LR-N17-0034, Salem Generating Station, Units 1 & 2, Revision 29 to Updated Final Safety Analysis Report, Section 3.9, Mechanical Systems and Components
(Redirected from ML17046A298)
Jump to navigation
Jump to search
Text
- *
- 3.9 MECHANICAL SYSTEMS AND COMPONENTS The original design basis analysis of the Unit 1 and 2 Reactor Coolant Loop/Supports System was conducted based on an integrated analytical model which included the effects of the supports and the supported equipment. A three-dimensional, multi-mass, elastic-dynamic model was constructed to represent the Reactor Coolant Loop/Supports System (The model is as shown on Figure 3. 6-2) . The seismic floor spectrum at the internal concrete to support interface, obtained from an elastic-dynamic model of the reactor containment internal structure, was used as input to the piping analysis. The dynamic analysis employed displacement method, lumped parameter, stiffness matrix formulations, and assumptions that all components behave in a linearly elastic manner. The proprietary computer code WESTDYN was used in this analysis. The reanalysis of the reactor coolant loop, which addresses the elimination of snubbers from the steam generator upper supports, is described in Section 3.9.1.8. Normal Operating Loads System design operating parameters were used as the basis for the analysis of equipment, coolant piping, and equipment support structures for normal operating loads. The analysis was performed using a static model to predict deformation and stresses in the system under normal operating conditions. The analysis with respect to the piping and vessels was in accordance with the provisions of USAS B31.1 and ASME Section III. Results of the analysis gave six generalized force components, three bending moments, and three forces. These moments and forces were resolved into stresses in the pipe in accordance with the applicable codes. Stresses in the structural supports were determined by the material and section properties assuming linearly elastic small deformation theory. 3.9-1 SGS-OFSAR Revision 23 October 17, 2007 I I Seismic Loads Analysis for seismic loads was based on dynamic analyses. The appropriate time history motions at the basemat elevation were used as input forcing functions to the detailed dynamic models. The loads developed from the dynamic models were incorporated into a detailed support model to determine the support member stresses. Analysis of blowdown loads resulting from a loss-of-coolant accident (LOCA) was based on the time response of simultaneously applied blowdown forcing functions on a single broken loop dynamic model. The forcing functions are defined at points in the system loop where changes in cross section or direction of flow occur such that differential loads are generated during the blowdown t*ransient. The loads developed from the dynamic model were incorporated into a detailed support model to determine the equipment support member stresses. The stresses in components result from normal loads and the worst case blowdown analysis *were combined with the worst case seismic analysis to determine the maximum stress for the combined loading case. This is considered a very cons*ervati ve method since it is highly improbable that both maxima will occur at the s'ame i.nstant. These stresses were combined to determine that the Reactor Coolant I.oop/Supports System will not lose its intended functions under this highly improbable situation to assure the station capability for safe shutdown. The limiting stress criteria to be used in the analysis are defined in Section $.2. 3.9-2 SGS-UI?SAR Revision 23 October 17, 2007 * *
- 3.9.1 Dynamic Systems Analysis and Testing 3.9.1.1 Vibration Operational Test Program Despite the lack of formal codes pertaining to vibration testing, it has long been considered good engineering practice by Public Service Electric and Gas (PSE&G) observe for unanticipated transients during system testing. The observations include, but are not necessarily limited to, vibration, thermal growth, hydraulic transients, leakage, and the like. Directives to this effect are made part of general test documents rather than of each specific test procedure. The net result is that observations for excessive unanticipated transients will be made during operational and startup testing. Additionally, confirmatory component vibration testing will be performed as part of the preoperational test program in the form of the specific transients listed below. These tests should confirm that the piping and piping restraints within the reactor coolant pressure boundary have been designed to withstand the dynamic loadings from operational transient conditions that will be encountered during service. In addition to visually checking all systems for excessive vibration during the normal course of the preoperational test program, specific attention will be directed to evaluating possible vibration problems during the performance of the specific transients listed below: 1. Reactor Coolant System Heat up SGS-UFSAR fie Trans Operational Tests of Centrifugal Charging Pump (Step Changes) Reactor Coolant Pump Start Operation of Pressurizer 3.9-3 Revision 6 February 15, 1987
- 2. Reactor Coolant System at Temperature 3. Reactor Coolant System Cooldown 4. Emergency Core Cooling Full Flow Test 5. Chemical and Volume Control System Test Power-Operated Relief Valves Operation of Pressurizer Spray Valves Operation of Letdown Isolation Valves Operation of Pressurizer Power-Operated Relief Valve Reactor Coolant Pumps (Stopping and Starting) Initiation of Residual Heat Removal Initiation and Termination of the Following: A. Safety Injection (Safety Injection Pumps) B. Boron Injection (Centrifugal Charging Pumps with Primary Water) C. Safety Injection (Residual Heat Removal Pumps) Operational Tests of Positive Displacement Charging Pump (Stop and Start) Systems and components will be visually examined for the following types of deficiencies which are indicative of possible vibration problems: 1. Cracks in the grouting of equipment foundations 2. Leaking gaskets in piping systems and pump seals 3. Leaks from flanged connections in piping systems 3.9-4 SGS-UFSAR Revision 6 February 15, 1987
- 4. Metal-to-metal contact indications on piping system restraints 5. Water hammer "noises" during transient operations If any of the above types of indications are observed, or the response to any transient is deemed excessive by the cognizant test engineer, the response will be measured, evaluated, and documented. Methods available to measure the magnitude of piping system vibrational responses include: direct observation and measurement by scale for low frequency, large amplitude responses, and the use of stroboscopic equipment and inertial vibration sensors for high frequency, low amplitude responses. No speci fie "go-no go" criteria have been established for determining the acceptance of piping sys terns or components in terms of vibration requirements. Confirmation of structural integrity of vibration systems will be based upon favorable comparison of vibration-induced and code allowable stresses. In the event any structural or piping modifications become necessary as a result of excessive vibration responses, the specific transient(s) will be repeated to assure vibrational acceptability. 3.9.1.2 Testing Procedures Safety-related mechanical equipment was analyzed for seismic and accident loading conditions. However, seismic qualification type testing as recommended in WCAP -7558, "Seismic Vibration Testing With Sine Beats," by A. Morrone, October 1971, is performed for equipment which is too complex for analysis. The integrity of the RCS for normal vibratory loading conditions is verified by preoperational flow tests. Quality Assurance Engineers perform examinations of selected components. Particular attention is given to selected areas of the reactor vessel internals. The operating mode is evaluated in the analysis or testing of 3.9-5 SGS-UFSAR Revision 6 February 15, 1987 safety-related mechanical equipment whenever the component is required to function following the vibratory loadings. Safety-related air-operated valves supplied by PSE&G were either seismically tested or analyzed. Those valves which were vibration tested were shown to be operable during a seismic event through cycling of the valve. These tests demonstrated operability of the actuator and other vital appurtenances physically attached to the valve which are necessary for its proper operation. Those valves which were analyzed were demonstrated to have natural frequencies above at least 30 Hz and were therefore considered to be rigid components without amplification due to a seismic event. Motors associated with safety-related pumps provided by PSE&G have been tested or analyzed for operability under postulated seismic conditions as specified in Section 3.10. For pumps and valves in the Westinghouse Scope of Supply, vital appurtenances which are attached to an active pump and motor or an active valve operator were designed and installed using the applicable standards and engineering practice in use at the time of ordering equipment. The system containing the active equipment will undergo cold hydro testing and hot functional testing prior to startup. After startup, the components will be inservice inspected and tested as required at regular specified intervals to assure continued functional availability. I 3.9.1.3 Dynamic Testing and Examination of Reactor Internals A program consisting of extensive design analysis, model testing, and post hot functional inspection was used to establish internals integrity. Additionally, Westinghouse has instrumented full size reactors to measure the dynamic behavior of the of each size plant and has compared measurements with predicted values. This program was instituted as part of a basic philosophy of westinghouse to instrument the internals of the first-of-a-kind of 3.9-6 SGS-UFSAR Revision 20 May 6, 2003 the current nuclear steam supply system design. The magnitude of this test program was much greater than the intent of the philosophy, and was established as part of an extensive plan to develop theories and basic concepts related to internals vibration under various operating conditions. Thus, not only is added assurance obtained that all of the hardware will operate in the manner for which it was designed, but these data also assist in the development of increased capability for the prediction of the dynamic behavior of Pressurized Water Reactor (PWR) internals. The previous "first-of-a-kind11 plants that were instrumented were Zorita, one loop; Ginna, two loop; and Robinson, three loop. Indian Point 2, a four loop plant, has been the most thoroughly instrumented plant to date, for the reasons presented above. The Indian Point 2 reactor was established as the prototype for the Westinghouse four loop plant internals verification program. Subsequent four loop plants are similar in design. Past experience with other reactors indicates that plants of similar designs behave in a similar manner. For these reasons a comprehensive instrumentation program was conducted on the Indian Point 2 Plant to confirm the behavior of the reactor components. The main objectives of this test were to increase confidence in the adequacy of the internals by determining stress or deflection levels at key locations and to obtain data that can be used to develop improved analytical tools for prediction of internals vibration. In the final analysis, the proof that the internals are adequate, free from harmful vibrations, and perform as intended is through component observations and examinations during service. With this thought, Indian Point 2 (the 5 loop prototype) was subjected to a thorough visual and dye penetrant examination by a qualified Westinghouse Quality Assurance Engineer before and after the hot functional test. This inspection was in addition to the normal inspection of the internals in the shop, and before and after shipment. 3.9-7 SGS-UFSAR Revision 6 February 15, 1987 Safety Guide 20, Paragraph D, "Regulations for Reactor Internals Similar to the Prototype Design," is satisfied for subsequent four loop plants in the following The internals were subjected to a thorough examination prior to preoperational flow tests. This examination included the 35 points shown on Figure 3.9-2 (Sheet 2). These 35 points included the following: 1. All major load bearing elements of the reactor internals were relied upon to retain the core structure in place 2. The lateral, vertical, and torsional restraints were provided within the vessel 3. Those locking and bolting devices whose failure could adversely affect the structural integrity of the internals. 4. Those other locations on the reactor internal components which were examined on the Prototype Indian Point design. The interior of the reactor vessel was also examined for evidence of loose parts or foreign material. Specifically, the inside of the vessel was inspected before and after the hot functional test, with all the internals removed, to verify that there were no loose parts or foreign material. Lower Internals A particularly close inspection was made on the following items or areas, using a SX or lOX magnifying glass or PT where applicable. The locations of these areas are shown on Figure 3.9-2 (Sheet 1). 1. Upper Barrel Flange and Girth Weld 3.9-8 SGS-UFSAR Revision 6 February 15, 1987
- 2. Upper Barrel to Lower Barrel Girth Weld 3. Upper Core Plate Aligning Pin. Examine for any shadow marks, burnishing, buffing, or scoring. Check for the soundness of lockwelds. 4. Irradiation Specimen Basket Welds 5. Baffle Assembly Locking Devices. integrity. Check for lockweld 6. Lower Barrel to Core Support Girth Weld 7. The Flexible Tie Connections (Flexures) at the Lower End of the Thermal Shield 8. Radial Support Key Welds to Barrel 9. Insert Locking Devices. Examine soundness of lockwelds. 10. Core Support Columns and Instrumentation Guide Tubes. Check all the joints for tightness and soundness of the locking devices. 11. Secondary Core Support Assembly Welds 12. Insert Locking Devices. Examine soundness of lockwelds. 13. Lower Radial Support Lugs and Inserts. Examine for any shadow marks, burnishing, buffing, or scoring. Check the integrity of the lockwelds. These members supply the radial and torsion constraint of the internals at the bottom relative to the reactor vessel while SGS-UFSAR permitting axial growth between the two. One would expect to see, on the bearing surfaces of the key and keyway, burnishing, buffing, or shadowing marks which would indicate pressure loading and relative motion 3.9-9 Revision 6 February 15, 1987 between the two parts. Some scoring of engaging surfaces is also possible and acceptable. 14. Bearing Surfaces of Upper Core Plate Radial Support Key 15. Mounting Blocks Thermal Shielding to Core Barrel. Examine the connections for evidence of change in tightness or lockweld integrity. 16. Gaps at Baffle Joints. Check for gaps between baffle and top former and at baffle to baffle joints. Upper Internals A particularly close inspection was made on the following items or areas, using a magnifying glass of SX or lOX magnification where necessary. The locations of these areas are shown on Figure 3.9-2 (Sheet 1). 1. Thermocouple Conduits, Clamps, and Couplings 2. Guide Tube, Support Column, and Thermocouple Column Assembly Locking Devices 3. Support Column and Conduit Assembly Clamp Welds 4. Radial Support Keys and Inserts Between the Upper Core Plate and Upper Core Barrel. marks, burnishing, buffing integrity of lockwelds. Examine for any shadow or scoring. Check the 5. Connections of the Support Columns and Guide Tubes to the Upper Core Plate. Check for tightness. 6. Thermocouple Conduit Gusset and Clamp Welds 3.9-10 SGS-UFSAR Revision 6 February IS, 1987 -
- 7. Thermocouple End-Plugs. Check for tightness. 8. Guide tube closure welds, tube-transition plate welds, and cadwelds Acceptance standards were the same as required in the shop by the original design drawings and specifications. During the hot functional test, the internals will be subjected to a total operating time at greater than normal full flow conditions (four pumps operating) of at least 10 days or 240 hours0.00278 days <br />0.0667 hours <br />3.968254e-4 weeks <br />9.132e-5 months <br />. This provides a cyclic loading of greater than one million cycles on the main structural elements of the internals. In addition, there will be some operating time with only one, two, and three pumps operating. Therefore, when no signs of abnormal wear are found or of harmful vibration being present in the core support structures, and with no apparent structural changes taking place, the four loop core support structures are considered adequate. They perform their function as intended and are free from harmful vibrations. This program of pre-and post-hot functional examination is identical to that which will be used for all subsequent four loop core support structures. 3.9.1.4 Correlation of Test and Analytical Results Results of measurements taken in models and prototypes are used to refine the methods of analysis and to establish inputs to the LOCA computations. Measurements taken when performing shaker tests on the prototype at the shop and results obtained during hot functional tests ensure that the natural frequencies, normal and damping values used in the analysis can be justified in a realistic manner. Specifically, this was done to confirm the analysis results for the guide tube beam modes, and barrel beam and shell modes. By comparing the results obtained in water and 3.9-11 SGS-UFSAR Revision 6 February 15, 1987 air, the effect of immersion is taken into consideration. Results of the prototype tests also allow the effect of the inertial mass of water to be incorporated into the analysis. The distribution of the mass of the water on the internals components is found from the results of the behavior of the components during vibration measurements. I 3.9.1.5 Dynamic Analysis Methods for Reactor Internals A 0.10 inch clearance is intentionally provided between the upper core plate and guide tube as specified in paragraph 3.2.1.4 of WCAP-7332-L. The highest vertical loads acting on the internals during a hot leg break are due to impacting as the internals are deflected by the transient hydraulic forces. Table 1-1 of WCAP-7332-L shows the upper core plate maximum deflection under these loads to be 0.014 inch; well below the 0.10 inch allowable limit. Many subsequent analyses on other internals show no deflection greater than o. 025 inch. Since a clearance is always present, no significant direct axial loads are induced in the guide tube. An elastic system dynamic analysis and an elastic component analysis were used for the reactor internals analysis under blowdown and seismic excitation. 3.9-12 SGS-UFSAR Revision 20 May 6, 2003 (THIS PAGE IS IN'I'ENTIONALL BLANK) 3.9-13 SGS-UFSAR Revision 20 May 6, 2003 The reactor internals are analyzed for postulated double-ended pipe rupture of the main coolant loop (DBA), for both cold and hot leg breaks. For vertical excitation, the reactor internals are represented by a multi -mass system connected with springs and dashpots simulating the elastic response and the viscous damping of the components. The effects of clearances between various internals, snubbing action caused by solid impact, and preloads in hold down springs have been incorporated in the analytical model. Various reactor internal components are also subject to transverse excitation during blowdown. Specifically, the barrel, guide tubes, and upper support columns are analyzed to determine their response to this excitation. A blowdown digital computer program which is developed for the purpose of calculating local fluid pressure, flow and density transients that occur during a LOCA is applied to the subcooled, transition and saturated two-phase blowdown regimes. This code is based on the method of characteristics wherein the result of ordinary differential equations, obtained from the laws of conservation of mass, momentum, and energy are solved numerically using a fixed mesh in both space and time. Predictions of this code have been compared with numerous test data and the results show good agreement in both the subcooled and the saturated blowdown regimes. The appropriate dynamic differential equations for the models of the internals are formulated and the responses are determined by performing a time-history analysis using the forcing functions from the aforementioned pressure and velocity transients. The results obtained from the linear analysis of the vessel internals indicate that during blowdown, the relative displacement between the components will close the gaps and consequently the structures will impinge on each other, making the linear analysis unrealistic and forcing the application of nonlinear methods to study the problem. It is clear that linear analysis will not provide information about the impact forces generated when 3.9-14 SGS-UFSAR Revision 6 February 15, 1987 components impinge each other, but can, and is, applied prior to gap closure. The effects of the gaps that would exist between vessel and barrel, between fuel assemblies and baffle plates, and between the control rods and their guide paths are considered in the analysis. Further details of the method of analysis are given in "Reactor Internals Response Under a Blowdown Accident," First International Conference on Structural Mechanics in Reactor Technology, Berlin, September 20-24, 1971 by G. J. Bohm and J. P. LaFaille. 3.9.1.6 Analytical Methods for ASME Code Class 1 Components The ASME Section III Nuclear Power Plant Components Code is inapplicable to the Salem Station; hence, the normal, upset1 emergency, and faulted conditions terminology does not apply. However, since the RCL loop vessels (reactor vessel, pressurizer, and steam generators) are basically standard components, analysis of these vessels with the more recent ASME Code conditions (normal, upset, emergency1 and faulted) have been performed with the .load combinations and associated stress limits for ASME Code Class 1 components and supports given in Section 5.2. Load combinations and stress limits for ASME Code Class 2 and 3 and balance-of-plant components and piping are combined as discussed in Section 3.9.2. As-built safety-related piping system stress analysis calculations and pipe support re-evaluation is discussed in Section 3.9.3. For the RCS, the square-root-sum-of-the-squares (SRSS) method of load combination has been utilized to combine the effects of LOCA and design basis earthquake. For the integrated head assembly, SRSS was utilized for combining the LOCA and design basis earthquake loads. Justification of the SRSS method is provided in Westinghouse Topical Reports1 WCAP-9283 and WCAP-9279. A plastic instability analysis of the support and supported system was not needed since the adequacy was proved by elastic analysis. 3.9.1.7 Component Supports Information on component supports is included in Section 3.9.3 and in applicable sections covering the component and its support as parts of an integral system in terms of design and analysis. Design information concerning bolted connections for linear component supports is presented in Appendix 3.9A. 3.9-15 SGS-UFSAR Revision 22 May 5, 2006 *I The report, "Evaluation of the Reactor Coolant System for Salem Unit No. 1 and 2," as amended by PSE&G letter dated March 6, 1979, (R. L. Mittl too. D. Pan) provides the buckling design criteria which was used for all ASME Class 1 component supports subjected to faulted condition loading combinations and justification if criteria exceeds the limits of Paragraph F-1370(c) of the ASME Section III Code, Appendix F. The PSE&G report, "Evaluation of the Reactor Coolant System Considering Subcompartment Pressurization Following a LOCA for Salem Units 1 and 2," was submitted on March 6, 1979, and revised on March 29, 1979. Section 6.1 of this report states that the computer program WESAN was used to analyze the steam generator and reactor coolant pump supports for the effects of asymmetric pressure loads combined directly with LOCA loop depressurization loads. The design control measures as required by Appendix B of 10CFR50, that were used to demonstrate the applicability and validity of the WESAN program are given in the I?SE&G submittal entitled "Explanation and Verification of the Computer Program WESAN," dated January 8, 1980. 3.9.1.8 Dynamic Analysis of the Reactor Coolant Loop In order to eliminate the maintenance and testing required for steam generator snubbers, all of the steam generator snubbers at Salem Unit 1 and Unit 2 have been deactivated. The two snubbers on the reactor side of each steam generator have been removed. Each of the two backside snubbers on each steam generator has been converted to a rigid, single-acting compression strut, via the addition of a compression collar clamped to the snubber body. A reanalysis of the reactor coolant loop (RCL) and primary equipment supports was performed to evaluate the revised configuration of the steam generator upper supports. This reanalysis includes the parameters and characteristics of the Model F (Onit 1) and Model 61/19T (Unit 2) steam generators to demonstrate that the structural criteria will be met with the respective steam generators in place. A non-linear, time history analysis was performed to evaluate earthquake loadings. Time history analysis has also been performed for the analysis of postulated pipe breaks. Revised loads and stresses were evaluated for the loop piping1 primary equipment supports, primary equipment nozzles, main steam and feedwater line piping/ auxiliary piping attached to the RCL, and building structure embedments. The application of leak-before-break was also verified to remain applicable to the primary loop piping. 3.9-16 SGS-UFSAR Revision 24 May 11, 2009 The model used in the original seismic analysis was modified to include all four primary loops, the primary (reactor vessel, steam and reactor coolant pumps) and the primary equipment supports. The steam generators and reactor coolant pumps are represented in the model as discreet mass models. A simplified model of the concrete internal structures of the containment building was generated and is coupled to the RCL/supports model. The reactor vessel is also represented by a discreet mass model with the masses lumped at various locations along the length of the vessel and along the of the of the core internaJ.s. The primary supports are included as stiffness matrices. The component upper and lower J.ateral supports are inactive plant cool down, and normal conditions. However, these restraints become active under the rapid motions of the reactor coolant components that occur from the dynamic loadings of postulated earthquake and pipe break events, and are represented by stiffness matrices and/or individual tension or compression spring members in the dynamic model. The analyses are performed at the full power condition. The time history analysis is performed using the WECAN computer code (Reference 4), and employs non-linear modal super-position techniques. From the mathematical description of the system, the overall stiffness matrix is developed from the element stiffness matrices, from which a reduced stiffness matrix associated with mass of freedom is From the mass matrix and the reduced stiffness matrix, the natural and the normal modes are determined. Time history input motions are at the basemat elevation for input to the model. used in the model is consistent with the values in Section 3.7. The modal superposition method is then used to generate a time history solution for the response of the reactor coolant loop. Three time history motions are applied individually at the basemat elevation representing the north-south, east-west, and vertical earthquakes, for both the OBE and SSE. The results are then combined by adding the vertical earthquake response absolutely with the worst of the two horizontal earthquake responses. 3.9-16a SGS-UFSAR Revision 20 May 6, 2003 The mathematical model used in the static is modified to the tine history pipe break analyses. The natural and are determined fran this loop model. The time history hydraulic forces at the node points are combined to obtain the forces and moments acting at the corresponding structural lumped-mass node points. The dynamic structural solutions are obtained by using a modified predictor-corrector-integration technique and normal mode theory. The postulated pipe breaks included in the take of the of leak-before-break in the primary (Section 3.6.4.2), as well as the elimination of intermediate breaks (Section 3.6.2.1). The breaks considered in the of the RCL are breaks at the loop nozzles of the accumulator lines, RHR lines, and the pressurizer surge line. Also, breaks at the steam generator nozzles of the main steam line and feedwater line are evaluated. In order to determine the thrust and reactive force loads to be applied to the reactor coolant loop during the postulated LOCA, it is necessary to have a detailed description of the t:ydraulic transient. Hydraulic forcing functions are calculated for the ruptured and intact reactor coolant loops as a result o: a postulated LOCA. These forces result from the transient flow and pressure histories in the reactor coolant system. The calculation is in two steps. The first step is to calculate the transient pressure, mass flow rates, and as a function of time. The second uses the results obtained from the hydraulic with of areas and direction coordinates, and calculates the time history of forces at appropriate locations (e.g., elbows) in the reactor coolant loop. The model the behavior of the coolant fluid within the entire RCS. Key parameters calculated by the hydraulic model are pressure, mass flow rate, and density. These are input to the thrust calculation, together with plant layout information, to determine the time-dependent loads exerted by the fluid on the loops. In evaluating the hydraulic forcing functions during a postulated LOCA, the pressure and momentum flux terms are dominant. The inertia and terms are evaluation of the local fluid in the 3.9-16b SGS-UFSAR taken into account model. Revision 25 October 26, 2010 in The blowdown hydraulic analysis, performed with the MULTIPLEX computer code, provides the basic information concerning the dynamic behavior of the reactor core environment for the loop forces, including predictions of the flow, quality, and pressure of the fluid throughout the reactor system. The hydraulic analysis considers a coupled fluid-structure interaction by accounting for the deflection of the core support barrel. The depressurization of the system is calculated, using the method of characteristics applicable to transient flow of a homogeneous fluid in thermal equilibrium. The system geometry is represented by a network of one-dimensional flow passages. In the second step, the transient (blowdown) hydraulic loads resulting from a LOCA are calculated using the THRUST code. In the model used to compute forcing functions, the reactor coolant loop system is represented by a model similar to that employed in the blowdown analysis. Each node is fully described by: 1) blowdown hydraulic information, and 2) the orientation of the streamlines of the force nodes in the system, which includes flow areas and projection coefficients along the three axes of the global coordinate system. Each node is modeled as a separate control volume, with one or two flow apertures associated with it. Two apertures are used to simulate a change in flow direction and area. The force components are then summed over the total number of apertures in any one node to give the total forces. These thrust forces provide the input to the RCL piping/support dynamic analysis. When elements of the system can be represented as single acting members (tension or compression members) , they are considered as nonlinear elements, which are represented mathematically by the combination of a gap, a spring, and a viscous damper. The force in this nonlinear element is treated as an externally applied force in the overall normal mode solution. The time history solution is performed using the WESTDYN computer program (Reference 5)
- The time history displacement solution of all dynamic degrees of freedom is obtained based on 4-percent critical damping. The time history displacement response of the loop is used in computing support loads and in performing stress evaluation of the reactor coolant loop piping. The support loads are computed by multiplying the support stiffness matrix and the displacement vector at the support point. The support loads are used in the evaluation of the supports. The time history displacements are used to 3.9-160 SGS-UFSAR Revision 20 May 6, 2003 determine the internal forces, deflections, and stresses at each end of the piping elements. For this calculation, the displacements are treated as imposed deflections on the reactor coolant loop masses. The results of this solution are used in the piping stress evaluation. in the evaluation of the RCL piping is the same as that presented in Table 5.2-13. The stress limits for normal and upset condition loadings are taken from the 1967 Edition of the USAS B31.1 Power Piping Code. The loading combinations applied are presented in Table 5.2-12. In the original design, the faulted loading conditions used the Design Limit Curves discussed in Table 5.2-13. For the steam generator snubber elimination analysis, the faulted condition stress criteria has been modified to be more consistent w:Lth industry standards. A 1970 interpretation for the 831.7 piping code {ANSI 831.7, Case 70) established that the faulted allowable stresses should be limited to 2 times l. 2S, or 2. 48, where S is the allowable stress defined in 831.1. 'J'his allowable value is in line with industry practice and has been used in the reactor coolant loop reanalysis. The structural integrity of combinations defined in 'I'able the RCL supports is verified using the load 5. 5-3. The stresses are compared to allowable values, also defined in Table 5.5-3, which are based on the American Institute of Steel Construction (AISC) Specification. The supports evaluated in the reanalysis include the steam generator upper and lower supports, reactor coolant pump supports, and reactor vessel supports. Also included is an analysis of the converted backside snubbers. The compression collars are analyzed to demonstrate that stresses are within allowable limits for all applied loads. The snubber bodies are also reanalyzed to address the modified load path resulting from the addition of the compression collar. Results of Analyses The analysis results for the reactor coolant loop demonstrates that the piping stresses remain within the allowable limits. In all support member stresses have been shown to remain within the appropriate allowable values for each loading combination. Also, the primary equipment nozzle loads have been shown to be less than the allowable nozzle loads as specified in the appropriate design specifications. 3.9-16d SGS-UFSAR Revision 23 October 17, 2007 * *
- Evaluations of the nozzle displacements resulting from the RCL analysis for the main steam and feedwater lines, and the auxiliary piping attached to the reactor coolant loop, and the assessment of the embedment loads at the primary equipment support attachment locations show that the stresses in these lines and the embedment loads remain within their respective allowable limits. 3.9.2 ASME Code Class 2 and 3 Components Analyses of Westinghouse supplied equipment were performed in accordance with the codes and standards requirements in effect at equipment order date. The codes used for the components of the RCS and Emergency Core Cooling System (ECCS) are given in Sections 5.2 and 6.3, respectively. unlike the 1971 Edition of the ASME Boiler and Pressure Vessel Code,Section III, "Nuclear Power Plant Components Code," these codes did not have design condition categories of "normal, 11 "upset," "emergency," and "faulted11 conditions and associated stress limits. However, seismic analyses performed for selected Seismic category I components using response spectra that envelop the vertical and horizontal floor response spectra show that the stress for the Operating Basis and Design Basis Earthquake are generally a small fraction of the upset limits of ASME section III, 1971 Edition, or the faulted condition limits of the soon to be published Appendix F of the same code. Similarly, analysis of the Reactor Coolant Loop/Support System shows that the stresses for the combined loading of Design Basis Earthquake (DBE) and DBA are within the Faulted Condition limits given in Section 5.2. All nuclear piping comparable to ASME Code Classes 2 and 3 and the then current Atomic Energy commission Quality Group Classifications was designed to stress criteria conforming to the design philosophy of ANSI B31.1.0, Code for Pressure Piping, Power Piping. As this piping was designed and purchased at a later point in time *than the aforementioned westinghouse-supplied equipment, it was, however, possible to extend the basic design philosophy and stress criteria of ANSI 831.1.0 in a manner paralleling the newer codes. The results of classifying the design stress limits (using ANSI B31.1. o terminology) into 11Design Condition Categories" (using present day ASME Section III terminology) are given as follows: 3.9-17 SGS-UFSAR Revision 20 May 6, 2003 I Design Loading Component Combinations class 2 and Normal pressure, and 3 piping weight and external Where: loadings Maximum (short time) pressure and weight and external loadings and OBE Maximum (short time) ..t pressure and weight and external loadings and DBE OBE = Operating basis earthquake DBE = Design basis earthquake Design Condition Category normal upset faulted Design Stress Limit Pc.:::_Sh Pl.:::_Sh P .:::_l.2S e h p .:::_1.2S 1 h p ..:::_1.28 c II p ..:::_1.5(1.2)8 = 1 h Sh "" Allowable "hot" stress from ANSI Bll.l.0-1967 P Circumferential (hoop) stress due to loadings listed c P1 = Longitudinal stresses due to internal pressure and various external loadings listed 3.9.2.1 Analytical and Emperical Methods for Design of Pumps and Valves The terminology "active components"* and "faulted conditions" is not applicable to the Salem station. This terminology was introduced in codes (ASME Section III, Summer 1968 Addenda) and standards after the code applicability date for the Salem Station. Valves were designed in accordance with USAS Bl6. 5, in general, and ASME Section VIII for flange connections. Valves were
- Active components of fluid system (e.g., valves, pumps) are those whose operability is relied upon to perform a safety function such as safe shutdown of the reactor or mitigation of the consequences of a postulated pipe break in the reactor coolant pressure boundary. 3.9-18 SGS-UFSAR Revision 20 May 6, 2003 hydrostatically tested to USAS B16.5 hydrostatic test followed by a seat rating table with the leak test to MSS-SP-61 criteria, except that the seat leakage criterion was as specified in the Preliminary Safety Analysis Report. Leak testing is performed to assure that no gross deformation was caused by the hydrostatic test. The components of the reactor coolant boundary which are required to function following a postulated pipe rupture (DBA) are the valves which isolate the normally high pressure systems from the low pressure safety systems. These components are generally designed for deflection limitation and for stress under normal operating conditions which is generally low. The differential pressure loadings on these valves decreases following a postulated pipe rupture of the RCS and, hence, the loadings on the valves due to system depressurization will decrease. The stresses will likewise decrease. The allowable stress that was used for the design of valves was 7000 psi as required by USAS-B16 .5. ASHE Section VIII was employed in the design of gaskets for flange connections only as required by USAS-Bl6.5. 3.9.2.2 Design and Installation Criteria, Pressure-Relieving Devices The design criteria applicable to mounting of nuclear class pressure-relieving devices provides for the system to withstand valve reaction forces due to the concurrent discharge of all valves on any given header in combination with pressure, weight, and seismic forces. The primary longitudinal stresses developed as a result of these combined loads are limited to the values listed in Section 3.9.2. Design procedures valve systems to include an analysis of determine dynamic fluid the safety and relief reaction thrusts. The combination of reaction thrust with pressure, weight, and seismic forces are then used to compute the stresses imposed on valves and piping in accordance with the design philosophy of ANSI B31. 1. 0. 3.9-19 SGS-UFSAR Revision 6 February 15, 1987 The computed stresses are compared with the design criteria limits to determine acceptability of the design. -Installation include the provisions to meet the aforementioned stress criteria addition of piping reinforcement, piping restraints, or combinations of both. The design criteria for safety and relief valves in the primary coolant pressure boundary provide for the system to withstand maximum loads due to combinations of valve reaction, weight, and seismic force. Maximum load, as determined in the design analysis of the piping and valves, is absorbed by a system of braces and shell girth bands rigidly connected to the pressurizer shell. The relief valve loads are actually transmitted to this system via a rigid connection on the relief valve body. 3.9.2.3 Field Run Piping Systems Field running of small diameter piping, i.e., complete assembly at the erection point without reference to design drawings, was not permitted for essential systems. Therefore, no special quality assurance measures or performance tests were required. 3.9.3 Seismic Analysis of As-Built Safety-Related Piping Public Service Electric and Gas has conducted the re-evaluation of safety-related piping required by IE Bulletins 79-14 and 79-07. Modifications shown to be necessary by this evaluation have been made and the capability for safe shutdown in a seismic event has been demonstrated. The re-evaluation of pipe stresses and supports was based on current standards and was performed subsequent to verification of "as-built" piping using the piping isometric drawings. 3.9-20 SGS-UFSAR Revision 6 February 15, 1987 --
The 15 safety-related systems that encompass the total scope of the evaluation are listed in Section 3.7.3.9.1. These systems are those required for safe shutdown of the plant. The re-evaluation of pipe stresses was performed in accordance with the criteria previously submitted to the NRC as part of the responses for Salem 1. This included additional information supplied as a response to IE Bulletin 79-07 (1), a guideline for implementing IE Bulletin 79-14 (2) and a description of the program for the reanalysis and modification of supports (3). The results indicated there are not overstressed pipes. The was reanalysis based on method used to determine pipe support capabilities seismic loads resulting from the three-dimensional SRSS method. This was performed in accordance with procedures submitted to the NRC as part of the r-esponses for Salem Unit 1. The reanalysis resulted in certain modifications of supports. The reanalysis of pipe stresses 1n the 15 safety-related systems indicated no overstressed conditions. However, the pipe supports within those stress calculations resulted in several modifications which have been completed. The majority of the modifications were to the U-bolt and strap anchor supports. These were basically designed to function as a six-way restraint and the pipe stress calculations were performed considering these conditions. However, when individual hanger details were being designed, fabricated, and installed, they did not fulfill the requirements of a rigid anchor as was assumed for pipe str-ess calculations. Hence, theoretically, they did not function as was originally intended. These U-bolt type anchors and strap anchors were redesigned to comply with the original intended functions. 3.9-21 SGS-UFSAR Revision 6 February 15, 1987 3.9.4 Inservice Testing of Pumps and Valves Inservice testing of pumps and valves is performed in accordance with the program originally transmitted to the NRC on April 29, 1981, and as subsequently amended. 3.9.5 References for Section 3.9 1. Letter from Schneider to Grier, "NRC IE Bulletin No. 79-07, supplemental Response, No. 1 Unit, Salem Generating Station,11 Sept. 21, 1979. 2. Letter from schneider to Grier, 11NRC IE Bulletin No. 79-14, Supplement 2, No. 1 Unit, Salem Generating Station, 11 Oct. 30, 1979. 3. Letter from Librizzi to Schwencer, "NRC IE Bulletin No. 79-07, Supplemental Response, No. 1 Unit, Salem Generating Station, Docket No. 50-272," Oct. 18, 1979. 4', WCAP-8929, Benchmark Problem Solutions Employed for Verification of the WECAN Computer Program, April 1977. s. WCAP-8252, Revision l, Documentation of Selected Westinghouse Structural Analysis Computer Codes, May 1977. 3.9-22 SGS-UFSAR Revision 20 May 6, 2003