipe and tube purchased according to a material specification. Likewise, impact testing is required when the thickness of welds made without filler metal exceeds 5/16 in. (8 mm) and the MDMT is colder than 50°F (10°C). Weld heat-affected zones (for welds made with or without filler metal) shall be impact tested when any of the following provisions apply: (1) The base metal requires impact testing. (2) The welds have any individual weld pass exceeding 1/2 in. and the MDMT is colder than 70°F (21°C). (3) The base metals are exempt from impact testing and the MDMT is colder than ­55°F ( 48°C). It is noted that the provisions of paragraph UG-84(i) generally require impact testing of the actual production welding to be performed by Vessel (Production) Test Plates anytime the welding procedures require impact testing by paragraph UCS-67; however, there are exceptions to the production impact tests as given in paragraph UCS-67(d). Specific design requirements apply when a vessel is to be operated at cold temperatures. These requirements are given in paragraph UCS-68. For vessels with a MDMT colder than­55°F ( 48°C), the following requirements of paragraph UW-2(b) apply (unless the coincident ratio of Fig. 21.19 is less than 0.35): (1) All Category A welded joints shall be Type 1 of Table 21.1. (2) All Category B welded joints shall be Type 1 or Type 2 of Table 21.1.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 76

76 · Chapter 21

1.00 Ratio: tr E /(tn ... c); See Nomenclature for Alternative Ratio

0.80

0.60

0.40 0.35 See UCS-66(b)(3) when ratios are 0.35 and smaller 0.20

0.00 0 20 40 60 80 100 120 140 F ( C) [See UCS-66(b)]

Nomenclature (Note reference to General Notes of Fig.UCS-66-2.) 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)], in. (mm) tn = nominal thickness of the component under consideration before corrosion allowance is deducted, in. (mm) c = corrosion allowance, in. (mm) E = as defined in General Note (3). Alternative Ratio = S E 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 E and E are as defined in General Note (3).

FIG. 21.19 REDUCTION IN MINIMUM DESIGN METAL TEMPERATURE WITHOUT IMPACT TESTING (Source: Fig. UCS-66.1 of Section VIII Div. 1 of the ASME Code)

(3) All Category C welded joints shall be full penetration welds extending through the entire section at the joint. (4) All Category D welded joints shall be full penetration welds extending through the entire thickness of the vessel wall or the nozzle wall. If the MDMT of a vessel made of UCS material is colder than ­55°F ( 48°C), then PWHT is required unless the MDMT was established using a coincident ratio value less than 0.35. Paragraph UCS-68(c) allows a 30°F (17°C) reduction in the impact test exemption temperature given in Fig. 21.16 when PWHT is performed but is not otherwise required by other parts of Division 1. This rule applies to P-No.1 material only. For

example, a 3/4-in. thick vessel made of SA-516 Grade 60 material does not require PWHT by the provisions of 21.3. This is a Curve C material, and Fig. 21.16 gives an impact test exemption temperature of ­15°F ( 26°C). If this vessel were postweld heat treated (not otherwise a requirement of Division 1), the impact test exemption can be reduced by 30 (17°C) to ­45°F ( 42.8°C). This means that the MDMT could be as cold as ­45°F ( 42.8°C) and no impact testing is required. This relaxation of the exemption temperature recognizes the beneficial effect that heat treatment provides for protection against brittle fracture. PWHT provides a tempering effect of the hardened heat-affected zone and reduces the residual stress, both of which improve resistance to brittle fracture.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 77

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 77

21.6.1.5 Fabrication Part UCS provides supplementary fabrication requirements for carbon and low alloy steels. Paragraph UCS-79 defines the requirements for formed sections. The most significant forming requirement is given in paragraph UCS-79. This paragraph defines the need for subsequent heat treatment after cold forming when the extreme fiber elongation is more than 5% from the preformed condition. For P-No. 1, Gr. Nos. 1 and 2 materials, the extreme fiber elongation may be as great as 40% with no postforming heat treatment when none of the following conditions exist: (1) (2) (3) (4) The vessel is in lethal service [see paragraph UW-2(a)]. The material is not exempt from impact testing. The preformed thickness is greater than 5/8 in.. The reduction in thickness by the forming operation is more than 10% at the location where the extreme fiber stress exceeds 5%. (5) The metal temperature during forming is in the range of 250­ 900°F. The strength of carbon and low alloy steel depends on the thermal heat treatment the material has received. Accordingly, it is important that test specimens, required to demonstrate the material's physical properties, accurately represent the material properties of the finished vessel. For example, it is possible that the material test coupons for normalized and tempered low alloy material show the specified strength is adequate, but subsequent heat treatment of the actual vessel during fabrication could result in a decrease in the strength of the material. Paragraph UCS-85 provides rules to assure that the test specimens used to validate the mechanical properties of the supplied material, as given on the material test reports, accurately reflect the properties of the finished vessel. The basic requirement of this paragraph is that the material test coupons must have been subjected to the same heat treatment that the vessel will experience during its fabrication. Fabrication heat treatments that must be replicated for the test coupons are those heat treatments where the temperature exceeds 900°F. Heat treatments, in the context of this paragraph, do not include thermal cutting, preheating, welding, or heating pipe or tubing for bending, provided that the bending is done at a temperature that does not exceed the material's lower transformation temperature. The manufacturer must define to the material supplier the temperature, time, and cooling rate to which the material will be subjected during fabrication. The material supplier is required to heat treat, at the identified temperature and hold time, the material from which the test specimens are taken. The hold time of the specimens shall be at least 80% of the total time at temperature expected during the actual vessel fabrication. The simulated PWHT requirement of the test specimens becomes very important when relatively thick normalized and tempered or quenched and tempered material is used. Any PWHT done at a temperature above or near the tempering temperature may reduce the strength of the material, and normalized and tempered material subjected to long hold time and/or high temperature during the PWHT may not meet the minimum strength required by the specification. For these cases, it may be required to use the annealed grade of the material. As an example, it is common to PWHT SA387 Grade 11 (11/4 Cr­ 1/2 Mo) material at a temperature of 1200­ 1250°F. If the material is normalized and tempered with a tempering temperature of 1200°F, it is likely that the material supplier will not be able to guarantee the mechanical properties required by the specification because the PWHT will reduce the material's strength. The annealed grade of this material could be

considered as an acceptable alternative. The simulated test specimens required by UCS-85 will assure that the strength of the material in the finished vessel is representative of the strength required by the specification. Paragraph UCS-85(f) provides exemptions for the simulated PWHT for mill specimens for P-No. 1, Gr. Nos. 1 and 2 material. Likewise, any carbon steel and low alloy material that is used in the annealed condition do not require simulated PWHT for the test specimens. Thus, material that is furnished in the annealed condition does not require simulated mill test specimens. These exemptions recognize that the strength of a material that has not been heat treated to enhance its mechanical properties is not significantly adversely affected by heat treatment during fabrication.

21.6.2

Part UNF: Requirements for Pressure Vessels Constructed of Nonferrous Materials

21.6.2.1 General Nonferrous materials are used to resist corrosion in the petroleum/chemical processing industry and for cleanliness in the food and pharmaceutical processing industry. Nonferrous materials may also be used for the high temperature applications where resistance to scaling and creep resistance is required. Likewise, these materials may be used for their notch toughness properties at cold temperatures. Nonferrous material includes copper and copper alloys, nickel and nickel alloys, aluminum, titanium, and zirconium. The rules in Part UNF apply to vessels and parts of vessels that are constructed of nonferrous materials. These rules are used in conjunction with the general requirements of Subsections A and B that pertain to the method of fabrication used. Unless superseded by Part UNF, all applicable requirements of Subsections A and B are mandatory for nonferrous materials. Because nonferrous material is generally used for specific corrosion resistance or other specific service related requirements, the code does not provide specific recommendations related to chemical compositions, heat treatments, unique fabrication requirements, and any additional testing that may be necessary for a specific application. The user must assure himself that the selected alloy will be suitable for the intended service, considering the fabrication methods, examinations, and heat treatments that the material will experience. 21.6.2.2 Materials All pressure containing components of nonferrous material must conform to one of the materials listed in Table UNF-23 of Section VIII, Division 1. The material shall also comply with the requirements of the applicable material specification of Section II, Part B. Except for bolting material, the allowable stress values for nonferrous material are found in Table 1B of Section II, Part D; allowable stresses for nonferrous bolting material is given in Table 3 of Section II, Part D. Nonferrous material may be used in combination with other material; however, the user must assure that the combination of materials will have no detrimental effect on the corrosion resistance of the material. For example, some material combinations may be incompatible because of galvanic cell corrosion potential. An experienced materials engineer should be consulted when consideration is given to using material combinations. Paragraph UNF-19 has specific requirements pertaining to welded joints in specific nonferrous material. For titanium, zirconium, and UNS N06625 (Inconel 625) material, all Categories A and B joints shall be butt welds of Type 1 or Type 2 as defined in Table UW-12 (see Table 21.1). Also, titanium and zirconium cannot be welded to other materials. When UNS N06625 material is used in a vessel with a design temperature

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 78

78 · Chapter 21

equal to or greater than 1000°F, all Categories C and D welded joints are also required to be either Type 1 or Type 2 of Table UW-12 (Table 21.1). 21.6.2.3 Design Considerations The allowable stress values for nonferrous materials are listed by the product condition, that is, allowable stresses are given for material in the annealed-, cold-, and hot-worked conditions. If welding or brazing is done on a material that has an increased tensile strength produced by cold or hot working, the allowable stress value must be taken for the material in the annealed condition. For example, if a cold-worked plate is used in a welded pressure vessel made of UNF material, the allowable stress for the annealed condition must be used in the design calculations. The designer needs to consider this requirement when selecting the appropriate allowable stress. For material that has an increased strength resulting from heat treatment, the allowable stress for welded or brazed construction shall also be based on the annealed condition unless the completed welds are subjected to the same heat treatment that resulted in the increased strength of the base metal. When the finished welds are heat treated, the Manufacturer shall confirm that the weld and base material are similarly affected by the heat treatment. 21.6.2.4 Fabrication Paragraph UNF-56 provides requirements that experience has shown to be necessary for postweld heat treatment for some specific nonferrous alloys. Generally, postweld heat treatment of nonferrous material is not necessary or desirable. Except for the specific requirements for certain alloys, no PWHT of nonferrous material shall be conducted unless there is an agreement between the user and Manufacturer. Specific PWHT requirements are given for SA-148, Alloy CDA 954; zirconium Grade R60705; and UNS Nos. N08800, N08810, and N08811 alloy material. Paragraph UNF-57 provides radiographic examination requirements for specific nonferrous material. The requirements of UW11 are generally applicable; however, welded joints in some alloys require full radiography. Categories A and B welded joints in titanium and zirconium (and their alloys) shall be fully radiographed. Likewise, all butt joints of thickness greater than 3/8 in. in vessels made of nickel and high nickel alloys (except alloy 200, 201, 400, 401, and 600) shall be fully radiographed. Paragraph UNF-58 specifies liquid penetrant examination requirements for specific nonferrous material. All welds (including groove and fillet) in vessels made from UNS N06625, N10001, and N10665 material shall be examined by the liquid penetrant examination method. Likewise, all welded joints in titanium and zirconium vessels shall be examined by the liquid penetrant method. All welded joints in vessels made of nickel and high nickel alloys (except alloy 200, 201, 400, 405, and 600) shall be examined by the liquid penetrant method when they are not required to be fully radiographed. 21.6.2.5 Notch Toughness Requirements Nonferrous materials do not exhibit a nil-ductility transition at cold temperatures, and no additional requirements are specified for low-temperature operation for UNF material within the following temperature limits: (1) wrought aluminum alloys, down to ­452°F (2) copper and copper alloys, nickel and nickel alloys, and cast aluminum alloys, down to ­325°F (3) titanium and zirconium and their alloys, down to ­75°F

UNF material may be used at lower temperatures if the user is assured (by suitable test results) that the material is suitable for the intended operation. Appendix NF provides nonmandatory guidelines relating to material properties and manufacturing methods that are applicable for nonferrous material. This appendix to Part NF is intended to be informative.

21.6.3

Part UHA: Requirements for Pressure Vessels Constructed of High Alloy Steel

21.6.3.1 General High alloy steels are used to resist corrosion in the petroleum/chemical processing industry and for cleanliness in the food and pharmaceutical processing industry. High alloy steel materials may also be used for high temperature applications where resistance to scaling and creep is required. Likewise, these materials may be used for their notch toughness properties at cold temperatures. The rules in Part UHA apply to vessels and parts of vessels that are constructed of high alloy steel material. These rules are used in conjunction with the general requirements of Subsections A and B that pertain to the method of fabrication used. Unless superseded by Part UHA, all applicable requirements of Subsections A and B are mandatory for high alloy steel material. Because high alloy steel material is generally used for corrosion resistance or other service-related requirements, the code does not provide specific recommendations related to chemical compositions, heat treatments, unique fabrication requirements, or any additional testing that may be necessary for a specific application. The user must assure himself that the selected alloy will be suitable for the intended service, considering the fabrication methods, examination, and heat treatments that the material will experience. 21.6.3.2 Materials All pressure containing components of high alloy steel material must conform to one of the materials listed in Table UHA-23 of Section VIII, Division 1. The material shall also comply with the requirements of the applicable material specification of Section II, Part A. Except for bolting material, the allowable stress values for high alloy steel material are found in Table 1A of Section II, Part D; allowable stresses for high alloy steel bolting material is given in Table 3 of Section II, Part D. 21.6.3.3 Fabrication Paragraph UHA-32 provides the requirements for post-weld heat treatment of UHA material. These requirements are summarized in Table UHA-32. PWHT is required for all welds in UHA material when the nominal thickness [as defined in UW-40(f)] exceeds the thickness limits given by the notes of Table UHA-32 for the material being considered. A review of Table UHA-32 shows that PWHT is generally required for ferritic and martensitic chromium stainless steel (such as Types 405, 410, and 410S stainless steel). Austenitic stainless steels and duplex (austenitic­ferritic) stainless steels are neither required nor prohibited to be postweld heat treated. When the user specifies that duplex stainless steel is to be heat treated, the material shall be heated to the applicable temperature given in Table UHA-32 followed by rapid cooling or quenching. When heat treatment is conducted on austenitic or duplex stainless steels, the temperature is generally set such that a solution annealing occurs. Solution annealing requires a temperature be reached that assures that any carbide or other precipitates at the material grain boundary of the material are put back into "solution" within the material's grain microstructure. The material is then rapidly cooled or quenched from the solution annealing temperature. The required

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 79

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 79

temperature for solution annealing depends on the alloy and the material specification should be consulted if heat treatment is required for service conditions. Solution annealing for austenitic stainless steel is normally conducted by heating the material to temperatures in the range of 1950­ 2000°F followed by rapid cooling. It should be noted that when a welded vessel is made of material having a P-No. that requires PWHT, the exemptions provided by the notes to Table UHA-32 are not applicable if the vessel is a service requirement of UW-2. When welds are made between materials with differing P-Nos., the PWHT temperature and holding time shall be established by the most severe requirement of either material. The construction of some vessels may require welding ferritic steel parts to austenitic stainless steel or duplex stainless steel. However, ferritic steel parts should not be exposed to the solution- annealing temperature that may be required by the austenitic or austenitic/ferritic stainless steel material because the solution-annealing temperature may degrade the ferritic steel mechanical properties. Paragraph UHA-33 provides radiographic examination requirements for specific, high alloy steel material. The requirements of paragraph UW-11 are generally applicable; however, welded joints in some alloys require full radiography. All butt welds in vessels made from Type 405 material that are welded with straight chrome electrodes shall be fully radiographed. Likewise, all butt welds made in Types 410, 429, and 430 material shall be fully radiographed in all thickness. These are the straight chromium grades of stainless steel that are susceptible to weld and HAZ cracking. When a ferritic stainless steel is used for welded construction, higher alloy material is often used for the weld filler metal to minimize potential cracking of the welds. For example, it is common practice to weld a Type 410S ferritic stainless steel with a high alloy nonferrous material filler material such as Inconel. For these alloys, when radiography is conducted, as required by UW-2, UW11, or UHA-33, it shall be done after any heat treatment. Paragraph UHA-34 requires that all austenitic chromium­nickel alloy steel and all duplex steel welds that exceed a 3/4-in. nominal thickness [as defined in UW-40(f)] must be examined by the liquid penetrant examination method. If heat treatment is conducted, this examination shall be done after heat treatment. 21.6.3.4 Toughness Requirements Paragraph UHA-51 provides requirements for impact testing of UHA material. Except for specific exemptions, all UHA materials require impact testing. The impact testing procedure requires a set of three Charpy V-Notch specimens to be taken from each of the base material, weld metal, and heat-affected zone. These specimens shall be subjected to same thermal treatments as the material in the finished vessel is expected to receive. (Only thermal treatments greater than 600°F need to be considered and thermal cutting or welding is not considered a thermal treatment.) Each specimen of each set is impact tested and the lateral expansion opposite the notch is measured and forms the basis of the acceptance criteria. When the MDMT of the vessel is ­320°F or warmer, the lateral expansion of all specimens must be 0.015 in. or more. Lateral expansion of the impact test specimens is used in lieu of absorbed energy because it is a better measure of the ductility of the austenitic material at low temperatures. Provisions are given in paragraph UHA-51(a)(3) for retesting when one of the three specimens fails to meet the required lateral expansion value. When impact testing is required for the base material or weld material, the Welding Procedure Qualification shall also have impact tests conducted on the weld material and the heat-affected zone in accordance with the above.

New toughness requirements were introduced in the 2005 Addenda when the MDMT is colder than 320°F ( 196°C). For this condition, the production welding processes shall be limited to shielded metal arc welding (SMAW), gas metal arc welding (GMAW), submerged arc welding (SAW), and gas tungsten arc welding (GTAW). When using Type 316L weld filler metal, notch toughness testing may be conducted using a test temperature of 320°F ( 196°C) so long as heat of each filler metal used in production has a Ferrite Number not greater than 5. When using filler metals other than Type 316L, notch toughness testing shall be conducted at a test temperature not warmer than the MDMT, using the ASTM E1820 JIC method. When UHA material is exposed to certain temperature ranges, embrittlement can occur. In order to assure that the material has sufficient ductility after potentially damaging thermal treatments, impact testing is required as defined in UHA-51(c) after the thermal treatment. When UHA material is thermally treated, impact tests are required at a temperature that is the colder of 70°F or the MDMT of the vessel as described below: (1) Austenitic stainless steel material thermally treated between 900 and 1650°F. Types 304, 304L, 316, and 316L material that are thermally treated between 900 and 1300°F do not require impact testing if the vessel MDMT is ­20°F or warmer and production impact test specimens are taken from the thermally treated weld metal. (2) Austenitic­ferritic duplex material thermally treated at temperatures between 600 and 1750°F. (3) Ferritic chromium stainless steels thermally treated at temperatures between 800 and 1350°F. (4) Martensitic chromium stainless steels thermally treated at temperatures between 800 and 1350°F. When impact testing is required because of thermal treatment, impact test exemptions given elsewhere are not applicable. Paragraph UHA-51(d) provides exemptions from impact testing of base material and heat-affected zones. If the nominal thickness is less than 0.099 in., impact testing is not required. Likewise, austenitic chromium­nickel stainless steel base material and HAZ do not require impact testing if for (1) Types 304, 304L, 316, 316L, 321, and 347, the MDMT is 320°F ( 196°C) or warmer. (2) For other chromium­nickel stainless steels having a carbon content not exceeding 0.10%, the MDMT is ­320°F ( 196°C) or warmer. (3) For chromium­nickel stainless steels having a carbon content greater than 0.10%, the MDMT is ­ 55°F ( 48°C) or warmer. (4) For castings, the MDMT is ­20°F ( 29°C) or warmer. Austenitic duplex, ferritic chromium, and martensitic chromium stainless steels do not require impact testing if the thickness does not exceed 3/8 in. (10 mm), 1/8 in. (3 mm), and 1/4 in. (6 mm), respectively, and the MDMT of the vessel is 20°F ( 29°C) or warmer. Paragraph UHA-51(e) provides exemptions for impact testing of the Welding Procedure Qualification (WPQ). For austenitic chromium­nickel stainless steel base materials with a carbon content not exceeding 0.10% and welded without the addition of filler metal, the WPQ does not require impact testing if the MDMT is ­155°F ( 104°C) or warmer. For austenitic weld metal, impact test qualification of the WPQ is not required if the MDMT is ­155°F ( 104°C) or warmer and the carbon content of the base materials

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 80

80 · Chapter 21

do not exceed 0.10%, and the filler metals are supplied in accordance with SFA-5.4, 5.9, 5.11, 5.14, or 5.22 (see Section II, Part C). If the base material has a carbon content greater than 0.10%, impact testing of the WPQ is not required if the filler metal is supplied in accordance with SFA-5.4, 5.9, 5.11, 5.14, or 5.22 and the MDMT is ­55°F ( 48°C) or warmer. The WPQ for duplex, ferritic chromium, and martensitic chromium stainless steels is exempt from impact testing if the vessel MDMT is ­20°F ( 29°C) or warmer and the base material is exempt from impact testing. A major revision to Paragraph UHA-51(f) concerning production impact test plates took place in the 2007 Edition. The focus of this paragraph has been changed to address preuse testing of welding consumables for MDMTs colder than 155°F ( 104°C). The requirements for vessel (production) impact tests that were originally published in UHA-51(f) have now been moved to UHA-51(h). Preuse testing of each heat /lot/batch of consumables is the only way to ensure acceptable toughness at cryogenic temperatures. In the previous rules a fabricator could opt to conduct a production impact test in lieu of preuse testing.

This would allow a fabricator to theoretically test only one heat of consumable while actually using numerous heats of consumables that may or may not exhibit acceptable lateral expansion. Unless ENiCr-FE-2, ENiCr-FE-3, ENiCr-Mo-3, ENiCr-Mo-4, ENiCrMo-6, ERNiCr-3, ERNiCr-Mo-3, ERNiCr-Mo-4, or E310-15/16 filler metal is used, preuse testing of the welding consumable will be required. Vessel production impact tests rules are provided in UHA51(h). Vessel production impact tests are required for welded construction of duplex stainless steels, ferritic stainless steels, and martensitic stainless steels if the Weld Procedure Qualification requires impact testing. Welding Procedure Qualification with impact testing is required for these materials for MDMT is colder than 20°F ( 29°C). For austenitic stainless steels with MDMT colder than 320°F ( 196°C), vessel (production) impact tests or ASTM E 1820 JIC shall be conducted. The impact test exemption rules for austenitic stainless steel material are difficult to follow by just reading the provision as stated in Part UHA. The flow charts given in Fig. 21.20 provide an easy way to understand the

Start

UHA-51(d)(1)(d) Is the material a casting? Yes Impact test if MDMT is colder than ­20°F (­29°C)

No

Is the material thermally treated as defined in UHA-51(c)?

Yes

See UHA-51(c) for exemptions

No

UHA-51(d) and (e)

Is the MDMT colder than ­55°F (­48°C)?

No

Impact testing is not required

Yes

Base Material and HAZ Requirements

Welding Procedure Qualification Requirements

Welding Consumable Pr-Use Testing Requirements

Production Impact Test Requirements

FIG. 21.20

ASME SECTION VIII DIV. 1 AUSTENITIC STAINLESS STEEL IMPACT TEST REQUIREMENTS (Source: Fig. JJ-1.2-1 thru 1.2-6 of Section VIII, Div. 1 2007 Edition of the ASME Code)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 81

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 81

Base Material and HAZ Requirements

UHA-51(d)

UHA-51(d)(1)(a) Is the base material 304, 304L, 316, 316L, 321, or 347 SS? No Is the base material carbon content < 0.10%

UHA-51(d)(1)(b) No

Yes

Yes

Is the MDMT is colder than ­320°F (­196°C)

Yes

Impact testing is not required

Yes

Impact testing of base material and HAZ is required

FIG. 21.20

(CONTINUED )

exact requirements applicable to austenitic stainless steel when used in cold temperature applications. Unless the material requires impact testing because of thermal treatment, vessels made from high alloy steel material do not require impact testing if the general primary membrane stress in the vessels does not exceed 0.35 times the material allowable stress value taken from Section II, Part D.

(2) 160 psi at temperatures not hotter than 375°F for vessels containing liquids (3) 250 psi for liquids at temperatures less than their boiling point at the design temperature, but not to exceed 120°F (4) 300 psi at temperatures not hotter than 450°F for bolted closures, heads, and covers that are not a major component of the pressure vessel Exceptions to these limitations are allowed for stress-relieved casting material. Materials made to Classes 40 through 60 of SA-278 and stress relieved may be used for design pressures up to 250 psi at temperatures up to 650°F, if it can be shown that the pressure-containing walls of the vessel have a uniform material distribution. Likewise, stress-relieved cast iron material in accordance with SA-476 may be used at pressures up to 250 psi at temperatures up to 450°F. Cast iron flanges and flanged fittings that are in conformance with ASME/ANSI B16.1 and Flanged Fittings, Classes 125 and 250, may be used at pressures that do not exceed the applicable pressure/temperature ratings up to a temperatures of 450°F. Cast iron is not allowed for bolting material. The design of cast iron components shall be in accordance with the applicable design requirements of Subsection A.

21.6.4

Part UCI: Requirements for Pressure Vessels Constructed of Cast Iron

Part UCI provides rules for pressure vessels and parts that are made from cast iron, cast nodular iron having an elongation of less than 15% in a 2-in.-gauge length, or of cast dual metal. The applicable requirements of Subsection A apply, as appropriate, for cast material. Cast iron vessels and parts shall not be used in lethal service for unfired steam boilers or for vessels that are subject to direct firing. All materials used for cast iron construction shall be in accordance with one of the classes listed in Table UCI-23. Paragraph UCI-3 provides the following pressure temperature limitations: (1) 160 psi at temperatures not hotter than 450°F for vessels containing gases, steam, or vapor

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 82

82 · Chapter 21

Welding Procedure Qualification Requirements

UHA-51(e)

UHA-51(e)(1)

Do all base materials joined have carbon content < 0.10%? Yes

No

UHA-51(e)(2)(b) Does filler metal have > 0.10% carbon and conform to SFA.5.4, 5.9,11, 5.14, or 5.22?

UHA-51(e)(2)(a) No Does filler metal have < 0.10% carbon and conform to SFA.5.4, 5.9,11, 5.14, or 5.22? Yes Yes No

Welded with filler metal?

Yes

No Yes UHA-51(e)(2)(b)

Yes

UHA-51(e)(2)(a) Is MDMT colder than ­155°F (­104°C)?

Is MDMT warmer than ­55°F (­48°C)? Yes

No

No Impact testing of welding procedures is not required

Impact testing of welding procedures is required

FIG. 21.20

(CONTINUED )

Table UCI-23 provides the allowable stress in tension to be used in the design of cast iron components. The allowable stress of cast iron material is based on a design margin of 10 applied to the specified minimum tensile strength of the material. For components subjected to primary bending stresses, such as a flat head or cover plate, the maximum allowable stress is 11/2-times that permitted for tension. In recognition of the fact that cast iron is stronger in compression than in tension, the maximum allowable stress in compression is two times that allowed for tension. However, shells under external pressure must meet the requirements of UG-28 using the appropriate figures in Section II, Part D. Openings in cast iron vessels shall meet the dimensional requirements of UG-36 through UG-46 for reinforcement and the reinforcement shall be integrally cast with the vessel or part. The thickness of the cast iron part in the vicinity of an opening, including the thickness of the vessel wall, shall not exceed twice the nominal thickness of the vessel wall. Because of the reduced ductility of cast iron, care must be taken to assure that generous fillet radii are used at changes in contour and wall thickness. High peak stresses in cast iron can result in fracture of the material. Accordingly, paragraph UCI-37 requires that all transitions between the pressure vessel wall and integral attachments have a fillet radius not less than one-half the vessel wall thickness at the location of the attachment. Because of the casting process, imperfections may exist that can result in leakage through the cast iron material. These

imperfections may be repaired by the use of threaded plugs, provided the requirements of paragraph UCI-78(a) are satisfied. Surface imperfections in the casting that do not result in leakage may be repaired by the use of driven plugs, provided the requirements of paragraph UCI-78(b) are satisfied. Weld repair of surface imperfections that do not cause leakage is allowed for cast iron vessel or parts containing liquids where the pressure is limited to 250 psi and the temperature does not exceed the liquid boiling point or 120°F. When a weld repair is made, the weld and the adjacent base material shall be examined by either the magnetic particle or liquid penetrant method. Cast iron vessels shall be hydrostatically tested as described in UG-99. The test pressure shall be determined as follows: (1) For cast iron vessels with a MAWP equal to or less than 30 psi, the test pressure is 21/2-times the MAWP but not to exceed 60 psi; (2) For cast iron vessels with a MAWP greater than 30 psi, the test pressure shall be twice the MAWP. Paragraph UCI-101 provides rules for proof testing of identical cast iron vessels or parts by hydrostatically testing one of them to destruction. The MAWP of the vessel is determined by dividing the burst pressure by 6.67 and multiplying by ratio of the specified minimum tensile strength divided by the actual tensile strength. This procedure assumes that the failure of the cast iron component is in bending. Since the allowable stress in bending is 11/2 times that for tension, the 6.67 factor results from the basic

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 83

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 83

FIG. 21.20

(CONTINUED )

design margin of 10 on tensile stress divided by 11/2 to account for the bending allowable stress.

21.6.5

Part UCL: Requirements for Welded Pressure Vessels Constructed of Material with Corrosion-Resistant Integral Cladding, Weld Metal Overlay, or with Applied Linings

When exposed to a very corrosive material, the inner surfaces of the vessel need to be corrosion resistant. One option is to construct the pressure vessel entirely from high alloy steel or nonferrous, corrosion-resistant material. However, it is often economically advantageous to make the main pressure-resisting components of the pressure vessel from less expensive carbon or low alloy steel with an inner layer of high alloy material to provide a protective boundary between carbon or low alloy steel and the process fluid. The rules of Part UCL apply to pressure vessel or vessel parts that

are constructed using corrosion-resistant linings that are integral or welded to the vessel wall. These rules are used in conjunction with the requirements given in Subsections A and B of Section VIII, Division 1. Because clad or overlay material is used for corrosion resistance or other service-related requirements, the Code does not provide specific recommendations related to chemical compositions, heat treatment condition, unique fabrication requirements, or any additional testing that may be necessary for a specific application. The user must assure himself that the selected alloy will be suitable for the intended service considering the fabrication methods, examination, and heat treatments that the material will experience. There are three basic types of corrosion-resistant linings. The "roll bonded" or "explosively bonded" techniques produce integral cladding where the corrosion-resistant material is integrally attached to the vessel base material. The weld overlay method uses deposited high alloy weld metal as the corrosion barrier to the base

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 84

84 · Chapter 21

FIG. 21.20

(CONTINUED )

material. The applied lining method uses strips or plates of corrosion-resistant material that are attached to the vessel wall by welding. If the cladding thickness is not considered in the design calculations, then the integral cladding, weld overlay, or lining may be any material of weldable quality that the user deems suitable for the intended service. Applied (loose) linings may not be considered as contributing to the required pressure thickness in the design calculations. If the cladding or overlay thickness is not included in the design thickness, no specific design provisions of Part UCL are applicable. When cladding is included in the design calculations, it may be done in one of the two ways. The first option does not include the full cladding thickness in the design calculations. The design may be based on the actual clad thickness less the specified minimum thickness. A reasonable excess thickness of actual cladding thickness may be included in the design calculations as if the excess cladding material were base material. The allowable stress considered in the design is that of the base material. The second method is to include the full thickness of the cladding in the design calculations. When the cladding thickness is considered to be part of the design thickness, the clad material, if roll bonded or explosion bonded, must be in accordance with one of the following specifications: (1) SA-263­Corrosion-Resisting Chromium­Steel Clad Plate, Sheet, and Strip (2) SA-264­Corrosion-Resisting Chromium­Nickel Steel Clad Plate, Sheet, and Strip (3) SA-265 ­ Nickel and Nickel-Base Alloy Clad Steel Plate

Also, the cladding must be subjected to the bond shear test provisions given in the above specifications, and must show a minimum shear strength of 20,000 psi. The shear or bond strength test is not required for weld overlay cladding. The design thickness may based on a thickness equal to t = tb + where: t tb tc Sc Sc t Sb c (21.13)

Sb

design thickness nominal thickness of base material nominal thickness of cladding (in the corroded condition) maximum allowable stress of the integral cladding material at the design temperature or, for weld overlay, the allowable stress at the design temperature of the wrought material whose chemistry most closely approximates that of the cladding maximum allowable stress of the base material at the design temperature

If Sc is greater than Sb, then Sc /Sb shall be taken as 1.0. A clad or overlay vessel that is built in accordance with Table 21.1, Column (c) (no radiographic examination), shall not have any cladding included in the design thickness of the vessel. When weld overlay is deposited and its thickness is included in the design, the weld metal overlay thickness must be verified by electrical or mechanical

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 85

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 85

FIG. 21.20

(CONTINUED )

means for each component that has weld overlay applied. The Inspector will determine the location for the weld overlay thickness inspection. When integral clad or applied linings are used, welds are required at the base material welds and/or between the applied lining and the base material. Care must be taken to assure that the type of joint and welding procedure do not produce brittle welds resulting from the mixture of the corrosion-resistant material and the base material. Typically, clad plate has the cladding stripped away from the base material weld seams at least 1/4 in. from the edge of the weld preparation. The base material seam is then welded and the corrosion-resistant weld material is deposited over the base material weld seam and connects with the adjacent cladding material. If the cladding material is not stripped back from the seam, it is likely the base material weld will be contaminated with high alloy material and become brittle. Clad, weld overlaid, and lined vessels are required to be postweld heat treated when the vessel base material is required to be heat treated by the requirements of Subsection C. The clad

thickness shall not be considered when determining the thickness to be used for the PWHT requirements. Vessels that have chromium stainless steel cladding, weld overlay, or applied linings shall be postweld heat treated because of the hardenability properties of the welds. Exceptions are allowed for Type 405 or Type 410S material that is welded with an austenitic or nonairhardening nickel­chromium­iron material. Paragraph UCL-35 defines the radiographic examination considerations for clad or lined vessels. When the base material seams are covered by alloy weld overlay, the radiographic examinations required elsewhere in Division 1 shall be made after the weld, including the weld overlay, is completed. However, if the thickness of the base material at the welded joint is not thinner than the required thickness, the corrosion- resistant alloy is nonair-hardening, and the completed weld overlay is spot examined for cracks, the base material weld may be radiographed prior to the deposition of the weld overlay. Because of their crack sensitivity, chromium stainless steel filler metal used for weld overlay of main joint welds must be fully examined. When the weld is covering the base

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 86

86 · Chapter 21

material welds, radiographic examination is required. If chromium cladding or liners are welded with a chromium­nickel steel or nonair-hardening nickel­chromium­iron filler material, then spot radiography is required. Applied linings may be attached to the base material by any method and welding process permitted by Division 1. The method and procedures to be used to make the attachment and the methods to determine the soundness of the attachment shall be by contractual agreement between the user and the Manufacturer. It is cautioned that application of applied, loose linings must be carefully considered. Usually, the fluid inside the vessel is very corrosive to the base material and failure of the lining can result in its accelerated corrosion. Unless an applied lining is installed by an experienced fabricator, fit-up and welding are very difficult to accomplish without leaving gaps between the lining and base material. Upon the application of pressure, the linings may flex that can result in cracking of the attachment welds. If the attachment welds crack, in addition to the corrosion considerations, it is possible that pressure can build up behind the lining. When this happens and the vessel is rapidly depressurized, the lining can be blown off the wall of the vessel. There have been numerous instances where catastrophic failures of applied linings have occurred. Certainly, in critical applications, the use of an integral lining or weld overlay is preferred. There are other general considerations regarding clad and lined vessels. The attachment of supports for internal components requires proper consideration. If the support will transmit a significant load into the vessel wall, then it is important to assure that the strength of the cladding or weld overlay bond is sufficient to resist the applied load. For clad plate that has been bond shear tested, sufficient strength may exist or the cladding may be stripped so that the support can be welded directly to the base material. If the cladding is not stripped back, it may be desired to ultrasonically examine the clad area for bond in the vicinity where the attachment will be made. Also, if an integrally clad vessel is in high temperature, high-pressure hydrogen service, it is possible that hydrogen can diffuse into any lack-of-bond area in the clad plate. This has the potential for the creation of bulged areas in the cladding as a result of a pressure buildup of the hydrogen. Careful consideration is also required for a vessel with an applied lining that will experience vacuum conditions because the lining may become detached. The user should consider if any operating conditions could occur that would result in vacuum in the vessel. If such loadings are possible, the designee should consider the effects of vacuum on the lining system. Often, there is an economic trade-off between using the solid corrosion-resistant alloy material and clad material. Obviously, the thicker the vessel, the more economically attractive clad or weld overlay becomes. However, as the thickness of the vessel decreases, there is a break point where the clad vessel will cost as much as or more than the vessel made of solid alloy material. As a rule of thumb, when the wall thickness is thinner than about 3/4 in., a clad vessel will cost more than a solid stainless steel vessel (other alloys may have a different break point because of the relative value of the corrosion resistant alloy). As the vessel wall thickness decreases, cladding and weld overlay become progressively more difficult to fabricate because of the weld distortion associated with high alloy weld overlay required at the base material welds seams.

21.6.6

Part UCD: Requirements for Pressure Vessels Constructed from Cast Ductile Iron

Part UCD is applicable for vessels and vessel parts constructed of cast ductile iron. Cast ductile iron has greater ductility than

cast iron and many of the limitations that are applicable to cast iron may be relaxed for cast ductile iron. If the cast ductile iron has an elongation of less than 15% in a 2-in. gauge length, the rules of Part UCI shall be applied. The applicable requirements of Subsection A apply, as appropriate, for cast material. Cast ductile iron vessels and parts shall not be used in lethal service, for unfired steam boilers, or for vessels that are subject to direct firing. All materials used for cast ductile iron construction shall be in accordance with one of the classes listed in Table UCD-23. Paragraph UCD-3 limits the maximum design temperature for UCD material to a temperature of 650°F and the pressure is limited to 1000 psi. If the casting is to contain liquid only and is examined in accordance with UG-24 such that a casting quality factor of 90% may be used, the above pressure limit may be exceeded. Cast ductile iron flanges and fittings covered by ASME/ANSI B16.42 may be used at the pressure­temperature ratings listed therein. Cast ductile iron flanges and fittings, meeting the dimensional requirements of ASME/ANSI B16.5, may be used at the pressure­temperature ratings for carbon steel, material category 1.4; however, the design metal temperature cannot be colder than ­20°F nor warmer than 650°F and the pressure is limited to 1000 psi. Acceptable cast ductile iron materials for use in pressure vessels and parts are given in Table UCD-23. Cast ductile iron may not be used for bolting. The maximum allowable stress in tension for cast ductile iron is given in Table UCD-23. The allowable stress is based on a design margin of five applied to the specified minimum tensile strength of the material. For components subjected to primary bending stresses, such as a flat head or cover plate, the maximum allowable stress is 11/2 times that permitted for tension. The allowable stress for compression is the same as that for tension. In addition, the allowable stress given in Table UCD-23 shall be multiplied by the applicable casting factor as determined in paragraph UG-24. Openings in cast ductile iron vessels shall meet the dimensional requirements of UG-36 through UG-46 for reinforcement and the reinforcement shall be integrally cast with the vessel or part. The thickness of the cast ductile iron part in the vicinity of an opening, including the thickness of the vessel wall, shall not exceed twice the nominal thickness of the vessel wall. Because of the reduced ductility of cast ductile iron compared to wrought material, care must be taken to assure that generous fillet radii are used at changes in contour and wall thickness. High peak stresses in cast ductile iron can result in fracture of the material. Accordingly, paragraph UCD-37 requires that all transitions between the pressure vessel wall and integral attachments must have a fillet radius not less than one-half the vessel wall thickness at the location of the attachment. Likewise, transitions between pressure parts of different contours shall have a radius not less than three times the thickness of the thinner part. Because of the casting process, imperfections may exist that result in leakage through the cast ductile iron material. These imperfections may be repaired by the use of threaded plugs, provided the requirements of paragraph UCD-78(a) are satisfied. Surface imperfections in the casting that do not result in leakage may be repaired by the use of driven plugs, provided the requirements of paragraph UCD-78(b) are satisfied. No provisions are provided for weld repair of imperfections in cast ductile iron. Cast ductile iron vessels shall be hydrostatically tested at a pressure that is twice the MAWP. Paragraph UCD-101 provides rules for proof testing of identical cast ductile iron vessels or parts by hydrostatically testing one of them to destruction. The MAWP of the tested vessel is determined by dividing the burst pressure by 5 and multiplying by ratio of the specified minimum tensile

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 87

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 87

strength divided by the actual tensile strength. The MAWP of the tested vessel must be multiplied by the appropriate casting quality factor as defined in UG-24 in order to determine the MAWP of the vessels that are to be put into service.

21.6.7

Part UHT: Requirements for Pressure Vessels Constructed of Ferritic Steels with Tensile Properties Enhanced by Heat Treatment

The scope of Part UHT applies to vessels constructed from ferritic steel material that is suitable for welding, but has had its tensile properties enhanced by heat treatment. This part is not intended to apply to those UCS materials that may be furnished in a heat-treated condition (such as accelerated cooling or quenched and tempered) that may be required to satisfy the mechanical properties of the material specification. For example, very thick sections of carbon and low alloy steel material, as allowed by Part UCS, may require quenching or accelerated cooling from the austenizing temperature (with a subsequent tempering heat treatment) in order to satisfy the yield and tensile strength requirements of the material specification. Part UHT does not apply to these materials. Likewise, Part UHT does not apply to quenched and tempered integrally forged vessels that do not contain welded seams (see Part UF). Steels that are covered by Part UHT are given in Table UHT-23. These are high strength steels that have increased yield and tensile strength as a result of quenching and tempering heat treatment. They generally have a very high yield to tensile strength ratio. Because of the pronounced effect that the material's thickness has on the heat treatment operation, it is not allowed to exceed the thickness limitations of the material specification. All materials listed in Table UHT-23 require Charpy V-notch impact testing in accordance with paragraph UHT-6. The impact testing is to be conducted at the minimum design metal temperature or 32°F, whichever, is colder. As allowed for Part UCS material in paragraph UCS-160, Part UHT materials may be used at temperatures colder than the minimum design metal temperature. If the coincident ratio of the applied stress to the allowable stress is 0.35 or less, the material may be used at a reduced temperature not colder than 155°F. When the coincident ratio is greater than 0.35, the procedure of Fig. 21.19 may be used to determine the allowed temperature reduction from the MDMT. At the coincident pressure, the metal temperature may be no colder than the impact test temperature minus the allowed temperature reduction taken from Fig. 21.19. All impact test specimens shall be prepared from material that is representative of the final heat-treated condition of the vessel. The test procedure of UG-84 shall be used, and paragraph UHT-6 provides requirements for impact testing. The impact testing procedure requires a set of three Charpy V-Notch specimens to be taken from each of the base material, weld metal, and heat-affected zone. Each specimen of each set is impact tested and the lateral expansion opposite the notch is measured and forms the basis of the acceptance criteria as given in Fig. UHT-6.1. For material thickness less than 11/4-in., the lateral expansion of all specimens must be 0.015 in. or more. When the material thickness is 3 in. and greater, the lateral expansion of each specimen shall not be less than 0.025 in. Between 11/4 and 3 in., linear interpolation is done. Provisions are given in paragraphs UHT-6(a)(4) for retesting when one of the three specimens fails to meet the required lateral expansion value. When SA508, SA-517, SA-543, and SA-592 are to be used for vessels with a MDMT colder than 20°F, and SA-645 is to be used with a MDMT colder than 275°F, drop weight tests in accordance with ASTM E208 shall be performed. The drop weight specimens shall meet the "no-break" criterion at the test temperature. This

requirement assures that the nil-ductility transition temperature of the steel is below the minimum design metal temperature. Part UHT places severe limitations on the type of construction that is allowed for these high strength materials. Except for specific material as defined in UHT-17(b) and shown below, all Categories A, B, and C joints must be full penetration, double butt welds (Type 1 from Table 21.1). This requirement also applies other pressure boundary welds, such as the joint attaching the side plates of noncircular vessels that do not have a joint designation defined by paragraph UW-3. Category D joints (nozzle-to-shell welds), must be Type 1 butt joints when the vessel wall thickness is 2 in. or less. When the vessel wall thickness is greater than 2 in., full penetration corner joints are allowed. In accordance with UHT-17(b), vessels made of SA-333 Grade 8, SA-334 Grade 8, SA-353, SA-522, SA-553, and SA-645 may have welded joints as defined below: (1) Category A joints shall be Type 1 butt welds. (2) Category B joints shall be Type 1 or Type 2 butt welds. (3) Category C joints shall be full penetration welds through the entire thickness of the joint. (4) Category D joints shall be full penetration welds through the joint, reinforcing pads are allowed. Paragraph UHT-20 specifies joint alignment tolerances that supersede those of paragraph UW-35. Generally, nozzles and other attachments to vessels made from UHT material shall be made from material that has a specified minimum yield strength within 20% of that of the vessel shell to which they are attached. Refer to paragraphs UHT-18 and UHT28 for exemptions to this rule. Conical sections in vessels made from UHT material are required to have a knuckle at each end. Also, the knuckle must have a straight skirt with a length not less than 0.5 1rt (where r is the inside radius of the adjacent cylinder and t is the cone thickness). This requirement assures that high localized stresses will not exist at the cone-to-cylinder junction. Postweld heat treatment requirements for UHT material is given in paragraph UHT-56 and Table UHT-56. The PWHT is to be done in accordance with UCS-56 as modified by Table UHT56. The PWHT temperature shall not exceed the tempering temperature. For material that has been quenched and tempered, subsequent heating to a temperature that is greater than the tempering temperature will result in a degradation of the material's mechanical properties. It is acceptable to perform the tempering heat treatment and the PWHT at the same time. When this is the case and accelerated cooling from the tempering temperature is required by the material specification, the PWHT procedure must incorporate the required cooling rate. Paragraph UHT-57 defines the examination requirements for vessels made from UHT material. All Type 1 butt welds require 100% radiography. In addition, the nozzle-to-shell attachment welds require full examination. All butt-welded Category D joints must be fully radiographed and full penetration corner joints attaching nozzles having an inside diameter greater than 2 in. must be fully radiographed. Category D joints made with full penetration corner welds that attach nozzles with an inside diameter of 2 in. or less must be examined by the liquid penetrant or magnetic particle method. All welds made on vessels made from UHT material must be examined using the magnetic particle method. This examination must be done after the hydrostatic test. Weld surfaces not accessible after the hydrostatic test shall be examined at the last stage of fabrication where they are accessible. Care must be taken when conducting the magnetic particle examination

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 88

88 · Chapter 21

to not to introduce arc strikes on the material. Arc strikes can be very detrimental to Part UHT material because they may result in a site for crack initiation. Liquid penetrant examination must be used for nonmagnetic material and may be substituted for the magnetic particle method for magnetic materials. Since the material properties used in the design of a vessel constructed in accordance with Part UHT depend on the heat treatment of the material, it is necessary to assure that the strengthening heat treatments have been properly conducted on the actual vessel that is constructed. Paragraph UHT-81 defines the requirements for heat treatment verification tests. Test coupons from each lot of material and welds shall be quenched with the vessel or vessel component and the coupons must be included in the subsequent tempering heat treatments done on the vessel or part. One tensile test and one impact test is required for each lot of material used in the vessel. The test coupons must satisfy the tensile strength requirements of the material specification and the impact toughness requirements of Part UHT. Special precautions are needed for the fabrication of Part UHT material. The high tensile strength of UHT material makes these materials susceptible to hydrogen embrittlement and welding must be done with filler material that has no moisture. This is accomplished by the use of low-hydrogen electrodes, storing the welding material in drying ovens, and maintenance of preheat of the parts prior to welding. The specific requirements are given in paragraph UHT-82. When material is removed by chamfering or beveling the plate edges in preparation for welding, it shall be removed in a manner that is not detrimental to the material. The preferred method of metal removal is to use machining, grinding, chipping, and other methods that do not involve thermal cutting or melting of the material. When thermal cutting methods are used, and subsequent welding does not eliminate the cut surface, the heat-affected zone shall be removed by grinding or machining followed by magnetic particle or liquid penetrant examination.

the inner shell or head must be designed to resist the vacuum pressure without any consideration of the layers and wraps. For axial compression, the total thickness of the layered shell may be used. The inner shell may be made of a material that is weaker than the layer materials. In order to include the thickness of the inner shell in the required pressure-resisting thickness, the allowable stress of the inner shell or head must be at least 50% of the allowable stress of the material used for the layers. The effective thickness of the inner shell or head is teff = tact where: teff tact S1 SL S1 SL (21.14)

effective thickness of the inner shell or head nominal thickness of inner shell or head design stress of inner shell or head design stress of layers

21.6.8

Part ULW: Requirements for Pressure Vessels Fabricated by Layered Construction

A layered vessel is defined in Appendix 3 as " . . . a vessel having a shell and/or heads made up of two or more separate layers." The most common types of layered vessels are shown in Fig. 21.21. Four types of layered construction are covered by Part ULW; they are concentric, wrapped, coil wound, shrink fit, and spiral wrapped. These types of layered construction represent the types of construction where there is extensive, documented construction and operating experience. Layered vessels are normally used for high-pressure applications where layered construction of thick shells is more economical than solid wall or mono-block construction. The rules of Part ULW shall be used in conjunction with the applicable rules of Subsections A, B, and C. When a layered vessel is used in lethal service, the requirements of UW-2(a) apply only to the inner shell and inner head. The inner shell or head is the inner cylinder or head that forms the pressure tight membrane. The material used for the pressure parts of layered vessels shall conform to one of the specifications permitted in the applicable parts of Subsections A, B, and C; however, 5%, 8%, and 9% nickel steels are permitted to be used only for the inner shells and heads. Generally, the rules of Part UG are used for the design of layered vessels except as modified by paragraph ULW-16. The reinforcement of openings in layered vessels shall be in accordance with Fig. ULW-18.1. When a layered vessel is in vacuum service,

All Category A weld joints for the inner shell or head shall be Type 1 and all Category B joints must be Type 1 or Type 2 of Table UW-12 (see Table 21.1 of this chapter). For the layered sections, Category A joints in layers greater than 7/8 in. wall thickness shall be Type 1. Category A joints in layered sections with a thickness of 7/8 in. and less shall be Type 1 or Type 2. However, the outside final weld seam of spiral-wrapped layered vessels may be a single lap weld. Category B welds of layered sections to layered shell sections or layered sections to solid sections shall be Type 1 or Type 2 of Table UW-12 (see Table 21.1 of this chapter). The transition between layered sections to layered sections or layered sections to solid sections shall be as shown in the applicable sketch of ULW-17.1 or ULW-17.2. Attachment details for joining layered sections to flat heads and/or tubesheets are shown in Fig. ULW-17.3. Some acceptable flanges for layered vessels are given in Fig. ULW-17.4. When the nondestructive examination requirements of Part ULW have been satisfied, the weld joint efficiency for layered vessels shall be 100%. All openings greater than NPS 2 shall meet the reinforcement requirements of Part UG. Openings that are NPS 2 and smaller do not require added reinforcement if they are welded to the inner shell. The nozzle thickness of unreinforced openings shall not be thinner than that required by UG-45 or Schedule 80, whichever is greater. Nozzle reinforcement may be provided by excess metal in the nozzle neck, excess thickness of the layered section, or by the addition of full circumferential reinforcement layers installed over the layers required for pressure loading. Acceptable methods or reinforcement for layered pressure vessels is given in Fig. ULW-18.1. Special precautions are warranted when welding attachments to layered pressure vessels. Unless provisions are made to transfer the load to other layers, only the thickness of the layer to which any support or attachment is made shall be considered when calculating the stress in the layer near the attachment. When required, layered pressure vessels shall be postweld heat treated in accordance with the provisions of the applicable sections of Subsections A, B, and C. If the vessel is in lethal service, the postweld heat treatment requirements of UW-2(a) are applicable only to the inner shell and heads. However, welds attaching a layered section to a layered section do not require postweld heat treatment if the thickness of the individual layer does not exceed the thickness that requires heat treatment as defined in Table UCS-56 or UHT-56. Thus, it is the thickness of each layer that determines the need for PWHT and not the total thickness of the

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 89

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 89

FIG. 21.21

SOME ACCEPTABLE LAYERED SHELL TYPES (Source: Fig. ULW-2.1 of Section VIII Div. 1 of the ASME Code)

layered section. When welding a layered section to a solid section where the solid section requires heat treatment by Table UCS-56 or UHT-56, heat treatment is not required if (1) the solid section has a weld overlay buttering, using a material that does not require heat treatment, deposited on the solid piece; (The overlay buttering is not required for P-No. 1 material.) (2) PWHT is conducted on the solid piece after buttering;

(3) multipass welding is used and the thickness of any pass does not exceed 3/8-in.; (4) the thickness of each layer is less than that defined in Table UCS-56 or UHT-56 where PWHT is required. Nondestructive examinations of layered vessels require special considerations. First, Categories A and B joints in the inner shell and head of layered vessels must be fully radiographed. Category A joints in layers 1/8-in.to 5/16-in.thickness that are welded to

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 90

90 · Chapter 21

the previous layer require 100% of their length to be examined using the direct current magnetic particle method. Category A joints in layers with a thickness greater than 5/16 in. up to 5/8 in. that are welded to the previous layer shall be 100% examined by the direct current magnetic particle method and 10% of their length, at random, inspected by ultrasonic examination. Category A joints in layers with a thickness greater than 5/8 in. up to 7/8 in. that are welded to the previous layer, shall be 100% examined by ultrasonic examination. Category A joints in layers that are not welded to the previous surface (e.g., layers for shrink fit construction) shall be fully radiographed. Step-welded Category B joints in layers 1/8 in. through 5/16 in. thickness shall have 10% of their length examined using the direct current magnetic particle method. Step-welded Category B joints in layers over 5/16 in. through 7/8 in. shall be 100% examined by the direct current magnetic particle method. Step-welded Category B joints in layers over 5/8 in. through 7/8 in. thickness shall be 100% examined using the direct current magnetic particle method and shall have 10% of their length, taken at random, ultrasonically examined. Step-welded Category B joints in layers over 7/8 in. thickness shall be 100% ultrasonically examined. All Categories A, B, and D butt joints that are made between a layered and solid section shall be fully radiographed. Because of a phenomenon known as layer wash, special angle radiographic techniques may be required in order to distinguish defects from gaps in the layers and slight weld penetration at the layer interface. Categories A and B butt joints that attach layered sections to layered sections do not require radiography when the inner shell or head has been fully radiographed. Step-welded Category C welds made between a layered section and a solid flat head or tubesheet shall be examined using the same rules that apply to stepwelded Category B joints. Butt-welded Category C welds shall be fully radiographed. Nonbutt-welded Category D joints shall be examined by the magnetic particle method. Vent holes shall be provided for all layered sections of a layered pressure vessel. The purpose of the vent holes is to detect leakage of the inner shell or head and to prevent a buildup of pressure between the layers. Some layered vessels have the inner shell/head made from a corrosion-resistant material, and an undetected failure of the inner shell can result in rapid deterioration of the noncorrosion-resistant layers. It is most important that the vent holes must be left unobstructed. Some installations use a monitoring system to detect leakage through the inner shell. These systems shall not allow a buildup of material or pressure within the layers. There have been documented failures of layered pressure vessels where the vent system has not been operational and leakage of the inner shell resulted in accelerated corrosion of the layers. The installation and operation of any layered pressure vessel must be done in strict accordance with the manufacturer's recommendations. Failure to properly maintain the vent system of a layered vessel can result in a catastrophic failure of the vessel. One measure of the effectiveness of the layers of a layered vessel to resist the internal pressure as an integral shell is the amount of gap between individual layers. The larger the gap between adjacent layers, the less effective each individual layer is for resisting the applied pressure with a uniform load distribution between the layers. In order to control the manufactured quality of layered sections, paragraph ULW-77 requires that radial gaps between adjacent layers be limited. The radial gap is measured at the ends of layered sections. A gap is relevant if it exceeds 0.010 in. The estimated area of the gap, expressed in square inches, shall not be greater than the thickness of the layer. The maximum length of any

relevant gap is the inside diameter of the vessel. As an alternative to measuring the gaps between adjacent layers, paragraph ULW-78 allows for the measurement of the dilation of the layered section during the hydrostatic test. The measured dilation of the layered section will be reduced if adjacent layers have excessive gaps or if the layers are not in intimate contact with one another. The circumferential expansion at the design pressure must be at least half of that of calculated for a solid vessel of the same dimensions and material. If the measured dilation is less, then the vessel is to be rejected.

21.6.9

Part ULT: Alternative Rules for Pressure Vessels Constructed of Materials Having Higher Allowable Stresses at Low Temperature

The provisions of Part ULT permit increased allowable stresses for vessels that operate at cryogenic temperatures (in contact with a cryogenic fluid as cold as ­320 F) in recognition that the material's tensile strength increases substantially at these temperatures. The rules of Part ULT supplement the general requirements of Subsection A and Part UW of Subsection B. The requirements of Subsection C, Requirements for Classes of Materials, do not apply unless specifically referenced in Part ULT. The vessel and system design must be such that the pressure does not result in a stress greater than the maximum allowable stress at the coincident temperature. This means that the characteristics of all fluids to be stored in the vessel must be known. The operating temperature of a vessel storing a cryogenic fluid is the liquid saturation temperature at the MAWP of the vessel. If a vessel is to store more than one cryogenic fluid, then the properties of all the liquids must be considered in order to establish the vessel-operating temperature used to establish increased allowable stress. Since the operating temperature is set at the saturation temperature at the MAWP of the warmest fluid to be contained, the pressure relief device provides a limitation of the warmest temperature that the vessel in contact with the cryogenic material may experience. If the pressure relief device is set at a pressure less than the MAWP, then the cryogenic liquid will not be able to cool to its saturation temperature; thus, any increase in material strength cannot be realized. Use of Part ULT requires that the external parts of the vessel be insulated. Table ULT-23 provides a listing of the materials where higher allowable stresses have been established for vessels operating at cold temperatures. These materials include 5%, 8%, 9% nickel alloy steel, Type 304 stainless steel, and 5083 aluminum alloys. The increased allowable stress given in Table ULT-23 may only be used for those components that will be in contact with the cold cryogenic liquid. Components that are not in contact with the cryogenic liquid may be made of any material allowed by other sections of Division 1 and the design of such components shall be based on the allowable stress at 100°F for the material. The 5%, 8%, 9% nickel alloy steels shall be tested for notch ductility as required in Part UHT-5(d) at the coldest temperature to be marked on the vessel. For aluminum alloy material, the provisions of paragraph UNF-65 that allow wrought aluminum alloys to be used at temperatures as cold as ­425°F without impact testing are applicable. For vessels that are made from Type 304 stainless steel, the welds are required to be impact tested, and the weld metal impact test exemptions provided in UHA shall not be used. When the weld material is different from the base material, the thermal stresses resulting from the difference in coefficients

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 91

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 91

of thermal expansion shall be considered in the design. This requirement is given in paragraph ULT-16; however, no criterion is given for the thermal stresses. Thermal stresses are secondary stresses that are self-limiting. As such, any thermal stresses resulting from differential thermal expansion resulting from low temperature operation should be considered the same as thermal stresses resulting from elevated temperature. Likewise, paragraph ULT-16 sets limits on the thickness of vessels made from 5%, 8%, and 9% nickel steel; the minimum thickness is 3/16 in. and the maximum thickness of the base material at welds is 2 in. When the provisions of Part ULT are used, all pressure boundary welds are required to be full penetration. Butt welds with one edge offset [see Fig. 21.9(k)] are not allowed, and the longitudinal welds in adjacent shell sections must be offset by at least five times the thickness of the thicker section. The requirement for full penetration welds minimizes the highly localized stresses (especially at Categories C and D joints) that can have a deleterious effect on the vessel's ability to resist brittle fracture. The required offset of longitudinal joints implements a "crack arrest" feature in order to preclude a "zipper" effect where a running failure cannot easily propagate from one shell section to the adjacent section along the weld seam. Paragraph ULT-57 requires that all butt joints be 100% examined by radiography [ultrasonic examination as allowed by paragraph UW-11(a)(7) is permitted]. All attachment welds and any pressure boundary welds not radiographed shall be examined by the liquid penetrant examination method. The design of vessels and vessel parts for internal pressure is given by paragraph ULT-27. The allowable stress to be used in the design of parts made from materials covered by Part ULT is given in Table ULT-23. Allowable stress is tabulated for temperatures from 100 to 320°F, and interpolation between temperatures is allowed. For 5%, 8%, and 9% nickel steels, allowable stresses are tabulated for welded and nonwelded construction. The allowable stress values for welded construction only apply for butt-welded construction. For example, the "Nonwelded Construction" allowable stress applies for the design of a seamless formed head even if it has a Category D, full penetration, nozzle attachment weld. The required thickness of the vessel at any location is the largest of that required for the conditions described below: (1) The MAWP plus any other applicable loads (see UG-22), including the hydrostatic head of the most dense cryogenic liquid that the vessel is expected to contain. The allowable stress value is taken from Table ULT-23 at the operating temperature that corresponds to the warmest saturation temperature (at the MAWP) of the cryogenic fluids the vessel is expected to contain. The allowable compressive stress is determined as described in UG-23 at 100°F. (2) The MAWP plus any other applicable loads (see UG-22), but not including any hydrostatic head of liquid. The allowable stress value is taken from Table ULT-23 at 100°F. This design method is consistent with the scope of Part ULT in that the increased allowable stress for cold temperatures may only be applied to the design of vessel parts that are exposed to the hydrostatic head of the cryogenic fluid. When Part ULT is used, special provisions are required for the welding procedure qualifications. Tensile tests must be conducted at room temperature and the temperature that is used to determine the allowable stress should be taken from Table ULT-23. The tensile strength for all specimens must meet the minimum

requirements as defined by Table ULT-23 for the applicable material. For Type 304 stainless steel, a limitation on the delta ferrite is imposed on each lot of filler material used in the vessel. Delta ferrite is a normal component found in stainless steel microstructure. High levels of delta ferrite may result in a loss of notch toughness of stainless steel, and low levels of delta ferrite may result in hot cracking during welding. Using Fig. ULT-82, the delta ferrite in each lot of filler metal shall not be greater than 14 FN and shall not be less than 6 FN. Special provisions for pressure testing are imposed for ULT vessels by paragraph ULT-99. The test pressure at every location should be at least 1.6 times the MAWP plus the hydrostatic head of the test liquid based on the operating position of the vessel. If the general membrane stress of any component in the vessel exceeds either 95% of the specified minimum yield strength or 50% of the specified minimum tensile strength, then the pressure test may be limited to a pressure where these limits are not exceeded. However, if these limitations result in a pressure less than 125% of the MAWP (using the allowable stresses at 100°F) in a pressure test, then a pneumatic test must also be conducted in accordance with UG-100 except the temperature correction is not required. When Part ULT is used, the Manufacturer's Data Report and vessel marking must comply with paragraph ULT-115. The vessel marking and data forms must define each liquid and its saturation temperature at the vessel MAWP that is to be contained in the pressure vessel.

21.6.10

Part UHX: Rules for Shell and Tube Heat Exchangers

Part UHX provides design rules for shell and tube heat exchangers. Primarily, the rules encompass the tubesheet design. These rules are somewhat different than the guidelines provided by Tubular Exchanger Manufacturer's Association [25]. The rules in Part UHX are now mandatory; thus, a designer does not have the option of using other rules (such as those in TEMA) when the exchanger is within the scope of Part UHX. Part UHX provides rules for the design of various types of U-tube, floating head, and fixed tubesheet heat exchangers. With publication of Part UHX as mandatory rules, numerous inquiries on the scope/applicability of Part UHX have been received by the Committee. Paragraph UHX-10 describes the general conditions of applicability for designing tubesheets as per Part UHX. Tubesheets satisfying the conditions described below fall within the scope of Part UHX: · The tubesheet shall be flat and circular and of uniform thickness, except that the flange extension may differ in thickness as determined by UHX-9. · The tubesheet shall be uniformly perforated over a notably circular area, and either equilateral, triangular, or square patterns. Untubed lanes for pass partitions are permitted. · The channel component integral with the tubesheet shall be either a cylinder or a hemispherical head. · The tube side and shell side pressures are assumed to be uniform. The rules do not cover weight loadings or pressure drop. Tubesheets for which these conditions of applicability are not satisfied shall be designed as per the provisions of U-2(g). The reader may want to review several published Interpretations on Part UHX scope/applicability: VIII-1-04-52, VIII-1-04-53, VIII1-04-54, VIII-1-04-61, VIII-1-04-71, and VIII-1-04-81.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 92

92 · Chapter 21

For the tubesheets of U-tube heat exchangers, the rules of Part UHX are applicable for flat, fully tubed, circular tubesheets for any of the edge conditions given below: (1) Tubesheet is simply supported (clamped between two flanges). (2) Tubesheet has a channel and shell welded such that they are integral with the tubesheet. (3) Tubesheet has either the shell or channel welded to be integral and the nonintegral side is gasketed with the tubesheet extended to act as a flange. For each geometry, the derivation of the analysis includes the effects of the applied loads (shell and tube side pressure and bolting load as applicable), the stiffening effect of the unperforated ring at the tubesheet edge, and the stiffening effect of the integrally attached channel or shell. The analysis was based on classical discontinuity analysis methods to determine the moments and forces that the tubesheet must resist. A typical free-body diagram required for the derivation is shown in Fig. 21.22. The perforated region of the tubesheet is considered to be a flat plate with effective elastic properties (Modulus of Elasticity and Poisson's Ratio) to account for the perforations. The effective elastic properties depend on the ligament efficiency of the tube array and the thickness of the tubesheet and are determined from Figs. UHX-11.3 and UHX-11.4 as applicable. A complete description of the stress analysis methods used for tubesheets may be found in Ref. [24]. There are several key issues and innovations presented in Part UHX that the tubesheet designer should understand before using these rules. The ligament efficiency of a tubesheet is expressed as m = where p dt p - dt p (21.15) where:

p* =

a1 -

4 min 3(AL), (4Do p)4 pD2 o

p

b

1>2

(21.16)

AL Do

total area of all the untubed pass partition lanes) diameter of the center of the outermost tube hole

The effective pitch accounts for any pass partition lane in the tube pattern by increasing the effective pitch. The effective tube diameter accounts for the effects of tube expansion and includes consideration of the strength and stiffness differences between the tube and tubesheet material. The effective tube hole diameter is given as d * = max e cdt - 2t t a E t St b a br d, [dt - 2t t] f (21.17) E S

Note that d* cannot exceed (dt ­ 2tt); where: Et E St modulus of elasticity of tube material at its design temperature modulus of elasticity of tubesheet material at its design temperature allowable stress of tube material at the tubesheet design temperature tube expansion depth ratio (depth of tube roll or expansion/tubesheet thickness)

tube pitch diameter of the tube hole that is usually taken as the outside diameter of the tube

It has been demonstrated and accepted by the committee that an expanded (roller expanded or hydraulically expanded) tube-totubesheet joint effectively increases the ligament efficiency. Likewise, any untubed area of the tubesheet for tube pass plates will affect the effective ligament efficiency of the equivalent perforated plate. These effects are accounted for by determining an effective ligament efficiency based on an effective tube pitch and an effective tube hole diameter. The effective pitch,

FIG. 21.22 FREE BODY DIAGRAM FOR TUBESHEET SHELL

As may be seen, when the tube is expanded to the full depth of the tubesheet thickness, the total tube wall thickness is considered to be effective in stiffening the tube hole unless a reduction results from the stiffness or strength difference between the tube and tubesheet materials. Use of the equivalent ligament efficiency has a significant effect on the required tubesheet thickness. For a tubesheet that is extended as a flange to which a channel or shell is to be bolted, the bolt load causes an additional moment in the tubesheet. The moment caused by the bolt load may either increase or decrease the total stress in the tubesheet depending on the direction that the moment is acting with respect to the moments caused by pressure. Accordingly, the procedure given in UHX-12 carries the appropriate sign of moment due to the flange design bolt load when determining the bending moment at the center of the tubesheet. The method presented in UHX-12.5 includes consideration of the interaction effect of the integral shell on the tubesheet. This interaction effect is accounted for by the term s given in UHX12.5.4. The inclusion of this interaction effect makes the tubesheet solution iterative. The tubesheet thickness must be assumed and the corresponding tubesheet stresses are calculated. If the stresses are too large, then the design must be revised. It may be possible to decrease the required tubesheet thickness by increasing the thickness of the attached cylinder because of the stiffening effect of the thicker cylinder. When this method is used to design the tubesheet, it is important to understand that the bending stress at the integral cylinder-to-tubesheet juncture has to be categorized as a primary bending stress and limited to 1.5 S for the shell or tubesheet material, whichever is lower. If the integral cylinder is not assumed to provide support for the tubesheet, then the bending stress at the cylinder-to-tubesheet junction may be considered to be a secondary stress and could be limited to SPS. Thus, when the strengthening effect of the integral cylinder is included in determining tubesheet thickness, the bending stress at the shell must be calculated and compared with the appropriate primary bending stress limit. If the bending stress at the junction

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 93

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 93

does not satisfy the primary bending stress limit, the designer may do one or more of the following: (1) Increase the tubesheet thickness. Increasing the tubesheet thickness will generally result in smaller bending stress at the shell to tubesheet joint. (2) Increase the thickness of the attached cylinder. The bending stress decreases as the shell thickness increases; however, it is possible that the moment causing the stress will also increase because of the larger relative stiffness of the cylinder. (3) Increase both the tubesheet thickness and the cylinder thickness. (4) Perform a simplified elastic-plastic analysis to validate the design. The first three items are self-evident; however, additional insight is required for the last option. The analyses used in the derivations of Part UHX are based on elastic action of the material. In the analysis of tubesheets, it could be considered that an integral cylinder (shell or channel) either provides no rotational support (cylinder attached with pinned connection) or provides rotational support to the tubesheet assuming that there is no yielding at the connection. Neither of these cases represents reality. It is certain that the integral cylinder provides some rotational support at the connection. Likewise, if the bending stress at the integral cylinder-to-shell junction exceeds the material yield strength, the elastic analysis will overestimate moment generated at the junction and will overestimate the strengthening effect of the integral cylinder. In order to apply a practicable solution to this seemingly "all or nothing" consideration regarding the appropriate cylinder stiffness, it was decided that the bending stress at the junction could exceed the yield strength (but cannot exceed the primary plus secondary stress limit, SPS). However, in such cases, the moment at the junction is set to the moment to cause fully plastic action at the junction. For an elastic perfectly plastic material, the fully plastic moment is 1.5 times that moment that causes initial yielding at the surface of the shell. It is not possible to directly use this moment in the given procedure; however, the equivalent effect can be obtained by reducing the cylinder's modulus of elasticity when the cylinder-to-tubesheet stress is excessive. In the development of the rules, a parametric study found that reducing the cylinder's modulus of elasticity by the square root of the ratio of its yield strength divided by the stress at the junction (calculated on an elastic basis) was a good approximation of using the fully plastic moment. The rules for the design of fixed tubesheets incorporate many of the features used for U-tube tubesheets. Note that provisions for effective ligament efficiency are included in the same manner as for U-tubes, and the same effective elastic properties for the perforated portion of the tubesheet are used. However, the design method for a fixed tubesheet heat exchanger is considerably more complex than required for a U-tube tubesheet. This is because of the interaction between the tube bundle, the tubesheet, and the integrally attached cylinders. The origin of analysis methods of fixed tubesheet heat exchangers may be traced back some 50­60 years to the work of Gardner, Miller, Yi Yan Yu, and others. For a comprehensive explanation of the procedures used to design fixed tubesheets, see Ref.[24]. In the analytical procedure, the tubesheet is considered as a flat plate supported by an elastic foundation. The analysis requires structural continuity in the axial direction between the tube bundle, the shell, and the tubesheet. The analytical solution results in the use of Bessel functions. Originally, graphs and tabular values provided the factors involving the Bessel functions; however, the

geometry of the exchanger meant that these factors were often "off the chart" and the designer would not be able to proceed. This problem was resolved by the addition of Table UHX-13.1 that provides the derivation of all the factors used by the procedure, including polynomial expansions for the Bessel functions. The design of fixed tubesheets requires seven loading cases to be considered for each operating condition. All expected operating conditions, including start-up, shutdown, and any upset conditions need to be evaluated in addition to normal steady-state operation. The seven load cases are as under: (1) (2) (3) (4) (5) (6) (7) Tube side pressure acting alone Shell side pressure acting alone Tube side and shell side pressures acting concurrently Thermal expansion acting alone (no pressure considered) Tube side pressure plus thermal expansion Shell side pressure plus thermal expansion Tube side pressure, shell side pressure, and thermal expansion

The complexity of the analysis, the number of load cases, and the number of operating conditions all combine to make it essential that computer tools be used for the design of fixed tubesheet heat exchangers. The rules include interaction of the cylinders that are integral to the tubesheet. The analysis is similar to the U-tube tubesheet rules in this respect. However, it is often found that the thickness of a fixed tubesheet can be dramatically affected by the stiffness of the integral cylinders. It is often possible to decrease the tubesheet thickness by increasing the cylinder thickness locally at the tubesheet. The supplemental procedure in paragraph UHX13.6 allows the designer to implement this option. Because of the significance of the interaction between the integral cylinder and tubesheet, the same provisions as those in the U-tube design methods for limiting the bending stress at the cylinder apply. There is a subtle difference between the pressure load cases and the load cases that include the effects of differential thermal expansion. When considering the pressure load cases, the stress at the cylinder-to-tubesheet juncture is considered to be primary bending and is limited to 1.5 S. The stresses resulting from differential thermal expansion are strain induced and are considered to be secondary in nature. In the analysis procedure, the allowable stress for the load cases involving thermal expansion is limited to SPS, which is basically twice the value used for the load cases involving pressure alone. This means that the allowable bending stress in the tubesheet and at the cylinder-to-shell junction may be at least twice as large as when pressure alone is considered. The philosophy that permits the allowable primary bending stresses to be exceeded by using a method that accounts for plasticity of the cylinder-to-tubesheet joint is included for fixed tubesheets by paragraph UHX-13.7. This procedure is not intended to apply when the allowable stresses of the cylinder or tubesheet material are controlled by time- dependent properties (when creep may occur). Section II, Part D, designates temperatures where time-dependent properties govern the allowable stress by showing the allowable stress value in italic print. This same limitation should be applied to the elastic­plastic procedure for U-tube tubesheets. It should be noted that U-tube tubesheet rules of Part UHX do not include the effect of dead weight. This effect may be easily accounted for by adding an equivalent pressure to the design pressure. Likewise, fixed tubesheet rules do not consider dead weight or pressure drop through the exchanger. The designer is cautioned that it is not correct to add an equivalent pressure to

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 94

94 · Chapter 21

either the shell side or tube pressure order to account for weight and pressure drop effects in fixed tubesheet heat exchangers. This is because the tubes stay the tubesheet and the pressure loads are balanced between the tube bundle and shell. The loads from weight and pressure drop are unbalanced and are not stayed by the tube bundle. In effect, when fixed tubesheets are exposed to unbalanced loads such as dead weight and pressure drop, they react similar to U-tube tubesheets. This effect can be very significant for large vertical heat exchangers and must not be ignored. No article on the ASME fixed tubesheet rules would be complete without some discussion of how they may compare with the rules of TEMA [25]. The committee has found the rigorous analysis procedures of Part UHX that will result in tubesheet thickness sometimes larger than TEMA, sometimes thinner than TEMA, and in many cases about the same as TEMA. The key parameter that defines if the rules yield comparable results between the two methods is the term Xa. This term, which is essentially the ratio of the tube bundle stiffness to the shell thickness, is expressed by the following: Xa = c2411 - n*22Nt where:

*

21.7

21.7.1

MANDATORY APPENDICES

Appendix 1: Supplementary Design Formulas

Ettt (dt - tt) a2 o E*Lh3

d

1>4

(21.18)

Nt Et tt dt E* ao L

effective Poisson's Ratio of the perforated region of the tubesheet number of tubes modulus of elasticity of tube material nominal thickness of tubes nominal outside diameter of tube effective modulus of elasticity of the perforated portion of the tubesheet radius to the outermost tube center plus dt /2 tube length between the inner faces of the tubesheet

Appendix 1 provides design formulas and rules that supplement the rules contained in Part UG. The use of these rules is mandatory when referenced by the other sections of the book. Likewise, if design rules are given in Appendix 1 for a geometry that is not covered in the other sections of Division 1, then the rules given in Appendix 1 must be used. Paragraph 1-1 provides equations to determine the required thickness and maximum allowable working pressure using the outside diameter of cylinders and the outside radius of spherical segments. These formulas may be used in lieu of those in UG- 27 when the designer chooses to do so. Using the equations based on the outside dimension is convenient when the pressure part is a piping component with the specified size based on the outside surface dimensions. For example, NPS 16 pipe has an outside diameter of 16 in., and it becomes very convenient to determine its required thickness on the basis of the outside diameter. The use of paragraph UG-27 formulas would require a trial and error solution when designing piping components in order to hold the OD constant. The use of the equations in paragraph 1-1 yields equivalent results to the equations of paragraph UG-27. For example, the maximum allowable working pressure for a cylinder is given in paragraph UG-27 (see Eq. 21.1) as, P = SEt R + 0.6t

When expressed in terms of the outside radius, this equation may be rewritten as SEt (Ro - t) + 0.6t SEt P = Ro - 0.4t P = (21.19) (21.20)

When the value of Xa is less than about 3.0, the ASME rules generally result in a thicker tubesheet than required by TEMA. In the range of Xa from 3 to about 9, the results of the two methods are comparable, and when Xa is greater than about 9, the ASME rules will generally result in thinner tubesheets than TEMA. Part UHX-15 The requirements of Paragraph UW-20 for tubeto-tubesheet welding had been relocated to UHX-15; however, the committee has voted to return the tube-to-tubesheet welding requirements to Part UW because they may be applicable to vessels other than heat exchangers. The technical requirements have not been revised and are the same whether they are given in UW20 or UHX-15. Paragraph UHX-19, dealing with heat exchanger marking and data reports, was extensively revised in the 2006 Addenda. These revisions introduced additional marking requirements for heat exchangers with differential pressure and/or mean metal temperature design conditions that are less severe than the adjacent chambers. For fixed tubesheet heat exchangers, the nameplate shall be marked with the following cautionary note: Caution: The code required pressures and temperatures marked on the heat exchanger related to the basic design conditions. The heat exchanger design has been evaluated for specific operating conditions and shall be reevaluated before it is operated in a different operating condition.

This becomes

which is the same as given by Equation 1 in paragraph 1-1. The rules of paragraphs 1-2 and 1-3 must be used in lieu of those in paragraph UG-27 when the limits of paragraph UG-27 are exceeded. In summary, the rules of paragraph 1-2 or paragraph 1-3 must be used when, Cylindrical Shells (Circular Stress) Thickness Pressure 1/2 Radius 0.385 SE Cylindrical Shells Longitudinal (Stress) 1/2 Radius 1.25 SE

Spherical Shells 0.365 Radius 0.665 SE

The formulas of paragraphs 1-2 and 1-3 account for the nonlinear distribution of stresses resulting from pressure in thick shells (commonly referred to as the Lame' effect). The maximum allowable working pressure of a vessel designed in accordance with the rules of paragraphs 1-2 and 1-3 are P = SE a Ro 2 Z - 1 b for cylindrical shells, where Z = a b Z + 1 R (21.21)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 95

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 95

P = 2SE a

R + t 3 Y - 1 b for spherical shells, where Y = a b Y + 2 R (21.22)

The equations of paragraphs 1-2 and 1-3 can be duplicated using the derivations given in Ref. [16] for thick cylinders. Paragraph 1-4 provides rules for formed heads under internal pressure. The equation given in paragraph UG-32(e) is only applicable to torispherical heads with a knuckle radius 6% of the crown radius and a crown radius equal to the outside skirt diameter. Likewise, the ellipsoidal head equation given in paragraph UG-32(d) is only applicable to 2:1 ellipsoidal heads. If heads of different proportions are used, they must be designed in accordance with the rules of paragraph 1-4. Paragraph 1-4(f) provides a recent addition to the code that provides rules for thin formed heads with t/L ratio less than 0.002. These rules provide assurance that the knuckle region of a formed head under internal pressure does not "wrinkle." The membrane stress in the knuckle area of a formed head under internal pressure generally is compressive. If the compressive membrane stress is too large, the knuckle can experience elastic instability (buckling). When this occurs, the head "wrinkles" in the knuckle area and the head may become unusable. The mandatory rules in paragraph 1-4(f) are meant to assure that this failure mode does not occur for thin formed heads. It is noted that this methodology is applicable only when the temperature is not in the creep regime for the head material and when the t/D ratio is greater than 0.0005. When the temperature is in the creep regime (as noted by Table 1-4.3), or when the t/D ratio is less than 0.0005, then the provisions of U-2(g) of the code shall be applied. Paragraph 1-5 provides rules for the reinforcement of conical reducer sections and conical heads under internal pressure. Additionally, the effects of axial loads resulting from dead weight, overturning moments from wind or seismic conditions are accounted for in the procedure. A free body diagram of a conical transition to shell connection (see Fig. 21.23) shows internal radial forces and moments are generated at the small and large end junctions. These radial forces and moments result in primary local membrane stress and secondary bending stress where the cone attaches to a cylinder. In order to assure that localized stresses at cone-to-cylinder junctions are not excessive, proper reinforcement must be provided in the vicinity. The rules for determining the need for and the required size of reinforcement for conical transitions under internal pressure are given in paragraph 1-5. The rules are only applicable for cones with a half-apex angle, , less than or equal to 30°. It is noted that these rules assure that in the vicinity of the cone-to-shell junction, the membrane stress does not exceed 11/2 times S and the membrane plus bending stress does not exceed SPS. Paragraph 1-5(g) allows the use of cones with half-apex angles greater than 30°, provided that a detailed stress analysis is conducted. When a detailed stress analysis is conducted, the stresses are limited to the same criteria that form the basis of the "design-by-rule" requirements; that is, the local membrane stress is limited to 11/2 S and the membrane plus bending stress is limited to SPS. Although reference is made to closed form classical discontinuity analysis method, the use of computer programs such as finite element is acceptable. It is recommended that Section VIII, Division 2, Part 5, be reviewed in order to understand the significance of the "local primary membrane stress" and the "membrane stress plus secondary

FIG. 21.23

CONE-TO-CYLINDER FREE BODY DIAGRAM

bending stress" and their appropriate allowable values as they pertain to areas of discontinuity. Paragraph 1-6 provides rules for spherically dished covers that are bolted to an adjacent component. The rules consider the interaction effect of the ring and dished portion of the head and the effect of moments generated by the attachment bolts (when used with ring-type gaskets). Provisions are given to design such covers with the bolt holes slotted through the edge of the head. Such designs are required when swing bolts are desired to provide quick opening of the cover. Rules are also provided for full-face gaskets for which the bolt load does not result in a substantial moment in the cover. It should be noted that the rules presented by Fig. 1-6 a apply to all bolted heads and the weld spacing provisions of this figure must be satisfied. Paragraph 1-7 provides requirements for large openings in cylindrical shells. These requirements are mandatory when the dimensional limits defined in paragraph UG-36(b)(1) are exceeded. Paragraph UG-36(b)(1) invokes the 1-7 requirements for openings in cylindrical shells when (1) The shell diameter is 60 in. or less and the opening diameter is greater than either 1/2 the vessel diameter or 20 in. (2) The shell diameter is greater than 60 in. and the opening diameter is greater than either 1/3 the vessel diameter or 40 in. Paragraph 1-7(a) is commonly referred to the "close-in" nozzle reinforcement requirement because most (2/3) of the required area of replacement must be located within reduced limits of reinforcement. Paragraph 1-7(b) provides additional requirements for nozzles with an inside diameter greater than 40 in. and 3.4 1Rt, where R is the vessel inside radius and t is the vessel wall thickness, when they are in cylindrical vessels greater than 60-in. diameter. The provisions of 1-7(b) are meant to control the amount of deformation at the nozzle-to-shell junction for large openings caused by an out-of-plane bending moment. The basis of the requirements of 1-7(b) is taken from the work of Jacobs and McBride [3]. The rules in 1-7 are only applicable to large openings in cylinders and do not apply to large openings in spherical or dished heads. Paragraph 1-8 provides rules for conical-to-cylinder transitions under external pressure (and axial loads due to weight, and wind or seismic moments). These rules are valid for cones with a halfapex angle of 60° and less. It is very important to understand that there are two different issues for which this paragraph provides rules. The first is to limit the discontinuity stress at the cone-tocylinder juncture in a similar manner as paragraph 1-5 does for

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 96

96 · Chapter 21

internal pressure. Again, a free body diagram shows internal radial forces and moments are generated at the junction, which results in localized stresses resulting from external pressure. ArL and ArS give the needed area of reinforcement to assure that the local stresses at the junction are within acceptable limits. These reinforcement rules do not apply when the cone has a knuckle at the cone-to-shell transition. However, when a cone-to-cylinder junction is considered as a line of support (as shown in Fig. 21.1 b, e, and f), then it must be demonstrated that the moment of inertia at the cone-to-cylinder junction satisfies the required moment of inertia (either Is or I's). If the moment of inertia available at the junction does not satisfy the required moment of inertia, then the cone-to-cylinder junction cannot be considered as a line of support. In that case, the unstiffened length must not consider the junction as a line of support (see Fig. 21.1a-2and c-2). The moment of inertia requirements at the junction apply to cones with and without knuckles. Experience will show that the available moment of inertia at a cone-to-shell junction (with or without a knuckle) will rarely satisfy the required moment of inertia without additional stiffening elements. From a practicable point of view, it is prudent to place stiffening rings near the cone-to-cylinder junctions for vessels designed for external pressure. Similar provisions to those given in paragraph 1-5 are provided for a detailed stress analysis to demonstrate that the discontinuity stresses are satisfied when the half-apex angle exceeds 60°.

load from the bolts to the gasket. Appendix 2 completely addresses only the third issue, that is, flange proportions. The required design bolt load is addressed to some extent by the design considerations in Appendix 2 and the actual bolt load that may be required in service is addressed in Appendix S. Section VIII does not provide guidance related to the selection of gasket type, material or facing. However, once these factors are determined, the appropriate design parameters to be used are provided. The rules of Appendix 2 are not limited to diameter, pressure and temperature; however, special precautions are related to flange rotation and high temperature designs are given in Appendix S. The basic assumptions in the development of the flange design rules were as under: (1) The flange is made of a homogeneous material having stable elastic properties such that creep at high temperatures and plastic yielding at lower temperatures do not occur. (2) The load imposed on the flange by tightening the bolts is assumed or determined. (3) The lever arm of the bolt has been determined. (4) The effect of the moment resulting from the bolt load and its appropriate lever is independent of the location of the bolt circle and the forces balancing the bolt load. For a tapered hub flange, a free body diagram of the geometry may be made with three components: the ring, the hub, and the shell (see Fig. 21.24). Each component may be studied as a separate unit with undetermined boundary conditions (displacement, rotation, moment, and shear) at each of the junctions where the free body elements are connected. By writing equations of continuity, that is, the rotation and radial deflection of adjacent components must be the same at the point where they are connected, the internal moments and forces can be determined. Once the internal moments and forces are determined, it is straightforward to determine the resulting stresses in the flange assembly. The analysis assumes that the ring has negligible deformation at its center and that dishing or rotation of the ring is small so that the effects of the pressure and bolt loading are linearly related and the problem may be solved by superposition. It is also assumed that the hub and shell may be treated as membranes subject to tension and bending where radius to the midsurface may, in general, be used interchangeably with the inside radius. The analysis develops the deflection and rotation of each of the elements of the free body acted upon by the imposed and internal loads. Since the deflection and rotation of adjacent components must be identical at the juncture, the free body deflection and

21.7.2

Appendix 2: Rules for Bolted Flange Connections with Ring-Type Gaskets

Appendix 2 provides rules for the design of bolted flanged connections that use gaskets that are entirely within the bolt circle. Unless a flanged connection is used in accordance with the provisions of UG-44 (e.g., B16.5, B16.47, and others), all flanges with ring-type gaskets (e.g., the gasket is within the bolt circle) must be designed in accordance with Appendix 2. There are a number of handbooks that delineate how the design method is to be used and a detailed description will not be repeated here. However, it may be useful to understand the basis used for developing these rules. In 1935, an effort was made by Waters et al. [17] to develop a rational design analysis for flanges over the practicable range of application. As noted in this development, flange design consists of three fundamental issues. The first is the selection of the proper gasket (size, type, material, and facing). The second is the determination of the bolt loads required to hold the joint tight under pressure, and the third is to design the flange to adequately transmit the

FIG. 21.24

FREE BODY DIAGRAM OF BOLTED FLANGE

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 97

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 97

rotation of each component are set equal to that of the other. For this model, the continuity analysis results in four equations with four unknown variables that may be solved to determine the shear and moments acting at each juncture. Using these shear and moment values, the stress in the ring, the hub, and the shell may be determined. The authors of the original paper on flange design did a remarkable job in simplifying the highly complex, tedious rules into a series of charts and factors that have been used for well over 50 years. These are basically the factors that are presently given in the charts and equations of Appendix 2. Some specific comments regarding Appendix 2 are appropriate. Appendix 2 provides rules for integral-, loose-, and optional-type flanges. An integral flange ring, hub (if any), and shell are connected in such a manner that they may be considered to be integral. This includes forged and cast welding neck flanges and long weld neck flanges and ring flanges that are welded to the shell where the shell acts as a hub. See Fig. 21.25 (5) through (7). Loose-type flanges fall into two categories, the first type is where the flange is not connected to the shell such as lap joint flanges and screwed flanges. The second type of loose flange includes those that are welded to the shell, but the weld detail is not considered to give the mechanical strength equivalent to an integral flange. See Fig. 21.25 (1) through (4a) for examples of loose flanges. Optional flanges have the flange ring attached to the shell such that the assembly may be considered to be integral; however, the designer may choose to design the flange as an integral flange or a loose flange, at his option. For loose flanges, the shell is not considered to contribute any strength to the flange ring even though the flange may be attached. Optional flanges have a weld connection that may be considered integral, and the calculations may be done according to the rules applicable for integral flanges; however, for simplicity of the calculations, the flange may be designed as a loose flange. Table 2-5.1 provides factors to be used in the design of flanges that depend on the type of gasket selected. The Gasket Factor, m, and Minimum Design Seating Stress, y, are listed. It should be noted that the values of m and y listed in this table are not mandatory. The value of these parameters to be used in the flange design should be in accordance with the gasket supplier's recommendations. Large diameter flanges (greater than NPS24) may not have sufficient rigidity to maintain the needed compressive load on the gasket to prevent leakage. The rigidity index of Appendix 2 provides a method of accounting for the flange flexibility in the design. The rigidity factor J is a measure of the flange rotation under the design loads. The rigidity indices are defined by the following expressions: For integral- and optional-type flanges, J = 52.14 MoV LEg2 hoKI o (21.23)

where:

E J KI KL

modulus of elasticity for the flange material at design temperature rigidity index rigidity factor for integral-or optional-type flanges 0.3 rigidity factor for loose type flanges 0.2

All other terms are as defined in Appendix 2 of Section VIII, Division 1. The rigidity index shall be equal to or less than 1.0. The specified limits are empirical in that they are experience based. It is noted that the rigidity factors of Appendix 2 may control the design of large diameter body flanges for vessels, heat exchangers, and large diameter openings. It should also be noted that flanges supplied to the ASME B16.47 [23] standard may not satisfy the rigidity factors of Appendix 2. The rigidity factors, J, are actually a calculation of flange rotation expected under the action of the flange design loads. Thus, the rotation of an integral flange should be limited to 0.3° and the rotation of a loose flange should be limited to 0.2°. Since it is relatively simple, it is useful to consider the derivation of the J factor for loose flanges without hubs. The rotation of a ring is given by Ref. [16] as u = where: M a K 12 Ma Et3 ln (K) (21.26)

ring rotation in radians moment per inch of centerline length radius of ring centerline ratio of outside radius to inside radius

Since Mo is the total moment acting on the ring, M o = 2paM or M = Mo 2pa (21.27)

In order to convert radians to degrees, 180/ . Thus, Mo a 2pa 180 u = 3 Et ln (K) p 12

must be multiplied by

or

u =

109.4 Mo Et3 ln (K)

(21.28) 0.2. If both

If the maximum rotation is limited to 0.2°, let max sides of the above equation are divided by max, then 109.4 Mo u = 3 u max Et ln (K)u max

(21.29)

For loose-type flanges with hubs, J = 52.14 MoVL LEg2 hoKL o (21.24)

For loose-type flanges without hubs and optional flanges designed as loose type, J = 109.4 Mo Et3 ln (K)KL (21.25)

The rigidity index J is / max and must be less than 1.0. The rigidity factor KL is the allowable rotation ( max) in degrees. While this derivation is for loose flanges, the rigidity index for the other types of flanges can be developed in a similar manner. It is noted that Code Committee action has recently moved the flange rigidity criteria into mandatory Appendix 2 from nonmandatory Appendix S, which will makes the flange rigidity requirements mandatory and not optional. However, a revision in the 2007 Edition will permit a Manufacturer to use other values of KI and KL in the flange rigidity check based on past experience, with the concurrence of the user.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 98

98 · Chapter 21

FIG. 21.25 TYPES OF FLANGES (Source: Fig. 2-4 of Section VIII Div. 1 of the ASME Code)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 99

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 99

FIG. 21.25

(CONTINUED)

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 100

100 · Chapter 21

21.7.3

Appendix 3: Definitions

(9)

Appendix 3 provides definitions of general terms that are used throughout Division 1. Definitions relating to specific application of the rules may be given in the applicable part of the code. For the purpose of illustration, some selected definitions from Appendix 3 are discussed here. (1) "acceptance by the inspector"­ When an Inspector is required to review and accept specific data, tests, or examinations, he is only required to review the subject in accordance with the provisions of the code and he is not assuming any of the Manufacturer's responsibilities. As an example, the Inspector is required to assure that design calculations have been conducted; however, he is not responsible for the accuracy of the design calculations. (2) "ASME Designated Organization"­ This is an organization that ASME has authorized to provide administrative functions for ASME. In the past, the National Board of Boiler and Pressure Vessel Inspectors was the only organization recognized to carry our audits and surveys of certificate holders or applicants for accreditation; however, now, there may be other organizations that ASME may authorize to act on its behalf other than the National Board. (3) "ASME Designee"­ This is an individual that ASME has authorized to provide administrative functions for ASME. (4) "butt joint"­ A joint between two members located in intersecting planes between 0 and 30°, inclusive. It should be noted that if two members with intersecting planes greater than 30° are welded, the weld joint is not a butt weld. If the service provisions of UW-2 require a butt joint for Category B or Category C joints, then a knuckle or another type of transition must be used. (5) "angle joint"­ If the intersecting planes between two members is greater than 30° but is less than 90°, then this is defined as an angle joint. It is noted that radiographic examination of such welded joints requires special considerations and may not yield results that can be properly interpreted. As such, angle joints are not allowed when the service conditions require butt joints or radiography. (6) "corner joint"­ When the intersecting planes of two members is approximately 90°. Corner joints are not intended to be radiographed, and they do not have any associated design joint efficiencies. Corner joints cannot be used when the service requirements mandate a butt joint. (7) "calculated test pressure"­ This is the alternative test pressure specified in paragraph UG-99(c) for a hydrostatic test and UG-100(b) for a pneumatic test. The calculated test pressure is the "new and cold" maximum allowable working pressure of the vessel. It is the allowable pressure at ambient temperature using the uncorroded thickness of the vessel. (8) "normal operation"­ A pressure vessel is to be operated within the design limits for which the vessel was stamped. Any coincident pressure and temperature are acceptable, provided they do not constitute a more severe condition than that used for the design of the vessel. For example, the temperature of a vessel may be allowed to exceed the design temperature if the coincident pressure is reduced such that the vessel thickness is shown to satisfy combination of pressure­temperature conditions. It is noted that abnormal conditions may also be considered in the design of a vessel where the design conditions may be selectively exceeded; for example, the pressure predicted for an internal

(10)

(11)

(12)

deflagration (an abnormal event) is allowed to exceed the vessel MAWP (see nonmandatory Appendix H). "required thickness"­ It is the thickness computed by the equations and requirements of the code and does not include any provisions for corrosion, erosion, or any other thinning allowances. "design thickness"­ The required thickness when the effects of corrosion, erosion or other thinning allowances are added to the "required thickness." "nominal thickness"­ It is generally the thickness that is selected as commercially available and supplied to the manufacturer. For example, if the design thickness of a vessel shell is 0.42 in.; but the plate is supplied in a commercially available thickness of 1/2 in., then the nominal thickness is 0.5 in. It should be noted that the "nominal thickness" used for postweld heat treatment requirements are as defined in paragraph UW-40(f). "vessel Manufacturer"­ It is any Manufacturer who constructs a vessel or part in accordance with the rules of the code and holds an ASME Certificate of Authorization to apply the Code Symbol Stamp to the item or component.

21.7.4

Appendix 4: Rounded Indication Charts Acceptance Standard for Radiographically Determined Rounded Indications in Welds

Radiographs will reveal indications in welds that are not cracks or lack of penetration/fusion. Porosity, slag, or tungsten in the weld may cause such indications. Appendix 4 provides guidance for the acceptability of rounded indications that are discovered by radiographic examination of welds. A rounded indication is one with a length that does not exceed three times its width. They may be circular, elliptical, conical, or of irregular shape. Table 4-1 provides examples of when rounded indications are considered to be relevant, and it provides the maximum acceptable size of rounded indications. A sequence of four or more rounded indications is considered to be aligned when they touch a line parallel to the length of the weld that is drawn through the center of the two outermost indications. Figure 4.1 defines the criteria for aligned rounded indications. The rounded indication charts, Fig. 4-3 through 4-8, show various types of rounded indications that may be randomly dispersed or clustered for different weld thicknesses greater than 1/8-in. These charts represent the maximum acceptable concentration of rounded indications in a welded joint. Rounded indication charts of Appendix 4 are empirically based and represent a quality standard for welding that should be achievable by normal welding methods and processes.

21.7.5

Appendix 5: Flanged and Flued or Flanged-Only Expansion Joints

Appendix 5 provides rules for flanged-only or flanged and flued expansion joints. These types of expansion joints are fabricated from heads that are flanged-only or flanged and flued (see Fig. 21.26). Flanged and flued expansion joints are thicker (greater than 1/8-in. thickness) than bellows joints and offer several advantages as mentioned below: (1) The joint is less prone to dents and damage. (2) The joint is not prone to vibration. (3) Weld repair is more easily accomplished if the expansion joint is damaged. (4) The material of construction is commonly the same as that of the vessel for which it is installed and not required to be made of high alloy material.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 101

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 101

FIG. 21.26 TYPICAL FLANGED AND FLUED OR FLANGED ONLY FLEXIBLE ELEMENTS (Source: Fig. 5-1 of Section VIII Div. of the ASME Code)

However, these joints normally have much more stiffness than the bellows-type joints and their ability to absorb large thermal expansion deflections is limited. It is possible to use more than one expansion joints in series to accommodate large differential thermal expansion. The rules in Appendix 5 apply to single layer flexible elements subjected only to axial motion. Appendix 5 contains rules for the common types of "thick" wall expansion joints; however, the rules do not preclude the use of other types of expansion joints that are not covered by this appendix. Joints that differ from the basic concepts of Appendix 5 shall be designed in accordance with paragraph U-2(g). It is noted that the minimum thickness, in the corroded condition, allowed for Appendix 5 expansion joints is limited to 0.125 in. (Appendix 26 provides rules for expansion joints with a wall thickness equal to or less than 1/8 in.) This thickness limitation separates the scope given in Appendix 5 from that of bellows expansion joints covered by Appendix 26. It is also noted that the minimum thickness requirements of paragraph UG-16 do apply for flanged and flanged and flued expansion joints. Generally, provisions are required for components in the assembly other than the expansion joint to resist the hydrostatic end loads caused by pressure since the expansion joint is normally not able to resist this load without becoming overstressed. For example, the tube bundle or the inner or outer pipe of a double pipe heat exchanger normally resists the pressure end load. The stress in the member that resists the pressure end load must be limited to the maximum allowable stress of the material from Section II, Part D (see paragraph UG-23(c). Appendix 5 is unusual among other Section VIII, Division 1, requirements in that it is based on a "design-by-analysis" concept. This means that detail design formulas are not provided, but rather the allowable stress criteria are defined. The designer is free to calculate the stress in the expansion joint using any method deemed appropriate. Acceptable methods would include finite element analysis, plate and shell computer programs, and closed form solutions. It is noted that a method to determine stresses for flexible elements in heat exchangers is given by Tubular Exchanger Manufacturer's Association [25]. Once the

stresses are determined, Appendix 5 provides the applicable criteria. Since the load in an expansion joint depends on the relative stiffness of the connected parts (including the stiffness of the joint itself), it is possible that the maximum stress in the joint can occur in the uncorroded condition. Because of this possibility, Appendix 5 requires that an expansion joint meet the allowable stress criteria in both the corroded and uncorroded condition. Paragraph 5-3(a)(1) limits the maximum stress in the expansion joint resulting from the application of pressure acting directly on the joint (shell side pressure in a shell and tube heat exchanger) to 1.5 S. This limit is based on the consideration that in the flat plate portion of the joint, the bending stress resulting from direct application of pressure is a primary bending stress. Consistent with other sections of Section VIII, primary bending stress is limited to 1.5 S. However, this paragraph also limits the stress in the knuckle or flue section of the expansion joint to 1.5 S. When an elastic stress analysis is performed, the stress is limited such that a plastic hinge is not developed in these areas. If a plastic hinge develops, then the bending stress in the flat section of the joint is underestimated by an elastic analysis. Appendix 5 conservatively limits the stress to 1.5 S in all areas of the joint considering the application of direct pressure. However, if it can be demonstrated that the stresses in the flat section of the joint still satisfy their allowable stress when considering any plasticity effects, then stresses at the knuckle or flue section may be limited to SPS. One way to show this would be to assume that the flange and flue sections are "simply supported," and if the resulting stresses in the flat section of the joint still meet 1.5 S, then the stresses in the flange and flue section could be compared to SPS and not 1.5 S. The tube side pressure in a shell and tube heat exchanger with shell side expansion joint results in a stress in the joint that depends on the stiffness of all the connected members. The tube side pressure results in a differential motion between the tube bundle and the shell that depends on their relative stiffness. This motion is absorbed by the expansion joint (plus deformation of the other components). By definition, this motion on the expansion joint is strain induced and the resulting stress in the joint is a secondary stress. However, there may be instances where the

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 102

102 · Chapter 21

stiffness of the joint allows the stress for the pressure-restraining elements (such as the tube bundle) to be reduced. If these restraining elements rely on the stiffness of the expansion joint to satisfy their maximum allowable stress under pressure loading, the stress in the expansion joint should be considered to be primary and is limited to 1.5 S. If the stiffness of the joint is not required to limit the stress in the pressure load resisting elements to S, then the stress in the expansion joint may be considered secondary and is combined with the thermally induced loads. A simple way to validate this requirement could be to analyze the heat exchanger for tube side pressure with the stiffness of the joint set to zero. If the stress in the tube bundle and tubesheet did not exceed the allowable primary stress limit, then the stiffness of the joint is not a factor in the pressure-restraining design. For these cases, the stresses introduced into the expansion joint are secondary stresses and need only to be combined with the thermally induced stresses. The stresses in the expansion joint due to axial deflection, including differential thermal expansion, are considered to be secondary stresses. As such, the allowable maximum stress is consistent with the primary plus secondary stress limits of Section VIII, Division 2, Part 5. Paragraph 5-3(a)(2) limits the pressure plus axial deflection stress to SPS [where SPS is defined in UG-23(e)]. Appendix 5 previously required the designer to determine the allowable cycle life of the expansion joint and show that the allowed number of cycles was greater than that required. However, recent revisions to Appendix 5 have deleted the specific requirement for cycle life determination. This was done because the primary plus secondary stresses in the expansion joint are limited to values consistent with the stress expected in other parts of the equipment (i.e., SPS). Thus, if the expansion joint requires consideration of cyclic loading in the design, then it is likely that other parts of the vessel require some consideration of cyclic loading. The deletion of specific requirement for cyclic loading does not mean that cyclic loading can be ignored. Paragraph UG22(e) requires consideration of cyclic loadings and the designer must consider such effects. Cyclic loading will have a more significant effect on flanged-only expansion joints than for flanged and flued joints. The geometric stress concentration at the flanged-only joint-to-shell weld is significantly greater than for the flanged and flued joint. As such, flanged-only joints are not recommended when significant thermal or pressure cycles are expected during operation. It is possible that the thickness required to satisfy the stress criteria of Appendix 5 for an expansion joint may be less than the shell thickness required by paragraph UG-27. It should be recognized that the flat portion of the expansion joint stiffens the adjacent area of the shell, and the stress is limited in the vicinity of where the joint attaches to the shell to 1.5 S. To not penalize an expansion joint when its thickness can be less than the adjacent shell, Appendix 5 allows an extended straight flange at the inner and outer torus of the flexible elements to be thinner than the value obtained by UG-27. The length of the thinner straight flange cannot exceed 0.5 1Rt, where R is the inside radius of the expansion joint straight flange at the point of consideration , and t is the uncorroded thickness of the straight flange. Appendix 5 also provides general fabrication and inspection requirements. When a flanged and flued joint or a flanged-only expansion joint is welded by a manufacturer other than the vessel Manufacturer, a U-2 Partial Data Report must be provided with the information required by paragraph 5-6.

21.7.6

Appendix 6: Methods for Magnetic Particle Examination (MT)

Appendix 6 defines the procedures to be used whenever examination by the magnetic particle method is specified in the other sections of Section VIII, Division 1. This appendix must be used in conjunction with Article 7 of Section V of the ASME Boiler & Pressure Vessel Code for the detailed requirements and procedures to be followed when performing a magnetic particle examination. Each magnetic particle examiner must have sufficient vision to be able to read a Jaeger Type No. 2 Standard Chart at a distance of not less than 12 in. and be able to discern the contrast between the colors used. The examiner must have his vision checked annually. The examiner must also be competent in the techniques used for magnetic particle examination. The Manufacturer must certify that each examiner meets these qualifications. An indication revealed by the magnetic particle examination is considered to be relevant if it has any dimension greater than 3/16-in. A linear indication is one where its length is greater than three times the width. Any relevant linear indication detected by magnetic particle examination is not acceptable. Relevant rounded indications greater than 3/16 in. are not acceptable, and four or more relevant rounded indications in a line separated by 1/16 in. or less are not acceptable. Any unacceptable indication shall be removed or reduced to an acceptable size. If repair welding is not required, the repaired area shall be blended into the surrounding area to avoid sharp corners, notches, or crevices. If weld repair is required, the area shall be cleaned and welded with a qualified welding procedure. However, before welding, removal of the defect shall be confirmed using a suitable method. After a defect has been repaired, the surface shall be reexamined by the magnetic particle method and by all other examination methods originally required for the repaired area.

21.7.7

Appendix 7: Examination of Steel Castings

Appendix 7 provides the requirements applicable to the examination of steel castings for which a 100% quality factor is to be applied. Castings made in accordance with an accepted standard such as ASME/ANSI B16.5 are not required to meet the requirements of this appendix unless the castings are used for lethal service applications. Two levels of examination are provided. The first level applies to castings that have a maximum body thickness less than 41/2-in. The second level is applicable to castings that are intended for severe service applications or that have a maximum body thickness greater than 41/2-in. It is the responsibility of the design engineer to determine if the service conditions warrant the more restrictive inspection requirements for castings with a body thickness less than 41/2 in. For castings with a maximum body thickness less than 41/2 in., all critical sections of the casting must be examined by radiography. The radiographs are compared to Standard Reference Radiographs for Steel Castings found in ASME Boiler & Pressure Vessel Code Section V using the maximum severity level for the imperfection category as defined in paragraph 7-3 of this appendix. In addition, either the magnetic particle or liquid penetrant method must be used to examine all surfaces. For castings in severe service applications and for castings with a maximum body thickness greater than 41/2-in., all sections must be examined visually and by radiography. For castings with a maximum wall thickness greater than 12 in., ultrasonic examination is substituted for radiography.

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 103

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 103

When an imperfection is to be repaired, the excavated area of the casting must be examined by either the magnetic particle or the liquid penetrant method to assure the defect has been removed or reduced to an acceptable size. If no more than 5% of the intended design thickness is removed at a location, then repair welding is not required if the repaired area is blended into the surrounding area. Castings made of material that cannot be welded and contain imperfections that are greater than the acceptable limits defined in paragraph 7-3 shall be rejected. The purchaser of the casting shall approve repairs if the repair of any casting involves removal of more than 75% of the wall thickness or a length of 6 in. or more.

inner vessel will produce a total external pressure greater than 15 psi on the inner vessel, the entire jacket falls within the scope of the code. For convenience, jacketed vessels are categorized into following types. (1) Type 1­Jacket of any length confined entirely to the cylindrical shell (2) Type 2­Jacket covering a portion of cylindrical shell and one head (3) Type 3­Jacket covering a portion of a head (4) Type 4­Jacket with addition of stay or equalizer rings to the cylindrical shell portion to reduce effective length (5) Type 5­Jacket covering cylindrical shell and any portion of either head These types are schematically shown in Fig. 21.27. The design of jacketed vessels shall satisfy all the applicable requirements of Subsection A unless otherwise stated in Appendix 9. Jacketed vessels may be designed as braced and stayed members in accordance with paragraph UG-47. The shell and head thickness shall be determined by the formula given in the applicable sections of Subsection A with appropriate consideration of the loading conditions defined in paragraph UG-22. Where vessel supports are attached to the jacket, consideration must be given to how the loads are transmitted between the support and the inner vessel and its contents. The inspection opening requirements of paragraph UG-46 apply to jackets; however, the maximum size of the opening need not exceed NPS 2 regardless of the vessel size. Paragraph 9-5 defines rules for the design of jacket closure that connects the jacket to the inner vessel. Depending on the geometry of the jacket, closure members are designed as cantilever or guided cantilever beams. The weld size and details of a jacket and closure shall meet the requirements of the applicable sketch of Fig. 21.28. Closures of geometries other than those shown in Fig. 21.28 may be used if they satisfy the proof test requirements of paragraph UG-101. Paragraph 9-6 provides rules for the design of openings or penetrations through jackets. These rules require that the opening in the inner vessel be designed in accordance with paragraphs UG-36 through UG-45. However, for the most common type penetration details through the jacket, as shown in Fig. 21.29 , is considered as stayed by the nozzle or closure member and reinforcement calculations are not required. The jacket penetration closure member is sized considering only pressure membrane loading; however, other applicable loadings shall be considered. Paragraph 9-6(d)(6) recognizes that localized and secondary bending stresses exist in closure members, and acknowledges that these stresses may exceed the basic code allowable stress. It is generally accepted that the stress categorization and stress criteria of Section VIII, Division 2, Part 5, is appropriate when considering localized and secondary stresses resulting in such configurations. Partial jackets do not fully encompass the circumference of the inner vessel. The rules of Appendix 9 apply to partial jackets and their closures; however, stayed partial jackets must also comply with the stayed construction rules of paragraph UG-47. Partial jackets that do not use staybolt construction may be used if the geometry meets the proof test requirements of UG-101. The fabrication and inspection of jacketed vessels shall be in accordance with the applicable requirements of Subsections A and B. If only the inner vessel is subjected to lethal service, the requirements of UW-2 shall apply to the welds of the inner vessel

21.7.8

Appendix 8: Methods for Liquid Penetrant Examination (PT)

Appendix 8 defines the procedures to be used whenever examination by the liquid penetrant method is specified in the other sections of Section VIII, Division 1. This appendix must be used in conjunction with Article 6 of Section V of the ASME Boiler & Pressure Vessel Code for the detailed requirements and procedures to be followed when performing a liquid penetrant examination. Each liquid penetrant examiner must have sufficient vision to be able to read a Jaeger Type No. 2 Standard Chart at a distance of not less than 12 in. and be able to discern the contrast between the colors used. The examiner must have his vision checked annually. The examiner must also be competent in the techniques used for liquid penetrant examination. The Manufacturer must certify that each examiner meets these qualifications. An indication revealed by the liquid penetrant examination is considered to be relevant if it has a major dimension greater than 1/16 in. A linear indication is one where its length is greater than three times the width. Any relevant linear indication detected by liquid penetrant examination is not acceptable. Relevant rounded indications greater than 3/16 in. are not acceptable, and four or more relevant rounded indications in a line separated by 1/16 in. or less are not acceptable. Any unacceptable indication shall be removed or reduced to an acceptable size. If repair welding is not required, the repaired area shall be blended into the surrounding area to avoid sharp corners, notches, or crevices. If weld repair is required, the area shall be cleaned and welded with a qualified welding procedure. However, before welding, removal of the defect shall be confirmed using a suitable method. After a defect has been repaired, the surface shall be reexamined by the liquid penetrant method and by all other examination methods originally required for the repaired area.

21.7.9

Appendix 9: Jacketed Vessels

Jacketed pressure vessels are used to provide a space for the circulation of a fluid on the outside of the vessel for heat transfer into or out of the primary vessel. Jackets are commonly used for vessels that have an exothermic or endothermic reaction occurring within the primary vessel where heating or cooling is required for the vessel contents. Jacketed vessels may also be used to provide a sealed insulation chamber for the inner vessel. The rules for jacketed vessels include the outer shell of the jacket, the shell of the inner vessel, the jacket closures, stiffeners and bracing which carry pressure loads, and openings through the jacket. Pressure vessel jackets with an internal design pressure of 15 psi or less are not required to be included within the scope of Division 1. However, if any combination of jacket internal pressure and vacuum in the

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 104

104 · Chapter 21

FIG. 21.27 SOME ACCEPTABLE TYPES OF JACKETED VESSELS (Source: Fig. 9.2 of Section VIII DIV. 1 of the ASME Code)

and the closure welds attaching the jacket to the inner vessel. The welds that attach the jacket to the inner vessel do not require radiography and may be fillet welded as allowed in Fig. 21.28. Postweld heat treatment shall be done if required by paragraph UCS-56.

21.7.10

Appendix 10: Quality Control System

It is a requirement that any Manufacturer with an ASME Code Certificate of Authorization must have a written quality control program that document how the code requirements will be met. The quality control program must be tailored to each Manufacturer's circumstances and operation. As part of the ASME accreditation process, the Manufacturer's quality control program shall be available for review by the Authorized Inspector, the ASME Designee, or an ASME-designated organization. Appendix 10 provides definition of features to be included in the written description of the Manufacturer's quality control system. The required information shall include (1) the authority and responsibility of those in charge of the quality control program; (2) how drawings and design calculations are controlled; (3) description of material receiving, identification, and segregation program;

(4) description of examination and inspection program; (5) description of how nonconformities are corrected; (6) confirmation that welding is conducted in accordance with Section IX requirements; (7) description of nondestructive examination procedures used; (8) description of controls that are used to properly control postweld heat treatment; (9) description of the calibration program for examination and test equipment; (10) description of records retention program for the maintenance of radiographs and Manufacturer's Data Reports; and (11) examples of forms used in the quality control program with appropriate reference to these sample forms. With the 2007 Edition, manufacturers will now be required to retain extensive construction records for a period of 3 years. The purpose of this revision according to 10-13 was to accommodate a request from ASME Team Leaders that conduct the triennial audits of certificate holders to have audit access to a minimum of 3 years of construction records. Prior to this revision, a simple demonstration vessel and supporting documentation were the only material readily available to a Team Leader to conduct

ASME_Ch21_p027-111.qxd

2/20/09

2:04 PM

Page 105

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 105

FIG. 21.28

SOME ACCEPTABLE TYPES OF JACKET CLOSURES (Source: Fig. 9-5 of Section VIII Div. 1 of the ASME Code)

his audit. The records that must now be maintained include the following: (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) Manufacturer's Partial Data Reports Manufacturing drawings Design calculations Material Test Reports WPS and PQR records Welder requalification records RT examination and UT reports Repair procedures and records Process control sheets Heat treatment records and test results Postweld heat treatment records Nonconformances and dispositions Hydrostatic test records

year for a period of 1 year. For manufacturer of mass-produced vessels under UG-90(c)(2), the records for six representative vessels produced per year shall be retained for a period of 3 years.

21.7.11

Appendix 11: Capacity Conversions for Safety Valve

For manufacturers of UM-stamped vessels, the records listed above need only to be retained for six representative vessels per

Relief valves and safety values typically have their capacity rated with steam or air as the flowing medium. The capacity of the safety or relief device in terms of a gas or vapor actually present in the intended service (if different than that which the device was originally rated) shall be determined by the capacity conversion equations of Appendix 11. For example, a relief valve to be used in a hydrocarbon service may have its capacity determined with steam. The rated capacity, in terms of steam flow, does not correctly define the amount of hydrocarbon that the valve is capable of discharging. The rated steam capacity must be converted into the rated capacity for the service fluid. Capacity conversion equations are provided for devices that are rated on the basis of

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 106

106 · Chapter 21

FIG. 21.29 SOME ACCEPTABLE TYPES OF PENETRATION DETAILS (Source: Fig. 9-6 of Section VIII Div. 1 of the ASME Code)

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 107

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 107

air or steam flow. The capacity conversions are a function of the valve's coefficient of discharge (K), a constant (C) that is a function of the ratio of specific heat at constant pressure (cp) to the specific heat at constant volume of the gas (cv), the molecular weight (M), actual discharge area (A), and the absolute temperature of the gas at the inlet (T). Interested readers are referred to Appendix 11 where examples are provided for how the capacity conversions are used.

21.7.12 Appendix 12: Ultrasonic Examination of Welds (UT)

Appendix 12 defines the procedures to be used whenever ultrasonic examination is specified in the other sections of Section VIII, Division 1, for welds. This appendix must be used in conjunction with Article 5 of Section V of the ASME Boiler & Pressure Vessel Code for the detailed requirements and procedures to be followed when performing ultrasonic examination. Each person who performs or evaluates ultrasonic examinations must be qualified and certified in accordance with their employer's written practice. SNT-TC-1A [33] shall be used as a guide for the qualification and certification of personnel doing ultrasonic examinations. Alternatively, the ASNT Central Certification Program (ACCP) or CP-189 may be used to fulfill the examination and demonstration requirements of SNT-TC-1A and the employers written practice. Any indication that is characterized as a crack, lack of fusion, or incomplete penetration is not acceptable regardless of length. Other imperfections are not acceptable if they exceed the reference level amplitude and have a length exceeding 1/4 in. for weld thickness up to 3/4 in., or 1/3 of the weld thickness for thicknesses from 3/4 to 21/4 in., or 3/4 in. for a weld thickness greater than 21/4-in. The Manufacturer is required to prepare a report of ultrasonic examination. This report must be retained as per the requirements of Appendix 10, 10-13. This report shall contain all the information required by Section V. Any repaired areas shall be noted along with the results of the reexamination. A record must be maintained of all reflections that exceed 50% of the reference level.

21.7.13 Appendix 13: Vessels of Noncircular Cross Section

Appendix 13 provides rules for the design of noncircular pressure vessels. These rules are applicable for vessels with rectangular or obround cross sections that may, or may not, have their sidewalls stayed, stiffened, or reinforced. All applicable requirements found in the other sections of Division 1 apply to noncircular vessels. The walls of noncircular vessels are subjected to the combined action of tensile and bending stresses resulting from the applied pressure. The design rules are derived from a frame analysis. By equating the rotation and deflection of each of the sidewalls at their junction, the resulting equations can be solved for the shear bending moments at the corner of the vessel. These shear loads and bending moments may then be used to determine the stresses in the sidewalls. The stresses in the flat sidewalls of a noncircular vessel consist of bending and tension resulting from pressure loads. The general membrane stress in the noncircular vessel shall not exceed the maximum allowable stress times any applicable butt weld joint efficiency, or 1.0 SE. The allowable stress for combined membrane and bending stress is limited to 1.5 SE in the flat plate section of rectangular vessels, where S is the maximum allowable stress at design temperature and E is the butt weld joint

efficiency when applicable. It should be noted that the value of E applies only to the butt welds in the side plates and is determined in accordance with Table 21.1 based on the extent of radiography of the joint. Joint efficiencies do not apply to nonbutt-welded Categories C and D joints. (Corner joints that attach two sides of a rectangular vessel are Category C weld joints.) The applicable weld sizing rules control the stresses in these joints and the joint efficiency is not a consideration. Because the stresses in noncircular vessels consist of bending stresses and membrane stresses, the proper algebraic sign of the stresses must be used for their proper combination. The sign convention used in Appendix 13 is defined in paragraph 13-4(c)(1). The rules for the design of end closures of noncircular vessels are not provided in Appendix 13. Other applicable sections of Division 1 shall be used to design these closures. Rules for the design of closures may include those presented in paragraph UG34 for flat covers, paragraph UG-101 for proof testing, or paragraph U-2(g). Openings in noncircular vessels are designed using the ligament efficiency method. The area replacement method is not described or considered in Appendix 13. The membrane and bending ligament efficiencies are applied for those sections with holes. If the ligament efficiencies are equal to or greater than the joint efficiency of a butt joint in the section, then the stress in the section is based on its gross area using the joint efficiency of the butt weld. If the ligament efficiencies of the openings are less than the joint efficiency of a butt joint in the section, then the stresses in the section are determined using the gross area, using E 1.0. These stresses then are divided by the ligament efficiencies to derive the stresses based on the net area. If there is no butt weld in the section, then the bending stress and membrane stress, based on the gross area, are divided by their respective ligament efficiency to compare with the allowable stress. If sidewall openings have multiple diameters, then equivalent ligament efficiencies must be determined. Paragraph 13-6 provides rules for the determination of ligament efficiencies for multidiameter holes subject to membrane and bending stresses. When subjected to external pressure, rectangular vessels with square corners are not reinforced, stayed, or stiffened and must comply with the rules of paragraph 13-14. All other noncircular vessels shall be designed for external pressure by using the provisions of paragraph U-2(g). Paragraphs 13-15 through 3-18 provide example problems that demonstrate the use of the design rules of Appendix 13 for noncircular vessels. These examples should be carefully reviewed for the proper application of the rules of this appendix.

21.7.14

Appendix 14: Integral Flat Heads with a Large, Single, Circular, Centrally Located Opening

Flat heads that have a single, circular, centrally located opening that exceeds one-half of the head diameter must be designed in accordance with Appendix 14. The rules of this appendix apply to shell-to-flat head joints of integral construction. This means that the joint shall be a butt weld or a full penetration weld through the shell or head thickness. The flat head may have a nozzle integrally attached at the central opening or cover may close the opening. (Alternatively, a nonintegral nozzle connection may be used if the design rules for an opening without an attached nozzle or hub are used.) The procedure requires that the design methods of Appendix 2, with some modification of certain terms and definitions, be used to determine the stresses of the assembly. Then, the

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 108

108 · Chapter 21

stresses that are determined from the Appendix 2 rules are appropriately modified to be applicable to this specific type geometry. The modified stresses in the flat head with the central opening must meet the stress criteria of paragraph 2-8 of Appendix 2. Thus, the analysis parallels the analysis of flanges given in Appendix 2, and Appendix 2 must be used to determine the required information used in Appendix 14. The analysis procedure of this appendix ignores stiffening effect of the attached shell in determining the stresses in the assembly. This assumption results in a conservative estimation of the stresses in the assembly.

(3) gas­metal arc spot welding with added filler metal made without a hole in either member (4) gas­tungsten arc seam welding without added filler metal; (5) plasma arc seam welding without the addition of filler metal; (6) submerged arc seam welding with filler metal obtained from the electrode and shielding provided by the flux (7) laser beam seam welding without the addition of filler material The welds in dimpled or embossed assemblies are considered as Category C joints for determining special requirements and the degree of inspection. Dimpled or embossed assemblies that are attached together by using fillet welds around holes or slots in one of the members are not included in the scope of Appendix 17. Paragraph UW-19(c) provides rules for the construction of filletwelded embossed assemblies. The maximum allowable working pressure of embossed or dimpled assemblies is determined from a proof test conducted in accordance with paragraph UG-101 (with a value of E 0.8 for the weld joint efficiency). For assemblies using a plain plate (not dimpled or embossed), the maximum allowable pressure is also calculated on the basis of stayed construction, and the MAWP is the lesser of the calculated value and that obtained from a proof test. The quality control of dimpled or embossed assemblies is achieved by (1) limiting the variation of the fabricated assemblies with respect to the geometry that was proof tested; and (2) conducting destructive tests, peel tests, or tension tests, or etch tests, on samples representing each production run.

21.7.15

Appendix 16: Submittal of Technical Inquiries to the Boiler & Pressure Vessel Committee

This appendix defines the procedures to be used for code users to communicate with the ASME Boiler & Pressure Vessel Committee. Submittals to the committee may be made to request a revision to existing Code rules, to request new or additional Code rules, to request a Code Case, or to request a Code Interpretation. Code Cases represent alternatives or additions to existing Code rules. The most common example of a Code Case is the adoption of a new material specification for Code construction. Code Interpretations provide clarification of the meaning of published Code rules. Interpretations are applicable for the Code Edition and Addenda in force at the time of its publication, and remain valid so long as the code paragraph being interpreted is not later revised. However, users should employ caution when using older interpretations to be sure that it is still applicable to the current code. In cases where the code text does not fully convey the meaning that was intended, an Intent Interpretation will be issued. Any Interpretation with the word "intent" in its Question means that a subsequent revision was made to support the answer provided in the Intent Interpretation. The committee publishes Code Cases and Code Interpretations. All inquiries to the committee should follow the format and guidelines presented in Appendix 16 in order to receive proper consideration. It should be noted that the committee does not endorse, approve, or certify any proprietary or specific design.

21.7.17

Appendix 18: Adhesive Attachment of Nameplates

21.7.16

Appendix 17: Dimpled or Embossed Assemblies

Dimpled or embossed assemblies (some types are shown in Fig. 21.30) represent a particular type of jacketed vessel used in applications where heat transfer is required between a fluid flowing through the dimpled assembly and a fluid outside of the assembly. Embossed or dimpled assemblies constructed in accordance with Appendix 17 shall not be used in lethal service, or in vessels that are unfired steam boilers or subjected to direct firing. The dimpled or embossed assembly may be made from sheets or plate formed prior to welding the pieces together, or may be formed by hydraulic or pneumatic pressure after the sheets are welded. Note that the 1/16 in. (1.5 mm) minimum plate and sheet thickness specified in UG-16(b) does not apply to dimpled or embossed assemblies. As per 17-4, the nominal thickness for plate shall not be less than 0.045 in. (1.1 mm). The rules of Appendix 17 are applicable to welding processes that are "weld through" in which the welding is done by penetrating through one or more members, but not penetrating through another. The welding processes that are covered by this appendix include those made with automatic or semiautomatic machines using the following processes: (1) resistance spot welding (2) resistance seam welding

There are pressure vessels where it is not desirable to attach a code nameplate or nameplate bracket by welding. For example, welding a nameplate or a bracket for a nameplate to the shell may damage pressure vessels made of very thin material. In such cases it may be possible to attach the code nameplate by using an adhesive system. When code nameplates are to be attached to the vessel using adhesives, the restrictions and procedures given in Appendix 18 shall apply.

21.7.18

Appendix 19: Electrically Heated or Gas Fired Jacketed Steam Kettles

Appendix 19 provides additional requirements for pressure vessels that have jackets containing steam that is generated electrically or by gas firing. The steam generated in the vessel jacket shall not be withdrawn from the jacket for external use, and the operating pressure of the jacket shall not exceed 50 psi. Any stainless steel components in contact with the products of combustion shall be either low carbon (e.g., TP 304L) or stabilized grade (e.g., TP 321 or TP 347) in order to minimize the effects of high temperature carbide precipitation at the grain boundaries. Categories A and B welded joints shall be Type 1 of Table UW-12. The steam jacket must have a pressure relief device that has a capacity that is based on the maximum possible heat input into the system. The jacket must be furnished with appurtenances to assure proper control and monitoring of the water/steam system. These include (1) a pressure gauge; (2) a water gauge glass or, for immersion electrical heaters, a low water level warning light;

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 109

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 109

FIG. 21.30

TYPES OF DIMPLED AND EMBOSSED ASSEMBLIES (Source: Fig. 17-3 of Section VIII Div. 1)

(3) a connection, fitted with a stop valve, for venting and/or adding water to the jacket; (4) automatic controls for maintaining the heat input to keep the steam pressure in the jacket below the pressure relief device setting; and

(5) a low water cutoff to shut off the source of heat if the water in the jacket drops below the minimum specified level; Use of this appendix shall be noted in the "Remarks" section on the Manufacturer's Data Report.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 110

110 · Chapter 21

21.7.19

Appendix 20: Hubs Machined from Plate

If a lap joint stub end, tubesheet or flat head is to be welded to its adjacent component using a butt weld, then the tubesheet or flat head is required to have a hub as shown in Fig. 21.11. A weld between a cylinder to tubesheet or flat head is a Category C joint. Certain service conditions require Category C joints to be butt welds [see UW-2(a)]. Normally, lap joint stubs, tubesheets, and flat heads with hubs for butt-welded connections are forged because of the superior through thickness properties of forged products compared to plates. This is important because the stresses imposed by the hub are acting in the through thickness direction of the material. Ordinary plate material is weaker and less ductile in the through thickness direction than in the transverse directions. Because the hub adjacent to the butt weld can be highly stressed in its longitudinal direction, plate material is not allowed for this type of construction unless the special provisions of this appendix are satisfied. When plates with hubs are used, Appendix 20 requires the plate material to have through thickness properties that are equivalent to those specified by the plate material specification. This means that the through thickness strength and ductility must satisfy the same requirements as a specimen taken from the transverse directions. This is normally achieved only by using special methods to produce the plate material. Electroslag remelt and vacuum arc remelt are steel production methods that result in enhanced through thickness properties of plate. The through thickness mechanical properties are determined using specimens taken from the plate to be used as defined in paragraph 20-2. It is possible for plate material to have laminations introduced by the rolling process. Laminations would significantly reduce the strength of the hub section of a tubesheet or flat head and must be avoided. Accordingly, stringent nondestructive examinations are required for hubs machined from plate. The plate must be ultrasonically examined before and after machining using straight beam UT from two directions. The hub must have its surfaces examined using liquid penetrant or magnetic particle inspection, as applicable, before welding. After welding, at least 1/2 in. of the hub and the attachment weld must be 100% radiographed or ultrasonically examined. It is only after all these stringent requirements are satisfied that hubs for butt welds attaching lap joint stub ends, tubesheets, and flat heads can be machined from plate material. Use of this appendix shall be noted in the "Remarks" section on the Manufacturer's Data Report.

Use of this appendix shall be noted in the "Remarks" section on the Manufacturer's Data Report.

21.7.21 Appendix 22: Integrally Forged Construction

This Appendix is applicable to the design, fabrication, and inspection of specially integrally forged vessels that use a higher allowable stress value than allowed for vessels under Part UF. The appendix is only applicable for the following material: SA-372 Grade A, B, C, or D SA-372 Grade E, Class 55, 65, or 70 SA-372 Grade F, Class 55 or 70 SA-372 Grade G, Class 55 or 70 SA-372 Grade H, Class 55 or 70 SA-372 Grade J, Class 55, 65, or 70 SA-372 Grade L SA-372 Grade M, Class A or B. The maximum allowable stress may be taken as the minimum tensile strength specified in the material specification for the grade used in the vessel divided by three (as opposed to 31/2 for normal design applications). The inside diameter is limited to 24 in. The maximum design temperature shall not exceed 200°F. A forged vessel using the enhanced allowable stresses allowed by this appendix shall be of a streamlined design. This means that the shell portion of the vessel cannot have any stress risers, such as openings, welded attachments, or stamped indentations. The heads shall be hot formed and shaped and contoured similarly to that as shown in Fig. 21.31. The center opening in the head shall not exceed the lesser of 50% of the vessel inside diameter or NPS 3. If other openings are required in the head, they shall not exceed NPS 3/4 and placed at a point where the calculated membrane stress is not greater than Su /6. The vessel shall not be welded other than seal welding of threaded connections. After heat treatment, either the magnetic particle or liquid penetrant method shall be used to examine the outside surface of each vessel. The MAWP determined using the increased allowable stress shall be stamped on the thickened portion of the head followed by

21.7.20

Appendix 21: Jacketed Vessels Constructed of Work Hardened Nickel

This appendix applies to jacketed pressure vessels that have the inner shell constructed of nickel sheet or plate meeting SB-162 that has been work hardened by a planishing operation over its entire surface during fabrication. The planishing operation results in an increase in the collapse pressure of the inner shell. Work hardened nickel inner shells of jacketed vessels designed in accordance with this appendix shall meet the following requirements: (1) The maximum size is limited to 8 ft. inside diameter. (2) The maximum operating temperature shall not exceed 400°F. (3) The thickness of the inner shell must be able to withstand a hydrostatic test of three times the jacket MAWP, and the outer shell and head must be able to accommodate this increased test pressure. (4) The thickness of the inner shell must not be less than that determined from the external pressure chart NFA-4 of Section II Part D.

FIG. 21.31 TYPICAL SECTIONS OF SPECIAL SEAMLESS VESSELS (Source: Fig. 22-1 of Section VIII Div. 1 of the ASME Code)

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 111

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 111

the words "Appendix 22." Additionally, the MAWP determined using the allowable stresses from Section II, Part D, shall also be stamped on the thickened portion of the head.

separately on the Manufacturer's Data Report and vessel nameplate? Reply (2): Yes. The procedure requires testing to failure of three full-sized specimens. The minimum pressure that causes visible collapse of any of the three specimens is denoted as B, minimum collapse pressure. Since the actual yield strength of the tested specimens may be greater than the material's specified minimum yield strength, it is possible that the results of the proof test may overestimate the collapse pressure for the material actually used. To account for this possibility, the actual yield strength of the tested specimens must be determined, and the allowable external pressure is reduced by the ratio of SMYS to actual yield stress. Accordingly, the maximum allowable external pressure is determined by Ys B P = Fa b a b 3 Ya where: F (21.30)

21.7.22 Appendix 23: External Pressure Design of Copper, Copper Alloy, and Titanium Alloy Seamless Condenser and Heat Exchanger Tubes with Integral Fins

Paragraph UG-8 requires the maximum allowable external pressure for fined tubes to be based on the minimum thickness at the root diameter of the finned portion of the tube. This is a very conservative requirement for finned tubes since integral fins provide some degree of stiffening against collapse of the tube. Appendix 23 permits the maximum allowable external pressure to be determined by proof test for integrally finned copper, copper alloy, and titanium alloy tubes that comply with SB-359, SB-543, SB-861 or SB-862, respectively. This appendix commonly establishes the maximum allowable external working pressure (MAEWP) for finned tubes used in evaporators in air-conditioning applications. Paragraph 23-4 establishes the criteria for use of these alternative rules. Of note are the temperature limits given for copper and copper alloy finned tubes (150°F, 65°C) and titanium alloy finned tubes (600°F, 315°C). As stated in UG-20(a), the design temperature for a pressure component shall be based on the mean metal temperature expected under operating conditions. For shell and tube heat exchangers, the designer will often use the highest temperature from either chamber as the design temperature for the tubes. But for many applications, this would then preclude the use of Appendix 23, given the temperature limits specified there. Alternatively, the heat exchanger designer may use a mean metal temperature for the tube based on actual operating conditions and determined either via a heat transfer calculation, or direct measurement of temperature for similar hardware and service. Interpretation VIII-1-04-68 illustrates this point further: Interpretation: VIII-1-04-68 Subject: Section VIII, Division 1 (2001 Edition, 2003 Addenda), Appendix 23 Date Issued: July 26, 2005 File: BC03-1826 Question (1): A shell-and-tube heat exchanger utilizes tubes that fall under the scope and other provisions of Appendix 23 of Section VIII, Division 1. As permitted by UG-20(a), the tubes in the heat exchanger are designed based on the maximum mean metal temperature (taken through the tube wall at any point in the bundle and not along the tube length) that is expected under operating conditions. Is it the intent of the code that the Appendix 23 rules can be applied if the heat exchanger is operated such that the tube design temperature so determined is within the maximum limits of the Appendix, but is less than the greater of the design temperatures of the shell side and tube side? Reply (1): Yes, provided either the shell side or tube side maximum design temperature does not exceed the tube design temperature. Question (2): If the reply to Question (1) is yes, shall the tube design temperature as described above be shown

B Ys Ya

factor to account for reduced strength at design temperature, F is the ratio of allowable stress at design temperature divided by the allowable stress at the test temperature. the minimum observed collapse pressure for each of the three specimens; specified minimum yield strength from Section II, Part D at room temperature average actual yield stress of the three specimens at room temperature

As may be seen, a factor of safety of three is applied to the actual collapse pressure. Use of this appendix in lieu of the UG-8 requirements will result in a significant increase in allowable external pressure. When this appendix is applied, the restrictions given in paragraph 23-4 must be satisfied and the Manufacturer's Data Report must indicate its use.

21.7.23

Appendix 24: Design Rules for Clamp Connections

Appendix 24 provides rules for the design of clamped connectors. These connectors are special, often proprietary, designed connections that use clamps held together by bolts to resist the pressure end loads. The design rules of this appendix shall be used for the design of all clamp connectors that fall within its scope. Typical hub and clamp details are provided in Fig. 21.32. The nozzle end or connection hub is machined with a shoulder that a clamp fits over. The clamp may be in two or more pieces connected by bolts at the clamp lug. The clamps must be provided with a bolt retainer or must have sufficient redundancy such that the failure of any one bolt will not result in connection coming apart. Friction between the clamp and hub is not considered to provide adequate redundancy or act as a separate retainer. In order to satisfy these requirements, multiple bolts (two or more bolts per lug) or separate independent pressure end load resisting elements must be utilized in the clamp connection. The design procedure considers the hydrostatic end loads and the loads resulting from assembly and gasket seating. Clamp connectors often utilize a self-energized style gasket that relies on the internal pressure to provide forces on the gasket necessary for maintaining a seal.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 112

112 · Chapter 21

1/ in. (6 mm) min. radius 4

Hub

1/ in. (6 mm) min. radius 4

h

T

h

T

g2 r hn go N g1

A

g2 r hn go N g1

A

B

(a)

B

(b)

hn We hD g1 = go HD

T HG C hT HT G B

(c)

hn We A hD Hp or Hm g1 = go HD

T hG C hT HG

A

Hp or Hm HT

N

N

(d)

G B

Clamp

B [see Note (1)] Bc La

Clamp lug

B [see Note (1)] A

Neutral axis Neutral axis

W/2

A Ci /2

W /2 Cg X C We r

m

Cw

Lh

eb

c

ci

Section A...A (e) (f)

Ct

FIG. 21.32

TYPICAL HUB AND CLAMP (Source: Fig. 24-1 of Section VIII Div. 1 of the ASME Code)

The design procedure for the clamp connector bolting requires consideration of hydrostatic end force resulting from design pressure plus a gasket compressive load to assure a tight joint. This load is designated as Wm1 = 0.637(H + Hp) tan (f - m) where: H Hp (21.31)

f m

clamp shoulder angle angle of friction between clamp and hub surfaces

The bolt load that is required to initially seat the gasket must also be considered in determining the required bolt area. This load is designated as Wm2. Wm2 = 0.637(Hm) tan (f + m) where: Hm (21.32)

total hydrostatic end force total load on the gasket required to maintain a seal (zero for self-energized gaskets)

total load required to initially seat the gasket

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 113

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 113

In order to assure proper preloading of the clamp connection against operating conditions, an assembly bolt load Wm3 shall be determined: Wm3 = 0.637(H + Hp) tan (f + m) (21.33)

Bolts that connect the clamps together must have an area that is at least equal to the greatest of Wm1 divided by the bolt allowable stress at design temperature, Wm2 divided by the bolt allowable stress at room temperature, or Wm3 divide by the bolt allowable stress at room temperature. A review of the equations for Wm1, Wm2, and Wm3 shows that Wm3 will normally govern the design. After the required bolt area is determined, the design bolt loads and moments may be determined for the clamps. The moments used in determining the hub stresses are illustrated in Fig. 21.32. The appendix provides the necessary equations to determine the stresses in the hub (paragraph 24-6) and in the clamp (paragraph 24-7). The allowable stresses that are to be used for the design of clamp connectors are given in Table 24-8 of Appendix 24.

21.7.24 Appendix 25: Acceptance of Testing Laboratories and Authorized Observers for Capacity Certification of Pressure Relief Valves

Paragraph UG-131 requires that the manufacturer of a pressure relief device must have the capacity of the device certified by capacity certification tests. Appendix 25 provides the requirements for ASME acceptance of testing laboratories and Authorized Observers for conducting the capacity certification tests. The test facility and the supervision of the test shall meet the requirements of ASME PTC 25, and the ASME Boiler and Pressure Vessel Committee must accept the testing facility. A representative from an ASME-designated organization (e.g., The National Board of Boiler & Pressure Vessel Inspectors) shall review and accept the laboratory's Quality Control Manual. The quality control manual must clearly establish the authority and responsibility of the personnel in charge of the quality control system; it must include a description of the testing facility, testing arrangements, pressure, size, and capacity limitations. The testing facility that is applying for certification must conduct flow tests of one or more pressure relief devices under the observation of the ASME-designated organization. These devices shall then be retested at a designated ASME-accepted testing laboratory to confirm the test results. Agreement between the two tests shall be within / 2%. If the tests and the comparisons are found to be acceptable, the ASME-designated organization will submit a report that recommends acceptance of the laboratory. If the laboratory is not accepted, the ASME-designated organization must provide a written report to ASME, defining the reasons for rejection.

included within the scope of Appendix 26 may still be used for pressure vessels; however, the provisions of paragraph U-2(g) will apply. A bellows-type expansion joint does not have the ability to resist the hydrostatic end force resulting from the pressure loading. When an expansion joint is used, the pressure end load must be resisted by other elements such as the tube bundle of a heat exchanger, the inner pipe of a double pipe vessel, or external pressure-restraining elements such as stays. The stress caused by the pressure end load in these restraining elements must satisfy the applicable allowable stress at the design temperature as given in Section II, Part D. The rules contained in Appendix 26 are very similar to The Expansion Joint Manufacturer's Association (EJMA) [18] design method. EJMA is an excellent reference for information related to the design and manufacture of bellows-type expansion joints. The design rules for determining the required thickness of reinforced and unreinforced bellows-type expansion joints are based on the EJMA equations. The minimum thickness of a bellows-type expansion joint is given as t Ú P(d + w) S a 1.14 + 4w b q (21.34)

for unreinforced type bellows, and t Ú P(d + w) S a1.14 + 4w b q c R d (R + 1)

(21.35)

for reinforced types. where: t nominal thickness of the bellows after forming Note The relationship between the minimum sheet thickness before forming (tm) and the nominal bellows thickness after forming (t) is given by t = tm c P

1>2 d d (d + w)

internal design pressure

21.7.25 Appendix 26: Pressure Vessel and Heat Exchanger Expansion Joints

Expansion joints are used in pressure vessels and heat exchangers to provide flexibility for thermal motion. Expansion joints must also provide pressure-containing ability. Thin wall bellowstype expansion joints that are used in heat exchangers and pressure vessels must comply with the requirements of Appendix 26. This appendix applies to single ply, bellows expansion joints with a thickness of 1/8 in. or less. Typically, such expansion joints are supplied with multiple convolutions. This appendix provides rules for axial deflection only; it does not provide rules for angular or lateral deflections of the joint. Expansion joints that are not

The other symbols are as shown in Fig. 21.33. When an expansion joint bellows has reinforcing rings to assist in resisting the internal pressure, the term R in the required bellows thickness expression is the ratio of the internal pressure force carried by the bellows to that carried by the rings. This is a function of the relative stiffness of the bellows and the elements comprising the reinforcing rings; thus, R is a function of the relative area and modulus of elasticity of the bellows and rings. Figure 21.34 provides typical details of how expansion bellows may be attached to the cylinder of the pressure vessel. Expansion joint suppliers will normally provide the bellows attached to weld ends so that the vessel Manufacturer may install the joint in the vessel without having to weld the thin bellows material to the pressure shell. The weld end to which the bellows is attached must satisfy all the code requirements, including the minimum required shell thickness including corrosion allowance. The purchaser of the bellows must define to the supplier those requirements that apply to the weld end if they are supplied with the expansion joint.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 114

114 · Chapter 21

FIG. 21.33

SOME TYPICAL BELLOWS-TYPE EXPANSION JOINTS (Source: Fig. 26-1 of Section VIII Div. 1 of the ASME Code)

In addition to the minimum thickness requirements given above, limits on stress are defined for pressure and thermal deflection effects. Equation (1) of paragraph 26-3(a): Scmp ... S (21.36)

assures that the circumferential membrane stress (Scmp) due to internal pressure is not be greater than the maximum allowable stress of the bellows material at the design temperature. This limitation on stress is the basis of the minimum thickness equations given above. Equation (2) of 26-3(a): Smmp ... S (21.37)

material taken at the design temperature. Equations that may be used for Scmp and Smmp are given in 26-3(h). These limits assure that the bellows can resist the primary membrane stress resulting from pressure with a design margin that is comparable to other parts of the vessel. Limits are also placed on the combination of the longitudinal membrane plus longitudinal bending stress (Smbp) resulting from the pressure load: Smmp + Smbp ... KS (21.38)

assures that the longitudinal membrane stress (Smmp) due to pressure is not greater than the maximum allowable stress of the bellows

This combination of stress is limited to KS where K is 1.5 for unreinforced bellows and 3 for reinforced bellows. The membrane plus bending stress in an unreinforced bellows is considered to be a primary stress. As such, the limit of 1.5 S assures that the plastic deformation and failure does not occur from the application of

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 115

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 115

FIG. 21.34 SOME TYPICAL FLEXIBLE ELEMENTS TO WELD END DETAILS (Source: Fig. 26-2 of Section VIII Div. 1 of the ASME Code)

pressure loads. In terms of the stress categories defined in Section VIII, Division 2, Appendix 4, Smmp plus Smbp is considered to be a primary membrane plus primary bending stress, and, consistent with Division 2, is limited to 1.5 S. However, when reinforcing rings are used, the shape of the ring limits the bending stress due to pressure loading. As such, the membrane plus bending stress due to pressure for bellows with reinforcement rings is considered to be a secondary stress and the limit is SPS, which is consistent with the primary plus secondary stress rules of Division 2, Appendix 4. The total stress in the expansion joint, including the longitudinal stress resulting from the deflection of the joint (Smbd), is limited by the required cycle life of the expansion joint. The total stress (Sn): 0.7(Smmp + Smbp) + Smmd + Smbd = Sn (21.39)

stainless steel and other high alloy unreinforced bellows expansion joints, the allowable number of cycles is given by T = P1r 0.85S1 - 0.6P1

(21.40)

If the cycle life is greater than 40,000 cycles, the allowable number of cycles is given by N ... a where: Kg

2.0 4.38 b 18.50KgSn/Eb - 0.02

(21.41)

must result in an allowable number of cycles, N, that is equal to or greater than the required number of cycles defined for the vessel. When the cycle life is less than or equal to 40,000 cycles for

fatigue strength reduction factor which is to account for any stress concentration effects. Kg may be determined by theory, photoelastic methods, or other experimental methods. If the bellows does not have a circumferential joint, Kg may be taken as 1.0.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 116

116 · Chapter 21

Eb

bellows material modulus of elasticity at the design temperature.

Bellows meridional bending stress due to deflection: Smbd = 5E¿ bte 3w2Cd (21.46)

The allowable number of cycles derived by the rules of Appendix 26 differ from the allowable number of cycles derived from EJMA. The fatigue criteria given in EJMA is a best fit of the fatigue failure data taken for cycle testing of actual expansion bellows, and it does not include design margins. The equations given in Appendix 26 are derived from the cycle life test data from the EJMA tests and from other sources. The equations were developed by regression analysis with a design factor applied to the lower bound of the failure points. It is noted that when a expansion bellows is to be installed in a code-stamped pressure vessel, the provisions of Appendix 26 are mandatory. Because the allowable cycle life from EJMA is less conservative than that given in Appendix 26, an expansion joint that complies with the EJMA requirements will not be acceptable for use in a codestamped vessel unless the rules of Appendix 26 are satisfied. It is very important that the user give careful consideration to the required number of cycles specified. The design cycle life should be a realistic estimation of number of operating cycles the bellows will experience. If an overly conservative number of required cycles is provided, the number of convolutions will need to be increased. An increase in the number of convolutions may introduce potential problems with the stability of the joint. Paragraph 26-3(c) requires that the design methods used to determine expansion joint stresses must be substantiated by test. The calculations must be validated by comparing the results to at least five separate tests of flexible elements of the same basic design. The tests must demonstrate that the actual rupture pressure of the bellows is at least three times the maximum allowable working pressure at room temperature. Likewise, the cycle life of the bellows must be demonstrated by test. The design cycle life may not be increased above that allowed by the equations in Appendix 26 regardless of the cycle life test results. These tests are intended to validate the supplier's design methodology and are not required for each joint that is manufactured. The testing procedures given in Appendix 26 are consistent with the requirements of EJMA. Equations for each of the component stresses in a bellows expansion joint are given in 26-3(h); however, the substantiation testing still must be conducted when these formulas are used. The stress equations for unreinforced bellows are given below: Bellows circumferential membrane stress due to internal pressure: Scmp = P(d + w) t a1.14 + 4w b q (21.42)

where Cp, Cf, and Cd are taken from Figs. 26-3, 26-4, and 26-5, respectively. All expansion joints must be free from injurious defects such as nicks, dents, dings, weld splatter, and so on. Visual inspection is required for all bellows to detect such defects. Any suspect area on the bellows shall be further examined using either liquid penetrant or magnetic particle examination. All full penetration butt welds shall be liquid penetrant or magnetic particle examined prior to forming the bellows. The examination shall be repeated after forming on all accessible surfaces. The attachment welds between the bellows and weld ends must 100% examined by liquid penetrant or magnetic particle inspection. Linear indications found by the examinations are considered to be acceptable if dimension does not exceed tm/4 but not more than 0.010 in. All expansion joints must be subjected to a hydrostatic test in accordance with paragraph UG-99. This hydrostatic test may be done as part of the vessel hydrostatic test if the bellows can be visually inspected during the test. Otherwise, the bellows shall be pressure tested prior to installation into the vessel. When the bellows expansion joint is supplied by an organization other than the vessel Manufacturer, a Partial Data Report is required from the bellows Manufacturer [see paragraph UG120(c)]. The bellows Manufacturer must hold a valid Certificate of Authorization from ASME. The required code stamping shall not be placed on the bellows element because of possible damage to the element; the required marking should be placed on the weld end.

21.7.26

Appendix 27: Alternative Requirements for Glass-Lined Vessels

Bellows meridional membrane stress due to internal pressure: Smmp = Pw 2t (21.43)

Bellows meridional bending stress due to internal pressure: Smbp = Pw2Cp 2t2 (21.44)

Glass-lined pressure vessels are required for some services because of the corrosive nature of the fluids being handled. The fabrication of glass-lined pressure vessels require that the vessel be heated to elevated temperatures multiple times during the "glassing" operation. Because of the problems introduced by the repeated heating to elevated temperature during the fabrication of glasslined vessels, Appendix 27 provide alternative requirements that may be used. All applicable requirements of Division 1 are mandatory except as modified by Appendix 27 for glass-lined vessels. Because of the time and temperature that a glass-lined pressure vessel experiences during the glassing operation, it is very difficult to maintain the roundness of the vessel within the permissible out-of-roundness tolerances of UG-80. If the out-of-roundness cannot be corrected, Paragraph 27-2 allows the out-of-roundness of cylindrical shells to exceed the value permitted by UG-80. (Once the glass lining is applied, it is not possible to jack or otherwise reshape the vessel without damaging the lining.) The maximum out-of-roundness, as measured by the difference between any two diameters at any cross section divided by the nominal diameter, is limited to 3% (as opposed to 1% for nonglass-lined vessels). However, the maximum allowable working pressure is reduced by a factor to account for the bending stress introduced into the vessel because of the out-of-roundness. The reduced pressure is determined from the following: P¿ = P 1.25 Sb + 1 S (21.47)

Bellows meridional membrane stress due to deflection: Smmd = E¿ bt e 2w3Cf

2

(21.45)

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 117

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 117

where: Sb = 1.5PR1t (D1 - D2) P t3 + 3 R1R2 a E (21.48)

21.7.27 Appendix 28: Alternative Corner Joint Detail for Box Headers for Air-Cooled Heat Exchangers

This appendix allows for alternative weld details from the requirements of paragraph UW-13(e)(4) and Fig. 21.10 for welded corner joints for box headers of air-cooled heat exchangers. Figure 21.10 requires corner joints, used in the manufacture of noncircular box headers, to have a weld preparation and size such that "a" plus "b" (as defined in Fig. 21.10) is not less than twice the plate thickness. The air-cooled heat exchanger industry was able to demonstrate to the committee that satisfactory fit-for-service welds that did not satisfy the sizing requirements given in Fig. 21.10 could be made using controlled procedures. This appendix replaces only the "a b not less than 2 ts" requirement of paragraph UW-13(e)(4) and the weld joint geometry of Fig. 21.10. All other rules of Division 1 apply to the fabrication of box headers. In order to use the alternative methods given in Appendix 28, the Manufacturer must prepare sample joint to qualify each weld procedure and each welder or operator. The samples are sectioned, polished, and etched to reveal the weld fusion lines. The lines of fusion are used in the calculations required by Fig. 21.35. This figure allows the weld penetration to be used to determine the weld size. Figure 21.35 shows the weld fusion line and geometry requirements when using this appendix. For some double beveled joints, it can be explicitly demonstrated that a2/ts > K without preparing a sample joint [see 28-2(f)(4)]. Explicit weld procedure requirements and essential variables are defined that control the use of the procedures that are used to qualify the corner joint alternative rules. It should also be noted that this appendix may only be used for the construction of box headers of aircooled heat exchangers.

P P E D1 and D2 Ra R1 S Sb t

reduced MAWP (if Sb is less than 0.25 S, then P MAWP) MAWP of the pressure shell modulus of elasticity of shell maximum and minimum inside diameters respectively at the critical section average radius to the middle of the shell wall at the critical section average inside radius at the critical section design stress value from Section II Part D bending stress resulting from the out-of-round condition nominal thickness of vessel shell

Paragraph 27-3 provides requirements for allowing tolerance deviations for hemispherical or 2:1 ellipsoidal heads. The inner head surface cannot deviate outside or inside the specified shape by more than 3% of the vessel diameter. However, a comparative analysis of the distorted shape versus the undistorted shape shall be done to validate that code design margins are maintained [see paragraph U-2(g)]. The reduced MAWP is the value that must be shown on the vessel nameplate and the Manufacturer's Data Report. Because of its fragile nature, the glass lining may be damaged if the vessel is exposed to pressures that are significantly greater than the design pressure. Accordingly, the hydrostatic test pressure for a glass-lined vessel must be at least equal to, but need not exceed, the maximum allowable working pressure to be stamped on the nameplate. Alternatives to the requirements for simulated PWHT mill test specimens of UCS-85 for glass-lined vessels is given in 27-4. (The alternatives in 27-4 do not apply if impact testing of the material is required.) The high temperature and long hold time required for glassing need to be taken into account when the material specification tests are conducted to qualify the material's strength. For example, SA-516 or SA-285 plate material is exempt from simulated PWHT test specimens if, (1) the temperature used in the glassing operation is between 1450 and 1700°F and at least one temperature cycle is above the upper transformation temperature of the material. For this cycle, the vessel is to be held at temperature for 1/2 h/in. of thickness and is cooled in still air to ambient temperature; (2) the carbon content shall not exceed 0.25%; (3) the tensile and yield strength of the material given on the material test reports is at least 10% greater than the minimum specified values by the material specification; and (4) impact testing is not required. Production of glass-lined vessels is a very specialized type of fabrication. The alternative requirements of Appendix 27 represent a collaborative effort between the glass-lined vessel manufacturers and the Code Committee. A vessel manufacturer experienced in the construction of glass-lined equipment is expected to be knowledgeable of the Appendix 27 requirements in addition to other specific techniques and practices required for such equipment.

21.7.28

Appendix 30: Rules for Drilled Holes Not Penetrating Through Vessel Wall

Some pressure vessels require holes to be drilled in the shell to accept instrumentation or other devices. The provisions of Appendix 30 apply to partially drilled (not penetrating through the vessel wall) radial holes in cylinders and spherical shells. The partially drilled holes (PDH) are limited to 2.0 in. in diameter in shells with a diameter to thickness ratio equal to or greater than 10. Figure 21.36 provides the acceptance criterion for the depth of a hole. The depth of the PDH is acceptable if the ratio of the remaining wall thickness to nominal thickness of the shell is above the curve for the specific ratio of the PDH diameter to vessel diameter. Additionally, the following requirements must be satisfied: (1) The minimum remaining wall thickness cannot be less than 0.25 in. (2) The calculated average shear stress in the remaining wall shall not exceed 0.8 S. (3) The center line between any two PDHs or between a PDH and an unreinforced opening must meet the requirements of UG-36(c)(3)(c) and UG-36(c)(3)(d). (4) The PDH shall not be located within the limit of reinforcement of a reinforced opening. (5) The outside edge of the PDH must be chamfered, and, for flat bottomed holes, the bottom corner radius shall be the lesser of 1/4 in. or the hole diameter divided by 4. It is noted that these rules do not apply to studded connections and telltale holes.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 118

118 · Chapter 21

FIG. 21.35 WELD FUSION LINE AND GEOMETRY REQUIREMENTS FOR USING APPENDIX 28 (Source: Fig. 28-1 of Section VIII Div. 1 of the ASME Code)

The rules for of Appendix 30 are based on work done by Sims et al. [19] to establish acceptance criteria for local thin areas of pressure vessels. This paper presents a method for evaluating local thin areas of vessels, piping, and storage tanks. A parametric study was done using a series of elastic/plastic finite element analyses that resulted in an empirical equation to evaluate local thin areas. The evaluation compared the collapse pressure of the shell with a local thin area to the collapse pressure of the shell without a local thin area. The ratio of the collapse pressure of the shell with a LTA divided by the collapse pressure of the shell without a LTA is called the remaining strength factor (RSF). Figure 21.36 is based on an acceptance

criteria of the RSF being 90% or greater. The remaining strength factor may be expressed as t min t 1 t min 1 a1 b m t (21.49)

RSF =

where:

m = 1 + 2.3

t min d 2.3 cAa b d t D

(21.50)

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 119

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 119

FIG. 21.36

THICKNESS RATIO VERSUS DIAMETER RATIO (Source: FIG. 30-1 of Section VIII Div. 1 of the ASME Code)

A = 55 a

t min 4 t min 3 t min 2 b - 168 a b + 189a b t t t t min - 100 a b + 25 t (21.51)

21.7.30

Appendix 32: Local Thin Areas in Cylindrical Shells and in Spherical Segments of Shells

where: tmin /t d /D the ratio of remaining thickness/nominal thickness diameter of PDH/vessel diameter

Figure 21.35 is a plot of the values of tmin /t and d /D, where the RSF value is equal to 0.9.

21.7.29 Appendix 31: Rules for Cr­Mo Steel with Additional Requirements for Welding and Heat Treatment

Appendix 31 covers special fabrication and testing requirements for specifically designated grades of 21/4Cr­1Mo and 3Cr­1Mo material. The appendix is applicable to the materials listed in Table 21.8. The required composition of the weld filler material used to weld these materials is listed in Table 21.9. These rules supplement the rules in the other parts of Division 1. This appendix requires two sets of material tension test specimens. One set of tensile specimens is heat treated to be representative of the maximum anticipated time and temperature of the postweld heat treatment expected for the vessel. The other set is heat treated to be representative of the minimum anticipated time and temperature of the postweld heat treatment expected for the vessel. Charpy V-Notch test specimens are required to be heat treated to simulate the minimum time and temperature expected for the vessel. Typically, Cv absorbed energy decreases as the PWHT time and temperature is reduced. The Charpy V-Notch absorbed energy is required to satisfy 40 ft-lbs average of the three specimens with the minimum of any one specimen not being less than 35 ft-lbs.

During fabrication of a vessel, there are times where the shell thickness may be compromised by gouges, dings, or by grinding. If the area of the reduced shell thickness is localized, then the structural integrity of the vessel is not compromised. Appendix 32 provides an analytical procedure that may be used to evaluate the acceptability of a cylindrical shell and spherical segments of shells that have local thin spots. This allows the Manufacturer to provide a basis for accepting vessels that have local thin areas without the need for weld repair. The methods in Appendix 32 are derived in a similar manner as the requirements that are given in Appendix 30 for partially drilled holes in a shell. This appendix may not be used for forged vessels constructed to Part UF, and may not be applied to vessel with a design temperature in the time-dependent (creep) temperature regime. The information and analysis of local thin areas shall be included in the manufacturer's design calculations, and the information shall be provided to the user when requested. Also, the use of this appendix must be noted on the Manufacturer's Data Report. It is important to have all this information clearly documented in order to facilitate evaluations that may be required for maintenance and in-service considerations.

21.7.31

Appendix 33: Standard Units to be Used in Equations

Design calculations may be made in either the U.S. customary or SI units. The standard dimensional units for each of these systems are given in Appendix 33. It is intent of the code that these units should be used in the equations used for all the design equations. It is the intent that the design should be done using the same unit system for all related calculations. Using the SI and the U.S. customary set of units interchangeably in the design calculations for a pressure vessel is not recommended.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 120

120 · Chapter 21

TABLE 21.8 MATERIAL SPECIFICATIONS (Source: Table 31-1 of Section VIII Div. 1 2007 Edition of ASME Code)

TABLE 21.9 COMPOSITION REQUIREMENTS FOR 21/4CR­1MO­1/4V WELD METAL (Source: Table 31-2 of Section VIII Div. 1 2007 Edition of the ASME Code)

Welding Process SAW SMAW STAW SMAW

C 0.05­0.15 0.05­0.15 0.05­0.15 0.05­0.15

Mn 0.50­1.30 0.50­1.30 0.30­1.10 0.30­1.10

Si 0.05­0.35 0.20­0.50 0.05­0.35 0.20­0.50

Cr 2.00­2.60 2.00­2.60 2.00­2.60 2.00­2.60

Mo 0.90­1.20 0.90­1.20 0.90­1.20 0.90­1.20

P 0.015 max. 0.015 max. 0.015 max. 0.015 max.

S 0.015 max. 0.015 max. 0.015 max. 0.015 max.

V 0.20­0.40 0.20­0.40 0.20­0.40 0.20­0.40

Cb 0.010­0.040 0.010­0.040 0.010­0.040 0.010­0.040

21.7.32

Appendix 34: Requirements for Use of High Silicon Stainless Steels for Pressure Vessels

Appendix 34 was first published in the 2006 Addenda, and reflects the incorporation of Code Cases 1953-4, 2029-3 and 2125-3 for high silicon stainless steel materials. The materials covered by this Appendix have silicon in the range of 3.7­6.0% (18Cr­15Ni­4Si, 18Cr­20Ni­5.5Si, 17.5Cr­17.5Ni­5.3Si). Specific heat treatment, weld procedure qualification, and toughness requirements are given for this material.

21.8

21.8.1

NONMANDATORY APPENDICES

Nonmandatory Appendix A: Basis for Establishing Allowable Loads for Tube-to-Tubesheet Joints

Appendix A provides a basis for establishing allowable loads for tube-to-tubesheet joints. The rules of this appendix are intended to apply for tube-to-tubesheet joints where the tubes act as stays that support or contribute to the strength of other pressure-resisting components. For such applications, the tube-to-tubesheet joint

must be capable of transferring the load from the stay to the member that is stayed. The rules of this appendix are not considered applicable to tube-to-tubesheet joints where the tubes do not act as stays. For example, Appendix A would not applicable to tube-totubesheet joints of U-tube type heat exchangers. Likewise, it is noted that this appendix may be applied to joints that rely solely on welds to resist the loads; however, tube-to-tubesheet joints that rely only on a weld must satisfy the mandatory provisions of paragraph UW-20 for full strength welds or partial strength welds. It is possible that a tube-to-tubesheet joint can resist the applied loads by tube expansion, tube rolling, or a combination of expansion and welding or brazing. It is intended that this appendix be applied when tube expansion or rolling is used either alone or in combination with welding or brazing of the tube-to-tubesheet joints. Some acceptable combinations of welded/brazed or mechanical joints are shown in Fig. 21.37. It is very difficult to analytically establish the strength of a tubeto-tubesheet joint. The strength of the joint depends on the tube and tubesheet material's relative strengths, moduli of elasticity and coefficients of thermal expansion, extent of tube expansion, the pitch of the tubes and the ligament between adjacent tube holes,

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 121

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 121

FIG. 21.37 SOME ACCEPTABLE TYPES OF TUBE-TO-TUBESHEET WELDS (Source: Fig. A-2 of Section VIII Div. 1 of the ASME Code)

coefficient of friction between the tube and tubesheet, and so on. Accordingly, the joint efficiency of the tube-to-tubesheet joint is established either by empirical, experience-based factors or mockup testing of the joint. Paragraph A-2 defines that the maximum allowable axial load in a tube-to-tubesheet joint is Lmax At Sa fr (21.52)

Where:

At is the nominal cross-sectional area of the tube. Lmax is the maximum allowable axial load (in either direction) on the tube-to-tubesheet joint. fr is a factor for the efficiency of the joint. The value of fr may be established by test or the value listed in Table 21.10 under the "no test" column may be used. fe is a factor to account for the depth of tube expansion in the tubesheet. fy is a factor to account the difference in strength between the tube and the tubesheet when tube expansion is used. fT is a factor to account for the increase or decrease of tube joint strength due to radial differential expansion between the tube and the tubesheet.

for joints that do not depend on tube expansion for strength, Lmax At Sa fe fr fy (21.53)

for joints that do depend partly on tube expansion for strength, and Lmax At Sa fe fr fy fT (21.54)

for joints that depend only on tube expansion for strength.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 122

122 · Chapter 21

The value of fT is determined by calculating the ratio of the interface pressure between the tube-to-tubesheet at the operating temperature divided by the initial interface pressure remaining after tube expansion during fabrication. It is noted that a tube-totubesheet joint becomes "tighter" at temperature if the tube has a greater coefficient of thermal expansion than that of the tubesheet. Conversely, the joint becomes weaker at temperature when the tube coefficient of thermal expansion is less than that of the tubesheet. Appendix A does not provide a method to determine the actual interface pressure but allows the designer to use analytical or experimentally derived values. In order to prevent unacceptable deformation of the tube or tube hole, the maximum interface pressure occurring during operation between the tube and tube hole shall not exceed 58% of the smaller of the tube or the tubesheet specified minimum yield strength. For a tube-to-tubesheet joint where all the factors are 1.0, the required value of Lmax is such that the joint must be capable of resisting the tube axial load that causes the tube stress to be equal to its allowable stress at design temperature. This assures that the tube-to-tubesheet joint can carry the load that develops the full strength of the tube. When the factors are not equal to 1.0, then the maximum allowable load of the joint is reduced according to the value of the appropriate factors. The efficiencies for "fr (no test)" given in Table 21.10 may be used without any qualification. However, if larger values of fr are desired, the values given by the "fr (test)" may be used if they are qualified by the shear load test described in paragraph A-3(a) through (j). This test requires that mock-up of at least three tubeto-tubesheet joints be tested to failure. The smallest load to cause

any of the three joints to fail is used to determine the joint efficiency as given by fr(test) or fr(test) as applicable; where: L(test) ST minimum axial load to cause failure of the specimen specified minimum tensile strength of the material L(test)/At ST fe fy (21.56) L(test)/At ST (21.55)

However, the joint efficiency used to determine the maximum allowable tube loading must be the smaller of the value determined by the shear load test or the value given in Table 21.10 under the "test" column, unless additional testing as per paragraph A-3(k) is done. The use of paragraph A-3(k) will allow larger values of fr to be used if at least nine joints are tested to failure, and statistical principles are used to condition the data. For these cases, the tube load used to determine the joint efficiency is taken as the­2-sigma standard deviation from the mean of failure loads for all the joints. A joint efficiency is never allowed to be greater than 1.00. It should be noted that the shear load test, in effect, requires the tube to fail before the joint fails if a joint efficiency of 1.0 is to be used. If the joint fails prior to the tube, the value of fr(test) will be less than one.

TABLE 21.10

EFFICIENCIES FR (Source: Table A-2 of Section VIII Div. 1 2007 Edition of the ASME Code)

Type Joint a b b-1 c d e f g h i j k

Description (1) Welded only, a Ú 1.4t Welded only, t ... a 6 1.4t Welded only, a 6 t Brazed, examined Brazed, not fully examined Welded, a Ú 1.4t, and expanded Welded, a 6 1.4t, and expanded, enhanced with two or more grooves Welded, a 6 1.4t, and expanded, enhanced with single groove Welded, a 6 1.4t, and expanded, not enhanced Expanded, enhanced with two or more grooves Expanded, enhanced with single groove Expanded, not enhanced

Notes (3) (3) (4) (5) (6) (3) (3)(7)(8)(9) (3)(7)(8)(9) (3)(7)(8) (7)(8)(9) (7)(8)(9) (7)(8)

fr (test) (2) 1.00 0.70 0.70 1.00 0.50 1.00 0.95 0.85 0.70 0.90 0.80 0.60

fr (no test) 0.80 0.55 ... 0.80 0.40 0.80 0.75 0.65 0.50 0.70 0.65 0.50

General Note: The joint efficiencies listed in this Table apply only to allowable loads and do not indicate the degree of joint leak tightness. Notes: (1) For joint types involving more than one fastening method, the sequence used in the joint description does not necessarily indicate the order in which the operations are performed. (2) The use of the fr (test) factor requires qualification in accordance with A-3 and A-4. (3) The value of fr(no test) applies only to material combinations as provided for under Section IX. For material combinations not provided for under Section IX, fr shall be determined by test in accordance with A-3 and A-4. (4) For fr(no test), refer to UW-20.2(b). (5) A value of 1.00 for fr (test) or 0.80 for fr (no test) can be applied only to joints in which visual examination assures that the brazing filler metal has penetrated the entire joint [see UB-14(a)] and the depth of penetration is not less than three times the nominal thickness of the tube wall. (6) A value of 0.50 for fr (test) or 0.40 for fr (no test) shall be used for joints in which visual examination will not provide proof that the brazing filler metal has penetrated the entire joint [see UB-14(b)]. (7) When d0>(d0 - 2 t) is less than 1.05 or greater than 1.410, fr shall be determined by test in accordance with A-3 and A-4. (8) When the nominal pitch (center-to-center distance of adjacent tube holes) is less than d0 + 2 t, fr shall be determined by test in accordance with A-3 and A-4. (9) The Manufacturer may use other means to enhance the strength of expanded joints, provided, however, that the joints are tested in accordance with A-3 and A-4.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 123

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 123

A tube-to-tubesheet joint may be roller expanded, hydraulically expanded, or explosion expanded (or a combination thereof). The depth within the tubesheet to which the tube is expanded has an influence on the strength of the joint. If the depth of expansion is less than the tube outside diameter, the value of fe is determined by the ratio of the expanded depth divided by the tube outside diameter. When the expanded depth is greater than the tube outside diameter, the expansion is considered to be fully effective and fe may be taken as 1.0. Also, if the tube holes are grooved or otherwise fully enhanced, fe may be taken as 1.0. When a tube is expanded into a tubesheet of weaker material than the tube, the joint cannot develop the full strength of the tube, accordingly, the allowable load, L max, is reduced by the ratio of the tubesheet yield stress divided by the tube yield stress. The value of fy cannot be greater than 1.0, and when fy is less than 0.6, a qualification test must be conducted. When a tube-to-tubesheet joint is expanded, temperature effects can decrease its strength. If the coefficient of thermal expansion of the tubesheet is greater than that of the tube, the interfacial pressure between the tube and tubesheet is diminished as the temperature increases, and this weakens the joint. Conversely, if the tube's coefficient of thermal expansion is much greater than the tubesheet, the interfacial pressure is increased resulting in a "tighter" joint as the temperature increases. However, if the interface pressure increases too much, the tube may yield at the operating temperature and the joint will not have the same strength when the temperature is reduced. High temperature can also have an effect of relaxing the residual stress in the joint if the temperature is in the creep regime of the tube or tubesheet material. All these factors must be carefully considered for tube-to-tubesheet joints at operating temperature. These factors are especially important when the joint is expanded only with no weld to carry a portion of the load. Rules that consider these effects are in development by the committee to establish allowable operating temperatures for expanded-only joints. There are provisions in Appendix A for determining the acceptability of an operating temperature for an expanded-only tube-totubesheet joint. This procedure requires two sets of mock-up specimens. One specimen is tested at the operating temperature, and the other is tested at ambient temperature after a heat soak at the operating temperature for at least 24 h. The operating temperature is acceptable for the expanded-only joint if failure load develops sufficient strength as determined below: A t ST fe fy , where L1(test) is the lowest axial load to L1(test) cause failure at the operating temperature, and A t ST fe fy , where L 2(test) is the lowest axial load to L2(test) cause failure of the heat soaked specimens.

temperature should be conducted over a sufficient length of time to accurately represent all significant operating conditions.

21.8.3

Nonmandatory Appendix D: Suggested Good Practice Regarding Internal Structures

Many pressure vessels contain internal components that are supported from the vessel wall. Some of the internal components must support large weights such as catalyst beds, packing material, trays, baffles, and so on. The design of the vessel and the internal components should consider the effects of these components (see paragraph UG-22). The user should specify an appropriate corrosion allowance for all internal components, their supports, and any welds that attach their supports to the vessel wall. The following suggestions are provided: (1) Connections to the vessel wall should not produce excessive tensile stress outward from the vessel wall. Plate material is generally weaker and less ductile in the through thickness direction than in the direction of rolling and transverse to the direction rolling. Loads that produce large tensile stress through the thickness of the plate can pull the plate apart resulting in failure of the vessel. (2) Structures should rest on their supports as opposed to being hung. If the components are hung, a failure of the connection will allow the component to fall. If the structure rests on its support, dependency on bolts or other connectors is not required. (3) Required corrosion allowance for the structure need not be the same as the vessel if structures and their supports can be replaced more readily than the vessel. (4) Corrosion resistant material may be used to fabricate structures and their supports. The user should evaluate the function of the structure and the consequences of its failure, the difficulty in replacement, and economic considerations of making the internals from corrosion-resistant material. The effect of any differential thermal expansion between the support and the vessel should be considered for vessels operating at elevated temperature.

21.8.4

Nonmandatory Appendix E: Suggested Good Practice Regarding Corrosion Allowance

21.8.2

Nonmandatory Appendix C: Suggested Methods for Obtaining the Operating Temperature of Vessel Walls in Service

Paragraph UG-20 allows the maximum design temperature for a pressure vessel to be established by measurement from similar equipment in service under equivalent operating conditions. This appendix provides guidance of how this may be accomplished. The suggested practice includes the use of three thermocouples (one of which is attached to a head with the warmest temperature) inserted into small holes drilled into the vessel. Likewise, it is acceptable and conservative to measure the operating fluid temperature and assume the maximum design temperature as the same as the maximum operating fluid temperature. Implicit in either of these approaches is that the measurement of the actual

As defined in paragraph U-2(a), the user is responsible to define the corrosion allowance to be applied in the design of the vessel. This appendix provides general corrosion allowance considerations. The need for corrosion allowance depends on the process fluids and the process conditions such as chemical concentration, temperature, pressure, use of inhibitors, and so on. Information in the public domain, such as published by the National Association of Corrosion Engineers (NACE), may be used to estimate corrosion allowance for materials in specific service. However, it should be noted that the published information only provides guidelines and may not reflect the effect of trace chemicals that may have a deleterious effect on the corrosion rate. The best indicator of expected corrosion is the corrosion experience and history of vessels in the same or similar service. When the corrosion effects are known to be insignificant, no corrosion allowance need to be specified for a vessel. If the rate of corrosion can be estimated, either by published information or by actual experience, then sufficient corrosion allowance should be specified to account for metal loss for the desired life of the

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 124

124 · Chapter 21

vessel. When the corrosion rate cannot be established from the literature or from experience, then special precautions may be required for inspecting the vessel at regular intervals to assure its safe operation. Consideration for corrosion should be applied for all parts and their attachment welds that are essential to the strength of the vessel such as stiffening rings that are in contact with the process. It is possible that external corrosion on a vessel can result from the environment. Often, external corrosion can be effectively prevented by paint or coating to protect the vessel from atmospheric corrosion. If a vessel is operating in an environment that is conducive to external corrosion and is not protected, then an appropriate corrosion allowance should be applied. Special provisions should be applied to prevent "corrosion under insulation" for vessels whose operating temperature may allow condensation to collect under the insulation resulting. Typically, this situation will result in significant external corrosion unless a high quality coating protects the vessel. It is strongly recommended that an experienced corrosion engineer be consulted for establishing the appropriate corrosion allowance to be specified for a vessel.

Additionally, consideration should be given to other loadings such as the weight of hydrostatic test water, loadings resulting from lifting and handling the vessel, and so on. Legs, lugs, brackets, or a support skirt may be used to support vertical vessels. Where legs, lugs or brackets are used, localized stresses shall be considered. The projection of the lugs or brackets should be kept close to the shell in order to minimize the bending moment in the shell. It may be required to thicken the shell in the vicinity of the support or attachment by the use of separate reinforcement elements or by locally increasing the shell thickness. Section VIII, Division 1, does not provide specific criteria for the categorization of localized stresses around legs, brackets, and clips. However, if the provisions of Section VIII, Division 2, are applied, it is concluded that the membrane stress in the shell local to the clip or bracket is classified as "local membrane" and the allowable stress is generally taken to be 1.5 S. The localized bending stress near a clip is generally categorized as a secondary stress and the maximum allowable stress range would be considered to be SPS. The following interpretation provides some insight into the code requirements for limiting stresses in the vicinity of brackets and clips. Interpretation: VIII-1-01-70 Subject: Section VIII, Divisions 1 and 2 (1998 Edition, 2000 Addenda), UG-22, UG-23(c), and Appendix 4, 4-112 Date Issued: June 15, 2001 File: BC01-380 Question (1): Does UG-22 of Section VIII, Division 1, require that the local stress in a vessel head or shell at a support gusset, resulting from seismic or wind loadings, be considered in the vessel design? Reply (1): Yes. Question (2): Is the local bending stress in a vessel head or shell at a vessel support gusset, resulting from a mechanical load such as seismic or wind condition, a primary stress as defined in Appendix 3-2 of Section VIII, Division 1? Reply (2): No. Question (3): Does UG-23(c) of Section VIII, Division 1 limit the combination of local bending stress, such as that described in Question (2), plus the general membrane stress resulting from pressure to 1.5 times the maximum allowable stress value in tension? Reply (3): No. Question (4): If the reply to Question (3) is no, what is the appropriate limit for the combination of the local bending stress, as described above, and the general membrane stress due to pressure? Reply (4): Section VIII, Division 1 does not provide rules for the combination of localized bending stresses resulting from mechanical loads at supports with the general primary membrane stress resulting from pressure. See U-2(g). Question (5): If the vessel, as described in the previous questions, was designed in accordance with the rules of Section VIII, Division 2, would the local bending stress at the support gusset be categorized as a secondary stress as defined by Appendix 4, 4-112 of Section VIII, Division 2? Reply (5): Yes.

21.8.5

Nonmandatory Appendix F: Suggested Good Practice Regarding Linings

In order to provide an economical design for vessels in corrosive service, a vessel may be lined with a corrosion-resistant material with the pressure-retaining shell made from a less expensive material that is not corrosion-resistant to the process. Typical types of corrosion-resistant lining include integrally bonded clad plate, weld overlay, spray metal overlay, loose liner or sheets welded to the vessel wall. When such linings are used, their thickness should be determined to give an estimated life of two times the length of the interval of the first inspection after the vessel has been put into service. The base material does not need to have a corrosion allowance applied when corrosion-resistant linings are applied. Special precautions are required for loose liners or strip lining applied by welding. The initial installation must be done on a surface that is free from foreign material, rust, scale, and moisture. Likewise, the fit up of the lining material must closely conform to the shape of the vessel to avoid cracking of the attachment welds during the hydrostatic test or in operation. The effects of vacuum, when in service condition, must be considered in selecting the lining thickness. If the attachment weld were to crack, then the process fluid can leak behind the liner and attack the base material. If there is any fluid trapped behind a strip-lined vessel, then heating the vessel above the trapped fluid's boiling point will result in the lining damaged by the pressure buildup. It is not recommended to consider any kind of paint as a permanent protection against internal corrosion. If a vessel has paint on the inside surface to protect against corrosion, a corrosion allowance should also be provided to the vessel as if the paint was not present.

21.8.6

Nonmandatory Appendix G: Suggested Good Practice Regarding Piping Reactions and Design of Supports and Attachments

Appendix G provides very general guidelines for the design of vessel supports and attachments. Vessels will normally have concentrated loads in the region where the supports are located. Likewise, attachments to the shell of a pressure vessel can result in localized stresses. The vessel design shall consider all loadings that the vessel may be expected to see during normal operation.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 125

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 125

The design of support skirts must consider where the skirt is attached to the shell. Localized stresses of large magnitude may exist in the shell and skirt material near the circle of skirt attachment if the skirt is not essentially tangent to the vessel wall. When the mean diameter of the skirt and shell are approximately equal, and the shell has a generous knuckle radius, such as a 2:1 ellipsoidal head, experience has shown that the localized stresses in the shell are not objectionable. If the skirt is "lapped" over the outside of the shell, it is also generally accepted that the localized stresses are satisfactory. However, if the skirt mean diameter is considerably less than that of the shell, then a more detailed analysis should be performed. Horizontal vessels may be supported by saddles or equivalent leg supports. Saddles shall extend over at least one third of the vessel, and should be as few in number as possible. It is good practice to limit the numbers of saddles supporting a horizontal vessel to two. Caution is warranted when more than two saddles are used because differential settlement of the individual saddles can result in high local loads on the other supports. When more than two saddles are used, provisions should be incorporated to minimize differential settlement of the multiple supports. One such consideration is to place all multiple saddles on a common concrete pad foundation. Special consideration is often required when rotating or reciprocating equipment is mounted on a vessel. It is important to avoid resonance between the cyclic load frequency and the natural frequency of the vessel. Also, it is important to minimize stress concentrations at the attachment locations to avoid potential fatigue cracks. Several references are provided for determining the localized stresses resulting from supports or attachments to pressure vessels. They include the following: (1) "Stresses in Large Cylindrical Pressure Vessels on Two Saddle Supports," Pressure Vessels and Piping: Design and Analysis, A Decade of Progress, Volume 2, published by ASME (2) British Standard BS-5500, Specification for Fusion Welded Pressure Vessels for Use in the Chemical and Petroleum and Allied Industries (3) Welding Research Council Bulletin #107, Local Stresses in Spherical and Cylindrical Shells Due to External Loadings (4) Welding Research Council Bulletin #297, Local Stresses in Spherical and Cylindrical Shells Due to External Loadings, Supplement to WRC-107 (5) Welding Research Council Bulletin #198, Part 1, Secondary Stress Indices for Integral Structural Attachments to Straight Pipes; Part 2, Stress Indices at Lug Supports on Piping Systems

deflagration event. As noted in paragraphs U-2(a) and UG-22, it is the user's responsibility to define any abnormal events, such as deflagration, that need to be considered in the design of the vessel. Appendix H provides two references that may be used to provide design criteria for overpressure resulting from a deflagration. These are NFPA-69 [27] and Section III, Nuclear Components of the ASME Boiler and Pressure Vessel Code [28, 29]. Each of the referenced documents provides two levels of design acceptability when considering abnormal events. One criterion is that the pressure vessel shall be able to withstand a deflagration without substantial deformation. The other criterion assures that the pressure vessel can withstand a deflagration without rupture. It is the user's responsibility to define which of these criteria are to be applied when a vessel is required to contain overpressure effects of a deflagration. When a vessel is to withstand an overpressure event without substantial deformation, the general primary membrane stress resulting from the pressure produced by a deflagration is limited to the yield strength of the vessel. It is anticipated that vessels designed to this criterion will be able to withstand a deflagration and still be operational afterwards (after appropriate inspections). However, if the deflagration design criteria is to prevent vessel rupture, the pressure resulting from a deflagration can be as large as 2/3 of the pressure required to rupture the vessel. When this criterion is used, it is expected that the vessel may not be usable after the event. NFPA-69 provides for a design factor on burst strength; however, it does not provide a method of determining the vessel burst strength. Methods for determining vessel burst strength are presented in Welding Research Council Bulletin 95 [30]. Either criterion allows a general primary membrane stress in the vessel resulting from this abnormal event (deflagration) that is larger than the basic code allowable stress given in Section II Part D. The decision of which criterion to use should be determined by of the likelihood of occurrence and the consequence of the occurrence. If a deflagration event is expected to be a very remote possibility and the consequence of permanent deformation is acceptable to the user, then the limit that permits the larger pressure may be appropriate. However, if the probability of the deflagration event is greater, or the consequences of permanent vessel deformation are not acceptable, then the criterion that limits the stress to the yield strength of the material is appropriate. When a vessel is specified to provide containment of a deflagration, it is important to avoid details that would result in stain accumulation or concentration. Partial penetration pressure boundary welds, cone-to- cylinder junctions not using knuckle transitions, and large openings in heads or cylindrical shells are details that should be avoided when deflagration is to be considered.

21.8.8

Nonmandatory Appendix K: Sectioning of Welded Joints

21.8.7

Nonmandatory Appendix H: Guidance to Accommodate Loadings Produced by Deflagration

There are some processes where it is possible that a vapor­air or a dust­air mixture can ignite and cause a deflagration within a vessel. A deflagration is the propagation of a combustion zone at a velocity that is less than the speed of sound in the unreacted medium. There are methods that accurately predict the pressure increase in a vessel resulting from a deflagration. This appendix presents guidelines that may be used when a vessel is to be designed to withstand overpressure conditions resulting from a

When the user and Manufacturer have agreed to examine welds by sectioning the joint, as provided for by paragraph UW-41, this appendix provides suggested etching solutions and some acceptable methods of closing the hole from which the sample was taken. It is noted in UW-41 that sectioning welded joints does not influence the joint efficiencies given in Table UW-12 and may not be used as a substitute for spot radiography.

21.8.9

Nonmandatory Appendix L: Examples Illustrating the Application of Code Formulas and Rules

There are numerous examples given in Appendix L that demonstrate the use of some of the formulas and rules of

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 126

126 · Chapter 21

Division 1. The examples and flow charts in paragraph L-1 show how to apply joint efficiencies in shells and heads with welded joints. The examples of L-2 demonstrate thickness calculations for shells under internal pressure with supplemental loadings. Paragraph L-3 presents examples regarding the use of the rules for vessels under external pressure, and paragraph L-4 demonstrates how to determine the maximum out-of-roundness for a cylindrical shell under external pressure. Paragraph L-5 demonstrates how to design a circumferential stiffening ring for a cylindrical shell under external pressure. Paragraph L-6 shows how to determine the required thickness for formed heads with pressure on the convex side. The L-7 examples demonstrate the use of the rules for openings and reinforcement, and L-8 shows how the rules regarding ligament efficiency are applied. Paragraph L-9 shows how to establish the minimum design metal temperature of a vessel. Paragraph L-10 provides examples for determining the size and allowable axial load of full strength tube-to-tubesheet welds for each of the joint types shown in Fig. UW-20.1. Selected examples will be annotated here. Figure 21.38 provides a very useful flowchart that demonstrates the correct application of joint efficiencies and joint types for welded joints in cylinders and cones. If full radiography is required because of the service requirements of UW-2 or because of the thickness requirements of UW-11(a)(2), then all Categories A, B, C, and D butt welds must be radiographed in accordance with UW-51 for their full length, and the joint efficiency is 1.0 for Type 1 welds and 0.9 for Type 2 welds (see Table UW-12). Note that Categories B and C butt welds in nozzles and communicating chambers equal to NPS10 and with a thickness 11/8 in. or less are not required to be radiographed and no penalty applies to their joint efficiencies. If full radiography is not required because of vessel service or weld thickness, the designer can elect the degree of radiography to be applied. If full radiography is selected for components that are not seamless, all Categories A and D butt welds must be examined for their full length in accordance with UW-51 and the joint efficiencies given in Column (a) for Full RT of Table UW-12 apply. However, an option is available for Categories B and C butt welds. Paragraph UW-11(a)(5)(b) makes a provision for Categories B and C welds that intersect the Category A joint to be spot radiographed in accordance with UW-52 without penalizing the joint efficiency of the Category A joint. This allows a cylindrical or conical vessel to be designed with a joint efficiency of 1.0 (when the thickness is governed by pressure) without performing a full radiographic examination of the Categories B and C butt welds. This is logical since the required thickness of a cylinder or cone due to internal pressure is governed by the circumferential stress acting normal to the longitudinal seam. The pressure longitudinal stress acting normal to the circumferential seam (Categories B and C) is one-half the stress acting on the Category A joint. Thus, the joint efficiency of these seams will not govern the thickness required for pressure. In order to assure an acceptable level of weld quality for these welds located in a vessel designed using a joint efficiency of 1.0, the spot radiograph of the Categories B and C welded joints was deemed appropriate by the Code Committee. When the spot radiographic provisions of UW11(a)(5)(b) are applied, the joint efficiency of the Categories B and C joints is taken from Table UW-12, Column (b) for Spot RT, that is, E 0.85 for a Type 1 butt joint. This lower joint efficiency may affect the design when external loads such as wind or seismic are considered. If the designer chooses spot radiography or no radiography, then the joint efficiencies of Table UW-12

Column (b), Spot RT, and (c), No RT, respectively apply to the design of the sections. When a vessel has seamless sections, such as a seamless head, the design requirements become subtler. Paragraph UW12(d) states that seamless vessel parts are to be considered equivalent to welded components with Type 1 Category A joints. The joint efficiency to be used in the design of the seamless component depends on the extent of radiography done on the weld attaching the seamless component to other parts of the vessel. If a seamless head is butt-welded to a seamless cylinder, this attachment weld must be spot radiographed (as a minimum) in order to use a E 1.0 for the design of the cylinder and head. If the joint is not radiographed, then the head and cylinder design must be based on a E 0.85. Example L-1.5.2 illustrates these requirements. Figure 21.39 provides a flow diagram that concisely summarizes the required degree of radiography based on service requirements. As may be seen from this figure and a review of UW-2 and UW-11, vessels designated to be in lethal service, unfired steam boilers, or if greater than the thickness defined in UW-11(a)(2) require full radiography. If the service conditions do not require full radiography and the designer elects to use the joint efficiencies of Table UW-12, Column (a), Full RT, for the design of the vessel components, then the Categories B and C butt welds must, at a minimum, be spot radiographed. If the Category B or Category C butt welds are not radiographed, the components must be designed using the joint efficiencies listed in Column (2), Spot RT, of Table UW-12 even though the Category A seams were fully radiographed. The examples of paragraph L-2 demonstrate use of rules that consider supplemental loadings. Example L-2.1 considers the design of a process column with a liquid hydrostatic head and exposed to an overturning moment due to wind loading. This example demonstrates a number of important points. The first is that the effects of hydrostatic head must be included in determining the required wall thickness due to internal pressure. Another point is that the worst combination of loads must be considered in determining the required thickness of a section. For example, the maximum longitudinal tensile stress at a section results from the tensile bending stress of the overturning moment and internal pressure acting coincidentally, but the maximum compressive stress results from the compressive stress due to the overturning moment with zero internal pressure. Also, the required thickness due to compressive stress resulting from an overturning moment is to be compared to the allowable axial compressive stress given by paragraph UG-23(b) using a weld joint efficiency of 1.0 for all butt welds in compression. There are a number of examples in paragraph L-7 that illustrate the rules applicable to openings and reinforcement. The examples show how to determine the required reinforcement and the amount of reinforcement available for various configurations. However, the most worthwhile illustrations offered by the examples of L-7.2 show how to determine the load strength path and strength of the connection elements. A vessel designer should review these examples in detail prior to conducting opening reinforcement calculations. Paragraph L-9 includes several examples to demonstrate how to determine the governing thickness and establish the MDMT of a pressure vessel. The examples include demonstration of how Figs. UCS-66 and UCS-66.1 are used to determine the coldest allowable MDMT of a carbon steel pressure vessel without the need to conduct impact tests. Likewise, the example presents appropriate considerations regarding the metal temperature

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 127

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 127

FIG. 21.38 JOINT EFFICIENCY AND WELD JOINT TYPE--CYLINDERS AND CONES (Source: Fig. L-1.4-1 of Section VIII Div. 1 of the ASME Code)

during a hydrostatic or pneumatic leak test based on the vessel MDMT.

21.8.10 Nonmandatory Appendix M: Installation and Operation

There are a number of issues that need to be considered when a pressure vessel is installed into the system where it will be in

service. The rules in Appendix M are suggested considerations; however, they do not supersede the requirements that may be imposed by the authorities at those locations where laws have been enacted that make special provisions for the construction, installation and operation of pressure vessels [see paragraph U1(c)(1)]. It is acceptable to use those requirements in Appendix M that may be different, as allowed in the mandatory parts of the

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 128

128 · Chapter 21

FIG. 21.39 JOINT EFFICIENCIES FOR CATEGORIES A AND D WELDED JOINTS IN SHELLS, HEADS, OR CONES (Source: Fig. L-1.4-3 of Section VIII Div. 1 of the ASME Code)

code, only if accepted by the authority having jurisdiction over the installation of pressure vessels. Paragraph M-2 presents general considerations regarding vessels subject to external corrosion. Painting or other appropriate coatings should be used to protect such vessels. If the vessel is not protected, then it should be installed so that all parts of the exterior are accessible for inspection. Vessels subject to internal inspection

should have all manholes, handholes, and inspection openings installed such that they are accessible. For vertical vessels subject to corrosion, the vessel should have a low point drain in the vessel or the piping system to assure that all liquid can be removed from the vessel. Paragraph M-3 requires that the code nameplate be located such that it is accessible and not covered with insulation or other

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 129

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 129

material that is not readily removable. Typically, nameplates are mounted on a bracket that extends beyond the insulation so that the vessel identification and code markings are readily viewed. Also, it is common practice to locate the code nameplate in the vicinity of the most accessible manway or inspection opening on the vessel. Appendix M provides guidance related to the installation and arrangement of stop valves for isolating the vessel and pressurerelieving safety devices. Since large, continuously operating processing facilities run for extended periods of time without a shutdown, it is sometimes necessary to isolate a relief device for its replacement, repair, and maintenance. Guidelines are also provided in M-5 and M-6 regarding the design of the inlet and discharge piping to the pressure relief device. Because the engineer must be able to fit the arrangement and proportions of the relief system to the specific requirements of the installation, the rules presented in Appendix M are general in nature. Paragraph M-5 provides rules for including stop valves between the pressure relief device and the vessel for which it is protecting. Paragraph M-5(a) discusses the situation where the service conditions of a vessel result in pressure being generated within the vessel, such as the pressure resulting from an exothermic reaction. According to the requirements of paragraph UG125 (g), in these instances, the pressure relief device must be located on the vessel. In order to be able to inspect, maintain, or replace the relief device, it is acceptable to provide a full-area stop valve between the vessel and the relief device, provided that the valve is under administrative controls. This means the valve must be locked or sealed open for normal operation. A person designated by the user can only close it. That person shall remain at the stop valve during the time the valve is closed, and he shall again lock or seal the stop valve open before leaving the station. These are very prescriptive requirements; however, following these guidelines is paramount to the safe operation of a pressurized system. When stop valves are placed between the pressure relief device and the vessel, it is possible to easily defeat the intended design function of the PRD unless the user's organization has operating procedures that are strictly enforced. There have been instances of vessel failures where the stop valve between the vessel and PRD was closed. Therefore, it is cautioned that any decision to add block valves under PRDs should be made with careful deliberation. The user or operator of the facility assumes the burden of implementing and controlling the appropriate administrative controls for the safe operation of such an arrangement. There are instances where the pressure originates outside the vessel and the PRD need not be located directly on the vessel [see paragraph UG-125(g)]. This is common arrangement for pressure vessels in a pumped system where the pump is the only source of pressure. It is not required that vessels in this system have a PRD mounted directly on the vessel, provided that there is a properly sized relief device(s) somewhere in the system providing overpressure protection. For such a system, it is acceptable to locate a pressure relief device at the outlet piping of the pump. Likewise, paragraph M-5(b) allows stop valves to be located between the vessel and the PRD and the stop valve is not required to be under administrative controls if the closing of the stop valve completely isolates the vessel from the source of pressure. It is a common misconception that a PRD need not be provided if the source of pressure cannot generate a pressure greater than the maximum allowed working pressure of the vessel. For example, if a pump provides the source of pressure to a vessel

and the pump "dead-head" pressure is less than the MAWP of the vessel, does the vessel have to be provided with a PRD? The answer is yes, all pressure vessels must be provided with a PRD; however, in this instance, the PRD may be located in the piping system or on a remote vessel and stop valves may be provided between the vessel and pump without sealing or locking the block valve open. See the following Code Interpretation that clarifies the intent. Interpretation: VIII-1-83-29 Subject: Section VIII, Division 1, UG-125, Relief Devices Date Issued: October 4, 1982 File: BC81-268 Question (1): UG-125(h) states, "The protective devices required in (a) need not be installed directly on a pressure vessel when the source of pressure is external to the vessel and is under such positive control that the pressure in the vessel cannot exceed the maximum allowable working pressure at operating temperature except as permitted in (c) (See UC-98)." A note states that control instruments, except for pilot operated valves, cannot be considered for such positive control. What constitutes "positive control" as used in UG-125(h)? Reply (1): Without excluding other possible methods of positive control, it is often taken to mean the maximum pressure that could be developed in a vessel based upon engineering calculations applicable to the system. One example might be the maximum shutoff head plus maximum suction pressure of a pump, which pressurized the vessel. Another example might be the pressure drop at maximum flow conditions when a vessel discharges through a system to a known pressure, such as the atmosphere. Question (2): If a vessel is pressurized by a centrifugal pump and the maximum allowable working pressure of the vessel exceeds the shutoff head of the pump plus its maximum suction pressure, must the vessel still be protected by a relief device(s)? Reply (2): Yes, see UG-125(a). However, this source of pressure need not be considered in determining the required capacity of the relief device(s). Pressure relief device discharges are commonly routed through a piping system to a flare. When it is necessary to do maintenance and/or replacement of the pressure relief device without shutting down the entire plant, block valves may be provided at the discharge side of the PRD. When stop valves are located at the discharge side of the PRD, the potential effect is the same as placing a stop valve at the inlet side. When a block valve on the discharge line is closed, the flow through the PRD is blocked and the vessel is not protected from overpressure. When block valves are placed on the discharge side of the PRD, administrative controls are required that are the same as described above for a stop valve at the inlet side of the PRD and the same degree of caution is required. Paragraph M-7 provides guidance regarding inlet pressure drop for pressure relief valves in compressible fluid service. The nominal pipe size of all vessel and piping components between the vessel and the PRV shall be at least equal to the nominal size of the PRV inlet. Likewise, the cumulative pressure drop from the vessel to the valve shall not exceed 3% of the valve set pressure.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 130

130 · Chapter 21

The calculated pressure drop shall be based on the flow based on the valve nameplate capacity that is adjusted to the characteristics of the fluid. This requirement is to assure that the valve operates properly and does not chatter. Paragraph M-8 provides guidance regarding discharge lines from PRDs. The most desirable configuration of PRD discharge piping is to use short sections of pipe with long radius elbows that discharge directly to the atmosphere. However, because of the toxic and flammable nature of many materials, this is not allowed. In these instances, the PRD discharge is into piping systems that route the material to a flare or vent system. These piping systems can become quite long and complex in a major processing plant. The proper operation of a pressure relief device depends on the amount of back pressure on the valve (i.e., the amount of pressure at the discharge side of the device). When relief devices discharge into a piping system, a hydraulic analysis is required to determine the amount of pressure that may be present at any PRD that may affect its proper operation. This information must be given to the supplier of the PRD so that these effects can be considered in the design of the PRD. Provisions shall be made to drain all discharge lines in the piping system. The accumulation of liquid in the discharge system can have significant adverse effects on the flow characteristics and may increase the back pressure in system resulting in improper operation of the PRDs. Paragraph M-11 provides general guidance for the margin between the pressure safety valve set pressure and the operating pressure of the vessel. The provisions of this paragraph need to be well understood by the user when the design pressure of the vessel is established [see paragraph UG-21]. For a pressure relief device to operate properly, a suitable margin must be established between the set pressure of the valve and the maximum expected pressure during normal operations. When considering normal operations, the user should consider start-up, shutdown, process upsets, instrument response times, pressure surges, and so on. If these considerations are not included when establishing the maximum operating pressure and, in turn, the design pressure, the PRD may be actuated more often than intended and may hamper proper operation of the unit. The user is responsible for establishing the design pressure [see U-2(2)(a)], and the MAWP of the vessel will be established by the design pressure. In order to avoid the nuisance of premature opening of the PRD, careful consideration is required of the margins when establishing the vessel design pressure early in the process system design. For safety relief valves, the blowdown characteristics of the valve are essential in establishing the required margins. When a PRV opens, it should reseat at a pressure that is larger than operating pressure (otherwise, it is a nonreclosing device). Using the example given in M-11, if a valve has a blowdown of 7% with a set pressure of 100 psi, the valve should open at 100 psi and will reseat at a pressure 7% less or 93 psi. If the operating pressure is greater than 93 psi, the valve will not reseat and will leak. Pilotoperated relief valves allow much greater flexibility in allowing the operating pressure to be closer to the valve set pressure without the concerns related to reseating. However, pilot operated valves should not be used in services that are abrasive, dirty, or where proper pilot operation is compromised by fouling, freezing, coking, polymerization, or corrosion. The following general guidelines are listed for establishing the difference between the valve set pressure (design pressure of the vessel) and the maximum anticipated operating pressure.

Valve set pressure (vessel design pressure) To 70 psi 71­1000 psi greater than 1000 psi

Differential between set and operating pressure 5 psi 10% of set pressure 7% of set pressure

These margins may be lower if the valve has been designed or tested in a specific or similar service or the valve manufacturer recommends smaller margins. When a PRD opens, there are flow-induced forces and reactions that must be considered and accounted for in the support of the relief device and its associated piping. Also, the loads resulting from the attached piping need to be considered in order to assure that thermal expansion, dead weight, and so on do not impose loads on the PRD that will affect its ability to operate as intended. Section VIII, Division 1, does not contain specific rules to size relief devices for fire conditions. Such relief devices are intended to reduce the pressure to a predetermined level in the event the vessel is exposed to a fire. Under fire conditions, heating of the vessel wall may result in a decrease in material strength that reduces the pressure level that may be safely contained. Several references are provided in paragraph M-14 for guidance when consideration of fire conditions is required. The more useful references include the following: (1) API RP 520 [20] (2) API Standard 2000 [21]

21.8.11

Nonmandatory Appendix P: Basis for Establishing Allowable Stress Values for UCI, UCD, and ULT Materials

The criteria used to establish the allowable stress values listed in Section II, Part D, for a material at a specific temperature for Section VIII, Division 1, application are established using following material property data: ST RT SY RY SRavg SRmin SC specified minimum yield strength at room temperature ratio of the average value of available data of tensile strength at temperature to the room temperature yield strength specified minimum yield strength at room temperature ratio of the average value of available data of yield strength at temperature to the room temperature yield strength average stress to cause rupture at the end of 100,000 h minimum stress to cause rupture at the end of 100,000 h average stress to produce a creep rate of 0.01%/1000 h

The stress criteria and the factors applied to establish the maximum allowable stress for materials listed in ULT-23, UCI-23, and UCD-23 are given by Table P-1 of Appendix P. The criteria and factors used for all other materials are given in Appendix 1 of Section II, Part D, Table 1-100 (Table 1-100 is reproduced here as Table 21.11 for convenient reference). Simply stated, the allowable stress for a material at a given temperature is the lesser of STRT/3.5, 0.67 SYRY, 0.67 SRavg, 0.8 SRmin, or SC. For some materials, such as stainless steel where strain hardening substantially increases the yield strength of the material, 0.9 SYRY is used instead of 0.67 SYRY. When the allowable stress is based on this criteria, a cautionary note is referenced

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 131

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 131

TABLE 21.11 CRITERIA FOR ESTABLISHING ALLOWABLE STRESS VALUES FOR TABLES 1A AND 1B (Source: TABLE 1-100 OF Section II Part D of the ASME Code)

Below Room Temperature Tensile Product/Material Strength Wrought or cast ferrous and nonferrous ST 3.5

2

Room Temperature and Above Tensile Strength ST 3.5 1.1 STRT 3.5

2

Yield Strength

Yield Strength

Stress Rupture FavgSR avg 0.8SR min

Creep Rate 1.0Sc

/3 Sy

/3 Sy

2

/3 Sy Ry

or 0.9 SyRy [Note (1)]

Welded pipe or tube, 0.85 S ferrous and 3.5 T nonferrous Ferrous materials, structural quality for pressure retention applications [Note (2)] Ferrous materials, structural quality for nonpressure retention function 0.92 S 3.5 T

2

/3

0.85Sy

(1.1 * 0.85) 0.85 ST STRT 3.5 3.5

2

/3

0.85Sy

2

/3

0.85) 0.85Sc 0.85SyRy (Favg 0.85) (0.8 Sp min or 0.9 0.85Sy Ry SR avg [Note (1)] 0.92SyRy NA NA NA

2

/3

0.92Sy

(1.1 * 0.92) 0.92 ST STRT 3.5 3.5

2

/3

0.92Sy

2

/3

ST 3.5

2

/3 Sy

ST 3.5

1.1 s 3.5 T

2

/3 Sy

2

/3 SyRy

NA

NA

NA

Notes: (1) Two sets of allowable stress values may be provided in Table 1A for austenltic materials and in Table 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 deformation is not in itself objectionable. These higher stresses are not recommended for the design of flanges or for other strain sensitive applications. (2) The value of ST shall be in accordance with the material specifications.

that says that use of the higher allowable stress values should not be used where permanent deformation is to be avoided. As an example, the "high stress" basis for stainless steel should not be used for the design of flanged connections since plastic deformation that may occur will make the flange more likely to leak. A joint efficiency factor of 0.85 is included in the criteria for establishing the allowable stress for electric resistance welded pipe and tube material. It should also be noted that the temperatures at which the timedependent properties (stress rupture or creep rate) govern the allowable stress basis are shown in italic type in the stress tables in Section II, Part D. This information allows the designer to understand when time-dependent properties govern the allowable stress. Some design details and design methods may need to be adjusted to account for creep and/or creep rupture.

in a weld and heat-affected zone that does not cool as rapidly. The slower cooling rate of the weld will have a positive effect on the microstructure of the HAZ, and will result in smaller residual stresses in the welded joint. The preheat recommendations of this appendix are not mandatory, and the appendix cautions, . . . preheating temperatures listed below do not necessarily insure satisfactory completion of the welded joint and requirements for individual materials within the P-Number listing may have preheating more or less restrictive than this general guide. Section IX of the ASME Code should be reviewed for weld procedure qualification requirements. Specific preheat requirements of Appendix R are summarized below: (1) For P-No. 1 material, a preheat of 50°F is recommended; however, if the material has a specified minimum carbon content greater than 0.30% and the thickness of the joint is greater than 1 in., the recommended preheat is 175°F. (2) For P-No. 3 material, a preheat of 50°F is recommended; however, if the material has either a specified minimum tensile strength greater than 70,000 psi or the thickness of the joint is greater than 5/8 in., the recommended preheat is 175°F.

21.8.12 Nonmandatory Appendix R: Preheating

Other than as required in Tables UCS-56 and UHA-32, Division 1 does not have mandatory weld preheat requirements. However, there are some factors where preheat of the weld will have beneficial effects on the completed welded joint. Some of these factors include the weld joint detail and the extent of constraint, the thickness of the members to be welded, and the composition of the materials to be joined. Preheating the weld results

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 132

132 · Chapter 21

(3) For P-No. 4 material, a preheat of 50°F is recommended; however, if the material has either a specified minimum tensile strength greater than 60,000 psi or the thickness of the joint is greater than 1/2 in., the recommended preheat is 250°F. (4) For P-No. 5 material, a preheat of 300°F is recommended. However, if the material has either a specified minimum tensile strength greater than 60,000 psi or has a specified minimum chromium content above 6% and the thickness of the joint is greater than 1/2 in., the recommended preheat is 400°F. (5) For P-No. 6 material, preheat of 400°F is recommended. (6) For P-No. 9A, Gr. No. 1 material, preheat of 250°F is recommended. (7) For P-No. 9B, Gr. No. 1 material, preheat of 300°F is recommended. (8) For P-No. 10A, Gr. No. 1 material, preheat of 175°F is recommended. (9) For P-No. 10B, Gr. No. 2 material, preheat of 250°F is recommended. (10) For P-No. 10C, Gr. No. 3 material, preheat of 175°F is recommended. (11) For P-No. 10F, Gr. No. 6 material, preheat of 250°F is recommended. (12) For P-No. 10D, Gr. No. 5 and P-No. 10E Gr. No. 5 material, preheat of 300°F with the interpass temperature maintained between 350°F and 450°F is recommended. (13) For P-No. 11A, Gr. Nos. 2 and 3, preheat of 300°F is recommended. However, if the material has either a specified minimum tensile strength greater than 60,000 psi, or has a specified minimum chromium content above 6% and the thickness of the joint is greater than 1/2 in., the recommended preheat is 400°F. (14) For P-No. 11A, Gr. No. 4, preheat of 250°F is recommended. (15) For P-No. 11B, Gr. No. 1­5, preheat of 50°F is recommended. However, if the material has either a specified minimum tensile strength greater than 70,000 psi or the thickness of the joint is greater than 5/8 in., the recommended preheat is 175°F. (16) For P-No. 11B, Gr. Nos. 6 and 7, preheat of 300°F is recommended. However, if the material has either a specified minimum tensile strength greater than 60,000 psi, or has a specified minimum chromium content above 6% and the thickness of the joint is greater than 1/2 in., the recommended preheat is 400°F. (17) For P-Nos. 11A and 11B, consideration should be given to limiting the interpass temperature during welding. When establishing the need for and the value of preheat for any material, it is strongly recommended that a competent welding engineer experienced with welding the specific materials being welded be consulted.

where maintaining a leak-free joint during operation is inherently more difficult. This appendix does not provide specific criteria where the provisions of the appendix should be applied; however, the following discussion provides some guidance. Generally, the flanges within the size and pressure­temperature limits of the ASME B16.5 [22] have operated satisfactorily. The ASME B16.5 standard is limited to flanges that are NPS 24 and smaller. Unless the flange connection is in very high temperature service, flanges supplied to the B16.5 standard (NPS 24 and smaller) should generally provide satisfactory leak-free service with normal make-up procedures. However, as the flange connection becomes larger and is operated at more severe conditions, the design and assembly of the joint become more important. Large diameter flanges may not have sufficient stiffness to maintain sufficient compression on the gasket to prevent leakage. Flanged connections operating in the creep regime or when the bolts have a different coefficient of thermal expansion than that of the flange will require special consideration as suggested in this appendix. The most important parameter regarding the sealing of a properly designed (including gasket selection) flange is the amount of bolt load that is applied during the assembly. If insufficient bolt load is applied, the flanged connection will leak during leak testing or operation. If excessive bolt load is applied, the bolts, flanges, and/or gasket may be overloaded to the extent they are damaged. The irony is that, although the bolt prestress is essential to obtaining a leak-free connection, the actual bolt preload value applied may not be specified, controlled, and measured. Using normal manual bolt tightening procedures, the probable bolt stress that is developed is given by S = where: S d 45,000 (21.57) 2d

the bolt stress the bolt nominal diameter

21.8.13 Nonmandatory Appendix S: Design Considerations for Bolted Flange Connections

The proper assembly of a flanged connection is key to its leakfree operation. There are several important points related to the flange design and assembly that are summarized in Appendix S. As noted, these considerations become more important for flanges of large diameter, high pressure, or high temperature service

As may be seen, as the bolt diameter increases, the amount of bolt stress obtained by normal manual bolt tightening procedures decreases rather dramatically. For example, the expected bolt stress for a 1-in. diameter bolt is 45,000 psi, but the expected bolt stress for a 2-in diameter bolt is only 31,800 psi when manual torque methods are employed. The rules for flange design that are given in Appendix 2 constitute a design procedure to be used to establish the required dimensions of the flange and bolting. It is not intended that the actual loads and stresses in actual operation be limited to the design values used in Appendix 2. Appendix S clearly states that the initial bolt load used to assemble a flange connection may be required to be larger than the design bolt load used in Appendix 2. If the bolt preload is greater than the design bolt load, then, obviously, the stresses in the flanges are greater than the stresses used in the design procedure. This should not be taken as a concern because the design procedure has some built-in safeguards that provide some protection. For example, it is common that location of the largest stress in a flange is at the hub. If the hub is "overstressed" and a plastic hinge develops, the flange ring will take more of the load. However, if the flange ring is yielded, then the flange may be damaged. The combined ring­hub stress limit of Appendix 2 [see paragraph 2-8 (a)(4)] provides some protection from this occurrence. Experience has shown that flanges designed in accordance with Appendix 2 are able to have an initial bolt prestress as high as

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 133

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 133

40% to 50% of the bolt yield stress without damaging the flanges. For low alloy bolting, this corresponds to an initial bolt stress of 40­50 ksi which is well above the ambient temperature design stress of 25 ksi. Special consideration should be given to flanges with low yield strength bolting (such as annealed stainless steel). Low yield strength bolting has a yield strength well below that for low alloy bolting material; thus, it is not possible to apply as much preload to the joint without damaging the bolt. This fact makes flanged connections with low yield strength bolts inherently more prone to leakage. It is good design practice to avoid using bolt material of low yield strength, if at all possible. Alternatives may include the use of strain hardened material or high alloy, high strength bolts, or to use lap joint flanges with low alloy, high strength bolting. The gasket should be selected and proportioned such that it will not be damaged as a result of the bolt preload. Some gasket types come with a compression stop that limits the amount of compression that the gasket can experience. For gaskets without compression stops, the gasket should be wide enough to avoid damage during bolt up. Experience has shown, as a rule of thumb, that if the compressive stress on the gasket resulting from the design bolt load of Appendix 2 is limited to twice the gasket y value, the gasket will perform satisfactorily. Likewise, it is important to consider what happens to a flanged connection at operating temperatures. The material selection for the operating temperature may have a strong influence on the ability to maintain a leak-tight joint. The temperature of the bolts will always lag the temperature of the flange during start-up and shutdown. If the flange connection is not insulated, the bolt temperature will be different that that of the flange even at steady state. For elevated temperature service, the bolts will be cooler than the flange. If the bolt material and flange material have similar coefficients of thermal expansion, this makes the joint tighter. For flange connections that operate at temperatures well above ambient, it is not recommended to use bolt material with a coefficient of thermal expansion that is greater than that of the flange because the bolt load will be diminished at steady-state temperature. It is possible to use bolts with expansion coefficients less than the flange; however, at extreme temperature, the resulting additional bolt load from differential thermal expansion effects may require consideration. Thus, differential thermal expansion between the bolts and flange is an important factor in the design of a flanged connection. The initial makeup of the flange assembly must be done correctly in order to assure that the joint will provide leak-free service. It is most important to assure that the preload is evenly and gradually applied such that excessive localize stresses do not occur in the flange or gasket. A proper bolt loading procedure will include a loading sequence that use a "crisscross" pattern where bolts on opposite sides of the assembly are tightened progressively. The initial loading sequences should start with small loads that are incrementally increased as the tightening proceeds. The amount of bolt preload needs to be established and a mechanism has to be made available to control the amount of bolt preload applied. This can be achieved by specifying the amount of torque to be applied to the bolts to achieve the required bolt prestress. It should be noted that this method is not an accurate method of assuring proper bolt preload unless the procedure is done in a very controlled manner. Bolt preload may be achieved by the use of hydraulic bolt tensioners, which provides more a accurate application of bolt stress. However, on critical joints, it may be required to have the ability to measure the actual amount of bolt prestress applied. This

can be done using ultrasonic extensometers or by using studs that are supplied with a measuring rod that allows measurement of actual bolt elongation. For further information on assembling bolted flange joints, see Ref. [34]. Information that follows has been moved to Appendix 2.

21.8.14

Nonmandatory Appendix T: Temperature Protection

Vessels with service conditions where damage can occur because of overheating should be provided with a means by which the metal temperature can be controlled within safe limits and/or provisions should be made to shutdown the operation of the vessel. Many vessels that operate at elevated temperatures in the petroleum refining and chemical processing industries use insulating refractory liners on the inside surface of the vessel. This allows the vessel wall to operate at temperatures considerably cooler than the process fluid temperature. Because the maximum design temperature is based on the mean metal temperature (see paragraph UG-20), this allows the use of metals that are much more economical which would otherwise not be appropriate for elevated temperature operation. Consideration must be given to the possibility of degradation or failure of the refractory lining for such vessels. Examples of some appropriate considerations include the following: (1) Use of heat sensitive paint on the vessel surface that will turn a different color when the temperature increases above the normal operating metal temperature. This allows detection of liner degradation. (2) Use of optical pyrometers to monitor (continuously or intermittently) the vessel's wall temperature. (3) Use of a jacketed vessel where the jacket contains a fluid that can remove the heat in the case of lining failure. Some vessels operating elevated temperature include a water jacket that allows the water to remove excessive heat in case of lining failure by boiling for a designated period of time. Any vessel that can be subjected to overheating should have some provisions for determining when the vessel is being overheated so that corrective actions may be implemented.

21.8.15

Nonmandatory Appendix W: Guide for Preparing Manufacturer's Data Reports

Appendix W provides guidance for the preparation of the Manufacturer's Data Form (MDF). The MDF is required by the requirements in paragraph UG-120. The forms that constitute the ASME MDR are not intended for use with pressure vessels that do not comply with all the provisions of the code, including the application of the code stamp. Sample forms are provided and instructions are identified for each required entry. Instructions are provided for the following forms. (1) Form U-1­Manufacturer's Data Report for Pressure Vessels. This form may be used for all pressure vessels, including those that are field assembled or erected. (2) Form U-1A­Manufacturer's Data for Pressure Vessels. This is an alternative form that may be used for single chamber, completely shop-fabricated pressure vessels. (3) Form U-2­Manufacturer's Partial Data Report. This form is used for a welded pressure part fabricated by one Manufacturer to be included in a code-stamped vessel fabricated by another Manufacturer.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 134

134 · Chapter 21

(4) Form U-2A­Manufacturer's Partial Data Report. This is an abbreviated alternative form that may be used for a welded pressure part fabricated by one Manufacturer to be included in a code-stamped vessel fabricated by another Manufacturer. (5) Form U-3­Manufacturer's Certificate of Compliance. This form is to be used for pressure vessels to be stamped with the Code UM symbol [See paragraph U-1(j)]. (6) Form U-4­Manufacturer's Data Report Supplementary Sheet. This form is used to provide information or remarks that will not fit on the applicable MDR.

21.8.17

Nonmandatory Appendix DD: Guide to Information Appearing on Certificate of Authorization

21.8.16

Nonmandatory Appendix Y: Flat Face Flanges with Metal-to-Metal Contact Outside the Bolt Circle

Appendix Y provides rules for circular, bolted flanged connections where the assembly is comprised of flat face flanges that are in uniform metal-to-metal contact across their entire face. The flanges of the assembly may be identical or nonidentical. The rules also apply to a pair of identical flat face flanges that are separated by a metal spacer located at the outer edge of the flanges. The rules assume that a self-sealing gasket is used that is approximately in line with the wall of the attached shell. Flat face flanges are classified in accordance with the following definitions. Class 1­The mating flanges are identical except for the gasket groove. Class 2­The mating flanges are not identical and one flange is a reducing flange with an inside diameter that is greater than one-half of the bolt circle diameter. Class 3­The flange is mated to a flat head or a reducing flange with an inside diameter that does not exceed one-half the bolt circle diameter. In addition to the classification of flat face flanges, the individual flanges are categorized as follows. Category 1­Internet flanges and optional flanges calculated as integral. Category 2­Loose-type flanges with hubs that are considered to add strength. Category 3­Loose-type flanges with or without hubs or optional-type flanges calculated as a loose-type flange with no credit taken for the hub. The analysis of a flat face flange is very similar to that of a raised face flange (see Appendix 2). However, since the pivot point between mating flat face flanges occurs outside the bolt circle, the application of pressure results in a prying action on the bolting. Appendix Y accounts for the additional bolt load and moment from the prying action of the pressure loading. Paragraph Y-10 provides the following references for additional guidance for the design of metalto-metal contact flanges: (1) Schneider, R. W. and Waters, E. O., The Background of ASME Code Case 1828: A Simplified Model of Analyzing Part B Flanges, Journal of Pressure Vessel Technology, Vol. 100, No. 2, ASME, May 1978, pp. 215­219. (2) Schneider, R. W. and Waters, E. O., The Application of ASME Code Case 1828, Journal of Pressure Vessel Technology, Vol. 101, No. 1, ASME, February 1979, pp. 87­94.

This appendix provides a guide regarding the information that is to appear on the Certificate of Authorization issued by the ASME to the Manufacturer. The required information includes the name and address of the Manufacturer. The certificate will define the type of code symbol stamp the Manufacturer is authorized to use, which defines the scope, and limitations of the Certificate of Authorization. The Manufacturer may have UStamp for the shop, field, or shop/field fabrication of pressure vessels. The UM Code symbol authorizes the Manufacturer to fabricate miniature pressure vessels [see U-1(j)]. The UV Code symbol authorizes the Manufacturer to fabricate or assemble pressure relief valves. Each certificate will have a unique number and will bear the signature of the current Chairman of the ASME Boiler and Pressure Committee and Director of ASME Accreditation.

21.8.18

Nonmandatory Appendix EE: Half-Pipe Jackets

Appendix EE provides a design method for vessels with halfpipe jackets (as shown in Fig. 21.40), and is applicable when the shell and jacket are under the influence of positive internal pressure. The procedure should not be used if there is vacuum in either the vessel or the jacket. The procedure given in Appendix EE is based on limiting the total longitudinal stress in the vessel, at the area under the jacket, to 1.5 S where S is the maximum allowable stress for the vessel material at the design temperature. The longitudinal bending

FIG. 21.40 (Source: Fig. EE-4 of Section VIII Div. 1)

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 135

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 135

The minimum thickness of the half-pipe jacket is given by T = where P1 r S1 P1r 0.85S1 - 0.6P1 (21.61)

jacket design pressure inside radius of the jacket allowable stress of the jacket material

FIG. 21.41

FREE BODY DIAGRAM--HALF-PIPE JACKET

stress in the vessel resulting from the jacket pressure is represented by the term F in paragraph EE-2. Thus, the permissible value of jacket pressure is based on the amount of reserve strength available to maintain the total stress to 1.5 S. The magnitude of the bending stress under the jacket may be developed from the theory of beams on elastic foundations. The free body diagram for the half-pipe jacket and shell is given in Fig. 21.41. Based on beams on elastic foundation theory, the deflection at any point A due to P, F1, and F2 is given by w = Pr2 12 - e - bx cos bx - e - b(l - x) cos b(l - x)2 2Et F1e-bx 8b 3D (sin bx + cos bx)

Likewise, when a fillet weld is used to attach the jacket to the shell, the throat thickness of the fillet weld shall not be less than the smaller of the jacket or shell thickness. It should be noted that half-pipe jackets attached to the shell by fillet welds may be susceptible to fatigue cracking due to thermal cycling of the jacket fluid. When such cyclic loading is a consideration, the jacket should be attached to the shell with a full penetration weld through the jacket wall with a covering fillet weld.

21.8.19

Nonmandatory Appendix FF: Guide for the Design and Operation of Quick-Actuating (Quick-Opening) Closures

-

F2e - b(l - x) 8b 3D B

4

(sin b(l - x) + cos b(l - x))

(21.58)

where: F1 and F2 D b =

PL/2 Et3/12(1 3(1 - m2) Rt 2

2

)

The axial bending stress in the shell is derived from the expression sb = D 02 w 6 0x2 t 2 = 6D 0 2w t2 0x2 (21.59)

The rules given in UG-35.2 for quick-actuating closures were extensively revised in the 2004 Edition. During the preparation of these revisions, it was recognized that these types of closures represent a significant risk in an operating plant if not properly maintained and operated. For these reasons, Appendix FF was developed to provide guidance to owners, users and operators in the form of recommendations for the installation, operation, and maintenance of quick-actuating closures . The appendix addresses the responsibilities of the User and Manufacturer of the quick-opening closure, specifically with regard to any safety devices supplied with the closure. The appendix also provides guidance for the designer of a quick-actuating closure, highlighting the need to build in a redundancy in the safety elements and to ultimately produce a closure that will guarantee fail-safe behavior. Consideration of cyclic loading is also emphasized, since by definition a quick-actuating closure is one where it is intended to be opened and closed with some regularity. Therefore, most closures would require a fatigue evaluation. Other topics covered in Appendix FF include installation, maintenance, inspection, and training. It is strongly advised that the readers review Appendix FF if they have any involvement with these types of closures from either the manufacturing end or as an owner/ user/operator.

21.8.20

The hoop stress in the shell is derived by sh = Ew R (21.60)

Nonmandatory Appendix GG: Guidance for the Use of U.S. Customary and SI Units in the ASME Boiler & Pressure Vessel Code

0.3 b. and the hoop bending stress is given by The above derivation may be found in Ref. [26]. The above equations were programmed on a computer and a parametric analysis was done to normalize the total longitudinal stress, due to the effects of the half-pipe jacket, to be expressed by the factor K. Thus, the allowed pressure in the jacket is P F/K.

With publication of the 2004 Edition of the ASME Boiler & Pressure Vessel Code in SI units, this nonmandatory appendix was included to provide guidance regarding the method of unit conversion (soft or hard) employed in the book. In addition, the appendix also contains recommended equivalents for U.S. fractions, nominal pipe sizes, pressures, and temperatures.

21.8.21

Nonmandatory Appendix HH: Tube Expanding Procedures and Qualification

Figures EE-1 through EE-3 plot the factor K for different sizes (NPS 2 through NPS 4) as a function of vessel diameter and shell thickness. (Fig. EE-1 is included as Fig 21.42 for reference.)

Nonmandatory Appendix HH provides Manufacturers with guidance in fulfilling the requirements of Appendix A (Basis for Establishing Allowable Loads for Tube-to-Tubesheet Joints) and Paragraph UHX-11.5.1 (ligament efficiency) for expanded-only tubes. The appendix provides the requirements for preparation and

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 136

136 · Chapter 21

FIG. 21.42

NPS 2 PIPE JACKET (Source: Fig. EE-1 of Section VIII Div. 1 of the ASME Code)

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 137

COMPANION GUIDE TO THE ASME BOILER & PRESSURE VESSEL CODE · 137

qualification of tube expanding procedures specification (TEPS), including Form QEXP-1 for documenting the TEPS.

14. Barsom, J.M. and Rolfe, S.T., Fracture and Fatigue Control in Structures: Applications in Fracture Mechanics, 2nd ed., PrenticeHall, Englewood Cliffs, NJ. 15. Pellini, W.S., Principles of Fracture Safe Design­ Part 1, 1971 Adams Lecture, Welding Journal Research Supplement, 1971. 16. Harvey, J.F., Theory and Design of Modern Pressure Vessels, Van Nostran Reinhold Co., New York, 1974. 17. Waters, E. O., Rossheim, D. B., Wesstrom, D. B., and Williams, F.S.G., Formulas for Stresses in Bolted Flanged Connections, Transactions of the American Society of Mechanical Engineers, Vol. 59, No. 3. [Note that this paper has been reprinted with the courtesy of Taylor Forge & Pipe Works, 1949, and entitled Development of General Formulas for Bolted Flanges.] 18. Standards of the Expansion Joint Manufacturers Association, 7th ed., Expansion Joint Manufacturers Association, Tarrytown, NY, 1998.

21.8.22 Nonmandatory Appendix JJ: Flowcharts Illustrating Impact Testing Requirements and Exemptions from Impact Testing by the Rules of UHA-51

This appendix provides guidelines for determining impact test requirements for austenitic, austenitic­ferritic duplex, ferritic chromium, and martensitic stainless steel vessels in accordance with the impact test rules in UHA-51. The guidelines are provided in the form of flowcharts. Copies of these flowcharts can be found in Fig. 21.20.

21.9

REFERENCES

1. The National Board Synopsis of Boiler and Pressure Vessel Laws, Rules and Regulations Arranged by States, Cities, Counties, and Provinces (United States and Canada), The National Board of Boiler and Pressure Vessel Inspectors, Columbus, OH, Updates, 2003 (see http://www.nationalboard.org for online information). 2. 29 CFR 1910.119, Process Safety Management of Highly Hazardous Chemicals, Title 29, Labor, Code of Federal Regulations, Chapter XVII, Occupational Safety and Health Administration (OSHA), Department of Labor, Part 1910, Occupational Safety and Health Standards. 3. Jacobs, W. S. and McBride, W. L., Design of Radial Nozzles in Cylindrical Shells for Internal Pressure, Vol. 102, Transactions of the ASME, February 1980. 4. API 510, Pressure Vessel Inspection Code: Maintenance, Rating, Repair, and Alteration, 8th ed., American Petroleum Institute, June 1997, Addendum 1 Errata, January 26, 1999. 5. National Board Inspection Code ANSI/NB-23, The National Board of Boiler and Pressure Vessel Inspectors, 1998. 6. API Standard 530, Calculation of Heater-Tube Thickness in Petroleum Refineries, 4th ed., American Petroleum Institute, October 1996. 7. ASME PVHO-1, Safety Standard for Pressure Vessels for Human Occupancy, American Society of Mechanical Engineers, 1997. 8. ASCE 7-95, Minimum Design Loads for Buildings and Other Structures, American Society of Civil Engineers, 1995. 9. Uniform Building Code: Volume 1-Administrative, Fire- and LifeSafety, and Field Inspection Provisions, Volume 2-Structural Engineering Design Provisions. Volume 3-Material, Testing and Installation Standards, International Conference of Building Officials, 1997. 10. Farr, J. R. and Jawad, M. H., Guidebook for the Design of ASME Section VIII Pressure Vessels, ASME Press, New York, 1998. 11. Corten, H.T., Fracture Toughness Considerations Underpinning New Toughness Rules in Section VIII, Division 1, of the ASME Code (unpublished). 12. Selz, A., New Toughness Rules in Section VIII, Division 1 of the ASME Boiler and Pressure Vessel Code: Mechanical Engineering, ASME, April 1988, pp. 84­87. 13. Yukawa, S., Brittle Fracture Margins in ASME Code Section VIII, Division 1, Prepared for MPC Program on Fitness for Service, February 23, 1995.

19. Sims J.R., Hantz, B.F., and Kuehn, K.E., A Basis for the Fitness for Service Evaluation of Thin Areas in Pressure Vessels and Storage Tanks, Pressure Vessel Fracture, Fatigue and Life Management, PVPVol. 323, American Society of Mechanical Engineers, 1992. 20. API Recommended Practice 520, Sizing, Selection, and Installation of Pressure-Relieving Devices in Refineries: Part I - Sizing and Selection, Part II - Installation, 6th ed., American Petroleum Institute. 21. API Standard 2000, Venting Atmospheric and Low-Pressure Storage Tanks: Nonrefrigerated and Refrigerated, 5th ed., American Petroleum Institute, April 1998. 22. ASME B16.5, Pipe Flanges and Flanged Fittings, NPS 1/2 Through NPS 24, American Society of Mechanical Engineers, 1996. 23. ASME B16.47, Large Diameter Steel Flanges, NPS 26 Through NPS 60, American Society of Mechanical Engineers, 1996. 24. Singh, K.P. and Soler, A.I., Mechanical Design of Heat Exchangers, Acturus Publishers Inc., Cherry Hill, NJ, 1984. 25. Standards of the Tubular Exchanger Manufacturer's Association (TEMA), 8th ed., Tubular Exchanger Manufacturers Association, Inc., Tarrytown, NY. 26. Jawad, M.H., Theory and Design of Plate and Shell Structures, Chapman and Hall Inc., 1994. 27. NFPA 69, Deflagration Pressure Containment, Chapter 5, Standard on Explosion Prevention Systems, National Fire Protection Association, 1997. 28. NB-3224, Level C Service Limits, Section III, Nuclear Components, Division 1, ASME Boiler and Pressure Vessel Code. 29. NB-3225, Level D Service Limits, Section III, Nuclear Components, Division 1, ASME Boiler and Pressure Vessel Code. 30. Langer, B.F., PVRC Interpretive Report of Pressure Vessel Research, Section 1­Section 1 Design Considerations, 1.4 Bursting Strength, Welding Research Council Bulletin 95, April 1964. 31. Bulletin on Stability Design of Cylindrical Shells, API Bulletin 2U, 1st ed., American Petroleum Institute, 1987. 32. AISC M016, Manual of Steel Construction Allowable Stress Design, 9th ed., American Institute of Steel Construction Inc., 1989. 33. Recommended Practice No. SNT-TC-1A, Personnel Qualification and Certification in Nondestructive Testing, American Society for Nondestructive Testing, Inc., Columbus, OH. 34. ASME PCC-1, Guidelines for Pressure Boundary Bolted Flange Joint Assembly, 2000.

ASME_Ch21_p027-111.qxd

2/20/09

2:05 PM

Page 138

Information

ASME_Ch21_p027-111.qxd

112 pages

Report File (DMCA)

Our content is added by our users. We aim to remove reported files within 1 working day. Please use this link to notify us:

Report this file as copyright or inappropriate

344086


You might also be interested in

BETA
01
Recommended Practice for Use and Procedures for Inspection, Maintenance, and Repair of Drilling and Well Servicing Structures
Chapter_51_pp285-308.qxd