Good and Tight

GOOD AND TIGHT FOR THE RIGHT TIGHT, TIGHTER IS BETTER: A DESCRIPTION OF ASME MODERN FLANGE DESIGN R obert W illiams WestermannBG August 21, 2017 Abstract The Design by Method approach found in the ASME Code allows for the selection of bolting materials with the intent of securing the joints as tightly as possible. During operation, stresses in the structure of piping runs will settle and the incurred strain will be transmitted to the weakest member of the assembly, the bolted flange. In such cases a tighter joint may not only reconcile those strains, but when employing a properly selected and fitted gasket, improve the fatigue resistance and volumetric laminar flow resistance of the joint.

I. B olted - flanged J oint Piping runs can be assembled in multiple ways; with fully welded assemblies, threaded pipes and connectors, or bolted flanges. The bolted flanged joint offers an unparalleled ease in the assembly and disassembly department of large bore piping systems, allowing for a variety of equipment and fittings to also be mounted. By its very nature, the flanged joint is the weakest link in a piping structure. When under normal operating conditions, a joint leak may be the result of inadequate design, materials, or assembly. On the other hand, leaks may also develop as a result of abnormal operating conditions, such as: A. Excessive working loads B. Fatigue caused by cyclic loads C. Sudden hammer loads D. Excessive dead loads E. Excessive or transient temperatures

F. Material creep

G. Corrosion

With the exception of fatigue and hammer loads, each of these conditions will produce a constant strain on the joint and its components causing a loss of contact load between the flange face and gasket. Fatigue and hammer loads cause a material to behave in a manner unlike those found when operating under laboratory conditions where the application of loads are slow and controlled. A flanged joint is composed of the following elements; flanges, bolts, a gasket, and (if part of the design) the wall of the pipe or vessel. A minimum axial load must be maintained on the joint to prevent unseating and leakage. With the use of a gasket between the flanges, a minimum axial load is necessary in keeping the gasket well seated.



y = 180 ( 2 m − 1 ) 2 III. G askets

In the 1930’s, British tests plotted axial load to a visible leak rate as internal pressure was added to a fixture. From such tests they concluded that an incremental relationship between axial load and internal pressure existed and that for a given gasket material at a given internal pressure there was a minimum gasket contact load. The contact load depends on the gasket material’s ability to conform to the imperfections of the flange face. Nowadays, the minimum contact load to conform to the flange face is empirically determined with the ASME gasket design constant - minimum seating stress, y. The minimum seating stress was once labeled as a yield value, this term was misleading and though it was abandoned, the variable remains ‘y’ to this day. Depending on their chemical makeup, elastomers and other gasket constructions whose behavior resembles elastomers, such as compressed fiber and spiral wound, ‘yield’ and neck according to their own natural draw ratio with an inclination toward brittle fracture at some precipitous point beyond the yield value. ‘Yield,’ in this case, might be best described as an intrinsic point prior to strain softening and orientation of the polymer chains or composite matrix. Today, researchers across multiple industries (medical, food, construction and aerospace) promote research to better understand the yield behavior of rubber and rubber like materials. The quadratic formula in Equation 1 defines the slope of the relationship between a gasket material construction’s minimum seating stress and the incremental load required to maintain a gasket seat when internal pressure is applied. This pressure creates an axial force in such a di- rection as to separate the joint. The coefficient ‘m,’ termed gasket design factor ‘m,’ is used as a multiple of pressure to satisfy the relation- ship. Its use in design is further described in the “Practical Application” section below.

Gaskets are chosen for their media compati- bility and temperature rating, and as such do not have pressure ratings. If flange faces were perfectly machined and perfectly aligned dur- ing assembly, a gasket would not be necessary. Gaskets do not leak — leaks are caused when system pressure is greater than the contact load required to keep the gasket seated. To satisfy the wide range of flange faces, media, and temperatures associated with pip- ing systems, gaskets come in a variety of con- structions such as elastomers, compressed fiber, and spiral wound. Their design constants, ‘m’ and ‘y,’ are published by the ASME and used in the minimum required bolt load formula. The ASME has divided gasket materials into groups to help follow bolting rules. An analysis of these constants will reveal that gasket constructions with lower ‘y’ values require smaller initial contact loads. IV. B olting Bolting should be capable of producing a mini- mum gasket contact load, be compatible with the operating temperature, and have suitable toughness when subjected to fatigue and ham- mer loads. Bolting is chosen for its elastic strain energy, a measure of toughness, as it provides a strong and ductile joint. V. L eak B efore B reak C riteria If the pressure load threatening to unseat the gasket is greater than the preload, the joint will leak, thereby relieving pressure on the system. This is a natural safety feature of the design. If the pressure load is great enough, the bolts will the carry the full imparted load. At this point, the bolts are subjected to both the initial preload and the system pressure load.



One needs to add both of these loads when making a minimum bolting strength decision.

the hydrostatic end force tending to unseat the gasket. Once the bolt load is known, the designer will go through an iterative process to determine bolting material, diameter, bolt pitch, and then design flange dimensions to include outside diameter, thickness, and hub dimensions. To utilize the full strength of the bolt or mod- ify it for a desired outcome, its drive style, body style, thread pitch, thread class, thread fit, thread protrusion, thread engagement, thread length, compatible nut, and selection of instal- lation method will have to be determined. VIII. S electing C omponents for S tandard ASME B16 F langes The ASME has standardized flange design under the general category specification B16. In these standards, one will find flange geometry, gasket suggestions, bolting rules, and pressure-temperature ratings. When one accepts the pressure-temperature rating of ASME B16.5 flanges and the allowable design stress philosophy of ASME Boiler and Pressure Vessel and B31 Piping Code, one accepts the use of listed bolting components. Following the design by method approach and choosing from the list is neither an option nor a suggestion if one desires to meet the ASME standards. If one chooses to deviate from the list and utilize a special or nonstandard material, the owner’s approval is required. Listed bolting materials are intended to meet the toughness criteria previously mentioned; providing a combination of strong and ductile bolts. IX. P ractical A pplication When a joint is assembled, the actual magni- tude of the load on the gasket is uncertain due to previously mentioned variances as well as other factors. When the piping system is pressurized, the structure will relax causing

A tough bolt, one capable of sustaining large strain energy, will allow the flange to leak and relieve pressure, so the risk of a cascade frac- ture propagating through the piping system is mitigated. The bolt should not break. VI. A ssembly Developing preload during assembly is critical. If the preload is too low, the joint will not seat the gasket. If the preload is too great there is little available strain, if any, remaining in the bolt to satisfy the leak before break criteria. Preload is the load initially produced by the torquing of the nut about the bolt and can often be given as a torque value. Preload is transferred from the mechanic’s pull on the wrench to the bolt, which will elongate the bolt as the torsional energy is transformed into an axial load. This axial strain of the bolts will compress the flanges and gasket. When installed by wrenches the exact magnitude and direction of the transferred preload within the joint and its components is uncertain. Some of these uncertainties are: A. The stiffness of gasket materials demon- strating elastomer-like behavior B. Load distribution at the bolt/nut/washer/flange interface C. The friction between the nut and bolt threads D. Friction between the nut and flange sur- face E. Tool accuracy VII. N ew F lange D esign B asics The design of flanges typically begins with selecting a gasket suitable for the operating conditions, then calculating a bolt load to produce a minimum gasket seat greater than



a reduction in the contact load on the gasket from that which was produced during the assembly. A tighter joint helps reconcile these strains. In the design of an ASME leak free joint, the ‘m’ factor is used to account for relaxations and variances. One will design for a higher, albeit uncertain, contact load knowing that as the structure relaxes the load will incrementally drop to a value no less than the minimum gasket contact load. For some gasket materials, the load to produce an initial gasket seat is sufficiently high to account for the contact load loss when the system is pressurized. For new ASME flange design, the minimum contact load on the gasket is solved using the minimum required bolt load equation found in ASME Boiler and Pressure Vessel Code Section VIII, Division 1, Appendix 2. ASME published gasket design ‘m’ factor and ‘y’ value are optional. Following the logic of ASME design, one will know the flange is properly designed for the relaxations and variances found in normal operating conditions if, after proper selection and assembly of gasket and bolting, and after pressure testing, the joint does not leak. The minimum required bolt load equation is not the source for calculating an installation torque value. The equation is used for calcu- lating a minimum bolt load. Once the bolting material has been selected, it is assumed that the full strength of the bolt will be utilized. An exception to this rule is when using high strength bolting and spiral wound gaskets that have a compression stop. The gasket’s compression stop limits the bolt load. Another exception is when the flange’s modulus of elasticity is substantially less than that of the bolting. When this occurs, the flange could be fractured by accidental over-tightening. The fracture initiation point is likely to occur around the bolt hole’s stress concentrations.

torque value is to use sixty-seven percent of the bolting’s yield or proof load. Importantly, with certain specialized applications, field ex- perience and environmental conditions may dictate otherwise. X. C onclusion The flanged joint, like any other structural joint, is a point of weakness subject to complex loads, but with proper design can serve as a failsafe point. As a practical matter, its design and the ability to apply its principles should be easy for the average person to follow. By using the right materials one can get it good and tight for the right tight; tighter is better. Codes 2007 ASME Boiler and Pressure Vessel Code: An International Code. VIII. Divi- sion 1, Rules for Construction of Pressure Vessels. New York, NY: American Society of Mechanical Engineers, 2007. Print 2015 ASME Boiler and Pressure Vessel Code, An International Code, Section II Materials, Part A, Ferrous Materials Spec- ifications, New York: American Society of Mechanical Engineers, 2015. Print Building Services Piping: ASME Code for Pressure Piping, B31. New York: Ameri- can Society of Mechanical Engineers, 2008. Print C.F. Braun and Co, Review of Piping and Pressure Vessel Code Design Criteria Tech- nical Report 217 LMFBR Piping Design Guide Project 4122-W, Alhambra, Ca, 1969. Print Companion Guide to the ASME Boiler & Pressure Vessel Guide, Criteria and Com- mentary on Select Aspects of the Boiler & Pressure Vessel and Piping Codes, Sec- ond Edition, Volume 1, New York: Ameri- R eferences

A good starting point for an installation



can Society of Mechanical Engineers, 2006. Print Pipe Flanges and Flanged Fittings: NPS 1/2 through NPS 24 Metric/inch Stan- dard. New York: American Society of Me- chanical Engineers, 2009. Print. Power Piping: ASME Code for Pressure Piping, B31. New York: American Society of Mechanical Engineers, 2012. Print. The American Society of Mechanical Engi- neers, Pressure Vessel and Piping Design , col- lected papers 1927-1959, New York, 1960. Print Brownell, Lloyd E, and Young, Edwin H., Process Equipment Design , New York: Wi- ley 1959. Print. Chambers, Jeffrey A, Preloaded Joint Anal- ysis Methodology for Space Flight Systems , Lewis Research Center, Cleveland, 1995. Print Commander, Naval Sea Systems Com- mand, Submarine Fastening Criteria (Non-Nuclear) S9505-AM-GYD-010 Rev. 2, Washington D.C., 2002 Gough, H.J. First Report of the Pipe Flange Research Committee , Pipe Flanges Research Committee, 1936. Print Ibrahim, RA, Pettit, C.L, Uncertainties and dynamics of bolted joints and other fasteners , Journal of Sound and Vibration 279 (857- 936), Elsevier, 2003. Print M. W. Kellogg Company. Design of Piping Systems, Revised Second ed. , John Wiley and Sons, 1956. Print. Rossheim, D.B, Markl, A.R.C, The Signifi- cance of, and Suggested Limits for, the Stress in Pipelines due to the Combined Effects of Pressure and Expansion , ASME Trans. Vol. 62. No. 5, New York, ASME,1940 Design

Walker, J.H, Crocker, Sabin. Piping Hand- book. First ed. New York & London: McGraw-Hill Book Company, 1930. Print.

Theory and Materials

Atlas of Stress-Strain Curves, Second Edition , Materials Park, Ohio, ASM International, 2002. Print. Blake, Alexander, Design of Mechanical Joints , New York, Marcel Dekker Inc, 1985. Print Blake, Alexander, Practical Fracture Me- chanics in Design , Marcel Dekker Inc, New York, 1996. Print Blake, Alexander, Practical Stress Analysis in Engineering Design, Second Edition , Re- vised and Expanded, Marcel Dekker, Inc, New York, 1990, Print Blake, Alexander, What Every Engineer Should Know About Threaded Fasteners, Ma- terials and Design , New York: Marcel Dekker Inc, 1986, Print Brinson, H.F, Brinson, L.C., Polymer Engi- neering Science and Viscoelasticity, An Intro- duction , Springer Science + Business Me- dia, LLC, 2008. Print Heyman, Jacques, The Science of Structural Engineering , London: Imperial College Press, 1999 Kurrer, Karl-Eugen, The History of the The- ory of Structures — From arch analysis to computational mechanics , Berlin: Ernst and Son, 2008. Electronic Lemaitre, Jean, Handbook of Materials Be- havior Models, Vol.1 Deformation of Mate- rials, San Diego, Academic Press, 2001 Roylance, David. Mechanics of Materials , New York: John Wiley and Sons, Inc, 1996. Print


Page 1 Page 2 Page 3 Page 4 Page 5

Made with FlippingBook - Online magazine maker