SECTION 33.0 3. 3.2 3.3 34 35 COADE Pipe Stress Analysis Seminar Notes Section 3 Table of Contents Modeling And Analysis Of The Piping System . see Computer Representation Of Basic Elements Piping Configuration Modeling Techniques ... Expansion Joint Modeling And Evaluation 3.3.1 Expansion Joint Stiffnesses .. 3.3.2 Evaluation of Expansion Joint Allowable Movements 8.3.3 Use of the ERATE Program .. 3.3.4 Modeling of Unbalanced Pressure Force 3.3.5 Modeling of Tie Rods . 3.3.6 Expansion Joint Assemblies. Piping Nozzle Evaluation 3.4.1 Equipment Nozzle Load Analysis 3.4.1.1 NEMA SM28 Standard for Steam Turbines 3.4.1.2 API610 Standard for Centrifugal Pumps 8.4.1.3 API 617 Standard for Centrifugal Compressors 3.4.1.4 API 661 Standard for Air Cooled Heat Exchangers. 3.4.1.5 HEI Standard for Closed Feedwater Heaters 3.4.2 Calculation of Vessel Stresses Due to Nozzle Loads... 3.4.2.1 Calculation of Vessel Stresses Due to Nozzle Loads .. 3.4.2.2 Running a Sample WRC 107 Calculation... 3.4.2.3 Evaluating Vessel Stresses .. 8.4.2.4 Completing the Sample Calculation 3.4.8 Estimation of Vessel Nozzle Flexibilities. 3.4.3.1 Use of WRC Bulletin 297 .. 3.4.3.2 Modeling Nozzles for Flexibility Calculations Restraint Modeling .. 3.5.1 Restraint Types 8.5.1.1 Anchor, 3.5.1.2 Restraint. 3.5.1.8 Spring Hanger 3.5.1.4 Hanger .. 3.5.1.5 Support ....3.6 1.7 Sway Brace .. 3.5.2 Non-linear Effects . Friction... One-Way Restraints Gaps Large Rotation Restraint 3.5.2.5 Bi-linear Stiffnesses 3.5.3 Evaluation of Restraint Stiffness ... 3.5.3.1 Use of the Structural Steel Modeler. 3.5.4 Use of CNODES When Modeling Restraints .. Miscellaneous Topics 3.6.1 Consideration of Cold Spring 3.6.2 Fiberglass Reinforced Plastic Pipe. 3.6.3 Underground Pipe. 3.6.8.1 Modeling Soil Restraint 3.6.3.2 Automated Underground Piping Modeler. 3.6.4 Jacketed Pipe 3.6.5 Flange Leakage Analysis 3.6.5.1 Equivalent Pressure Calculation 8.6.5.1 Flange Leakage Analysis Module ..COADE Pipe Stress Analysis Seminar Notes 3.0 Modeling And Analysis Of The Piping System ‘The first two sections of these seminar notes have served to give the user an overview of the requirements of pipe stress analysis. This section presents the task on a detailed level, presenting ideas for modeling of various piping configurations and explaining specific analyses which may be performed to evaluate individual piping components. Included in this section is information on the following subjects: 1 2 3 Computer representation of basic elements Piping configuration modeling techniques Expansion joint modeling and evaluation Nozzle evaluation, including evaluation of equipment loads, determin: nozzle/vessel stresses, and estimation of nozzle/vessel connection flexibi Piping restraints/structural modeling Miscellaneous topics (cold spring, underground pipe, plastic pipe, jacketed pipe, flange leakage analysis, etc.) 3-1COADE Pipe Stress Analysis Seminar Notes 3.1 Computer Representation Of Basic Elements Pipe stress analysis computer software algorithms are based upon certain assumptions. ‘These assumptions serve to make the computer model (and its corresponding analytical results) only an approximation of reality. In many cases this approximation may be sufficiently close to reality to fall within the tolerances, margins, and factors of safety of the problem tobe an adequate representation, while in other cases the user may find itnecessary torefine the model through more detailed modeling. This section describes the assumptions used in the computer algorithms in order that the user may more fully understand the limitations (and the potential work-arounds) of the system. The “stiffiess method” algorithm, which is used to perform the actual analysis done by CAESAR II and other prominent pipe stress/structural computer programs is described in detail in Section 6 of these seminar notes. Lo] a Lo ————_— © "Stick" Member Aibitrary Cross Section Pipe Cross Section Structural Cross Section H Os Figure 3-1 Piping basic elements are modeled as centerline, or “stick” members. These elements are defined by two node points (one at the “from” end, and the other at the “to” end), each with fixed spatial coordinates and six degrees of freedom (three translational and three rota- tional). The elements are further defined by a constant (non-varying along the element length) set of stiffness parameters (i.e., material and cross-sectional properties). Response of the elements under load is governed according to recognized strength of material 32COADE Pipe Stress Analysis Seminar Notes relationships (as described in Section 6 of these notes), subject to certain limiting assump- tions. These assumptions, described on the following pages, govern the relationship between the mathematical model in the computer and the actual pipe existing in the power plant or refinery. 1 - All elements remain stable under load (local buckling of cross-sections is ignored): f 48°00 0375" Wall Local Bucking of Cross Section Section AA Figure 3-2 2 - Plane sections remain plane: Center of Bending B D Figure 3-3 ‘The computer algorithm assumes that points A and B (of Figure 3-3) always lie on the same cross-sectional plane, whether in the deformed or the undeformed state. 3-3COADE Pipe Stress Analysis Seminar Notes Figure 3-4 In reality, the moment F x L (in Figure 3-4) does not produce a uniform “plane- sections-remain-plane” bending load at the cross-section A-B, since it causes Jocal warping. 3 - Hooke’s Law is applicable across all fibers of the cross-section: f3 CSS Compressive Normal Stress \ t Tensile ~~ Normal 7 Uniform Bending Stress Pipe Shoe W/Saddle ‘Saddle inhibits uniform bending and (Normal stresses very extension along all fibers at the cross lineerly from the neutral section. axis) Figure 3-5 4 - Hooke’s Law is applicable throughout the entire load range: Ay () Stress Distribution —— Not Plastic Remains LinearCOADE Pipe Stress Analysis Seminar Notes 5 - Moments and forces applied to the beam are assumed to act about the neutral axis: { : let tt. } Should not be modeled as: T 1 (Unless the F*L. moment can be assumed negligible) Figure 3-7 6 - Element cross-sections do not ovalize under load (except as adjusted for bend os | 3S SECTION A-A_ This ovalization will make the pipe more flexible, i.e. the pipe will bend easier. Ovalization of this type for straight pipes is not considered. Figure 3-8 The stresses at the ovalized section are intensified due to: 1 - reduction in section modulus, and 2 - added local plate bending in the top and bottom fibers. 35COADE Pipe Stress Analysis Seminar Notes 7 - Applied loads are not affected by the deformed state of the structure (P-delta effect): 1000 1» Figure 3-9 Inreality, there will be an additional moment applied to the system, equal to the load times its displaced distance from the neutral axis of the structure (i.e., 1000 pounds x 0.25 inches = 250 in-Ib). The computer software models this load as. strictly a force with no applied moment. 8 - Rotational deformations of the system are assumed to be small: Aja ak APA ok Figure 3-10 3-6COADE Pipe Stress Analysis Seminar Notes Node point rotations are added vectorially by the computer software. This is not a valid representation of reality for large rotations, as demonstrated in Figure 3-10 for three 90° rotations. Boundary conditions are assumed to respond in a linear fashion: linear. Nonstineae Restraint Response Restrain Constant Threugheut Ditferent for Up and {eed Range Dow Cocde (One Woy Restrain’) a Figure 3-11 4 The stiffness algorithm cannot solve for non-linear restraint conditions, such as one-directional restraints, bi-linear restraints (soil or bottomed out springs), friction, ete. However, CAESAR II does include a procedure which overcomes this limitation; see point 8 below. These limitations are of the most concern when modeling the following situations (pointers for increasing accuracy in each situation are also given): 1 Large Diameter/thin wall piping or ducts: In this case, itis advisable tominimize localized loadings by distributing them with pads or saddles, or do plate buckling analysis (preferably with finite element software) when the loads cannot be altered. Localized stress conditions not explicitly covered by an SIF, ie. a saddle: The portion of the pipe impacted by the saddle may be modeled as a rigid element, while saddle/piping local stresses may be estimated through the use of finite element analysis or through the use of Welding Research Council Bulletins, such as 107 and 198. Pipe connections to thin walled vessels: The flexibility of the connection may be modeled by a flexible element (such as that generated using Welding Research Council Bulletin 297), while stresses in the pipe and vessel may be estimated through the use of finite element analysis or through the use of Welding Research Council Bulletins 107 and 297. Highly corrosive systems (especially when subjected to cyclic loadings): Corro- sion of a pipe results in an irregular cross-section which is usually modeled by using the uncorroded cross-section for load generation (weight and thermal forces), and the fully corroded cross- section for calculation of the section modulus 3-7COADE Pipe Stress Analysis Seminar Notes (stress calculation). Corrosion is much more dangerous under fatigue loadings due to the fact that it provides many more opportunities for crack initiation; in order to compensate, a low cyclic reduction factor should be used to match the allowable expansion stress range to the fatigue curve for a highly corroded material. 5 - Elbows: Elbows ovalize significantly when subjected to bending loads. This can be accounted for by increasing the flexibility of the elbow element in the computer model and multiplying the calculated stress by a stress intensification factor (thisis done automatically by most programs such as CAESAR ID). Code defined “flexibility factors” for bends have been determined theoretically and verified experimentally. _— Weldline S Flange \ <4r «= moan radius ofthe pip cross ebaton (From BS 806-1975) Figure 8-12 The flexibility and stress intensification factors of bends must be reviewed in those cases where ovalization is inhibited (such as when the elbow is stiffened by flanges or welded attachments). The piping codes provide correction factors for bends with one or two flanges, but omit geometries such as shown in Figure 3-13. Zr Figure 3-13 ‘These attachments almost certainly affect the flexibility, and more importantly, the stress intensification factors for the bends. The factors for heavily stiffened bends, such as that shown in Figure 3-13(A), could be estimated using finite element analysis, or stiffness could be increased by modeling the elbows as 3-8COADE Pipe Stress Analysis Seminar Notes flanged, or simply as straight pieces of pipe (with increased stress intensification factors applied). In less pronounced cases such as those shown in Figures 3-13 (B) and (C), deviations from the response of an unstifiened bend is usually ignored. 6 - Loadings which produce stresses which are well outside of the code allowable ranges: These loads will tend to produce stresses well beyond the material yield stress, stresses in the buckling range, large displacements resulting in signifi- cant P-delta loads, or large rotations (leading to inaccurate results). This limits programs such as CAESAR II as accurate analysis tools throughout the full range of potential loadings. However accuracy is not affected for those loads which are of most interest to the engineer — code allowables are based upon the fact that the analysis being done assumes linear material response. 7 - Nonlinear boundary conditions: The effects of non-linear restraints must be simulated through an iterative process aimed at convergence of the non-linear restraints in legitimate states —for example, with the pipelifted offataone-way support (and with the support function removed from the analysis), or with the pipe sliding along a frictional restraint (and with an appropriate force applied along that line of action in the analysis). This process is activated (during static analysis) automatically when a non-linear effect is detected by CAESAR TL. 8 - Non-homogenous elements: As noted, piping elements are modeled as stick elements of constant cross-section and material properties. In certaincases, such as with reducers, which have a variable cross-section, this is not a valid representation. An element such as this is usually modeled as a single, or as a series of elements, each having average parameters. For example, a 12x8 standard wall reducer may be modeled as a 10-inch standard wall pipe (approxi- mately the average of the inlet and outlet pipes), or as two segments, with outer diameters and wall thicknesses interpolated between the two. When using those codes which define a stress intensification factor for reducers, one would have to bbe calculated and specifically applied at that location. 9 - Rigid elements: Rigid elements, such as valves and flanges are most difficult to model due to the inability to represent their geometry, and their stress distribu- tion with stick elements. Therefore, pipe stress software cannot be used to accurately determine the effects of the piping system on rigid elements. Analysis of these components is best left to finite element analysis, test, or other recognized methods. However, the effects of the rigid elements on the piping system can be simulated by providing an element ofhigh relative stifiness in the model (itis always more important to adequately model relative stiffnesses than absolute stiffnesses when constructing a model). This is done by providing an element with sufficiently large cross-section, and having the defined weight of the rigid item. In CAESAR Il, a rigid element is modeled as having a) aninsulation factor of 1.75 (compared tothe matching pipe), unless a zero weight rigid (a modeling construct) b) fluid weight of the matching pipe added, unless a zero weight rigid ©) thesame inside diameter and 10 times the wall thickness of the matching pipeCOADE Pipe Stress Analysis Seminar Notes 3.2 Piping Configuration Modeling Techniques Piping may be modeled in varying detail, depending upon how much accuracy is required. This section looks at the various ways (providing corresponding degrees of accuracy)in which sample piping configurations might be modeled. Consider the following geometry, of a large diameter pipe supported by a dual spring assembly: 50" diameter 41/2" wall — | eee # trunnion support shown. (Typ. both sides) Figure 3-14 ‘Simplest Method: Centerline 50" diameter pipe Flexible Restraint atcenterline = K K=2Kspring Figure 3-15 Limitations: 1 - local stress calculations not considered for 50" pipe 2 - stiffness of trunnions not considered 3 - torsional resistance due to the restraint pair is not considered (see Figure 3-16) 4 - local flexibility of the shell of the 50" pipe is not considered 3-10COADE Pipe Stress Analysis Seminar Notes 73 Resistance to t torsion Figure 3-16 More Accurate: Centertine 50° diameter pipe Rigid element - zoro weight Kspring Figure 3-17 Limitations: 1 - local stress calculations not considered for 50” pipe 2 - stiffness of trunnions not considered 4 - local flexibility of the shell of the 50" pipe is not considered ‘Most Accurate: Khel! local (WRC 297) Pipe element ‘madeling trurion Khel! local (WRC 297) wn Fligid element - zero weight Length = 172. 0D Kspring Figure 3-18 3-11COADE Pipe Stress Analysis Seminar Notes Limitations: local flexibilities and stresses only as close as WRC 297 and WRC 107 bulletins 1: (see discussion in this Section 3.4 of these seminar notes) ‘Looking at another configuration, a heavy-wall forged WYE fitting: + Extra heavy forged fiting (Wo tost sits or foxbities avaiable) Simplest Model: OS incoming pipe with branch ‘roperties coded to intersection point Apply SIPs for welding tee here. Incoming pipe with Properties coded to intersection point Figure 3-20 Limitations: 1 - weight of forged fitting probably underestimated considerably 2 - rigidity of forged fitting probably underestimated considerably 3 - stress intensification factors may be too conservative 3-12COADE Pipe Stress Analysis Seminar Notes More Accurate: Rigid elements whose ‘cumulative weights ile te equal that of the forged ‘iting and flanges. Figure 3-21 Limitations: 1 - no provision for stress calculations in forging, but this isn’t usually a problem, because of the extra heavy wall of the fitting would ensure that the connecting pipe would probably fail first. Any questions regarding load capacity should probably be directed to the fitting manufacturer ‘Most Accurate: Rigid elements modeling flanges Pipe with wall thickness equal to that offing Rigid element modeling ange. Figure 3-22 3-13COADE Pipe Stress Analysis Seminar Notes Comments: 1 - the flexibility of this model will be more accurate (but only marginally so for a heavy fitting) 2 - stresses (unintensified) will be computed at the crotch; however, there will be some unknown intensification factor existing at the crotch 3 - thismodel probably does not yield any significant improvement over the previous one One of the most common types of pipe support is shown in Figure 3-23: 0" dia. std. wall 6 dia. std. wall stanchion Figure 3.23 Simplest Model: *Y* direction restraint appliog at the bend point shown. Figure 3-24 3-14COADE Pipe Stress Analysis Seminar Notes Limitations: 1 - flexibility of the stanchion is not included in the model 2 = the point of application of the stanchion is not at the correct location on the bend curvature 3 - pipe may lift off of (or lock up with) modeled support due to thermal expansion ‘between centerline of horizontal run and point of application on riser 4 - stiffening effect on bend of stanchion not considered 5 - local stresses at stanchion not considered More Accurate: Figure 3-25 Limitations: 1 - stanchion doesn’t act at the proper point on the bend curvature 2 - stiffening effect on bend of stanchion not considered 3 - local stresses at stanchion not considered 3-15COADE Pipe Stress Analysis Seminar Notes Most Accurate: F(t - cose 41.4 deg. For Long Radius Bend = 48.2 deg. For Short Radius Bend Use WRC107 To Calculate Local Stresses Mode! as Single Flanged Bend Figure 3-26 Limitations: 1 - points A and B aren't exactly at the same location (this can be resolved using CAESAR II's “OFFSETS” feature, but other pipe stress software may have a difficult time with this) 2 - modeling the stiffening effect of the stanchion on the bend through the use of a single flange bend is an approximate solution 8 - local stresses at the stanchion are only as accurate as WRC 107 bulletin A few configurations which illustrate solutions to potentially tricky modeling situations follow below: The distance L in Figure 3-27 may become important if the gap on the guide closes and there is a horizontal restraint force which will cause a torsional moment to exist in both members. 3-16COADE Pipe Stress Analysis Seminar Notes -Centerline of the pipe "stck* model eccrine toe structural ‘stick’ model a) Figure 3-27 Because the elbow in Figure 3-28 connects directly to the equipment flange and the equipment flange is anchored, the stiffness of the model in this local region is very high. If the stanchion connects at A and the equipment centerline is at B, the differential thermal growth of the elbow between those points could put enormously high loads on both the stanchion and the equipment model. This is also in reality, a difficult problem to design for. Unless the user is willing to put a spring at the stanchion location, the differential thermal growth in this small area might result in large nozzle loads. ca aU ar Rotating Equipment Centerline — Figure 3-28 In the Figure 3-29, a small, but heavy process monitor and actuator is mounted on the line. The rigidity, weight, and moment due to the offset is best modeled using a weightless rigid element going from the centerline of the pipe out to the center of gravity of the process monitor, at which point a small rigid element with the weight of the equipment should be modeled. The rigidity of the body of the monitor (within the pipeline) should be modeled as 3-17COADE Pipe Stress Analysis Seminat Notes arigid as well. (Note that some engineers may prefer to model the effects of this equipment by applying a force equal to the weight and a moment equal to the weight times offset at the centerline of the pipe. This approach, although acceptable for static analysis, is absolutely incorrect for dynamic analysis, and should therefore be avoided since it cannot be promised that no dynamic analysis will be conducted on a system in the future.) —3 J Process Monitor nw ‘Spring Loaded Actuator Cylinder a ee Figure 3-29 In Figure 3-30, the large 18 inch line comes directly from a flue-gas furnace, passes through a small exchanger and enters a waste heat boiler. This is a very stiff system relative to the vessel connections. Therefore, instead of modeling the connections as rigid anchors (which would give the same relative stiffness to the restraints and to the piping), WRC Bulletin 297 should be used to estimate and model the nozzle flexibilities. This method will provide the best approximation of the distribution of the piping loads to the vessels. Boller le 24's EBoler AN 18% os Boiler and fumace nozzle flexibilities? (his is a very tight, sit system) Rigid Element Figure 3-30 3-18COADE Pipe Stress Analysis Seminar Notes In Figure 3-31, rectangular ducting connects the two separators, which are rigid relative to the ductwork. In order to size each spring for its share of the distributed weight of the whole assembly plus the connected piping, it is best to simulate the stiffness of the duct through the use of an equivalent structural member or piping element. User defined eross section in structural Steel preprocessor cova of 7 # foe She, tet fteraia fy oo ping modeled as pipe Figure 3-31 An angle valve could be modeled as shown in Figure 3-32. It may be necessary to model it as three rigid elements if the weight of the operator is significant in comparison to the valve body Figis element ‘2 wot ol valve rT ees Figure 3-32 ‘The following sections of these seminar notes provide more detailed methods for modeling and analyzing specific components of the piping model. 3-19COADE Pipe Stress Analysis Seminar Notes 3.3 Expansion Joint Modeling And Evaluation Expansion joints are used when it is necessary to provide a large amount of flexibility in a small space. Expansion joints are constructed out of sheet metal, which, after rolling and welding to for a cylinder, has convolutions (also called corrugations) formed in it via either hydraulic pressure or rolling. Expansion joints may vary in terms of the number and type of convolutions, the material, the number of plies, all of which effect the pressure capacity, the stiffiiess, and the allowable movement. Figure 3-33 For the most part, these details are taken care of by the expansion joint manufacturer. A typical expansion joint piping design proceeds: 1- ‘The decision is made to use an expansion joint in the piping system. (In many design problems the joint is used to protect a sensitive piece of equipment from excessive nozzle loads.) Based upon the design temperature and pressure, a standard expansion jointis selected from a manufacturer's catalog. The properties of that bellows are then inserted into the piping model. Ifthe bellows reduces loads and stresses as intended then the range of expansion movements on the bellows must be checked. For each bellows there is a limit to the cumulative axial, bending and lateral displacement that can be absorbed by the joint without excessively deforming the convolutions or causing fatigue failure. These limits are presented in different ways in different manufacturer's catalogs, but are always functions of the number of applied cycles, bellows material properties and convolution shape. Where excessive displacement is a problem, increasing the number of convolutions can be the solution. Once the bellows movement is within the allowable range of movements, the design is completed. A competent expansion joint manufacturer should be able to provide assistance throughout the design stage as required. 3.3.1 Expansion Joint Stiffnesses Each particular combination of material, thickness, and convolution geometry has a different axial spring rate (per convolution) associated with it. Bending and lateral convolution spring rates can be computed from the axial spring rate. 3-20COADE Pipe Stress Analysis Seminar Notes ‘The behavior of a bellows under load is described by the following equations: F = fey Where = axial force in each convolution (also the axial force throughout the entire bellows), Ib £ = axial stiffness per convolution, Ib/in = NKax N = _ number of convolutions in the joint Kox total expansion joint axial stiffness, ibfin x = axial displacement per convolution, in = X/N X = total axial displacement of joint, in M, = fDe,/4 Where: ‘M, = _bendingmomentin each convolution (also the bending moment supported by the entire bellows), in-Ib D = is the effective diameter of the joint (equal to the inside diameter plus the height of one convolution), in ¢r = arial displacement per convolution resulting from a rotation ofthe convolu ion,in = (rxD)/(2N) r = bending rotation of single convolution, radians vo = fDey/@) < " shear force in each convolution (also the shear force supported by the entire bellows), Ib 3-21COADE Pipe Stress Analysis Seminar Notes ey = axial displacement per convolution resulting from a lateral deflection of the convolution. = 3Dy/(Nl) y = total lateral displacement of the joint, in 1 = length of the bellows, in ‘These expressions can easily be converted into stiffness and flexibility coefficients: Axial Stiffness: Ke = FR Bending Flexibility: Myr = (1/8) (Kx) (D2) Lateral Stiffness: V/y = (8/2) (D2) (Rax) / (2) These stiffness values are provided in most manufacturer's catalogs. In the event that the manufacturer only gives axial stiffness, the other two can be calculated once the effective diameter and length are known. (Note that torsional stiffnesses are not usually provided, since unprotected expansion joints are not designed to carry torsional loads and may fail catastrophically if inadvertently exposed to even moderate torsional moments.) Note however that the bending flexibility coefficient should not be used in any piping program. The bending stiffness that should be used is exactly four times the bending flexibility. This is because the so-called bending flexibility is calculated by applying a moment (Mr) to the free end of an expansion joint and observing its end rotation (0). A computer model, however, expects a bending stiffness to be the ratio of the applied moment to the angular rotation at the end of an expansion joint that is fixed against translation — ie, a representation of guided cantilever. This angularsstiffness for a guided cantilever expansion joint model is calculated as: Myr = (ax) (D2)/2 9n 5 task Mr, _ FLEXIBILITY ‘STIFFNESS Figure 3-34 Some pipe stress programs only offer “point”, or zero-length expansion joint models. (In CAESAR II the user can define “finite length” or “point” expansion joints.) There is a difference in terms of how the two models are entered. As seen above, for finite length expansion joints, the lateral and bending stiffnesses are related by the equation: 3-22COADE Pipe Stress Analysis Seminar Notes 1 Bending Stiffness Myr = Kay x D9/2 = W/y x 19/3, Lateral Stiffness x 12/3 Because of this exact relation, and since the length is known, the user can only enter one of these two values. CAESAR II computes the other value using this equation. For a “point” expansion joint, the length is unknown, so all three stiffnesses must be defined for the model. Exampl Consider an expansion joint with the following parameters: Nominal diameter = 4 in Effective Area = 19.6 in? K,x (from manufacturer) = 316 Ib/in Bellows Length = 4.447 in ‘The expansion joint stiffnesses are calculated as: Myr = (1/8) (Kax) (D2) D = = [4519.6/pi]#2= 4.9955 in Myr = (1/8) (4.99552) (316) = 985.7 in-Ib/rad = 17.2 in-Ib/deg. ‘The bending stiffness to use in a piping program would be: AxMyr = 4x17.2= 68.8 in-lb/éeg Vy (8/2) (D2) Kax)/ (1) (8/2) (4.99552) (316) / (4.4472) = 598.14 Ib/in 3.3.2 Evaluation of Expansion Joint Allowable Movements Since the failure mode of expansion joints is fatigue, the relative expansion displacements ‘between thestart andend of the expansion joint must be checked against the manufacturer's allowables. Note that the allowables provided will not be absolute values, but will be based upon a specific number of cyclic applications. The manufacturer must always provide a fatigue curve or some other type of adjustment factor in order to determine the allowable displacement for a different number of cycles. Occasionally, themanufacturer provides allowable movements only for axial displacements. In this case, the equations given in Section 3.3.1 can be used tocalculate an equivalent axial displacement from lateral and rotational displacements: Er RD/2, or: Ey 0.00872665 6 D 3.23COADE Pipe Stress Analysis Seminar Notes Ey = 3DY/1 Where: E, = total equivalent axial displacement due to rotation, in R = total rotation on expansion joint, radians = effective diameter of expansion joint, in @ = total rotation on expansion joint, degrees Ey = total equivalent axial displacement due to lateral displacement, in te i = length of expansion joint, in = total lateral displacement on expansion joint, in Therefore, movements on an expansion joint are acceptable if: X+Ey+ By Compute <2>Help More Figure 3-35 3.26COADE Pipe Stress Analysis Seminar Notes Output for this problem is shown in Figure 3-36. EJMA EXPANSION JOINT RATING More output: AXIAL DISPLACEMENTS PER CORRIGATION Axial Displacement, 033 Axial Displacement due to Lateral 044 Axial Displacement due to Rotation -006 ‘Axial Displacement TOTAL +083 RELATIVE MOVEMENTS OF END “i” WITH RESPECT TO END “3” (Local Joint Coordinate System) Relative Axial Displacement, “x” 401 Relative Lateral Displacement, “ +188 Relative Bending, “theta” (deg) 1.517 Relative Torsion (deg) -.019 | RELATIVE MOVEMENTS OF END “1” WITH RESPECT TO END “j* (Global Piping Coordinate System) Relative X Displacement -.399 | Relative Y Displacement ~.132 Relative Z Displacement 095 Relative Rotation about X (deg) +000 Relative Rotation about Y (deg) 1.250 Relative Rotation about Z (deg) -860 To Exit <2>For Help Input Output. Nore Figure 3.36 Results — e(total) — are given in terms of equivalent axial displacement per convolution — in this case, 0.083 inches per convolution. The accuracy of the initial estimate (done without the benefit of the ERATE program) of an equivalent axial extension of 0.904 inches can be determined by multiplying 0.083 times 12 convolutions, for a total equivalent extension of 0.996 inches (an error of approximately 9%). It can be seen that the bulk of the error is due to the underestimation of the lateral displacement (see x, y, theta, tors —the total axial, lateral, rotational, and torsional displacements respectively on the bellows as a whole) on the expansion joint — the estimate was 0.13 inches, while ERATE calculated it as 0.158inches. This s important, since lateral loadingis very critical. Additionally, without the use of a calculational aid such as ERATE, the lateral displacement will be the most difficult to estimate, especially as the rotational angle increases. ‘Torsion on the expansion joint should be approximately zero. Ifnot, itis recommended that the acceptability of the bellows for the torsional load be verified by contacting the expansion joint manufacturer. 3-27COADE Pipe Stress Analysis Seminar Notes 3.3.4 Modeling of Unbalanced Pressure Force Normally pressure in a pipeline is absorbed by tension in the pipe walls: Figure 3-37 12 inch nominal diameter pipe, standard wall (D; = 12 in), P = 250 psi Axial Tension = Area x Pressure = pi/4 d2P = pi/4 (122) (250) = 28,274 Ib Due to the axial flexibility of expansion joints, they are incapable of carrying this large axial load. Due to the larger internal diameter of the convolutions of the expansion joint, the pressure thrust force is calculated from an effective diameter, which is: De = Di+w Where: Det = effective diameter of bellows, in Dj = _ internal diameter of expansion joint (or pipe), in internal height of one convolution, in The unbalanced pressure force is therefore: Fp = P(pi/4) Deg? Where: Fp = pressure thrust force, Ib P = system pressure, psi ‘The pressure force is actually developed at the point where the pressure encounters the first metal area perpendicular to the axis of the expansion joint — for example, a capped end or a change in direction. The distribution of the pressure thrust loads in a number of configurations is shown in Figures 3-38 through 3-42. 3-28COADE Pipe Stress Analysis Seminar Notes Pressure thrust loads in untied, unanchored expansion joints: Taw q q qi a GTi re UN eee eee End is blind flange T = tension in bellows walls (tending to blow the bellows apart), Ib = P(pi/d) Den? T, = tension in pipe wall =P (pi/4) D2 Figure 3-38 Pressure thrust loads in anchored expansion joint installation: — nnasw2 Th To I we 7 Flanged end anchored T = tension in bellows walls, lb = 0 T, = compression in pipe wall, Ib = P(pi/d) Dea? - D2) T2 = total load on anchor, Ib = P(pi/A) Deg? Figure 3-39 3-29COADE Pipe Stress Analysis Seminar Notes Pressure thrust in tied expansion joi I hae ye 5 tension in bellows, Ib 0 tension in pipe wall, Ib P (pid) DAP) tension in tie bars, Ib P (pid) D2 Figure 3-40 Pressure thrust loads in rotating equipment without tie bars (ends anchored): ‘Spinging Impeller 23 1% tension in bellows, Ib 0 ee ees iS ane * ei eee an a_/ vy compression in pipe on either side of expansion joint, 1b P (pi/4) Den? - Dj?) pressure load on impeller and on inside of elbow, Ib P (pid) D2 reaction load on pump base and on anchor, Ib P(pi/4) Deg? Figure 3-41COADE Pipe Stress Analysis Seminar Notes Pressure thrust loads on rotat g equipment with tie bars: ea nics? Fj ix to T aaa ee eT pS Ww = Tg 3 T = tension in bellows, lb = 0 T, = tension in pipe on either side of expansion joint, Ib = P(pid) D2 Tz = pressure load on impeller and on inside of elbow, lb = P(pi/4) D2 Ts = tie bar load, Ib = P(pi/4) Deg® Figure 3-42 Note that there are no reaction loads (T4) when tie bars are used. Most pipe stress programs such as CAESAR II automatically calculate the unbalanced pressure thrust load and simplify the model by applying the entire magnitude at either end of the bellows. In most cases, this is an adequate approximation of the actual situation. Greater modeling accuracy can be achieved by disabling the application of pressure load at the bellows (by defining an effective diameter as0.0),and calculating and applying the thrust load manually to the model as so: Apply the force T-'T; at the ends of the bellows. Apply the force Ty - T; at the locations identified by the Tz arrows. Expansion joint, tie rod, and reaction loads T, Ts, and T, will be calculated correctly by the program. In any event, even though the tension/compression in the pipe wall may not be completely accurate in the default computer model, the load tending to open the bellows will be; this is usually a much more critical detail to consider when designing a system to absorb the pressure thrust. It is left to the user to confirm that this is normally not a major design issue. 3-31COADE Pipe Stress Analysis Seminar Notes 3.3.5 Modeling of Tie Rods Ifrestraints are not used to absorb the unbalanced pressure load, it must be handled by tie rods. These are tension-carrying rods attached to either end of the expansion joint, which prevent the ends from pulling apart. Since tie bars assure that there is no pressure reaction from the expansion joint, tie rods can be modeled in two different ways: 1 - implicitly, by omitting both the tie-rods and the pressure load from the model, or 2 - explicitly, by including both the tie-rods and the pressure load in the model. Assuming that the tie rods absorb 100% of the load, the net effect of both of these models on the piping system are the same. ‘Implicit Model of Tie Rods: ‘The first case noted aboveis obviously the simpler ofthe two. Omitting the tie rodsis possible if there is no pressure load on the bellows; this can be omitted by defining an effective diameter equal to 0.0 for the expansion joint. ‘Tie rods, besides absorbing the pressure load, also prevent extension and compression of the bellows under piping operating loads. Therefore, when the user leaves the tie bars and pressure thrust out of the model, itis alsonecessary toset the axial stiffness of the expansion joint to be essentially rigid (or actually to the total axial stiffness of the tie rods, which is AE/ 1). If the axial load on the expansion joint is tensile then the surrounding pipe is trying to stretch the tie bars even further. If the axial load on the expansion joints compressive, then: 1 - Ifthe compression is less than the pressure thrust load, there is not a problem. 2 - Ifthe compression exceeds the pressure thrust load, then the tie rods will be in compression. The compression must be checked to ensure that it is not so great that it buckles the tie bars. If the tie rods are tension only (ie., lock nuts are placed only on the outside of the expansion joint flanges) then some redesign is required, either: a) put nuts on the tie bars on both sides of the expansion joint flanges, or b) redesign the piping system so that the compressive load is not so great. Explicit Model of Tie Rods: When explicitly modeling the tie rods, the pressure load is included in the model by defining an effective diameter for the expansion joint. ‘The tie rods can be modeled by using a structural element (of the same cross-sectional area as the tie rods) to connect the two ends of the expansion joint. The structural element used could be a pipe, a rigid element, or a user-defined structural steel element. 3-32COADE Pipe Stress Analysis Seminar Notes Inthe event that a single pipe elementis used torepresent the tierods, the best way to provide the same axial stiffness as the tie rods is to leave the diameter equal to the diameter of the attached piping and set the wall thickness approximately equal to: t (N/A) Dyoa2 / Do Where: N number of tie bars Dod = diameter or tie rod, in Dy outer diameter of pipe element, in Ifa rigid element is used to model the tie rods, again the diameter should be set to that of the attached piping; the wall thickness should be set to: t = (N/40) Dyog2/ Do The rigid element should be given a weight equal to the total weight of all of the tie rods, which, if made of steel, is approximately: W = N(0.283) L (Pi/4) Dyoa2 Where W = total weight of tie rods, Ib L = length of tie rods, in ‘The tierods are modeledin CAESAR II toresistonly axial loads through the use ofrestraints with “CNODEs” (other nodes in the system to which a restraint is connected), Consider the expansion joint/tie rod assembly shown in Figure 3-43. The bellows element is modeled as running between the two node points 5 and 10. The tie rod element is then run from node point 5 to node point 20, using the same delta-coordinates as for the expansion joint. This puts node points 10 and 20 at a coincident location, without any actual attachment. The attachment is provided by placing a restraint at the far end of the tie rod (node point 20) in the direction of the expansion joint axis, in this case the Y direction. Placement of arestraint here in this manner restrains node point 20 (the end of the tie rod) against a rigid point in space; this can be adjusted by defining the restraint node point 10 as a CNODE. This means that node point 20 is not restrained against a point in space, but rather that it cannot move in the global Y-direction relative to node point 10 — the end of the expansion joint — an effective representation of a tied expansion joint. 3-33COADE Pipe Stress Analysis Seminar Notes Tie bars designed to 2? take compressive or tensile loads axially. Figure 3-43 ‘The tie rods should be set to the ambient temperature, if they are outside of the piping insulation, or to a temperature closer to the operating temperature if they are inside of the insulation. ‘Tie rods may also be modeled in a more complex fashion, using multiple rigid elements, as shown in Figure 3-44. In this model, loosening of the nuts on the rods due to rotation of the expansion joint flanges will be simulated. Figure 3-44 As noted, tie rods must be checked for potential buckling loads afterthe analysisis complete. Oralternatively, they may be designed totake tension only. Thisis done by placing lock nuts only on the outside of the flange, as shown in Figure 3-45, In this case, the expansion joint, is prevented from extending by the nut, while the flange can move freely during joint, contraction. This configuration can be modeled in CAESAR II by using one-way restraints (or even gaps, if appropriate) between the end of the tie rod element and the CNODEs. For example, if the tie rod shown in Figure 3-42 was tension only, it would be modeled by placing a+Y restraint at node point 20, with 10 as the CNODE, indicating that the end of the tie rod cannot move down against the expansion joint (but can move up). 3-34COADE Pipe Stress Analysis Seminar Notes Possible free feterend "040 Rigid element trom ccenterfine of pipe to centerline of tie bars. TENSION ONLY TIE BARS. Figure 3-45 Complex expansion joint/tie rod models are cumbersome to build and check, but where hot, large diameter tight piping systems are being analyzed they yield the most accurate model. ‘This is especially true where tie rods are long and not designed for compression. In these cases a slight rotation of the expansion joint can put one side of the tie rods in compression and the other side in a greater tension. 3.3.6 Expansion Joint Assemblies: Expansion joints may be used in a number of different types of assemblies, based upon the application. Various assemblies are described and sample models thereof are shown. (Note that CAESAR II provides an expansion joint modeling feature which can automatically build many of the expansion joint assemblies shown here. It is accessed by pressing J at the piping input spread sheet.) 3-35COADE Pipe Stress Analysis Seminar Notes Hinge Joints: Hinge expansion joints are shown in Figure 3-46. SINGLE HINGE EXPANSION JOINT DOUBLE HINGED EXPANSION JOINT yee. ‘Center Hinge Anns Hinge Pin x my dlinge Pin Figure 3-46 The hinges restrict angular rotation of the bellows to a single plane, and may be used in a single- or double-hinged configuration, the latter of which comes asasingle unit, When using a pair of single hinged joints, the joints should be placed as far apart as possible to reduce angular rotations as well as forces and moments. In most cases, the hinges are designed to pass through the full pressure thrust load, so there is no need for tie rods. In some cases, the hinge connections may be slotted to permit axial displacement of the bellows, however, then the pressure thrust must be absorbed by adjacent anchors. A typical hinge application is shown in Figure 3-47. Note that the piping system requires bending in one plane only. 3-36COADE Pipe Stress Analysis Seminar Notes A computer model for a single hinge expansion joint is shown in Figure 3-48. Direction tor tional agar Leet Hinge axis Rigid elements, weight of each equal {0 1/2 tolal hinge assembly weight. o-——> Remember: Hinges are 40 46 a5 eA 0 almost always used in pairs Figure 3-48 Azero-length expansion joint is placed between node points 45 and 46. The axial, lateral and torsional stiffnesses of the expansion are set torigid (1E12), with the bending stiffness taken from the manufacturer’s catalog (or calculated from Deff” Kax /2. The effective diameter for the pressure thrust may be set to zero. The bending restriction is modeled by placing a rotational restraint, in the direction perpendicular to the axis of the pipe and the hinge, at node point 45, with a CNODE at 46. Half of the weight of the hinge hardware is assigned to each of the rigid elements, as it may be important for hanger design and/or equipment loadings. Gimbal Joints: Gimbal expansion joints are shown in Figure 3-49. Hinge Pin {],-Gimbal Ring 18:20 Hinge 25-30 Gimbal Hinge Pin 1035-1040 Bellows 1015-1020 Bellows ANGULAR AND TRANSVERSE Figure 3-49 3:37COADE Pipe Stress Analysis Seminar Notes Gimbals are designed to permit angular rotation in any plane. The hinges and gimbal ring are capable of absorbing axial pressure or vacuum loads, dead weight of adjacent piping, and torsional moments. Like hinges, if gimbals are used in pairs, they should be located as far apart as possible to maximize absorbed displacement, and reduce rotation and forces and moments. A typical gimbal joint application is shown in Figure 3-50. Note that the piping system requires bending in two planes. Figure 3-50 Acomputer model for a gimbal joint, such as the first one shown in Figure 3-49is very simple to build — one simply defines an expansion joint with rigid axial, lateral, and torsional stiffhesses, and a bending stiffness equal to that of the actual expansion joint used, since the gimbal is free to bend in all directions. (Rigid elements with weights equal to the weight of the gimbal assembly may be included as well.) Sometimes, however, a gimbal may be used inconjunction with hinges, as shown in the diagram of the angular/transverse jointin Figure 3-49. A computer model for something like this is more complex to build — one solution is shown in Figure 3-51. 10 15 % $ 2025 30 40 945 EO 10 1015"1020 10351040 45 Figure 3-51 ‘The hinges/gimbal are modeled as the rigid series of elements running from 10 to 45 along the top of the figure, while the bellows and spool piece are modeled as the elements running along the bottom of the figure. 3-38COADE Pipe Stress Analysis Seminar Notes Element 10 to 15 is a rigid element, having a length equal to the distance from the face of the inlet flange to the axis of the first hinge, with a weight equal to approximately one- quarter of the total weight of the hinge/gimbal hardware (note that fluid and insulation weightis automatically added tonon-weightless rigid elements, so that shouldbe considered when assigning a weight to these elements). The hinge (element 15 to 20) is modeled as a zero-length expansion joint with rigid (1E12) axial, lateral, and torsional stiffnesses, and a bending stiffness of 1 (which is effectively zero). The hinge is restricted to one-directional rotation by restraining node point 15 rotationally about the X-axis, with a CNODE of 20. Element 20 to 25 is another rigid element, having a length equal to the distance from the axis of the first hinge to the mid-point of the gimbal, and again a weight equal to one-quarter ofthe total hardware weight. The gimbal (element 25 to 30)is a zero-length expansion joint with rigid axial, lateral, and torsional stiffnesses, and a bending stiffness of 1 (unrestricted by any restraints). Element 30 to 35 is a third rigid element, having a length equal to the distance from the mid-point of the gimbal to the axis of the second hinge (again with one- quarter of the total hardware weight). The second hinge (element 85 to 40)is modeled in the same way as the first, except that the rotational restraint applied at node point 35 is about the Y-axis, with a CNODE of 40. The final rigid element, from node point 40 to 45, has a length equal to the distance from the axis of the second hinge to the face of the outlet flange, and provides the final quarter of the hardware weight. Since neither the hinges nor gimbals are internally pressurized, the expansion joints which are used to model them should be given effective diameters of zero as well. ‘The expansion joints and spool piece will be modeled from node point 10 to 45 as well, indicating that the centerlines of the two assemblies are coincident, but connections are present only at the end points. Elements 10 to 1015 and 1040 to 45 are rigid elements with the length and weight of the two end flanges. Elements 1015 to 1020 and 1035 to 1040 are finite length expansion joints modeled with the exact properties of the actual bellows used (including effective diameter). Element 1020 to 1035 is modeled as a normal pipe element, representing the spool piece between the two expansion joints, Universal Joints: A universal expansion joint is shown in Figure 3-52. Universal joints consist of two unrestricted expansion joints flanking a spool piece. They are usually used to absorb large lateral movements in any direction. By increasing the length of the center pipe the amount of lateral displacement absorbed can be increased, with a corresponding reduction in the lateral forces and bending moments. In most cases universal joints are tied to prevent the pressure from blowing apart the assembly. (When universal expansion joint must absorb axial movement other than its own axial growth, an untied universal should be used — in that case, adjacent restraints must be designed to handle the pressure thrust.) A center support may be provided on the tie rods to help support the weight ‘of the center piece, to provide limit stops for the displacement, and/or to reduce the length of compressive rods (and the corresponding tendency of the rods to buckle). 3-39COADE Pipe Stress Analysis Seminar Notes UNIVERSAL EXPANSION JOINT Reo Figure 3-52 A simple model of a universal expansion joint is shown in Figure 3-53. Ambient Temp., Zero length expansion lateral and torsional stiffnesses rigi and no pressure thrust. 3 4 5 1005 § qoos_Z g Expansion joint stitinesses inserted jeen the Zero length elements: 5-1005 & 6-106 Figure 3-53 3-40COADE Pipe Stress Analysis Seminar Notes Since the tie rods isolate the pressure loads and thermal axial growth of the joint from the remainder of the system, the assembly can be simply modeled by calling the effective diameter of the bellows zero, the axial stiffness of the bellows rigid, and the temperature of the assembly ambient. Elements 4 to 5 and 1006 to 7 are rigid elements (with the weight of the tie rods, etc.), elements 5 to 1005 and 6 to 1006 are expansion joints with effective diameters of zero, axial stiffness of rigid, and bending and lateral stiffnesses as determined from the manufacturer. Element 1005 to 1006 is simply a pipe element representing the spool piece. The total length of the elements from node point 4 to node point 7 should be the same as the length of the tie rods, and, as noted above, the assembly should all be set to ambient temperature. ‘More complex models involve entering as accurately as possible the bellows, tie rods, and all supporting mechanisms. These models are very cumbersome to build but will give the most accurate representation of the loads, movements, and other conditions in and around the joint. More complex universal joint models are shown in Figure 3-54. 3001 42 1042 too2 Use double acting x restraint with gap and a a connecting node: F Gap to imi lateral ct oes “tection = 15° ™ fe 1090 x "1003 30 Figure 3-54 3-41COADE Pipe Stress Analysis Seminar Notes Other expansion joint configurations may be modeled by modeling various combinations of bellows, hinges, gimbals, tie rods, limit stops, and other hardware as shown in these figures. Amore accurate (and more likely correct) representation ofthe real configuration can usually be achieved with a more complex model. When modeling the assembly, total hardware weight must be considered, including internal or external sleeves andbellowsend connection details, in addition to the items noted above. 3-42COADE Pipe Stress Analysis Seminar Notes 3.4 ping Nozzle Evaluation Often piping configurations that are acceptable by stress criteria are limited by the loads which they place on connected equipment and vessels, Piping loads on the nozzles of equipment such as pumps, compressors, turbines, and heat exchangers may have the tendency to deform or overstress equipment casings, overload bearings, or cause shaft binding. Normally manufacturers should provide allowable nozzle loadings to which their equipment may be subjected, or they may reference industry standards, such as NEMA SM-23 (Steam Turbines), API 610 (Centrifugal Pumps), API 617 (Centrifugal Compressors), API 661 (Air Cooled Heat Exchangers), or HEI (for Closed Feedwater Heaters). These standards provide look-up tables or simple calculations which serve as a common reference for equipment vendor and engineer. Piping attached to vessels induces stresses in the vessel walls, in the form of membrane and bending stresses. These stresses must be evaluated against the requirements of the ASME Boiler and Pressure Vessel Code, Section VIII, Division 2. Calculation of stresses in a vessel wall is difficult without a finite element analysis; the best means of doing a hand calculation is to use aids such as Welding Research Council (WRC) Bulletin 107 “Local Stresses in Spherical and Cylindrical Shells due to External Loadings”. Piping attached to vessels may also tend tobend, buckle, or otherwise deform the vessel wall, permitting some displacement or rotation of the connection under load. Therefore, totally rigid restraint models may not be accurate representations of piping to vessel connections, Flexibilities of the connection should be estimated when possible; this can be done with the assistance of literature such as Welding Research Council Bulletin 297 “Local Stresses in Cylindrical Shells Due to External Loadings on Nozzles —Supplement to WRC Bulletin No, 107”. ‘The three types of possible analyses which may be done on nozzles — evaluation of equipment loads, calculation of vessel stresses, and calculation of piping/vessel connection flexibilities — are described in this section. 3.4.1 Equipment Nozzle Load Analysis ‘The most accurate means of evaluating a piece of equipment for anticipated nozzle loads is to perform a test. In lieu of a test, the next best method may be a finite element analysis, if operability failure (as opposed to stress failure) can be accurately determined from the model. In the absence of either of these, the engineer can often specify that the equipment meet a recognized standard, which provides for evaluation of nozzle loads. This standard, which may provide look-up tables or simple calculations, becomes a common reference between the manufacturer and the engineer — a promise that the equipment can stand at least a certain set of loads, which the engineer can then ensure that the piping loads remain below. It should be noted that these loads are minimum loads — in most case, the standards do not provide a means of actually evaluating the capacities of individual pieces of equipment. Equipment can be modeled in the piping problem in a number of ways. The nozzles can be considered to be rigid anchors, or entire pieces of equipment can be built-up from an 3-43COADE Pipe Stress Analysis Seminar Notes assemblage of rigid elements, with varying degrees of complexity. In either case the forces considered when evaluating the equipment are those forces which the pipe stress analysis, shows are acting at the equipment connection. The load cases for which the nozzle loads are to be checked are the greater of those from the cold and the hot cases — that is, from the sustained and the operating load cases (except when cold spring is considered, in which case the cold case would be sustained plus the effects of cold spring). ‘Typically suction, discharge andextraction lines are includedin separate pipe stress models. Once all of the loadings on a particular piece of equipment have been computed, the equipment can be evaluated to determine whether these loads are acceptable (ie., in accordance with the governing standard). CAESAR II provides the ROT program, which may be accessed from the main menu, to automatically evaluate piping nozzle loads against the requirements of a number of these standards. Equipment (and standards) covered include: 1 - Steam Turbines — National Electrical Manufacturer’s Association (NEMA) Standard SM23 2 - Centrifugal Pumps — American Petroleum Institute (API) Standard 610, 6th and 7th Editions 3 - Centrifugal Compressors — API Standard 617 4 - Air Cooled Heat Exchangers — API Standard 661 5 - Closed Feedwater Heaters — Heat Exchange Institute (HED Standard Inorder to use this program, the user is required to enter some description ofthe equipment (geometry, nozzle sizes, etc.) and the applied loads. Specific requirements of these standards (and the corresponding use of the ROT program) are described below. 3.4.1.1 NEMA SM23 Standard for Steam Turbines NEMA Standard SM23 requires that steam turbines be evaluated by twosets of simple force/ moment calculations. The two types of computations are used to satisfy: 1. - individual nozzle allowables 2 - cumulative equipment allowables First, the loads on each individual suction, discharge, and extraction nozzle must satisfy the equation: 3F +M<500D Where: F = resultant force on the particular nozzle, Ib 3-44COADE Pipe Stress Analysis Seminar Notes M D resultant force on the particular nozzle, ft-lb nominal pipe size of the connection, in Ifthe loads on the individual nozzles are acceptable, the cumulative load on the equipment from all nozzles must be checked next. Cumulative equipment allowables require that forces and moments on all connections be resolved at the discharge nozzle; the absolute magnitude of each of the force and moment components must then fall below a multiple of De, as: Fx <= 50D, Fy <= 125 D, Fz <= 100 D, ‘Mx <= 250 De ‘My <= 125 De Mz 125 De F. + Me/2 <= 125 De Where: Fx total X-foree (from all nozzles) on the equipment (where the X-axis is defined as being parallel to the equipment centerline), Ib total Y-force (from all nozzles) on the equipment (where the Y-axis is coincident with the direction of gravity), Ib total Z-force (from all nozzles) on the equipment (where the Z-axis is defined ‘by the right hand rule form the other two), Ib total resultant force acting on the equipment, Ib total X-moment (from forces and moments on all nozzles) acting on the equipment, resolved about the discharge nozzle, ft-Ib total Y-moment (from forces and moments on all nozzles) acting on the equipment, resolved about the discharge nozzle, ft-lb total Z-moment (from forces and moments on all nozzles) acting on the equipment, resolved about the discharge nozzle, ft-lb total resultant moment acting on the equipment, resolved about the dis charge nozzle, ft-lb diameter of an opening whose area is equal to the sum of the areas of all individual equipment nozzles, in 8-45COADE Pipe Stress Analysis Seminar Notes Anexample ofa NEMA SM23 analysis is shown in Figure 3-55. In this example, the turbine ismodeled as a series of rigid elements, soboth the discharge and suction nozzle loads appear on the same pipe stress analysis output. 65 a FIELD ANCHOR ‘0 r a 10! T° . rape ot / Fy HEADER "Mes “, Bs 50 nesesengase THREE RIGID ELEMENTS 45% “Eqn Figure 3-55 ‘The forces and moments acting on the elements connected to the turbine (excerpted from the CAESAR I output for this piping system) are shown below: wove rx Y fz wx ny we 30 -108 “49 “93 ns 188 603 35 108 67 93 162 “47 -481 30 “192 7 uu 369 “522 39 55 192 “63 u 78 aw “56 Since the sign conventions of these forces is that of forces acting on the elements, it is necessary to reverse the signs to get the forces and moments acting on the turbine nozzles at points 35 and 50, or: NOZZLE LOADS AT TURBINE NODE x FY Fz x Hy nz 36 -108 -93 -162 a7 481 50 192 u -369 522 -39 3-46COADE Pipe Stress Analysis Seminar Notes When running the NEMA portion of the ROT program, there are two input spreadsheets upon which descriptive and load data must be entered, which in this case would be filled out as shown in Figure 3-56. NEMA SHZ3 Input Date Equipment ID = NEMATL Suction Nozzle Node Number Suction Nozzle Nominal Diameter . peewee (iat) cr Discharge Nozzle Node Number v0.20... Discharge Nozzle Noninal Diameter ......) int EE Factor for NEMA Allouables (1.85 for API 617) ... Equipment Centerline: 1-X, 2-2. we Per Extraction Nozzle i Node Number Extraction Nozzle #1 Nominal Diameter . Extraction Nozzle #2 Node Nurber pogo fT Extraction Nozzle #Z Nominal Diameter ......Cin.) PoUp/PgDn/Home/End-Change Page © CEscI-Exit_ [Fi] ~Execute NEMA SHZ3 Input Data Equipment ID = NEMATL X Distance From Discharge to Suction ......¢in.? ¥ Distance From Discharge to Suction |. 2 Distance From Discharge to Suction | X Force Acting on Suction Nozzle ¥ Force Acting on Suction Nozzle Z Foree Acting on Suction Nozzle X Mowent Aeting on Suction Nozzle ......¢ft.1b.) Y Moment Acting on Suction Nozzle ......Cft.1b.) Z Moment Retion on Suction Nozzle Seetl 1b» X Force Acting on Discharge Nozzle ........C1d.) ¥ Force Acting on Discharge Nozzle - Tap 2 Fores Acting on Discharge Nozzle . Tal X Moment Acting on Discharge Nozzle ....¢£t.1b.) Y Moment Acting on Discharge Nozzle ....¢ft.1b.) Z Moment Acting on Discharge Nozzle ....(ft.1b.) PgUp/PaDn/Home/End-Change Page CEsci-Fxit —‘CF1] ~Execute Figure 3-56 3-47COADE Pipe Stress Analysis Seminar Notes ‘The program then proceeds to check the loading on each nozzle, and then resolves the loads about the discharge and checks the cumulative loading on an equivalent nozzle, as described above. The two page NEMA output report for this particular example is shown in Figure 3- BT. The results first show that the discharge nozzle passed (only reaching 30.4% of the allowable), the suction nozzle passed (reaching 49% of the allowable), and the cumulative load passed, with the highest summation load (Z-moment) reaching only 56% of the allowable. If the turbine had failed, the symbol **FAILED** would have appeared in the “STATUS” column next to the load combination that was excessive. Nozzle Node Dianeter Diameter Distance Fron Discharge Nozzle Wom) (USED) x Y Z cin.) Gn.) Gn Gn ny DISCHARGE 50 8.000 8.000 000 000 000) sucTION 35 4.000 4.000 = 34.000, 000 15.000 Individual Nozzle Cateutations Nozzie Node Components Resultants —Values/Allowabies. (ab. fete. DISCHARGE so 18% Fm = 126 Ae 7 sz re ou s00*D(used) = 4000 Mee 369 Hof ALLOW. = 30.40 we 522 me st0 Me 39) ‘sucTToN 35 108 SF = 980 Ae -67 a7 = -93 s00*D(used) = 2000 Woe -262 % of ALLOW. = 49.00 MAT 509 M481 MxCAbout dis. noz-)= 83 Nv(About dis. noz-)= -298 Na(About dis. noz-)= 189 ‘Sunmation Calculations. Diameter Due to Equivatent Nozzle Areas, OC~ 8.94 in. Nozzle Loads Sumations ANionables Hof ALLOW. Status Cb. & fteIb.) sk = at soc- 447 18.79 sy = 8 asec = 18. 6.62 ssa aoc = 884 9.7 FOORSLT) = 138 hx = 47 2sornc = 2236 20.00 shy - waste = 118 15.25 she = ol asec =~ le 56.51 WeCRSLT) = 792 Fo+Ms2~ 535 asec = ae 47.85 3-48COADE Pipe Stress Analysis Seminar Notes 3.4.1.2 API 610 Standard for Centrifugal Pumps Analysis of pump nozzle loads is covered by API Standard 610 “Centrifugal Pumps for General Refinery Service”. The requirements of the 6th edition, issued in January of 1981, and those of the 7th edition, issued in February, 1989, can be evaluated using CAESAR II's ROT program. The requirements of the two editions differ slightly; those of the 6th edition are described first: API Standard 610 basically presents reference tables of nozzle loads which a specified size pump nozzle must be good for. The 6th edition of the standard provides separate allowables for two types of support/base plates: 1 - heavy duty 2 - standard ‘The “heavy duty” pumpis typically a reinforced standard pump which can withstand higher piping loads. (The criteria necessary to meet the requirements of a heavy duty base plate/ support are discussed in paragraph 2.4.6 of API 610, 6th Edition.) All force components acting on each nozzle of a heavy duty pump must be less than two times the value shown in ‘Table 2 for that specificload component. For pumps with heavy duty base plates, this is the only requirement. A copy of Table 2 (from the 7th edition) is shown in Figure 3-58. Table 2—Nozzie Loadings Note: Each value shown below indicates a range from minus that value to plus that value: for example, Y60 indicates a range from ~ 160 to +160. ‘Nominal Sa of Norse Flange (inches) po) Dn ee ee Each top nozzle x 10 240 2D 56 BSD 12001500 1600 190 rag 200 © 300-400 70011001300 1800 200 3K Fz quo 2026096070000 12001300150 FR 30) «5710101560200 2600 2900 33K Each side nezle Fx 10-240 30 SSD 115600150 24 330-200 260 460, 700000120300 150 4 20 ©3000 70011001800 3800 2000 BL FR 200 430-5010 1560 2200 2600 29 38K Each end nome 7x 20 © 3004007001100 15001800 2000230 7 130-200 260 4607003000 12001300150 rz 1 «0D 568512001500 1600190 FR 20 30-57 1010 15602002600 00 HK Each nozale Mx Mo 700-980 1700 2600-3700 45004700 Suk uy 2 530 74013001900 2800 34D 3500 aoe MZ Yo 350 S087 1300s) 092300 270 MR 460 9501330, 2810350 $000 1006300720 [M = moment, in foot-pounds; R= resultant. See Figures 1-5 for orientation 01 and 2). Figure 3-58 3-49COADE Pipe Stress Analysis Seminar Notes ‘The “standard” API 610 pump has a slightly more complex scheme for compliance 1 - Innocase shall any of the individual force or moment components on either the suction or discharge nozzles exceed twice the allowable value for the component as shown in Table 2. 2 - Providing that the first is satisfied, the pump passes if the resultant forces and moments on both the suction and discharge nozzles are within the allowables given in Table C1, shown in Figure 3-59 (note that resultant allowables are included in Table 2 of the 7th edition). Table C-1—Suggested Allowable Resultant Forces ‘and Moments (for Vendor's Standard Baseplates) —— ‘Nominal Size of Nowa Flange Gaches) Fowce/Momet 2 3 4 6 8 (10 1 e 430 640 860 1500 2300 2700 2900 M, 0 1400 2000 3500 5200 6100 $200 Nove: F; = resultant force. in pounds; M, = resulumt moment, in foot-pounds.. ‘For sies larger than 12 ioches, allowable reultants shouldbe mumally agreed upon bythe purchaser andthe vender. Figure 3-59 3. - lfthepump passes the first requirement, but fails the second, the pump may still pass if both conditions described below are satisfied as well: CONDITION 1 CRITERIA a) Theratioof the applied resultant foree to the allowable resultant force from Table C1 for each nozzle is less than or equal to 2. b) The ratio of the applied resultant moment to the allowable resultant moment from Table C1 for each nozzle is less than or equal to C; where C=2 for nozzles 6 in. and smaller, and C=(D+6)/D for nozzles larger than 6 in. ©) Foreach nozzle, the sum of the force ratio and the moment ratio found in (a) and (b) above must be less than 2. CONDITION 2 CRITERIA. ‘The summation of the forces and moments from both the suction and the discharge nozzles, taken about the base point (which is the intersection of the shaft axis and the pedestal centerline), must be less than the sum of the force and moment resultant allowables for both the suction and discharge taken from Table C1. ‘The 7th edition of this standard modified the requirements toasmall extent. The differences in the 7th edition are: 3-50COADE Pipe Stress Analysis Seminar Notes 1 thisedition does not recognize separate provisions for heavy duty baseplates;the methodology for evaluating any pump under this edition is the same as that for as standard base plate pump under the 6th edition 2 - different allowables are provided for vertical in-line pumps that are supported only by the attached piping 3 - there is no difference in the allowables for tension and compression loading on top nozzles 4 - the allowables for larger diameter nozzles have been increased in some cases = Tables 2 and C-1 have been combined into a single Table 2 An example of an API 610 analysis (if evaluating under the 6th edition, the pump is considered to have a standard baseplate) is shown in Figure 3-60. For this example, the suction and discharge piping were analyzed in separate models — the 8-inch suction piping was analyzed in a job called “INLETS” and the 6-inch discharge pipe was analyzed in a job called “DISCH3”. In both cases the pump nozzles were modeled as anchors. The forces and moments acting on these anchors, as printed in the CAESAR Ioutput report, are thenozzle forces and moments (with the same sign) that should be entered into the rotating equipment input spreadsheets. These reactions are shown below: REACTION LOADS AT PUMP NODE FX fv cy mx My Mz 305, 507 “1 -231 3950 28612189 50 105 30 430, 46 “51 =201 CONTINUE ON WITH t= JOB "DISCHS: CONTINUE ON WITH Soe "IMters Ls oiscHarce + i BASE POINT earacee we Es ar ages RY TAB LEOML NE, ERE Figure 3-60 3-51COADE Pipe Stress Analysis Seminar Notes When running the API 610 portion of the ROT program, there are three input spreadsheets upon which descriptive and load data must be entered, which in this case would be filled out as shown in Figure 3-61. API 610 Input Date Equiprent ID = 61971 6p oy = Pr Suction Nozzle Node Number ....-.+- ae rE Suction Nozzle Type: 1-Top. 2-Side, 3nd 1211111 of Suction Nozzle Nominal Diameter ... eins Base Point Node Nunber .. . Base Plate Type: 1-Standard, 2-Heavy duty. Discharge Nozzle Node Number ....-- : com Discharge Nozzle Type: 1-Top, Z-Side, 3-End ..... pace Discharge Nozzle Nominal Diareter . gins, Ber Factor for Table 2 Allouables . : Factor for Table C1 Allowables ... 5 Equipment Centerline: 1-%. 2-2, 34 ‘ Par API 610 Input Data Equiprent ID = 61073 SUCTION Nozzle Data Entry Nozzle Orientation : x Sor ( Base Point to Nozzle > = ¥, eace| ' z, om Forces on Nozzle : x, Bar : v Sree z, 2000) Moments on Nozzle x. ¥ z, aPI 610 Input Data Equipment ID = 610TL DISCHARGE Nozzle Data Entry Nozzle Orientation Cin? Base Point to Nozzle > fein Teint Forces on Nozzle rE tab. Ec Tab!) eee Moments on Nozzle X..CPE. 1b. cg yieel bl) rrr 2.ceel1bl) cre Figure 3-61 3-52COADE Pipe Stress Analysis Seminar Notes ‘The program then performs the compliance check, as described above. The two page API610 output report for this particular example is shown in Figure 3-62. The results first show the status of the individual nozzle checks — each of the load components on both the discharge and the suction are within twice the Table 2 allowable; however, the resultant moment on the suction nozzle exceeds the allowable from Table C-1. Because of this failure, Conditions and 2 must be checked. Since both of these conditions are satisfied, the pump meets the requirements of API 610 — despite the one failure. 38 939 |532 7 34a Ba|iee 2 BUR BER | Ske F Figure 3-62 3-53COADE Pipe Stress Analysis Seminar Notes 3.4.1.3 API 617 Standard for Centrifugal Compressors ‘The requirements of this standard are identical to those of NEMA SM23, except that all of the NEMA SM23 allowables are increased by 85%. Therefore it is not necessary to discuss this standard here. 3.4.1.4 API 661 Standard for Air Cooled Heat Exchangers ‘The API 661 Standard for Air Cooled Heat Exchangers for General Refinery Service covers the allowable loads on the vertical, collinear nozzles found on most single- or multi-bundled air cooled heat exchangers. The types of nozzles which may be checked are those which are labeled “9” in Figure 3-63. ES tion 2 Seethame $ Serer Hy Tateopace Pe 1 Teas pon commenter Figure 5—Typieal Construction of a Tube Bundle with Plug Headers Figure 3-63 API 661 has the following two requirements: 1 - each nozzle in the corroded condition shall be capable of withstanding the following moments and forces (referenced as Figure 8 values): Nomina apy (feb) Dianeter Ca FY rn mm ™ ” 1 100 150 100 50 7 50 2 150 200 150 7 120 70 3 150 250 300 200 300 200 4 500 400 500 ‘400 600 490 6 750 750 600 800 1500 1050 8 1200 2000 850 1100 3000 13500 10 3500 2000 3000 1280 3000 2000 2 2000 2000 1250 1500 3000 2500 “ 2500 200 3500 1750 3800 3000 3.54COADE Pipe Stress Analysis Seminar Notes 2 - the sum of the forces and moments on each fixed header (i.e. each individual bundle) will be less than 1500 pounds transverse to the bundle, 2500 pounds axial to the bundle, and 3000 pounds axial to the nozzle centerline, while the allowed moments are 3000, 2000, and 4000 foot-pounds respectively (referenced as Table allowables) ‘The input to the API 661 portion of the ROT program is relatively self-explanatory; typical input screens are shown in Figure 3-64 and the output report is shown in Figure 3-65. API 661 Input Data Equipment 1D = H661 Inlet Nozzle Node Nunber..¢ Optional >. Inlet Nozzle Nominal Diareter.............+-Cin.) Outlet Nozzle Node Number.( Optional ). Outlet Nozzle Nominal Dianeter. Figure 8 Force/Monent Multiplier...... Resultant Force/Monent Multiplier. Tube Bundle Direction ¢ 1-x, 2-Z )..... Peer API 661 Input Data Equipment ID = H661 SUCTION Nozzle Data Entry Nozzle Orientation Cin? Forces on Nozzle 2 Reb? : yeliiab : zilliial Moments on Nozzle X..ft.1b.? yeceti tb. 2.creltbl> Figure 3-64 3-55COADE Pipe Stress Analysis Seminar Notes API Standard 661 1987 2nd Edition File + H66t Date : DEC 16,1992 User Entered Description Tine + 11 6 pm Suction Nozzle Node Nunber 5 Discharge Nozzle Node Nurber 28 Suction Nozzle Nominal Diaveter = 8. Discharge Nozzle Nominal Diameter = 8 User Entered Figure 6 Allouable = — .a0 User Entered For/Mon. Allouable = 188 Tube Bundle Direction = x Units For This analysis = in. th. FID. in, Suction Figure 8 Force & Moment Status Values Ratios ¥ Distance = 18.0 KForee = 100.8 1228. PASSED Yrorce = Baza 20081 PASSED ZForee = 30.0 958, PASSED X Moment = 203.0 a1g2. a8 PASSED Y Horent = 309.0 ee, tae PASSED Z Morent = 7300.2 1500) 1133 FAILED Discharge ‘Figure 8 Fores & Moment Status Values Ratios ¥ Distance = -10.8 X Force 203.8 zea. 7 PASSED ¥ Force 23.8 zae3. ler PASSED 2 Foree 22218 25a, 25, PASSED X Moment 300.0 1198. zr PASSED Y Moment 3000.2 3e00. 1108 PASSED 2 Moment 302.8 1528. "8 PASSED CEnter-Continue Resultant Foroo/torent Check © Resultant Table Allovable Ratios Status X Force 303.0 2500. az PASSED Y Force 28.8 3200, tes PASSED 2 Force -172:8 1500. ta PASSED % Moment site za0e. 48 PASSED & Moment 3308.8 a8, lez PASSED 2 Moment 2756.5 3280, ts PASSED Overall Loading Status x FAILED >, LEnter}-Continue Figure 3-65 3-56COADE Pipe Stress Analysis Seminar Notes 3.4.1.5 HEI Standard for Closed Feedwater Heaters ‘The HET Standard for Closed Feedwater Heaters provides a method for evaluating the allowable loads on shell type heat exchanger nozzles. This method is a simplification of the WRC Bulletin 107 methodology (discussed in Section 3.4.2 of these seminar notes), in which the allowable loads have been linearized to show the relationship between the maximum permitted radial force and the maximum permitted resultant bending moment. If this relationship is plotted (using the moment as the abscissa and the force as the ordinate) a straight line can be drawn between the maximum permitted force and the maximum permitted moment. Then, plotting the actual combination of applied force and moment, if these loads fall outside of the line, the nozzle is considered to fail, while if it falls inside of the line, it is considered to pass. ‘The maximum permitted force and moment is calculated from dimensionless parameters based upon the shell geometry; these parameters have been linearized for lookup by CAESAR II's ROT program. ‘A sample input to the HET portion of the ROT program is shown in Figure 3-66. Note that the program automatically calculates and includes the pressure thrust load (internal area ofthe pipe times the internal pressure) whenever the entered design pressure is greater than zero. HEI Nozzle/Vessel Input Data Equipment ID = HEITL Design Pressure (P)... Nozzle Outside Diameter Shell Outside Dianeter . Shell Thickness (7). Material vield Strengi Material Allouable Stress (sa) Maximum Radial Force. . Maximum Cireunferential Moment. Maximum Longitudinal Moment. (ab. 7eq, in. eles ine > > > 3 > > b> cee PoUp/PaDn/Home/End-Change Page = CEsc]-Exit_ —«[F1] Execute Figure 3-66 ‘The output report corresponding to this input is shown in Figure 8-67. Note that the nozzle failed for this application, since the load combination fell outside of the allowable load combination line. 3-57COADE Pipe Stress Analysis Seminar Notes Heat Exchange Institute (Nozzle Loads 84) File = HEITL Date : DEC 16,1992 User Entered Description : Tine | 10°44 pm User Input Values = Design Pressure 150.00 1b./eq.in. Nozzle Outside Dianeter 30.08 in Shell Dutside Diancter 5 in: Shell Thickness 5 ins Material Yield Strength 31500102 Ib. /eq.ins Material Allouable Stress [email protected] 1b. 7aquin. Maxinun Radial Force zaa80..2e 1b. Maxinun Circunferential Moment ‘9333.33 fe.1be Haxinun Longitudinal Moment. 1.08 Ftibe Pressure Thrust Load 95698.95 ib tEnter] ~ Continue Dirensionless Parameters Selected = 440.08 135 x 50.08 348.08 1180:08 Computed Force 7 Moment allouablest Fre| Fee = 53213.78 1b. Mren = S8537.35 ft.1b. Mele = 86081,13 felb, Mee = $9337.35, t1b Mem The wax allowed force for the input moments aaao4. a. Note, applied force includes pressure thrust if P > 8.0. ‘The nozzle FAILED in accordance with this analysis procedure. fn "x" outside the shaded region chous failure, tEnter] ~ Continue Figure 3-67 3-58COADE Pipe Stress Analysis Seminar Notes 3.4.2 Calculation of Vessel Stresses Due to Nozzie Loads Loads from piping attached to vessels induce stresses in the vessel walls, in the form of membrane and bending stresses. These stresses normally must be evaluated against the requirements of the ASME Boiler and Pressure Vessel Code, Section VIII, Division 2. Accurate calculation of stresses in a vessel wall is difficult without a finite element analysis; the best means of doing a calculation otherwise is to use a reference which parameterizes results of finite element analyses. The most common reference of this type is Welding Research Council Bulletin 107. Section VIII Division 2 of the ASME Boiler and Pressure Vessel Code provides fairly detailed rales for allowed stress in nozzle junctions and vessels. A synopsis of the elastic code rules are outlined here in order to provide some rules of thumb by which to analyze stresses in the vessel, at a junction: 1 - Section AD-160.3 contains two conditions to determine if fatigue analysis may be ignored for nozzles. The second of these, Condition BP is summarized below: a) The expected design number of full-range pressure cycles doesnot exceed the number of allowed cycles corresponding to an S, value of 4S,, on the material fatigue curve, where S,, is the allowable stress intensity for the material at the operating temperature. b) The expected design range of pressure cycles other than startup or shutdown must be less than 1/4 the design pressure times (Sq/S,,), where S, is the value obtained on the material fatigue curve for the specified number of significant pressure fluctuations. ©) The vessel does not experience localized high stress due to heating. 4) The full range of stress intensities due to mechanical loads (including piping reactions) does not exceed S, from the fatigue curve for the expected number of load fluctuations. 2 - If fatigue analysis is not required, then Appendix 4 states that the following limits must be satisfied: a) General pressure membrane stress intensity must be less than Sp. b) Primary membrane plus primary bending stress intensity must be less than L5S_. ©) Primary plus secondary stress intensity must be less than 35;,. Note that the 3S, limit applies to the range of stress intensity. The quantity 38, is defined as three times the average of the tabulated S,, values for the highest and lowest temperatures during the operation cycle. In the determination of the maximum primary- plus-secondary stress intensity range, it may be necessary to consider the superposition of cycles of various origins that produce a total range greater than the range of any of the individual cycles. The value of 3S,, may vary with the specific cycle, or combination of cycles, being considered since the temperature extremes may be different in each case. 3-59COADE Pipe Stress Analysis Seminar Notes In pipe stress terminology, this ean be approximated as: 1 - The sum of the pressure stress intensity in the vessel and the local sustained stress intensity at the nozzle connection, computed using WRC 107, must be less than 1.55,,. 2 - The sum ofthe pressure stress intensity in the vessel, the local sustained stress intensity at the nozzle connection, and the local expansion stress intensity at the nozzle connection, computed using WRC 107, must be less than 3S,, (where Sm is the average of the Sm at the operating and installed temperature). Because it is often difficult to include pressure in the local loading condition in a WRC 107 analysis, and because the area reinforcement requirements are supposed to take care of the pressure stress requirement at the intersection, it may sometimes be convenient to simplify these requirements to the following: 1 - The local sustained stress intensity at the nozzle connection, computed using WRC 107, should be less than 0.58. 2 - The sum ofthe local sustained stress intensity at the nozzle connection and the local expansion stress intensity at the nozzle connection, computed using WRC 107, must be less than 2.08,,. This is based upon the worst case assumption that the full value of S,, is used to satisfy the pressure stress; this leaves 0.5Sm to satisfy the local stresses from the sustained external loads. The same rationale can be applied to the second requirement as well, leaving 2.0Sm, to satisfy the local stresses from the sustained and expansion external loads. Ifthese reduced allowables are not satisfied then the engineer should review the magnitude of the pressure loading and revert back to considering it within the full local stress analysis. 3.4.2.1 Calculation of Vessel Stresses Due to Nozzle Loads The Welding Research Council Bulletin No. 107 (WRC 107) has been used extensively since the mid 60’s by design engineers to estimate local stresses in vessel/attachment junctions. ‘Welding Research Council Bulletin 107 is a parameterization of the results of a set finite element analyses examining stresses in vessels due to loaded attachments. WRC 107 contains equations and non-dimensional curves (based upon parameters such as ratios of the nozzle to vessel diameter and the vessel diameter to vessel thickness) which are used to extract coefficients for the calculation of stresses in the vessel wall at the point ofattachment. Note that WRC 107 computes stresses in the vessel shell at the nozzle/vessel interface — stresses in the nozzle wall (which in some cases can be higher than the stresses in the vessel wall) are not computed. Stresses in the nozzle wall may become greater than the stresses in the vessel wall as the t/T (nozzle to vessel thickness) ratio becomes less than one. WRC 107 may be used to analyze cylindrical or spherical vessel at attachments. The attachments may be round, square, or rectangular; the round and square attachments may be solid or hollow (i.e., a round hollow attachment represents a pipe or nozzle connection), 3-60COADE Pipe Stress Analysis Seminar Notes while only solid rectangular attachments (Lugs) are considered — the nozzle thickness does not effect the calculation. Appendix B to WRC 107 provides information on weldment stress intensification factors. Applications include vessel nozzles, pipe junctions, welded lugs and trunnions. The techniques introduced in this bulletin were updated and extended in WRC Bulletin 297, which may be used in a similar way as that explained here, ‘The convention adopted by WRC 107 to define the applicable orientations of the applied loads and stresses for both spherical and cylindrical vessels are shown in Figure 3-68. Io Ts nN ip Mc ce tp eM.) \* Yew > Ve ‘Sve postions A 8, ¢, ‘Sues postons 4B. C.D on aie surase iron ade sacs ‘Lon nie srtace oninaesurtace LOAD AND STRESS ORIENTATIONS LOAD AND STRESS ORIENTATIONS ‘ON SPHERICAL SHELLS ‘ON CYLINDRICAL SHELLS Figure 3-68 WRC Local Coordinates ‘The WRC 107 bulletin is used as follows. Based upon various dimensional ratios of the vessel/ nozzle configuration, the engineer selects 12 dimensionless parameters from as many different figures in the bulletin. These 12 parameters are used with local loads in 15 equations to calculate 80 different stresses — circumferential membrane, circumferential bending, longitudinal membrane, longitudinal bending, and shear stresses (in two direc- tions) at each of eight locations in the vessel. These eight locations are the at the inner and outer edges (identified by the subscripts l and urespectively) of the vessel, at the 0°, 90°, 180°, and 270° azimuth (identified as Figure 3-68 locations A, D, B, and C respectively) around the nozzle. 3-61