ML20132D474
ML20132D474 | |
Person / Time | |
---|---|
Site: | FitzPatrick |
Issue date: | 11/30/1984 |
From: | TELEDYNE ENGINEERING SERVICES |
To: | |
Shared Package | |
ML20132D467 | List: |
References | |
JPN-85-60, TR-5321-2, TR-5321-2-R01, TR-5321-2-R1, NUDOCS 8508010059 | |
Download: ML20132D474 (97) | |
Text
._
ATTACHMENT 2 l
JPN-85-60 l dated J
l A
8500010059 850729 PDR ADOCK 05000333 p PDR NEW YORK POWER AUTHORITY James A. FitzPatrick Nuclear Power Plant i
! Docket No. 50-333 l
4 I
I 1
l I
i i
l l
TEl.EDYNE ENGINEERING SERVICES CONTROLLED "vtTELEDYNE occuteENT ENGINEERING SERVICES Tgs pn01. NO.aO DATE TECHNICAL REPORT TR-5321-2 REVISION 1 l .
l MARK 1 CONTAINMENT PROGRAM I
i PLANT-UNIQUE ANALYSIS REPORT OF THE TORUS ATTACHED PIPING FOR JAMES A. FITZPATRICK NUCLEAR POWER STATION
.- _ _ NOVEMBER 1984 -
)
~ ~ " ~
%h
- _ Jg.y.7--k ,
ENGINbbHING SERVICES 13] SE COND a.E %E
_, , wai' av vassacm $r- i C22%
~+ . . . m .m g , ,
hWh RECO. IDS MANAGEMENT July 9, 1985 5321-186 Mr. Leon Guaquil Director Project Engineering - BWR New York Power Authority 123 Main Street Whito Plains, NY 10601
Subject:
JAFNPP Torus Program - Final Piping Report TR-5321-2, Revision 1,
" Mark 1 Containment Program Plant Unique Analysis Report for the Torus Attached Piping for JAFMPP " dated November 1984
Dear Leon:
There are two minor errors in the subject report that have been brought to our attention by Mr. P. Okas of your staff. They are:
- 1. On page 59, the node listed as 370 should be 70.
- 2. On page 62, there is no indication for penetration x 212, node pt.
26 to show if it was modified or not. It was not modified and an - 1 "X" should appear in that column to indicate so.
These are both typographical errors that have no affect on the technical results or conclusions. Because of that, we will not be changing the report unless a revision is done for some other reason.
Should you have any questions, please contact us.
Very truly yours, TELEDYNE ENGINEERING SERVICES l'
- oz d3C Nicho s S. Cel a '
Manager, Engineering cc: P. Okas (NYPA)
a.s*
NEW YORK POWER AUTHORITY 123 MAIN STREET WHITE PLAINS, NEW YORK 10601 TECHNICAL REPORT TR-5321-2 REVISION 1 MARK 1 CONTAINMENT PROGRAM PLANT-UNIQUE ANALYSIS REPORT
, 0F THE TORUS ATTACHED PIPING FOR JAMES A. FITZPATRICK NUCLEAR POWER PLANT l
NOVEMBER 1984 WTELEDYNE ENGINEERING SERVICES 130 SECOND AVENUE WALTHAM, MASSACHUSETTS 02254 617-890-3350 j f
TN Technical Report TR-5321-2 -
Revision 1 -ii-RECORD OF REVISIONS REVISION PAGE DESCRIPTION 1 Cover and Title Add Revision 1 and change date from May 1984 toNovpmber1984.
10 In equation (1) change "1.85n" to 1.8 Sh".
" ... load combinations (15), (1)
Also change and (2)..." to "
... load combinations (14),
(15), (2), and (1)...".
19 Support PFSK-831, Y-Rigid, change "X" yes modified to "X" No modified and remove "Sup-port Reinforced".
41 Section 3.3.5, change "..., equal to the total TAP motion..." to "..., equal to twice the TAP motion...".
49 Section 3.5, fifth bullet, change " Relocate 10" condensate line" to "Resupport 10" con-densate line".
52-54 Table 3-3, Branch Line Pipe Stresses, change all maximum stress values to reflect updated analysis.
58 Support PFSK-2107, change Node number from 745 to 245.
Support PFSK-2477, change Node number from 169 to 160.
60 Node 85, change Spring to PFSK-1854 and add Spring under type of support.
Node 715, change Support number from PFSK-1986 to PFSK-1982 Support PFSK-2567, change Node number from 386 to 385.
Support PFSK-2446, change Node number from 750 to 250.
62 Support PFSK-1049, change Snubber to Spring. l l
Support PFSK-2223, change Node number from 186 to 190.
Y Technical Report TR-5321-2 Revision 1 -iii-RECORD OF REVISIONS (cont.)
REVISION PAGE DESCRIPTION 64 X-226, Node 100, change H23-62 to PFSK-983.
4 Support PFSK-1994, change Node number from 318 to 300.
X-227, Node 125, remove H14-27.
Note: Revision 1 changes are a result of incorporating NRC review comments which are documented in TR-5321-1, Revision 1, and to incorpoate up-dated analysis results as well as to correct typographical errors.
W TELEDYNE
-5321-2 -iv- S8WICES Revision 1 ABSTRACT The work summarized in this report was undertaken as part of the Mark 1 Containment Long Term Program. It includes the evaluation of all piping systems that are attached to the suppression pool (torus).
These piping systems include both Main Steam Safety Relief lines and piping attached to the torus shell.
Mark ,1 induced loads, as well as original design loads, are included in the evaluation. Necessary modifications have been completed and are summarized here.
l
Technical Report TE N TR-5321-2 -v- BlGNEstlNG SERVICES Revision 1 TABLE OF CONTENTS Page ABSTRACT ii 1.0 GENERAL 1 2.0 SRV PIPING ANALYSIS 2 2.1 Applicable Codes and Criteria 2 2.2 SRV Loads 3 2.2.1 SRV Gas Clearing Loads 3 2.2.2 SRV Water Clearing Loads 4 2.2.3 Pool Drag Loads 5 2.2.4 Thermal Expansion 6 2.2.5 Weight, Pressure and Seismic 6 2.3 SRV Analysis Method 7 2.3.1 Piping Analysis 7 2.3.1.1 Computer Model 7 2.3.1.2 Piping Analysis Method 7 2.3.2 Pipe Supports Analysis 8 2.3.3 SRV Main Vent Penetration Analysis 9 2.4 Evaluation and Results (SRV) 9 2.4.1 General 9 2.4.2 SRV Pipe Stresses 10 2.4.3 SRV Pipe Supports 11 2.4.4 Support Steel for SRV Supports 11 2.4.5 SRV Penetration 12 2.4.6 Valves 13 2.4.7 Fatigue Evaluation 14 2.5 Summary of SRV Line Modifications 14
Technical Report TR-5321-2 -vi- YE NE Revision 1 ENGNEERNG SERVICES TABLE OF CONTENTS (CONTINUED)
Page 3.0 TORUS ATTACHED PIPING (TAP) 32 3.1 Applicable Codes and Criteria 32 3.2 TAP Loads 33 3.2.1 Shell Motion Due to Pool Swell 34 3.2.2 Shell Motion Due to DBA Condensation Oscillation 35 3.2.3 Shell Motion Due to Chugging 35 3.2.4 Shell Motion Due to SRV Line Discharge 36 3.2.5 Loads on Internal Piping 37 3.2.6 Deadweight, Jhermal and Seismic Analysis 38 3.3 TAP Analysis Method 38 3,3.1 Representation of Torus Shell for Piping Analysis 38 3.3.2 Piping Analysis Method - Large Bore Systems 39 3.3.3 Piping Analysis Method - Complex Small Bore Systems 40 3.3.4 Piping Analysis Method - Simple Small Bore Systems 40 3.3.5 Piping Analysis Method - Branch Piping 41 3.3.6 Piping Analysis - Load Input for Computer Models 41 3.3.6.1 Mark 1 Loads Due to Shell Motion '41 3.3.6.2 Submerged Drag Loads on Internal TAP 42 3.3.7 TAP Penetration Analysis 44 3.3.8 Analysis Method for Piping Supports 44 3.3.9 Ve.cuum Breaker Analysis 45 3.3.10 Active Components 45
Technical Report TR-5321-2 Revision 1
-Vii- WM ENGNEERNG SERVICES TABLE OF CONTENTS (CONTINUED)
Page 3.4 Evaluation and Results 45 3.4.1 General 45 3.4.2 Piping Stress - Large Bore Systems 46 3.4.3 Piping Stress - Small Bore TAP Systems 46 3.4.4 Pumps and Valves 47 3.4.5 Piping Fatigue Evaluation 47 3.4.6 Torus Shell Penetration Evaluation 47 3.4.7 Piping Supports 48 3.5 Summary of TAP Modifications 49 REFERENCES 76 APPENDIX 1 - Use of C0 Load for Small Bore Piping Al-1 Appendix 2 - 32 Hz Cutoff for Condensation Oscillation Analysis A2-1 .
Technical Report Rev sion 1 Y
ENGNEERNG SERVCES FIGURES AND TABLES Page FIGURES:
2-1 SRV Line Routing - Typical 25 2-2 SRV Line Arrangement - Torus 26 2-3 SRV Line Routing - Typical 27 2-4 SRV Tee-Quencher and Supports 28 2-5 SRV Piping Model - Typical 29 2-6 SRV Seismic Spectra - Typical 30 2-7 Vent System Model 31 3-1 Shell Response from Pool Swell- Typical 66 3-2 Shell Response from Condensation Oscillation - Typical 67 3-3 Shell Response from SRV - Typical 68 3-4 Drag Load on Internal Piping 69 3-5 TAP Seismic Spectra - Typical 70 3-6 TAP Penetration Representation (Typical) 71 1-7 Detailed Shell Model 72 3-8a TAP Penetration Locations 73 3-8b TAP Penetration Locations 74 3-8c TAP Penetration Locations 75 TABLES:
1 Structural Acceptance Criteria for Class 2 & Class 3 Piping Systems 78 2-1 SRV Load Cases / Initial Conditions 15 2-2 SRV Pipe Stress 16 2-3 SRV Pipe Support Modifications 17 2-4 SRV Valve Evaluation 24 3-1 Large Bore TAP Results 50 3-2 Small Bore TAP Results 51 3-3 Branch Line Pipe Stresses 52 3-4 Pump and Valve Evaluation 55 3-5 TAP Pipe S~upports 58 3-6 TAP Penetration Stress Results 65
TN Technical Report TR-5321-2 N N ES Revision 1 s-1.0 GENERAL The purpose of the Mark 1 Containment Program is to evaluate the effects of hydrodynamic loads resulting from a loss of coolant accident and/or an SRV discharge on the torus structure.
Teledyne report TR-5321-1 (Reference 1) reported the effects of Mark 1 loads or, the FitzPatrick torus structure, support system and internals. Tiiis second report completes the work on the program by considering the effects of the Mark 1 loads on the p'i' ping-systems attached to the torus. Both the main steam relief lines and the piping connected to the torus shell are considered.
Also included is the evaluation of piping penetrations, supports and active components.
A summary of modifications made as a result of this analysis is included.
The report is separated into two major categories, one that deals with main steam relief lines (SRV piping) and one that deals with piping attached to the torus shell (TAP). Each of these sections is written to stand alone and includes a discussion of methods and results.
TM Technical Report MM TR-5321-2 Revision 1 2.0 SRV PIPING ANALYSIS There are eleven main steam relief (SRV) lines at FitzPatrick. These lines connect to the main steam lines in the drywell, extend down the main vents and penetrate the main vent into the torus (Figures 2-1 and 2-2). These lines penetrate the main vent pipe at the vent header intersection, run horizontally inside the vent header impact deflector, and enter the pool vertically over the ring girders. (Figures 2-3 and 2-4).
Analysis results for the discharge end of the SRV lines were previously reported in Reference 1. This referenced report includes SRV piping in the torus airspace, the submerged part of the SRV line, the tee-quencher and the quencher support beam. This report will cover the remaining portion of the line, which includes:
e The main vent penetration, e The SRV piping between the penetration and the main steam line, e SRV pipe supports between the penetration and main steam lines.
The analysis of SRV piping in this report accounts for the f act that some modifications have previously been made to these lines. These modifications are described in the Reference 1 report and consist of the addition of tee-quenchers and support beams (Figure 2-4) and the addition of one ten-inch vacuum breaker on each SRV line.
2.1 Applicable Codas and Criteria The SRV piping and pipe support analysis was performed in accor-dance with Section III of the ASME Code,1977 Edition, including Suniner 1977 Addenda (Reference 2).
In cases where modifications to SRV line supports were required, they were designed in accordance with Section III of the ASME Code (Reference 2).
Technical Report
'A V W NE
";53n;2 , BGNEERNG SERVCES Load combinations and stress levels were evaluated in accordance with Table 5-5 of the Mark 1 Containment Program Structural Acceptance Cri-teria Plant Unique Analysis Application Guide (Reference 5). Table 5-5 is reproduced in this report as Table 1.
2.2 SRV Loads The Mark 1 Program defined several new SRV line conditions. These conditions resulted from different drywell.and torus conditions and produced several different reflood heights and discharge pressures. The load cases considered are listed in Table 2-1.
The analysis and evaluation in this report considers all these SRV cases as well as seismic, weight, thermal and pressure effects.
The specific loads considered in this analysis include:
e Gas clearing (blowdown) loads, e Water clearing discharge loads, e Submerged structure drag on the SRV line, quencher and support due to pool motion.
e Thermal expansion of SRV line.
e Thermal expansion of containment structure.
e Seismic.
e Weight.
e Internal Pressure.
Calculational methods developed as a part of the Mark 1 generic program were used to the extent that they apply.
2.2.1 SRV Gas Clearing Loads Sudden pressurization of the SRV line, due to rapid opening of the safety relief valve, causes unbalanced dynamic forces on the SRV l
TF WE 3 - NO Revision 1 piping. These forces progress through the system as pressure waves, whose speed and amplitude depend upon the particular line conditions being con-sidered; the various SRV cases are listed in Table 2-1.
TES has evaluated the stresses resulting in various SRV pip-ing systems, due to the cases listed in Table 2-1, and has concluded that SRV Case A1.2 is the bounding case for gas clearing loads. Case A1.2 is a first actuation af ter an SBA/IBA break and is characterized by increased gas density in the line before valve actuation. This increased density is a consequence of increased drywell pressure which affects the internal line pressure and density through the vacuum breakers. This increased density produces higher thrust forces than the lower density cases. This load case was run for each of the eleven SRV lines.
The calculation of loads resulting from Case A1.2, as well as all other SRV cases, was based upon use of the " Computer Code RVFOR-04" (Reference 7), which is the property of General Electric Company.
Case A1.2 was run for each of the eleven SRV lines at Fitz-Patrick. Gas clearing loads associated with this case were used for all SRV cases and, therfore, produced conservative results for normal actuation, as well as other cases. In cases where this conservative condition exceeded the lower allowables associated with normal SRV actuation, Case A1.1 was also calculated.
2.2.2 SRV Water Clearing Loads Water clearing loads are produced in the SRV line as water accelerates under line pressure and is forced around the elbows at the quencher end of the line. These forces are very sensitive to reflood height, which varies for several of the second actuation cases. ,
I Maximum line reflood and water clearing are clearly associ-l ated with SRV Case C3.3. Case C3.3 is the second actuation after an IBA/SBA break with steam in the drywell. The high reflood is a consequence of l
l
Technical Report yg Rv n1 O
additional steam entering the line through the vacuum breaker af ter the first actuation (rather than air).
The high water clearing loads that result from this condi-tion affect the torus end of the SRV line, including the piping in the main vent. It has a negligible effect on piping loads in the drywell.
Water clearing for FitzPatrick was calculated for SRV Case C3.3, using G.E. programs RVRIZ and RVFOR-04 (References 7 & 14). These programs were run for the longest and the shortest SRV lines and maximum worst case water clearing loads were used on all eleven lines. (These lines are identical inside the torus, except some lines are mirror images of others).
The second valve actuation was assumed to occur at the point of maximum reflood.
Water clearing loads associated with SRV Case C3.3 bound all other cases and were used for all SRV analysis conditions.
2.2.3 Pool Drag Loads The torus end of the SRV line, including the tee-quencher and quencher support beam, are submerged in the torus pool. These components are subject to drag loads due to pool motion from the following loads:
Pool Swell - Jet Loads
- Bubble Loads Condensation Oscillation -
- Source Induced Drag
- Fluid Structure Interaction (FSI) Drag Chugging - Source Induced Drag
- FSI Drag SRV Discharge - Drag from Adjacent Quenchers (as applicable)
The drag loads associated with these events were calculated in the earlier part of the program and the methods are reported in Reference
1 Technical Report WWE i ev s on 1
- 1. At that time, the data drag loads were used to determine stresses in the SRV piping in the torus, the quencher and the support beam; these were all reported in Reference 1. The same drag load information was used as a part of this analysis to help determine stress in the penetration, and the SRV line and supports in the main vent pipe.
2.2.4 Thermal Expansion Two different load conditions were considered for thermal expansion stress.
The first assumed that the entire SRV line was at its maximum operating temperature (3500 ). It included maximum thennal motion of the connection at the main steam line and assumed the drywell and torus were at ambient.
The second case was like the first except the main vent pipe was assumed to be at 340 F. This has the effect of moving the penetration in the main vent pipe relative to the torus and quencher.
2.2.5 Weight, Pressure and Seismic Weight, pressure and seismic loads were also considered in the analysis. The seismic analysis was based on different spectra for OBE and l SSE response. Total seismic response was determined by the SRSS combination of each of the three response directions, in accordance with the FSAR.
Seismic end effects were considered for this analysis, but judged to be negligible.
A typical horizontal spectra is illustrated in Figure 2-6.
These seismic spectra were developed according to the procedure outlined in FSAR Section 12.5.4.
Technical Report Qg Rev sion 1 N
2.3 SRV Analysis Method 2.3.1 Piping Analysis 2.3.1.1 Computer Model Analysis of all SRV load cases was performed using computer models of the piping systems and the STARDYNE computer code (Reference 15). A typical computer model is illustrated in Figure 2-5.
Features of the model include: .
e Modeling of the main steam line with each SRV line.
e Representation of the stiffness of the main vent penetration by a set of six attachment springs, developed by computer analysis of the penetration area.
e Full representation of the tee-quencher and quencher support beam in the piping model, e Full representation of the brackets between the quencher and support beam which allow free torsional rotation of the quencher arms, e Two percent damping used for time history analysis.
2.3.1.2 Piping Analysis Method Analysis for SRV discharge cases was done by impos-ing individual time histories for water and gas clearing loads at each bend
Tcchnical Report Qg R n1 and elbow in the system and performing the dynamic analysis. Bounding analy-sis was perfonned for these cases by combining gas clearing loads from SRV Case A1.2 with water clearing loads from SRV Case C3.3 into a single load condition. This conservative combination was used to bound all discharge cases, including normal actuations. Different line-unique loads were applied
, to each of the eleven SRV lines for gas clearing; water clearing is the same for all lines and equals the maximum load for the longest line.
Damping for these time history analyses was taken at 2% of critical and calculational time increments for the solution were taken at .0025 seconds. All response frequencies to 50 Hz were considered in the solution.
Seismic analysis was done using the same model and static analysis. The seismic spectra were applied in the vertical and two horizontal directions and the results were combined by SRSS. Separate spectra were used for OBE and SSE analysis.
Analysis for thermal and weight conditions was done using static analysis. Calculations for internal pressure were done by hand.
2.3.2 Pipe Supports Analysis Analysis for SRV piping supports was done using both hand and computer analysis. The STAAD computer program was used for the analysis of complex supports (Reference 16).
The support analysis included the attachment weld to the supporting steel. In cases where there was a question regarding the ability of the support steel to carry the new loads, the steel was also analyzed.
In addition to the SRV line supports in the drywell, there are eleven supports in the main vent pipes.
TN Technical Report
. ENGNERNG SOMCES TR-5321-2 Revision 1 Analysis of these supports included a detailed evaluation of the stresses in the main vent wall, near the support. These stresses were calculated using a Bijlaard analysis (Reference 8) in combination with intensified free-shell stresses due to vent header loads. Free shell stresses were taken from work done in Reference 1 using the computer model illustrated in Figure 2-7 of this report (Figure 4-4 in Reference 1).
Support analysis was done to the ASE Code Section III, Subsection NF (Reference 2).
2.3.3 SRV Main Vent Penetration Analysis The SRV line penetrations of the vent pipe are illustrated in Figure 2-3. Analysis of these penetrations was done using a Bijlaard analysis (Reference 8), to determine local penetration stresses due to SRV line loads. These local stresses were added to intensified free shell stres-ses which occur in the vent pipe due to vent header loads. These were calculated using the finite element model illustrated in Figure 2-7. Devel-opment of these free shell stresses and a description of the model are given in Reference 1, Section 4.
2.4 Evaluation and Results (SRV) 2.4.1 General Combinations of the previous analysis cases were done to allow evaluation of the results in accordance with Table 1. This table lists a total of 27 different load combinations; of these,13 include an SRV event.
This evaluation is concerned with piping and supports from the main steam line to the vent pipe penetration; evaluation of piping and supports inside the torus is reported in Reference 1. This separation is important to the selection of the controlling load combinations that follow.
Technical Report EN TR-5321-2 N NES Revision 1 The results of a conservative load case (described below) were evaluated against level B allowables, without use of increased allow-ables, as allowed in Table 1. Where this load combination produced unaccept-able results, less conservative combinations were evaluated, as described below.
Thermal loads were considered differently for piping and supports as discussed below.
2.4.2 SRV Pipe Stresses Initial evaluation of SRV pipe stress was done as described in Section 2.4.1 above; that is:
P + DW + /(SSE)2+(Blowdown)2 ,= 1.2 S h
In cases where this conservative condition could not be met, the following four cases were evaluated:
(1) P + DW + (0BE)2 + (Blowdown)2 4 1.8 S h (2) P + DW + (SSE)2 + (Blowdown)2 7 1.8 S h (3) P + DW + OB E = 1.2 S h
(4) P + DW + Blowdown = 1.2 S h These four cases represent load combinations (14), (15),
(1) and (2) in Table 1, and are still conservative. No further reduction l in conservatism was necessary to qualify the SRV piping.
Thermal expansion stresses were evaluated for piping as a separate load condition, using ASE Section III, Subsection NC Code Equation 10.
Results of SRV pipe stress evaluation are listed in Table 2-2.
Technical Report WF WNE TR-5321-2 ENGNEERING SERVICES Revision 1 2.4.3 SRV Pipe Supports SRV pipe supports were evaluated in accordance with the ASME Code,Section III, Subsection NF (Reference 2).
A worst-case load condition was developed to include:
e The conservative A1.2/C3.3 blowdown case.
e SSE seismic.
e Worst case thermal load, e Deadweight.
Seismic and blowdown were combined by SRSS and added to the other loads. Allowable stress for this condition was maintained below yield to assure that pipe stress would not be effected by support motion. This stress criteria is consistent with the Case 15 allowables from Table 1.
Results of pipe support analysis are listed in Table 2-3.
2.4.4 Support Steel for SRV Supports Evaluation of drywell support steel for SRV supports was done in accordance with Subsection NF of the ASME Code, (Reference 2), as required.
Evaluation of local stress in the main vent pipe wall was done using the same method described for the SRV penetration except evaluation for the Nozzle Piping Transition, paragraph NE-3227.5 is not required. This evaluation was performed for all main vent supports.
Controlling stresses for the main vent pipe wall are:
Technical Report TN.
TR-5321-2 ENGNEER1NG SERVICES Revision 1 PRIMARY STRESS (Local Membrane Shell Stress Intensity)
Controlling Calculated Allowable Load Case Stress Stress Support Case 15 22,486 28,900 (Table 1) (1.5 Smc)
SECONDARY STRESS (Primary and Secondary Stress Intensity)
Support Case 15 63,366 69,900 9
2.4.5 SRV Penetration Stresses in the main vent pipe penetration area were evalu-ated in accordance with subsection NE of The ASME code, using the following paragraphs:
NE-3221.2 Local Membrane Stress Intensity NE-3221.3 Primary General or Local Membrane plus Primary Bending Stress Intensity NE-3221.4 Primary plus Secondary Stress Intensity NE-3221.5 Analysis for Cyclic Operation NE-3227.5 Nozzle Piping Transition (for vertical lines only) b
Technical Report YE N TR-5321-2 ENGNEERNG SERVICES Revision 1 Fatigue evaluation of the penett ation (paragraph NE-3221.5) showed that the maximum load could be cycled on the penetrations for at least 7500 cycles without exceeding code allowables. The major load component in this case is SRV Case C3.3, which can only occur for a few cycles (less than 50). Normal SRV actuations produce substantially less load for up to 4500 effective stress cycles (Reference 9). Since the 7500 cycles of maximum load bounds both of these by such a large margin and since no other significant loads are imposed on the line, the penetration was assumed acceptable for fatigue without further evaluation.
Controlling stresses in the SRV penetration follow:
PRIMARY STRESS (Local Membrane Shell Stress Intensity)
Controlling Calculated Allowable Load Case Stress Stress
, Case 15 8,370 28,900
~(Table 1) (1.5 Smc)
SECONDARY STRESS (Primary plus Secondary Stress Intensity)
Case 15 58,966 69,900 (3.0 Smi) 2.4.6 Valves Evaluation of the SRV valves was done on the basis of stres-ses in the adjacent piping for the combined load cases. Pipe stresses meeting level B criteria were considered adequate to insure proper operation of the device (Reference 5, Section 5.5).
Results of the valve evaluation are listed in Table 2-4.
Technical Report TR-53 1-2 WM ENGNEERNG SERVICES 2.4.7 Fatique Evaluation Fatigue evaluation of SRV lines was undertaken as a generic Mark 1 Program effort, using bounding assumptions. This effort is described and reported in Reference 9, and concludes that fatigue will not be a problem for Mark 1 SRV lines; this includes the SRV lines at FitzPatrick. No further plant-unique analysis is necessary.
Fatigue evaluation of the SRV penetration is discussed in Paragraph 2.4.5.
2.5 Summary of SRV Line Modifications Modifications to the SRV lines at FitzPatrick included the fol-lowing changes:
e Installation of tee-quencher discharge devices and quencher supports on all eleven lines (Figure 2-4).
e Installation of one ten-inch vacuum breaker on each SRV line.
e Modification to supports and supporting steel in the drywell
~
as listed in Table 2-3.
1 L:
Technical Report 5
SPTF1 WE Revision 1 ENGNEERING SERVICES TABLE 2-1 SRV LOAD CASE / INITIAL CONDITIONS Any One ADS
- Multiple .
Design Initial Condition Valve Valves Valves 1 NOC*., First Act. A1.1 A3.1 A 2 SBA/IBA,* First Act. A1.2 A2.2 A3.2 3 DBA,* First Act.1 A1.3 1 N0C, Subsequent Act. C3.1 SBA/IBA, Sub. Act.
C 2 Air in SRV/DL C3.2 3
SBA/IBA, Sub. Act.
Steam in SRV/DL C3.3 (1) This actuation is assumed to occur coincidently with the pool swell event. Although SRV actuations can occur later in the DBA accident, the resulting air loading on the torus shell is negligible since the air and water initially in the line will be cleared as the drywell to wetwell AP increases during the DBA transient.
NOC = Normal Operating Condition SBA = Small Break Accident IBA = Intermediate Break Accident DBA = Design Basis Accident
Technical Report '#PTri Fry (E Revisio 1 ICES TABLE 2-2 FITZPATRICK SRV PIPE STRESS Line Size &
SRV Max. Stress Sch. O Max. Maximum Allowable Line Location Stress Pt. Stress Stress 10" SSV-302-1A 171 10" Sch. 40 19,695 22,500 10" SSV-302-1B 292 10" Sch. 40 17,542 18,000 10" SSV-302-1C 362 10" Sch. 40 19,460 22,500 10" SSV-302-10 670 10" Sch. 40 17,122 18,000 10" SSV-302-1E 886 10" Sch. 40 17,355 18,000 10" SSV-302-1F 665 10" Sch. 40 19,461 22,500 10" SSV-302-1G 125 10" Sch. 40 22,383 22,500 10" SSV-302-1H 138 10" Sch. 40 17,990 18,000 10" SSV-302-1J 803 10" Sch. 40 20,438 22,500 10" SSV-302-1K 358 10" Sch. 40 21,743 22,500 10" SSV-302-1L 341 10" Sch. 40 21,144 22,500
(
i l
l l
i
TABLE 2-3 ,y e :o o FITZPATRICK N~y SRV PIPE SUPP0RT MODIFICATIONS E [0, Type of Modified Type of E SRV Piping Support 3 Line Node # Designation Support Yes No Modification PFSK-1298 Y-Rigid X Support Reinforcea A 125 124 PFSK-1286 X-Snubber X 195 H29-180 Spring X Replace Spring Can 145 PFSK-829 Y-Rigid X Support Reinforced 132 PFSK-1291 X-Rigid X 4 H29-177 Spring X Reset Existing Spring 115 New Support Lateral Rigid X Add New Support 160 i
119 New Support Axial Snubber X Add New Snubbers H29-163 Spring X e Reset Existing Spring B 219 X-Snubber X Remove Support Steel 545 PFSK-1236 H29-164 Y,Z-Rigid X Support Reinforced 225 232 PFSK-870 X,Y-Rigid X Support Reinforced H29-166 Spring X Reset Existing Spring Can 242 530 PFSK-1237 X-Snubber X l
247 H29-167 Y-Rigid X
TABLE 2-3 (CONTINUED)
E i' R FITZPATRICK ggg El3 s
R m
SRV PIPE SUPPORT MODIFICATIONS
- =
m Type of Modified Type of '8 SRV Piping Support Modification A Designation Support Yes No Line Node #
PFSK-822 Y-Rigid X Replace Rigid Struts C 155 270 PFSK-1427 X-Rigid X Spring X Replace Spring Can 286 H29-196 Y-Rigid X Support Reinforced 300 PFSK-832 Rigid X New Support Added 3 345 New Support Y-Snubber X Pcplace Snubber D 496 PFSK-1255 X,Z-Rigid X Support Reinforced 500 H29-154 Spring X Replace Spring Can 550 H29-153 570 H29-155 Spring X Snubber X Replace Snubber 575 PFSK-1263A 580 H29-156 Y-Rigid X 590 H29-157 Y-Rigid X Strut X Replace Existing Snubber 604 PFSK-1256A X,Y-Rigid X Support Redesigned 625 H29-158' 650 PFSK-826 Y-Rigid X Rigid X Add New Support 685 New Support
TABLE 2-3 (CONTINUED)
E' 2 2 FITZPATRICK 1&e i k! 3.
SRV PIPE SUPPORT MODIFICATIONS g {0
- o e
SRV Piping Support Type of Modified Type of '8 Line Node # Designation Support Yes No Modification ;!
E 760 H29-190 X,Y-Rigid -
X Support Redesigned 815 PFSK-1289 X-Rigid X Support Redesigned 841 PFSK-740 . Spring X Replace Spring Can 860 PFSK-831 Y-Rigid X 905 fiew Support Rigid X New Support Added 3 F 374 New Support Snubber X Support Added 395 H29-182 Spring X 430 PFSK-1926 Snubber X 445 H29-216 Rigid X 460 H29-183 Spring X Reset Existing Spring 461 PFSK-1238 Snubber X 471 H29-184 X,Y-Rigid X 505 H29-185 Spring X Replace Spring Can 512 PFSK-1251 Snubber X 513 H29-186 Y-Rigid X e
TABLE 2-3 (CONTINUED)
FITZPATRICK x g .a
< , o SRV PIPE SUPPORT MODIFICATIONS {y[
8 N"70 SRV Piping Support Type of Modified Type of x Line Node # Designation Support Yes No Modification j s
F 550 PFSK-200 Y-Rigid X Support Redesigned (continued) Support Added 635 New Support Rigid X 590 PFSK-830 Y-Rigid X G 134 New Support Snubber X Support Added 165 H29-160 Y-Rigid X Support Redesigned 206 H29-215 Z-Rigid -
X Support Redesigned 210 PFSK-1907 Snubber X Replace Snubber 230 PFSK-827 Y-Rigid X Support Reinforced 275 New Support Rigid X Support Added g H 113 H29-198 Spring X Reset Existing Spring 117 PFSK-1299 Y-Rigid X Support Reinforced 123 PFSK-1290 X-Rigid X Support Redesigned 130 PFSK-741 Spring X Replace Spring Can 132 PFSK-833 Rigid X Support Reinforced 14E New Support Rigid X Add New Support 159 PFSK-1287 Snubber X New Support Snubber X Add New Support 500
TABLE 2-3 (CONTINUED)
FITZPATRICK y{
m :r SRV PIPE SUPPORT MODIFICATIONS y.y.
[kE Piping Support Type of Modified Type of f SRV Designation Support Yes No Modification B Line Node # A New Support Snubber X Add New Support J 208 H29-203 Spring X Reset Existing Spring 800 801 PFSK-1275 Snubber X 216 H29-204 Y,Z-Rigid X Support Reinforced Snubber X 218 PFSK-1886 228 H29-205 Spring X Replace Spring Can b H29-206 X,Z-Rigid X Su'pport Redesigned 235 New Support Y-Rigid 243 H29-207 Spring X Replace Spring Can 248 PFSK-1258 X,Y-Rigid X H29-208 294 H29-209 Spring X 257 H29-210 Spring X Reset Existing Spring 249 PFSK-1268 Snubber X 295 PFSK-1267B Snubber X PFSK-1267A Snubber X 818 296 H29-211 Y-Rigid X
TABLE 2-3 (CONTINUED)
FITZPATRICK EY#
5a SRV PIPE SUPPORT MODIFICATIONS { y.
L%
SRV Piping Support Type of Modified Type of E Line Node # Designation Support Yes No Modification '8 A
J 264 H29-212 Spring X Reset Existing Spring (continued) 269 H29-213 X,Y-Rigid X Add Weld 270 PFSK-1259 Snubber X 275 PFSK-834 Y-Rigid X 284 New Support Rigid X Add New Support Y-Rigid Support Redesigned K 319 PFSK-1135A X h
338 PFSK-1135B X,Z-Rigid X Support Redesigned 314 New Support Y-Snubber X Add New Support 328 New Support X,Z-Snubber X Add New Support 361 PFSK-1135C Rigid X Support Reinforced 377 New Support Rigid X Add New Support L 320 PFSK-1156 Y-Rigid X Support Reinforced 331 PFSK-1158 X,Z-Rigid X Support Redesigned 346 PFSK-1159 Y-Rigid X Support Reinforced 357 New Support Snubber X New Support Added
TABLE 2-3 (CONTINUED) yMg FITZPATRICK SRV PIPE SUPPORT MODIFICATIONS
=ra N-E u
Piping Support Type of Modified Type of 8 SRV Line Node # Designation Support Yes No Modification 371 New Support Snubber X New Support Added L
(continued) New Support Added 372 New Support Snubber X 374 New Support Snubber X New Support Added 376 New Support Snubber X New Support Added 377 New Support Snubber X New Support Added I
I I
Technical Report gg Rev s on 1 TABLE 2-4 FITZPATRICK ,
SRV VALVE EVALUATION Component Component SRV Pipe Allowable Designation Type System Stress Pipe Stress 02RV/71A SRV SRV A 2710 18000 3"-V CF-3 A 3" Check SRV A 3364 18000 02-VB -1 10" Check SRV A 6157 18000 02RV/71B SRV SRV B 1414 18000 3"-V CF -30A 3" Check SRV B 3862 18000 02-VB -2 10" Check SRV B 4882 18000 02RV/71C SRV SRV C 3170 18000 3"-V CF -30A 3" Check SRV C 8568 18000 02-VB -3 10" Check SRV C 6072 18000 02RV/71D SRV SRV D 3092 18000 3"-V CF -30A 3" Check SRV D 9250 18000 02-VB -4 10" Check SRV D 5450 02RV/71E SRV SRV E 2880 18000 3"-V CF -30A 3" Check SRV E 5502 18000 02-VB -5 10" Check SRV E 6170 18000 02RV/71F SRV SRV F 3155 18000 3"-V CF -30A 3" Check SRV F 5593 18000 02-VB -6 10" Check SRV F 5902 18000 02RV/71G SRV SRV G 3248 18000 3"-V CF -30A 3" Check SRV G 4747 18000 02 -VB -7 10" Check SRV G 6180 18000 02RV/71H SRV SRV H 2955 18000 3"-V CF-30A 3" Check SRV H 3570 18000 02-VB -8 10" Check SRV H 5558 18000 02RV/71J SRV SRV J 2634 18000 3"-V CF -30A 3" Check SRV J 9050 18000 02-VB -9 10" Check SRV J 5336 18000 02RV/71K SRV SRV K 2870 18000 3"-V CF -30A 3" Check SRV K 5050 18000 02-VB -10 10" Check SRV K 6656 18000 02RV/71L SRV SRV L 2904 18000 3"-V CF -30A 3" Check SRV L 7846 18000 02-VB -11 10" Check SRV L 5462 18000
Technical Report NNM TR-5321-2 Revision 1 s
l MAIN STEAM % l LINE I
JET :
DEFLECTOR _ .;. _
\
N, I s ' ,%
/ p#
/
-=
t N WATER LEVEL x DISCHARGE TEE-QUENCHER
- g p;e- no- up ,, g - rg g _ _.
)I l ' l I ' l ' I' l
(
~ ~
FIGURE 2-1 SRVLIMEROUTING-TYPICA[
i
! Technical Report W1RENDENM TR-5321-2 Revision 1 0 '
14 15 ,
/
13 16
/ l 12 , \ 1 1
11 2 270* - -
-t-90' 10 x y 3 9 4 l
/
8 5 11 "T" QUENCHER 5~
7 6 180 FIGURE 2-2 SRV LINE ARRANGEMENT-TORUS
Technical Report 9pmm TR-5321-2 Revision 1 N /
\ U - - - - - --
VENT HEAD'ER' PENETRATION DEFLECTOR y
TO QUENCHER DETAIL"I"
\
\
\-
s ',%
\\ -;
%s s
\
's
~
[ g j Y SEE DETAIL "A"
- l -
~ TEE-QUENCHER e y gt= p tp wt En nt g;p fl{l'l'll' I'l l FIGURE 2-3 SRV LINE ROUTING-TYPICAL
Technical Report MMM TR-5321-2 Revision 1 l
4 VN , -
s .
. ./ /\.
- a f , .
3
, l r- h
.1 a E !!
I
~
N,
- T l
.< --l y l2 I
i f" TYPE 316LARMS '
t i l) \ ', f r;
4,.
s ....
iiii
~
,l
. r ....
i ....
!ili'
. ... g iii! t
- qi j
l i l
!!!!' i i i d fl lr!!!! i
. i!!! ", i iiii, i i N ' []! l[] i-o-7_. TO NEXT RING GIRDER 20" SCH 120 PIPE [
, r RING GIRDER i .
I )
TORUS
~
FIGURE 2-4 -~SRV TEE-QUENCHER Y 50PP0iiT
a d , Gi u
J i B, '
10" SSV-302-IJ a'
d ba
, la g y VENT PIPE PENETRATION y
n &
T SRV QUENCHER h
~
b y
< > 1 -
.1, _ mNe gg> j -3 QUENCHER SUPPORT '
'Jd FIGURE 2-5 SRVPIPINGMODEL, TYPICAL l l
W l
$h$
CS?
$ ~ X, si o i
,m-l OPERATING BASIS EARTHQUAKE < E A
HORIZONTAL EL 330.fL 4- ,
i l
l r
- E -
i ; 3-e
- E i w
o I 2 v
e
=
o C 2-( $'
I.
d.
O M,
1-l j .5-i i i i i i i l 20 25 30 35 O! 5 10' 15l FREQUENCY (HERTZ):
i FIGURE 2-6 SRV SEISMIC SPECTRA, TYPICAL'
WTERMEMM Technical Report s TR-5321-2 % Revision 1 d >
11N0 DES /
SPAN (TYP) s <>
RING GIRDER
' 5, d' 00WNCOMER VENT HEADER SUPPORT COLUMN VENT PIPE i >-
f s 1A B
<> \ >
\
\ -
I I
\ ,
L VENT HEADER INTERSECTION (STIFFNESS MATRIX INPUT) s c /
\
/
4 I 1 6 >
l i
db >
4 g SYMMETRY BOUNDARY CONDITIONS FIGURE 2-7 VENT SYSTEM MODEL
Technical Report TR-5321-2 WM Revision 1 gg 3.0 TORUS ATTACHED PIPING (TAP)
The torus at FitzPatrick has 17 piping systems attached to its outer shell. These systems connect to 29 penetrations and are listed in Tables 3-1 and 3-2, and Figure 3-8c. Analysis of the large diameter attached piping systems included all piping from the torus to the first anchor. Small diame-ter piping was analyzed to the first anchor or a distance where the torus loads could be considered negligible. ,
Also considered in this analysis are:
e o Branch piping connected to TAP systems, o Torus penetration stresses.
e Piping inside the torus attached to TAP systems.
e Pump and valve loads, e All pipe support and anchor loads.
The analysis method is different for large bore TAP systems (above four-inch diameter) and small bore systems (four-inch and below), as discussed in the following text.
Different organizations were involved in these analyses. TES performed piping analysis for torus dynamic loads for all systems. Stone & Webster
- had recently completed weight, thermal and seismic analysis for the same lines as a part of Bulletin 79-14 work; these results were used for these loads (Reference 18). In cases where a line had to be resupported, or where Stone &
Webster analysis was not available, TES analyzed for all loads. All piping supports were analyzed to the new combined loads by TES.
This report includes descriptions and results for all analysis.
3.1 Applicable Codes and Criteria Analysis and modifications to TAP piping and supports were in accordance with the following codes:
- Stone & Webster Co., New York, NY
Technical Report TN TR-5321-2 O Revision 1 Piping Analysis TES Analysis - ASE Section III,1977 Edition (Reference 2) and B31.1 1967 Edition (Reference 17).
S & W Analysis - (Bulletin 79-14), B31.1 1967 Edition (Reference 17 & 18).
Support Analysis All TAP and branch supports - ASE Section III,1977 Edi-tion, and including NRC Bulletin 79-02 requirements (Refer-ence 3).
Load combinations and stress levels were evaluated in accordance with Table 5-5 of the Mark 1 Containment Program Structural Acceptance Cri-teria Plant Unique Analysis Application Guide (Reference 5). Table 5-5 is reproduced in this report as Table 1.
Damping for all time history piping analysis was taken at 2% of critical for lines equal to or less than 12 inches in diameter, and 3% for larger lines. Seismic analysis used 2% damped spectra.
3.2 TAP Loads Loads applied to TAP systems include:
Mark 1 Loads Shell motion due to pool swall.
Shell motion due to SRV line discharge.
Shell motion due to condensation oscillation.
Shell motion due to chugging.
Pool drag and impact loads on internal piping.
Technical Report TN TR-5321-2 6 gg Revision 1 and Original Design Loads Deadweight.
Thermal expansion.
Seismic.
The Mark 1 loads, due to shell motion, were calculated based on plant unique shell response data developed during an earlier phase of this program and reported in the PUA report, Reference 1. Drag leads on internal piping were developed using generic methods from the Mark 1 Program as a part of this piping analysis work. These loads are described more fully in the Mark 1 Load Definition Report (Reference 10).
Analysis for seismic response was based on FSAR spectra.
3.2.1 _Shell Motion Due to Pool Swell TAP input loads, due to shell motion during pool swell, were based on data developed during the Plant Unique Analysis for the shell (Refer-ence1). The PUA shell analysis provided time history response information in five degrees of freedom for every point on the shell where large bore TAP was connected. This data consisted of three translations and two out of plane rotations (no torsion). Data for small bore piping was based on conservative bounding of the large bore data. Attachment points for large bore piping are illustrated in Figures 3-8a, 3-8b and 3-8c.
Data available from the plant unique shell analysis consists of time history displacements and rotations. These were converted to equiva-lent time history forces as described in paragraph 3.3.6.1.
A typical pool swell force time history is illustrated in Figure 3-1.
Technical Report 9
e n1 b 3.2.2 Shell Motion Due to DBA Condensation Oscillation The DBA condensation oscillation load definition is given in Reference 11 as a set of spectral pressures, from 1-50 Hz. Shell response due to this loading was calculated by applying each frequency in this band to the torus shell model shown in Figure 3-7 and calculating response for each sinusoidal excitation. (This work was done earlier to allow calculation of shell, stress for Ref erence 1). Shell response was calculated for frequencies up to 32 Hz; frequencies above 32 Hz were considered negligible as discussed in Appendix 2.
Shell responses for each of these frequency components were combined into an equivalent time history using random phasing of the indivi-dual components. Amplitudes of this equivalent time history were then increased by a f actor of 1.15 to allow for the in-phase response of the four peak frequency components. See Reference 12 for a further discussion of the factor and component phasing.
This method of combining frequency components and generating an equivalent shell response time history was repeated for each TAP penetra-tion for large bore piping. Responses for small bore piping were based on conservative bounding of the large bore data.
A typical DBA C0 shell response is illustrated in Figure 3-2.
i l
i 3.2.3 Shell Motion Due to Chugging Shell response during chugging was defined separately for pre-chug and post chug loads.
Pre-chug is a sinusoidal pressure load equal to + 2 psi on the torus shell; this load can occur at any frequency between 6.9 and 9.5 Hz
R P' TN N'E"d* 2 ENGDEERING SERVICES Revision 1 (Reference *0). Shell response for pre-chug was calculated by applying a continuous 12 psi sine pressure to the large torus model (Figure 3-7) in the specified frequency range. Maximum shell response in this range occurred at 9.5 Hz. This was considered as one of the inputs to TAP.
Post cheg is specified as a spectrum of pressures from 1-50 Hz. Shell response was calculated for each 1 Hz component in this spectrum, then all 50 components were combined into an equivalent time history using random phasing of all components. Amplitudes of this time history loading were multiplied by 1.15 to account for the f act that some elements of the spectrum are not randomly phased. Further discussion of this factor can be found in Reference 6. The resulting pressure time history was applied to the model in Figure 3-7 to calculate shell response.
3.2.4 Shell Motion Due to SRV Line Discharge TAP input loads, due to shell motion during SRV line dis-charge, were based on data developed for the PUA shell analysis (Reference 1).
This shell analysis was the result of a finite element analysis that was calibrated with in-plant SRV test data, as described in Reference 1. The data resulting from the shell analysis were time histories and were used to provide time history input functions for the TAP.
Section 5.2 in the Load Definition Report (Reference 10) requires that we allow for a 125 percent shif t in the SRV frequency for discharge through a cold line, and a 140 percent shif t for discharge through a hot line. This was considered by examining the response modes and frequencies of the TAP piping systems and then making adjustments within the specified ranges to force worst case input-response frequency pairing.
The strongest torus shell respense during SRV actuation is the result of simultaneous actuation of several SRV lines. These cases were considered by adding the shell pressures due to the individual actuations by absolute summation.
Technical Report WM TR-5321-2 ENGNEERING SERVICES Revision 1 A typical shell response due to SRV actuation is illustrated in Figure 3-3.
3.2.5 Loads on Internal Pipino.
1 Most of the large TAP systems extend into the torus. In the case of suction lines, the internal portions usually consist of a pipe fitting and strainer. For return lines, longer sections of pipe, up to approximately 20 feet, extend into the torus.
The internal portions of these systems are subjected to sub-merged structure drag if they are in the pool; or pool impact, if they are above the water level. In either case, the appropriate Mark 1 loads were calculated and considered during the piping evaluation.
Loads for piping in the pool and above the pool were calcu-lated in accordance with the methods of the Load Definition Report (Reference 10), NUREG 0661 (Reference 11) and Appendix 1 of Reference 1. All loads were considered, including: ,
For Submerged Piping:
o C0 Source and FSI Drag, e Post Chug Source and FSI Drag. ,
o Pre-chug Drag.
e SRV Bubble and Jet Loads, e Pool Swell Bubble Drag.
e Pool Swell Fallback.
For Structures Above the Pool:
o Pool Swell Water Impact and Drag.
e Froth, e Fallback.
1 _
TME
-5 -
Revision 1 A typical submerged structure load spectrum is shown in Fig-ure 3-4. This spectrum includes C0 and CH source and FSI drag.
3.2.6 Deadweight, Thermal and Seismic Analysis Analysis for all TAP systems was also done for deadweight, thennal and seismic conditions.
Thermal analysis was performed at the original design ther-mal conditions. Thennal displacement of the penetration was determined from the maximum operating temperature of the torus and applied for all cases.
Seismic analysis was done using the spectra from the FSAR.
The enveloped OBE spectra for a typical line is shown in Figure 3-5.
3.3 TAP Analysis Method The method for TAP pipe stress analysis varied for each of the following cases:
e Large bore piping (over 4" diameter),
e Small bore piping systems (4" and less), which could be reduced to single degree-of-freedom approximations.
e Small bore piping which could not be reduced to single dof systems.
e Branch piping off of TAP systems.
1 Analysis of supports, anchors and torus penetrations did not vary and was the same for all types of piping systems.
3.3.1 Representation of Torus Shell for Piping Analysis Because the larger TAP systems are stiff and heavy when compared to the torus shell, it is important that the piping ccaputer model
I Technical Report pg v n1 allows for dynamic interaction between the piping and the torus. This was done for all TAP piping systems by including a set of ground springs in the piping model to represent the torus connection, as illustrated in Figure 3-6.
Five ground springs were used to represent the torus shell; these represented stiffnesses associated with the three translations of the shell and the two out of plane moments on the shell. Torsional pipe loads were considered negligible.
The stiffness values of the ground springs were calculated by applying unit loads and moments to the large shell finite element model of the torus illustrated in Figure 3-7. Different attachment stiffnesses were calculated for each pipe penetration location, and then applied to the appro-priate piping system model.
3.3.2 Piping Analysis Method - Large Bore Systems Analysis of all large bore piping systems was done using finite element models of each system. These models included ground springs to represent the torus and also included piping inside the torus.
All analysis on these models was done using the STARDYNE computer code (Reference 15). Time history dynamic analysis used damping values of 2% of critical for all lines with 12 inch diameter or less, and 3%
for larger lines; OBE seismic used a 2% damped spectra. Analysis on these models included:
e Zero and Full AP Pool Swell Motion and Drag Loads, o Post Chug Shell Motion and Drag Loads.
e DBA C0 Shell Motion and Drag Loads, e SRV Shell Motion and Drag Loads, o Deadweight.
e Seismic.
e Thermal .
Technical Report itTF1 ETVNE 40 e s n1 ENGNEERING SERVICES Pre-chug was considered as a separate load condition, but it was determined that it would always be bounded by DBA C0. On that basis, pre-chug loads were not run for each TAP system.
All TAP response due to shell motion was done using time history analysis. Response due to drag loads on internal piping was calcu-lated by harmonic analysis for the spectral loads and hand analysis for transients. The effects of both shell motion and internal loadings were considered for all points in the piping system.
Pipe stress due to welded support attachments was considered by separate analysis and included in the pipe stress evaluation.
3.3.3 Piping Analysis Method - Complex Small Bore Systems Analysis of small bore piping systems that could not be reduced to single degree of freedom systems were treated identically to large bore systems, except for the loads considered. For these systems, the loads considered included:
e DBA C0.
e Deadweight, e Seismic.
e Thermal .
Consideration of Mark 1 dynamic loads was limited to DBA CO, based on experience with large bore piping analysis for five Mark 1 plants.
This experience showed that all high stressed lines were controlled by DBA CO, except in a few special cases. Appendix 1 discusses this further.
3.3.4 Piping Analysis Method - Simple Small Bore Systems Small bore piping systems that could be reduced to single mass approximations were analyzed using hand analysis. Torus shell stiffness
Technical Report TN TR-5321-2 N NES Revision 1 was included in these models to the extent that it affected first mode response, as a minimum. Higher modes were considered if they fell within the range of the input load. Typically, these systems consisted of a short length of pipe, terminating in a valve or tubing.
Shell input to these systems (for Mark 1 loads) was format-ted in the frequency domain to provide an input spectrum. This spectral data was used in combination with the hand analysis to calculate response levels.
Loads considered for simple small bore systems were the same as for the mere complex small bore systems, including seismic, weight and thennal, if .p,elicable.
3.3.5 Piping Analysis Method - Branch Piping Branch piping connected to TAP systems was modelled with the TAP systems if the ratio of their bending stiffness was greater than approxi-y mately 1:40.
Branch piping too flexible to meet this ratio.was considered by separate analysis. These systems were analyzed statically by placing a displacement at the connection point, equal to twice the TAP motion at the connection point. (except deadweight deflections, which were considered neg-ligible). The entire branch line was modelled for these analyses.
3.3.6 Piping Analysis - Load Input for Computer Models 3.3.6.1 Mark 1 Loads Due to Shell Motion Shell motion, due to internal Mark 1 loads, is due to pressures across broad areas of the shell, as opposed to concentrated forces at the penetrdtion. Because of this, the interactive effects of piping
Technical Report
"- WTF1 WE 0;i3 h2, ENOMERM SERVCES and shell should include allowance for local shell compliance in the force input to the piping system. The method of load input for TAP accounts for this. The method is illustrated in Figure 3-6.
The steps involved are:
e Extract displacement time history from large computer model for a shell without an attached TAP system. (Reference 1 and Figure 3-7).
e Determine local shell stiffness from large computer model (Reference 1 and Figure 3-7).
e Determine an equivalent force time history at the penetration by multiplying displacement by stiffness.
e Apply the force time history to the TAP as shown in Figure 3-6.
The use of forces, rather than displacements to drive the model, is necessary to accurately account for the inertial inter-action of the piping, since the available shell response data is for an unloaded shell (no piping). Use of forces as input will allow displacements at the penetration to increase or decrease in reaction to the inertial forces from the piping.
3.3.6.2 Submerged Drag Loads on Internal TAP Drag loads on internal piping during C0, CH, SRV and pool swell were evaluated using the same TAP piping models that were used for shell induced, seismic and other loads. Internal drag loadings were run as separate cases, with worst-case orientations, and then combined with other loadings to determine pipe stress, support loads and penetration stress. The effects of drag load on both internal and external parts of the TAP system were calculated and included in all evaluations.
Technical Report YE WE TR-5321-2 ENGNEERING SERVICES Revision 1 Loads were applied to the piping and evaluated by the following methods:
o Pool Swell Drag - Static Analysis x 2.
e Pool Swell Fallback - Static Analysis x 1.
e Pool Swell Impact - Static Analysis x 2.
e Pool Swell Froth - Static Analysis x 2.
e C0 Drag - Dynamic Analysis (Spectrum).
e Post Chug Drag - Dynamic Analysis (Spectrum).
e SRV Drag - Static Analysis x 1.
e Pre-chug - Bounded by DBA C0.
Piping response to C0 and post chug drag were eval-uated using dynamic analysis. These spectra, including their FSI components, were then enveloped to form a single spectrum that was used in this analysis.
Each frequency component in this spectrum was then applied to the CG of the submerged internal piping as a harmonic forcing function. The load in the pipe was calculated at a point just inside the penetration, in each of six degrees-of-freedom. These single-frequency piping loads were then combined into a single load at that point by absolute sum of the four largest compon-ents added to the SRSS of the balance. This was done for each degree of freedom. (The basis for this method of combining individual frequency com-ponents is discussed in Reference 12). The loads calculated in the pipe were then applied to the system as static loads; and pipe stress, penetration stress, and support loads were determined. A typical combined spectrum is
, illustrated in Figure 3-4.
TAP analysis for other loads noted above, was done by applying the appropriate load to the CG of the affected area and performing static analysis.
TME Technical Report N MES TR-5321-2 Revision 1 3.3.7 TAP Penetration Analysis Analysis of torus penetrations included the following loads:
e Loads from piping response due to shell motion (Mark 1 loads).
e Loads due to submerged drag and/or pool impact, on internal sections of TAP, as applicable.
e Loads from weight, seismic and thermal conditions on the attached piping.
e Shell loads which exist due to the Mark 1 and other loads, independent of piping (from Reference 1).
The calculation of stress from the loads was done using a Bijlaard analysis (Reference 8) to account for local penetration stress due to piping loads. These stresses were combined with free shell stresses in that area, intensified to account for the discontinuity. Free shell stress was taken from earlier containment analysis, as reported in Reference 1.
Penetration stresses were calculated for each load in each degree of freedom.
Stresses resulting from this analysis were combined to form the load cases defined in the PUAAG (Reference 8 and Table 1).
Stress in the piping within the limits of reinforcement was calculated by combining the stress in the pipe with the local shell stresses by absolute summation. This was also evaluated for each degree of freedom and each of the PUAAG load cases (Table 1).
3.3.8 Analysis Method for Piping Supports Analysis was done for all piping supports for all TAP and branch systems. Calculations were made using both hand and computer analy-sis, depending on the complexity of the individual support. Evaluation of baseplates and anchor bolts was included, using the current procedures devel-
Technical Report yg v n1 M SB{\/lCES oped in response to NRC Bulletin 79-02 (Reference 3). The STAAD computer program was used in most cases where computer analysis of supports was done (Reference 16).
3.3.9 Vacuum Breaker Analysis The torus TAP systems include a portion of the vacuum relief system which connect the main vent pipe to the the torus airspace, and which include the wetwell-to-drywell vacuum breakers. Analysis of these vacuum breakers was not within the scope of the Mark 1 Containment Program, but is
, addressed in Reference 13.
l 3.3.10 Active Components Active components en TAP systems include ten pumps and 68 valves, excluding the 11 vacuum breakers. Acceptability of these components was assured by limiting stresses at these locations, as described in the evaluation section. No analysis was necessary on these components.
3.4 Evaluation and Results 3.4.1 General Combinations of the previous analysis cases were done to allow evaluation of results in accordance with Table 5-5 of Reference 5.
(Table 1 in this report.) This table lists a total of 27 load cases for both essential and non-essential piping systems. For purposes of this evaluation, all TAP systems are classified as essential.
Technical Report WM TR-5321-2 Revision 1 N SS{\/ ICES The 27 load cases shown in Table 1 were reduced, by conserva-tive bounding, to the cases listed below:
Case No. Major (Table 1) Load (s) Allowable (Eq. 9) 1 2 SRV (C3.1) + OBE 1.2 S h 2 16 Zero A P 2.4 5 3 21 DBA C0/CH + SSE 2.4 S h 4 25 Pool Swell + SRV (A1.3) 2.4 S h 5 15 Post Chug + SRV 2.4 S h In these cases, the seismic stresses were combined with the absolute sum of the Mark 1 dynamic loads by the the SRSS method.
3.4.2 Piping Stress - Large Bore Systems Stress in all large bore TAP systems was combined and eval-uated in accordance with Section III of the ASME code for the five cases listed in Paragraph 3.4.1. These evaluations included the effects of local pipe stresses due to welded attachments at supports. Fatigue was considered as explained in Paragraph 3.4.5 & 3.4.6.
The large bore TAP systems are listed in Table 3-1 along with the maximum stress for the controlling load combination.
3.4.3 Pipe Stress - Small Bore TAP Systems Evaluation of small bore TAP systems was the same as for large bore systems, except that the only Mark 1 dynamic load considered was DBA CO. This approach was based on experience gained in large bore analysis and is discussed further in Appendix 1.
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - . - _ _ _ . _ _ - - . - - - _ _ _ _ _ _ - - - - - - - - - - - - -- J
Technical Report TR-5321-2 Y Revision 1 47_ MM Small bore systems are listed in Table 3-2.
3.4.4 Pumps and Valves Evaluation of pumps and valves was done based on stresses in the adjacent piping. Pipe stresses meeting Level B criteria were considered adequate to assure proper operation of the pumps or valve. (Reference 5, Section 5.5).
Results of the pump and valve evaluation are listed in Table 3-4.
3.4.5 Piping Fatigue Evaluation Consideration of the fatigue effects of cyclic loading is reported in Reference 9 for bounding Mark 1 plants. This reference defines bounding conditions and concludes that the stress levels and cycles involved in these systems will not produce a f atigue problem. The conclusions are applicable to the FitzPatrick Plant. No further plant unique evaluation was done to address f atigue considerations for piping. Fatigue for the penetra-tion is considered below.
l 3.4.6 Torus Shell Penetration Evaluation Evaluation of torus penetration stresses considered loads from the external and internal piping, as well as the loads that exist in the shell, due to the same event (s). Shell stress away from penetrations is reported in Reference 1. ,
l Stresses in the penetration area were evaluated in accord-ance with subsectin NE of The ASME code, using the following paragraphs:
NE-3221.2 Local Membrane Stress Intensity
\
2f-2 W Revision 1 NE-3221.3 Primary General or Local Membrane plus Primary Bending Stress Intensity NE-3221.4 Primary plus Secondary Stress Intensity NE-3221.5 Analysis for Cyclic Operation NE-3227.5 Nozzle Piping Transition Fatigue evaluation of the penetration (paragraph NE-3221.5) showed that the maximum load could be cycled on each penetration for at least 10,000 cycles without exceeding code allowables. The major loads that form these load combinations are pool swell (1 cycle), DBA.C0 (900 cycles), and SRV Case C3.3 (50 cycles). Other loads; normal SRV actuation, IBA C0, and chugg-ing, can produce up to 10,000 cycles, but only at greatly reduced stress levels. Based on this, the 10,000 cycles at maximum stress represents a conservative level of evaluation and the TAP shell penetrations are con-sidered acceptable for f atigue.
Controlling stresses in the TAP penetrations are listed in Table 3-6. Additional information of number of cycles for each condition can be found in Reference 9.
3.4.7 Piping Supports All piping supports on the TAP systems were evaluated for the same load ccmbinations as the piping (Table 1).
Evaluation was done in accordance with ASE,Section III, Division I, Subsection NF,1977 with 1978 Summer Addenda and included the following criteria:
e Expansion type anchor bolts and baseplates were evaluated in accordance with Bulletin 79-02 cri-teria (Reference 3).
1 TM Revision 1 _49 nom FS e No stresses in pipe supports were allowed to exceed yield, regardless of pipe stress allowables.
A listing of pipe supports and modifications is given in Table 3-5.
3.5 Summary of TAP Modifications Modifications to torus attached piping systens consisted of support changes, as well as modifications to internal piping.
Modifications to internal piping included shortening some lines to reduce subnergence and drag loads; rerouting one line and supporting it from the ring girder and resupporting one other. The following modifications were made; these are illustrated in Reference 1:
e Reroute RHR line and support from ring girder.
e Reinforce spray header supports on the ring girders.
e Modify HPCI line.
o Relocate RCIC line 8".
e Resupport 10" condensate line.
Modifications to external piping consisted of support and support steel modifications as summarized in Table 3-5 of this report.
TABLE 3-1 FITZPATRICK E A m#
LARGE BORE TAP RESULTS < 6, g.
'" M 3.
System Penetration Line Size Controlling Maximum Allowable O [$,
N ame - Number & Schedule Load Case Stress Stress '
O 30" Std. Seismic & SRV 24,029 27,000 g Vacuum Relief Line X-202A/F a 30" Std. DBA C0 27,926 36,000 Vacuum Relief Line X-202B/G Reactor Building X-205 20" Sch.10 Seismic & SRV 21,415 27,000 Normal Vent RHR Discharge X-210A & X-211A 24" Std. DBA C0 31,961 36,000 RHR Discharge X-210B & X-211B 24" Std. DBA C0 35,073 36,000 RCIC Turbine X-212 8" Sch. 40 (Std) DBA C0 32,608 36,000 /,,
Exhaust ?
Drain X-213A/B 3" Sch. 40 (Std) DBA C0 31,195 36,000 HPCI Turbine X-214 20 Sch. 10 DB A CO 21,139 36,000 Exhaust Vent Purge Outlet X-220 20" Sch.10 SRV 11,633 27,000 RCIC Pump Suction X-224 6" Sch. 40 (Std) SRV + PS2 25,146 36,000 RHR Pump Suction X-225A 20" Sch. 10 SRV + PS2 26,407 36,000 RHR Pump Suction X-225B 20" Sch. 10 DBA C0 35,643 36,000 HPCI Pump Suction X-226 16" Std. PS1 29,480 36,000 Core Spray X-227A 16" S td. DBA C0 33,381 36,000 Pump Suction Core Spray X-2278 16" Std. PS1 33,746 36,000 Pump Suction X-228 10" Sch. 40 (Std) SRV + PS2 29,447 36,000 Condensate Drain
Y$
705 EUX "L%
TABLE 3-2 x E
FITZPATRICK 5?
n SMALL BORE TAP RESULTS Line Size Type of Maximum Allowable Nax. Stress System Penetration Stress Location Name Number & Schedule Analysis Stress 1" Sch. 80 Computer 11,110 36,000 Node i Oxygen Analyzer X-203B ,
1" Sch. 80 Computer 25,738 36,000 Node 1 Liquid Level X-206A,B,C,D Indicator g 2" Sch. 80 Computer 25,545 36,000 Node 28 23 X-217 2" Sch. 80 Computer' 19,538 36,000 Node 24 Vacuum Pump X-221 1 Discharge 2" Sch. 80 Computer- 18,162 36,000 Node 58 Condensate Drain X-222 1
I I
TABLE 3-3 e :n m FITZPATRICK y. g ]3-BRANCH LINE PIPE STRESSES hhh
,m-E Branch Line TAP TAP Branch Line Maximum Allowable B Designation System Penetration Dia./Sch. Stress Stress ;;
1" W25-152-18 HPCI Pump Suction X-226 1" Sch. 80 (XS) 10,455 36,000 3/4" Vent HPCI Pump Suction X-226 3/4" Sch. 80 (XS) (1) 1" W23-152-22A Core Spray Pump Suction X-227A 1" Sch. 80 (XS) (2) 2" W23-152-16A Core Spray Pump Suction X-227A 2" Sch. 80 (XS) 12,700 37,500 2" W23-152-17A Core Spray Pump Suction X-227A 2" Sch. 80 (XS) 12,700 37,500 ,
1" W23-152-22B Core Spray Pump Suction X-2278 1" Sch. 80 (XS) (2) 2" W23-152-16B Core Spray Pump Suction X-227B 2" Sch. 80 (XS) 12,700 37,500 2" W23-152-178 Core Spray Pump Suction X-2278 2" Sch. 80 (XS) 12,700 37,500 g 1" WD-152-48 HPCI Turbine Exhaust X-214 1" Sch. 80 (XS) 15,748 36,000 3/4" Drain RCIC Pump Suction X-224 3/4" Sch. 80 (XS) (4) 3/4" Vent RCIC Pump Suction X-224 3/4" Sch. 80 (XS) (4) 3/4" Drain RCIC Pump Suction X-224 3/4" Sch. 80 (XS) (3) 2" W22-152-11 RCIC Pump Suction X-224 2" Sch. 80 (XS) (3) 1" W20-302-110 RCIC Pump Suction X-224 1" Sch. 80 (XS) (1) 1 " Drain RHR Pump Suction X-225A 14" Sch. 80 (XS) (2)
TABLE 3-3 (CONTINUED)
FITZPATRICK
- o -< -<
BRANCH LINE PIPE STRESSES $Q E M 3.
O s-* O Branch Line TAP TAP Branch Line Maximum Allowable~" " h E Designation System Penetration Dia./Sch. Stress Stress : o E
1 " Drain RHR Pump Suction X-2258 1h" Sch. 80 (XS) (2) g 1 " W20-152-46B RHR Pump Suction X-225B 1h" Sch. 80 (XS) 8,840 36,000 1 " W20-152-124B RHR Pump Suction X-225B 1h" Sch. 80 (XS) (3) 1" W20-152-45B RHR Pump Suction X-225B 1" Sch. 80 (XS) 6,200 36,000 4" W20-152-41B RHR Discharge X-210B & X-211B 4" Sch. 40 (Std) 6,572 36,000 4" W20-302-19B RHR Discharge X-210B & X-2118 4" Sch. 40 (Std) 31,482 36,000 h 3" W23-152-78 RHR Discharge X-210B & X-211B 3" Sch. 40 (Std) 7,147 36,000 RHR Discharge X-210B & X-211B 1" Sch. 80 (XS) 2,317 27,000 1" W23-302-298 3" W23-302-6B RHR Discharge X-210B & X-211B 3" Sch. 40 (Std) (1) 4" W20-302-35 RHR Discharge X-210B & X-211B 4" Sch. 40 (Std) 12,761 18,000 1 " AS-302-55B RHR Discharge X-210B & X-211B 1h" Sch. 80 (XS) 2,543 27,000 2" AS-302-55B RHR Discharge X-210B & X-2118 2" Sch. 80 (XS) 20,132 27,000 1 " SLP-152-51 RCIC Turbine Exhaust X-212 1 " Sch. 80 (XS) 4,392 27,000 1" SLP-152-25 RCIC Turbine Exhaust X-212 1" Sch. 80 (XS) (4) 2" SLP-152-49 HPCI Turbine Exhaust X-214 2" Sch. 80 (XS) 4,006 27,000 5
E' d #
TABLE 3-3 (CONTINUED) 569-FITZPATRICK b.
a h,
o BRANCH LINE PIPE STRESSES [
8 TAP Branch Line Maximum Allowable Branch Line TAP Stress Stress Designation System Penetration Dia./Sch.
X-210A & X-211A 4" Sch. 40 (Std) 6,572 36,000 4" W20-152-41A RHR Discharge RHR Discharge X-210A & X-211A 3" Sch. 40 (Std) 499 27,000 3" W22-152-16 RHR Discharge X-210A & X-211A 2" Sch. 80 (XS) 11,976 27,000 2" W22-152-15 X-210A & X-211A 4" Sch. 40 (Std) 32,110 36,000 4" W20-152-40A RHR Discharge ,
3" W23-152-7A RHR Discharge X-210A & X-211A 3" Sch. 40 (Std) 7,147 36,000 f RHR Discharge X-210A & X-211A lb" Sch. 80 (XS) 12,700 37,500 1h" W23-302-10A RHR Discharge X-210A & X-211A 1" Sch. 80 (XS) 3,586 18,000 1" W23-302-29A 3" W23-302-6A RHR Discharge X-210A & X-211A 3" Sch. 40 (Std) (1)
X-210A & X-211A 4" Sch. 40 (Std) 15,491 18,000 4" WLP-302-123 RHR Discharge RHR Discharge X-210A & X-211A 2" Sch. ,80 (XS) 20,122 27,000 2" AS-302-55A RHR Discharge X-210B & X-2118 4" Sch. 40 (Std) 33,750 36,000 4" W25-152-16 1 1/2" AS-302-55A RHR Discharge X-210B & X-211B 1 1/2" Sch. 80 (XS) 2,543 27,000 1 1/2" W23-302-10B RHR Discharge X-210B & X-211B 1 1/2" Sch. 80 (XS) 12,700 37,500 Notes: (1) Beyond Scope of Mark I Analysis.
(2) Nping Frequency 'Jreater than 50 Hz.
(3) Total combined displacement of less than 1/16" at branch line connection.
(4) Mark I stresses at branch line connection are less than 10% of allowable.
TABLE 3-4 FITZPATRICK (fk
- m =r PUMP AND VALVE EVALUATION hyh, o , a Max. Pipe -
Component Component TAP TAP Stress at Allowable E Designation Type System Penetration Component Pipe Stress VB-1 Valve Primary Cont. X-202A,F 11,771 18,000 Vacuum Brkr. Pip.
VGW-15A Gate Valve Condensate X-228 8,927 18,000 Drain Line VGW-15AN Gate Valve Steam Line & X-212 14,575 18,000 VCW-15AN Check Valve Vent From 14,320 18,000 VCW-15AN Check Valve RCIC Pump 16,799 18,000 13-TU-12 RCIC Turbine 8,680 18,000 a, MOV -7 B Mtr. Oper. Valve Core Spray X-227B 4,097 18,000 14P-1B Core Spray Pump Pump Sucticn 15,190 18,000 VGW-15AN Gate Valve (East Lead) 2,965 18,000 3" Globe Valve Globe Valve Drain Line X-213A/B 10,796 18,000 1" Globe Valve Globe Valve 1,459 18,000 27A0V-117 Air Oper. Valve Air Piping X-205 13,483 18,000 270A0V-118 Air Oper. Valve 12,537 18,000 VB -2 Valve Primary Cont. X-202B,G 13,294 18,000 A0V-101A Air Oper. Valve Vacuum Breaker 3,444 18,000 A0V-1018 Air Oper. Valve Piping 3,471 8,000 VB -6 Valve 1,163 l?,000 VB -7 Valve 1,171 18,000 VGW-15AN Gate Valve Steam Line & X-214 11,587 18,000 VCW-15AN Check Valve Vent HPCI Pump 10,488 18,000 VCW-15AN Check Valve 9,870 18,000 23TU-2 HPCI Turbine 664 18,000
TABLE 3-4 (CONTINUED)
FITZPATRICK , y _;
E "R PUMP AND VALVE EVALUATION g$g CUC Max. Pipe :'" 4 %
Component Component TAP TAP Stress at Allowable x Designation Type System Penetration Component Pipe Stress j 7,611 18,000 S MOV-151A Mtr. Oper. Valve RHR Piping X-225A MOV-13A Mtr. Oper. Valve 6,031 18,000 MOV -15A Mtr. Oper. Valve 4,555 18,000 MOV-13C Mtr. Oper. Valve 8,038 18,000 MOV -15C Mtr. Oper. Valve 4,223 18,000 10P -3C Pump 11,929 18,000 10P -3 A Pump 11,720 18,000 27A0V-116 Air Oper. Valve Air Cooling X-220 5,726 18,000 27A0V-115 Air Oper. Valve 6,158 18,000 MOV-58 Mtr. Oper. Valve HPCI Piping X-226 10,647 18,000 VCW-15AN Check Valve 3,266 18,000 MOV -57 Mtr. Oper. Valve 4,186 18,000 23P-1 Booster Pump 13,453 18,000 MOV -17 Mtr. Oper. Valve 8,279 18,000 VCW-15AN Check Valve 4,376 18,000 MOV-41 Mtr. Oper. Valve Suction Line X-224 13,308 18,000 VCW-15AN Check Valve to RCIC Pump 3,863 18,000 MOV-39 Mtr. Oper. Valve 6,824 18,000 VGW-15AN Gate Valve 5,171 18,000 13P-1 RCIC Pump 3,351 18,000 VCW-15AN Check Valve 2,806 18,000 MOV -18 Mtr. Oper. Valve 2,872 18,000 MOV -36 Mtr. Oper. Valve 79* 18,000 A0V -71 A Air Oper. Valve 5* 18,000 MOV -21 A Mtr. Oper. Valve 14* 18,000 Mark 1 dynamic stress only - values remote from torus - wt, thermal, & seismic not available
TABLE 3-4 (CONTINUED)
FITZPATRICK 5' $ #
5.M PUMP AND VALVE EVALUATION 5.M 3.
8 'i' O Max. Pipe f "'
Component Component TAP TAP Stress at Allowable %'
Designation Type System Penetration Component Pipe Stress B e
MOV-151B Mtr. Oper. Valve RHR Piping X-225B 8,498 18,000 MOV-13B M(r. Oper. Valve 12,663 18,000 10P-3B Pump Suction 11,423 18,000 MOV-13D Mtr. Oper. Valve 8,798 18,000 10P-30 Pump Suction 8,824 18,000 MOV-15D Mtr. Oper. Valve 5,167 18,000 F0V-15B Mtr. Oper. Valve 7,871 18,000 MOV-34A Mtr. Oper. Valve RHR Discharge , X-210A/ 9,400 18,000 MOV-39A Mtr. Oper. Valve Spray Header X-211A 7,190 18,000 4 MOV-26A Mtr. Oper. Valve 6,935 18,000 y VCW-30AN Check Valve 5,136 18,000 14P-1A Core Spray Pump 6,281 18,000 MOV -27A Mtr. Oper. Valve 11,162 18,000 MOV-25A Mtr. Oper. Valve 9,695 18,000 MOV-38A Mtr. Oper. Valve 2,719 18,000 MOV-34B Mtr. Oper. Valve RHR Discharge X-210B/ 13,263 18,000 MOV-39B Mtr. Oper. Valve Spray Header X-211B 6,636 18,000 MOV-26B Mtr. Oper. Valve 12,308 18,000 12" Valve Check Valve 5,136 18,000 14P-1B Core Spray Pump 6,281 18,000 MOV -26B Mtr. Oper. Valve 5,096 18,000 MOV-31B Mtr. Oper. Valve 7,827 18,000 MOV-27B Mtr. Oper. Valve 10,168 18,000 MOV-25B Mtr. Oper. Valve 14,880 18,000 F0V-38B Mtr. Oper. Valve 1,332 18,000 MOV-7 A Mtr. Oper. Valve Core Spray X-227A 5,747 18,000 14P-1A Core Spray Pump Pump Suction 13,163 18,000 VGW-15AN Gate Valve (West Lead) 3,214 18,000 VCW-ISAN Check Valve 2,759 18,000
TABLE 3-5' 5' d nT
+ <
TAP PIPE SUPPORTS T$5
'o Un'
=*
h *'.
Pipe System Support Type of Modified x Penetration Node Designation Support Yes No Type of Modification j 210B/ 240 H10-52A Rigid X 211B 190 PFSK-1949 Snubber X 210 PFSK-1953 Snubber X 715 PFSK-1956 Rigid X Support Reinforced 75 PFSK-2074 Rigid X Clips Added 245 PFSK-2107 Snubber X 780 PFSK-2220 Rigid X 775 PFSK-2225 Rigid X 440 PFSK-2265 Snubber X 445 PFSK-2392 Snubber X 105 PFSK-2437 Rigid X Support Redesigned a,-
m 755 PFSK-2457 Rigid X 160 PFSK-2477 Snubber X 720 PFSK-2487 Rigid X 306 PFSK-2534 Snubber X 251 PFSK-2558 Snubber X 308 PFSK-2535 Snubber X 175 PFSK-2570 Rigid X 205 PFSK-2573 Rigid X Clips Added i 156 PFSK-2042 Snubber X 133/126 PFSK-2047 Snubber X 337 PFSK-2052 Rigid X 1 605 PFSK-2161 Rigid X l 607 PFSK-2582 Rigid X i 610 PFSK-2597 Rigid X 616 PFSK-2548 Rigid X 640 PFSK-2397 Anchor X 275 PFSK-2434 Rigid X I
TABLE 3-5 (CONTINUED) ygg TAP PIPE SUPPORTS hhg F:3
,,Ka Pipe System Support Type of Modified Designation Support Yes No Type of Modification .g Penetration Node O S
X-225A 107 PFSK-2404 Spring X Y Rigid X Support Redesigned 80 PFSK-2470 (0.P. 80) X Support Reinforced 83 PFSK-2471 Snubber (D.P. 83) 330 PFSK-2513 Spring X 325 PFSK-2560 Rigid X Snubber X 81 PFSK-2237 245 H10-7 Spring X 275 H10-7A Soring X 145 H10-8 Spring X H10-13A Spring X $'
400 300 PFSK-624 Spring X 370 PFSK-1855 Spring X Snubber X Support Redesigned 358 PFSK-1971 Support Redesigned 415 PFSK-2053 Rigid X 390 PFSK-2084 Rigid X 335 PFSK-2110 Rigid X 355 PFSK-212if Snubber X Baseplate Stiffened j Rigid X Support Redesigned 50 PFSK-2238 Rigid X Shim Plate Added l 372 PFSK-2285 Snubber X 81 PFSK-2337 Anchor X 420 PFSK-2387 X
X-225B 450 H10-15 -
Rod Replaced With Struts 415 H10-16 Rigid X 320 H10-21A Spring X i
145 H10-23 Spring X 290 H10-22 Spring X l 570 PFSK-770 Rigid X 650 PFSK-1003 Spring X 535 PFSK-1005 Spring X Anchor X 725 PFSK-2302 1
l P
I TABLE 3-5 (CONTINUED)
TAP PIPE SUPPORTS khkw o-M o-n B.
Type of Modified = A 5" Pipe System Support Type of Modification x
Designation Support Yes No Penetration Node to PFSK-1854 Spring X 3" X-2258 85 615 PFSK-1923 Rigid X (continued) Rigid X 675 PFSK-1934 Snubber Replaced by 98 PFSK-1936 Snubber X Greater Capacity Snubber 715 PFSK-1982 Rigid X Rigid X Rigid Support Replaced 95 PFSK-2009 by Strut 365 PFSK.-941 Rigid X 360 PFSK-2020 Spring X Snubber X Replace Base Plate 96 PFSK-2072 and Anchor Bolts 380 PFSK-2077 Spring X 4 o
480 PFSK-2078 Rigid X 386 PFSK-2112 Snubber 435 PFSK-2134 Spring X 590 PFSK-2149 Spring X 530/532 PFSK-2187 Rigid X 488 PFSK-2260 Rigid X 695 PFSK-2281 Rigid X Anchor X 500 PFSK-2387 Snubber X 575 585 PFSK-2456 PFSK-2468 Rigid X 4 690 PFSK-2489 Rigid X Snubber X Support Redesigned 385 PFSK-2567 Support Redesigned l
65 PFSK-2270 Rigid X l
Snubber X X-210A/211A 345 PFSK-2343 625 PFSK-2398 Rigid X X
250 PFSK-2446 Snubber 655 PFSK-2449 Rigid X X
181 PFSK-2502 Snubber Snubber Replaced by 340 PFSK-2509 Snubber X Greater Capacity Snubber
. TABLE 3-5 (CONTINUED)
TAP PIPE SUPPORTS
' y;;j g hb$
. Pipe System Support Type of Modified Type of Modification
[NN" ,
Penetration Node Designation Support Yes No :o i 685 PFSK-2518 Snubber X $
X-210A/211A ;!
(continued) 485 PFSK-2000 Spring X 265 PFSK-2327 Rigid X
-269 PFSK-2317 Rigid X 210 PFSK-1947 Rigid X '
Snubber X 135 PFSK-1952 Snubber X 130/132 PFSK-1984 Snubber X 185 PFSK-2079 Support Reinforced, 65 PFSK-2085 Rigid X Clips Added l
85 PFSK-2128 Rigid X H10-42A Rigid X-170 Spring X &
970 H10-47 7 Snubber X 615 H10-388 800 ~H10-397 Spring X l X 1 255 PFSK-878 Snubber 190 PFSK-944 Rigid X 680 PFSK-1641 Rigid X 875 PFSK-1902 Spring X Snubber X 126/127 PFSK-1940 Snubber X 120/121 PFSK-1944 695 PFSK-2310 Rigid. X Anchor X 285 PFSK-2354 X-202A,F- 30 H27-8 Spring X Stanchion Changed to Struts X-2028,G 115 PFSK-2463 Rigid X Stanchion Changed to Struts 120 PFSK-2280 Rigid X Rigid X Support Redesigned 30 PFSK-1951-110 PFSK-2506 Rigid X l
I
TABLE 3-5 (CONTINUED)__
TAP PIPE SUPPORTS g g.
Pipe System Support Type of Modified EUT Penetration Node Designation Support Yes No Type of Modification "AE
- o X-205 135 BFSK-519 Spring X e Rigid X Removal 4 25 BFSK-695 "
90 BFSK-711 Rigid X BFSK-982 Rigid X Support Redesigned 40 5 BFSK-715 Anchor X BFSK-696 Rigid X Spacers Added 25 PFSK-1914 Rigid X Support Redesigned X-212 255 Support Redesigned 45 PFSk-1919 Snubber X i PFSK-1921 Snubber X Support Redesigned l 45 l 47 PFSK-1049 Spring X 26 PFSK-1963 Rigid PFSK-2384 Rigid X e 215 225 PFSK-2385 Rigid X Support Reinforced y i
New TES Rigid X Support Redesigned X-213A 15 Support 8332 New TES Rigid X Support Redesigned X-213B 15 Support 8333 X-214 205 H23-1 Spring X 190 PFSK-1955 Rigid X 180 PFSK-1958 Rigid X PFSK-1987 Rigid X Support Redesigned 115 PFSK-2223 Rigid X Support Redesigned 190 PFSK-2247 Snubber X Snubber Replaced 90 PFSK-2494 Rigid X Rod Replaced By Strut 75 BFSK-877 Anchor X Support Redesigned X-220 110 New Support Added 60 New TES Lateral X i Support 8362 Snubber New TES Axial Rigid X New Support Added 62 Support 8362
I TABLE 3-5 (CONTINUEul :o a a TAP _ PIPE SUPPORTS Uh J" Yr ?-
OWO Pipe System Support Type of Modified [E>E Penetration Node Designation Support _ Yes No Type of Modification y O
X-224 45 PFSK-840 Rigid X 3 590 PFSK-1055 Rigid X 152 PFSK-2101 Rigid X 315 PFSK-2164 Rigid X 58 PFSK-2218 Rigid X 25 PFSK-2237 Rigid X 550 PFSK-2381 Rigid X 360 PFSK-2383 Rigid X 660 PFSK-2465 Rigid X 177 PFSK-2467 Rigid X Support Made Double-Acting 183 PFSK-2473 Rigid X Support Made Double-Acting 325 PFSK-2481 Anchor X e
295 PFSK-2538 Rigid X 635 PFSK-2546 Rigid X y 395 H10-66A Rigid X l H10-66B Rigid X j 405 153 H13-3 Rigid X 134 H13-4 Spring X 92 H13-19A Rigid X 74 H13-20 Rigid X 52 H13-21 Spring X 172 H13-48 Rigid X -
4 X-226 310 PFSK-1950 Rigid X 485 PFSK-1959 Rigid X 465 PFSK-1995 Rigid X 455 PFSK-2118 Rigid X 305 PFSK-2242 Snubber X 106 PFSK-2248 Snubber X Support Redesigned 65 PFSK-2305 Rigid X Rod Hanger Changed to Strut 300 PFSK-2500 Snubber X 165 H23-30 Spring X 255 H23-31 Spring X 330 H23-33 Rigid X
TABLE 3-5 (CONTINUED)
'FMg TAP PIPE SUPPORTS.
are ro -
Pipe System Support Type of Modified &
Penetration Node ' Designation Support Yes No Type of Modification B A
X-226 350 H23-34 Rigid X (continued) 350 H23-35 Rigid X 375 H23-36 Rigid X 400 H23-37 Rigid X 100 PFSK-983 Rigid X H23-89 Snubber X Support Redesigned 106 X-227A 115 PFSK-2028 Spring X 250 PFSK-2122 Rigid X Support Redesigned 295 PFSK-2325 Anchor X Stanchion Replaced ,
by a Smaller Stanchion g 350 PFSK-2394 Rigid X Support Redesigned i 90 PFSK-2418 Rigid X Support Redesigned 280 PFSK-2508 Rigid X 45 PFSK-2511 Rigid X Support Redesigned 215 H14-8 Spring X 345 H14-28 Rigid X Rod Replaced by Strut 67 H14-40 Spring X X-227B 275 PFSK-2323 Rigid X 250 PFSK-2324 Rigid X 102 PFSK-2454 Rigid X Stanchion Changed to Strut 50 PFSK-2512 Rigid X Support Redesigned 300 PFSK-1994 Anchor X 220 H14-20 Spring X 125 H14-21 Spring X 75 H14-54 Spring X l
Technical Report TR-5321-2 WTri prVNE Revision 1 ,
ENGINEERING SERVICES TABLE 3-6 FITZPATRICK TAP PENETRATION STRESS RESULTS - LARGE BORE PIPING Primary Stress Secondary Stress Penetration Calculated Calculated Number _ _
Max. Stress Allowable Max. Stress Allowable X-202A & F 15,715 19.300 34,742 69,900 X-202B & G 16,200 19,300 22,642 69,900 X-205 12,939 19,300 28,646 69,900 X-210A 12,610 19,300 54,034 69,900 X-210B 12,536 19,300 56,616 69,900 X-211A 12,436 15,100 43,986 69,900 X-2118 12,436 15,100 43,986 69,900 X-212 13,728 15,100 45,137 69,900 X-213A & B 11,774 15,100 34,942 69,900 X-214 11,107 19,300 44,266 69,900 X-220 12,934 19,300 46,621 69,900 X-224 13,877 15,100 41,649 69,900 X-225A 13,828 19,300 58,526 69,900 X-225B 18,577 19,300 49,988 69,900 X-226 13,643 19,300 55,407 69,900 X-227A 13,818 19,300 57,105 69,900 X-227B 13,818 19,300 57,105 69,900 X-228 13,481 15,100 56,982 69,900
!1i g
g!oE mh h n \ ar
= 2,, "
0
,8 0
v 2
,7 0
4
,6 0
L A
C y I N P O 6 Y I
T v
,5 T,
A 0 L L
R T E E ( W N S E F P L 8 O L
( ,4 O
L P E f 0 H M S ) O C R R E F E
P 0 S E P (
,4 S U .
E N 0 MI O L P A T S C E I R P
Y L T 2 L
,3 E l
0 iS 1
3 4 E
,2 R U
0 G I
i F I
6
,1 I
0 8
,0 0
0 5 0 5 5 0 5 1 1 0 0 1 1 k! *se fllll1 l l' I ill l 1ll !
l ll 1 ll j1 l ll 11i l111.
2 0 e. ~ N s6e 5 ~ b' 22n0 j 0
- 5
,5 4
L A
C
,0 I 4 P Y
T N
O I
T
,5 A 3 L L
I C
S O
N 0 O
,3 I N T O A I S T N
_ A Z E R \
H D T N
_ E -
O N
,52 Y C E
P C M N O L E R L U F E Q H E S R E F SN R 0 E ,2 O P P P S U E R
L A L C L I - E P l i
Y i5 S
1
_ T 2
3 E
R U
,0 G 1 I F
,5
~ - _
0 6 5 4 2 j 0
i Gfg "o $*
l I\ \l
\\
4 h $'* . ??> A 4P,$k
- 4=,
,1 E "
3 0
8 2
w l7 l6 4
L A
o 3 56 C I
P Y
T V
R S
5 8 M 4 O
)
R F
N Z 1
O 1
(
E I S l
T Y l
A C 0 R 0 P T
l 4 N S E E E N U R E Q P E L R L L F E L l i
E S H B2 3 S 3 R 3 E
P P E U R
\ U L G A 54 I F
C 2 I
P Y
T B6 1
8 s
5 7
6 5
4 Ge8b p m E 0
3 5
1 sI 0
( !ll
s.
l Technical Report WN NN TR-5321-2 Revision 1
~
PENETRATION NO. X-2'25A & B CONDENSATION SCILLATION AND P0'ST CHUG 2000 -
E i i i l i l i i I
1500 - 1 I l g I ai m i L d :J l E I t
E . 000 3 . - g _
E .
I l
- n t I I
)~ I \
I I
~
500 -
' l I ~
(
is h \
g q
's s _
1
\
\
_ s j
0 l l .l . 1.
10 20 30 40
~
FREQUENCY (HERTZ)
~~ ~
FIGURE 3 DRAG LOAD ON INTERNAL PIPIrio
NeEO- @ $ g AaMTm 's E1EEm
,5 3
s
,0 3
L A
, 5' C 2 I P
A Y E
_R i
T, rT A E C A R
U P 0 T QS 2 ) C H 2, Z E T D ,
P 0 T S R E 5 ,2 R A P 0 E C
E O 2 H L X (
I S E M I V N Y S S N O C I
E A E I N S B T E H A U P G T R Q A N U T E I O E ,'5 R T T S N 1 F
A - E R H P 5 E T -
P R 3 O O E N
R U
G
, I
,0 F 1
l
,5 8
7 6
- L S
~
4 3
- b2 '
O
- g~>5oc n8 5Y&w uM
, i'l' j \ l
Technical Report WN NN TR-5321-2 Revision 1
=
TORUS EXTERNAL PIPE ION FORCE-IIME HISTORY P IhPUT AT PENETRATION
~
//
~
TORUS INTERNAL PIPE TORSION PIPE RIGID 5 DEGREES OF FREEDOM IN OTHER DIRECTIONS FIGURE 3-6 TAP PENETRATION REPRESENTATION (TYPICAL)
l Technical Report WlE.E79E N M TR-5321-2 Revision 1
\
\
b RING GIRDER
/ ?
/ \f x I N _A '
\ a "
, -y ;,
y .
\r i
'\ \
s
\
s
_t
~
si SUPPORT \
COLUMNS FITZPATRICK NUCLEAR PLANT FIGURE 3-7 DETAILED SHELL MODEL
WTE.EMEMM Technical Report ggg.68 TR-5321-2 / ,, -
Revision 1 l~ O O O s
o o CO m e m m e m e is e I * @ 2 se . W. CO.
- m m u 90 , o li 3 90
-I INSIDE- .r i
,sl I g., 4 4 - - -
80
<, I es 1 I 70 a f' I i K
s- -l- - - 60 m I , z
<s', o I -
1 ,b ,I- r zh -
50 xI I $
of,f ,n- f - 9 -
40 3
s- z l x
,'jY- h -tI - - -
30 E h I a g g j 3:f- - :- - h - - -
20~ $ b 1 I ,
w 5 o '
'\ $
I - - - l- i_ _ _ _ 16 m z c-8 mi I
C 0 , <a 360 $ u o
' ';ll- @f u- es- -- - e
=
i 0 $ M g E TOP
Y A g > 'g - - -h--
-. - 350' "o.
N w e ' lo I H m o .s st e :- - - - o e ,y - -
340 u- e w
t I
, 's a w u. I 9
,h- -e ,,- e -4i - - - -
330 0
- t o~,, ei I u n- 320
, , e -4 r - - f,e ,- e- o i
'. 'ez --+-d
, h'-
31 0 s' 300 j ,'f - - *',- - h* * * --
, cre o 7 - 3A' 9 r, '- -
3
290 y' 1 1 I I 280 f k' I N et' i 270 ' "aj '
I 270 OUTSIDE l'p.
1 o o o o No O e m
-' s s in m m @
CO CD IN T rs CT.;.
N. e W CO
'?l:- d u; a
-+-
e
1 l
Technical Report NMM TR-5321-2 Revision 1 o o
& % o m s a-m e
s m cn m 3 m r-
- o. e. m. < e o. m ,cn
- .% ; cd J e m.
I m OUTSIDE . L_. _ _.._. - o 3-
~
470 270a g, ,g,' ,W ,
il- ,M- - -, -j- - - -
260 v , ,ic,- i i t'. 9 _'r,Q _ _ _ J - - - - -
256 U::M'e',
3 Ja - ,+ - - L -- - -
i 240
.:K ' ' i $
_-_ig.- - -M - - 'c s t > -
235~ S r
.: y,' i $1e 8 o
, ,'d- ,2' , A - - ,i ---4>-
220 o, i egi 8 --.
z
,i---->t i -
' Sri 210 <
r
= o.
r z w,' : I g g E r'-A 2 34 - - 'r- - -- -
200 e d
, *i ,-- ii i z o.-. .
I w w r a m e p _- g _ _s# y _ _. r - - - -
190 o x m o a.
d BOTTOM -'T, , 'i
', E .-
d ;5 m g ,->r - '>r - T - - 4>- 180 g a n.
m 180, s 4- i i
--_g g r
s i o i , ,_i _, i% _ _ _ ,i___. -
170 < "
e . ' ' m S t* i i i --
y E 2
.-,i,si -->--
i ,- i
- - - - - 160 i u. e
',% 6 8 . la-
'g s, _ _ _ _ . -
-A ,-$ - g;- - - - 150 ;
Wi I
~
',5 - St - ',- - - - 14D
-k' i i
,i s i
- T - ; , - 130 1 - k ' i T,
-A 3'si g 3 ,%--p -- r -- - - .120- - - - -
f.g I I I.d.'_'5
', .j ..._ ~110 f-k' v' i ~~ i
' i i
i ;v - - - L--- -
-100 i i INSIDE +v. *i i i
,a a ,
o 90 90 l.- 0 O- o--
o o o CO Ch m m cn m m r- e m.
- e. e. m.
- e m m J h
. . . ..929'68 mit o i ra e reo s eis ori63iRb
1 ENN TORUS PENETRATIONS hh(
[b[
p -.
E' N0DE NO. LOCATION PENETRATION NO. ANGLE
}
37 A x220 0.00 73 B x205 0.00 108 C x202a-e 352.20 107 D x211a-b 340.20 106 E x218 330.00 106- F x219 330.00 159 G x200b 320.15 104 H x212 315.00 68 I x210a-b 315.00 33 J x228 315.00 i
L x200c 317.30 5'
-- M x200a 37.60 N x201a-h 49.20 0 x203a 293.008 P x203b 293.00 Q x206a/c 295.00
-- R 206/ 245.00 S x209a/c 293.00 g T x209b/d 293.00 154 U 21 4 273.90 V x216,x217 30.00 W x215 293.00 101 X x221 276.20 155 Y x222 276.20 l 24 Z x226 227.00 l 59 AA x225a-b 223.50 94 BB x224 215.00 23 CC x227a-b 220.00 19 DD x213a-b 180.00 l K x231a-b 320.00 FIGURE 3-8c TAP PENETRnTION LOCATIONS 1
Technical Report Y TR-5321-2 O Revision 1 REFERENCES
- 1. TES Report TR-5321-1, " Mark 1 Containment Program, Plant Unique Analysis of the Torus Suppression Chamber for James A. FitzPatrick Nuclear Power Plant", dated August 11, 1983.
- 2. ASit B&PV Code,Section III, Division 1, through Summer 1977.
- 3. USNRC IE Bulletin 79-02, dated November 8, 1979, (Revision 2), Pipe Support Base Plate Designs using Concrete Expansion Anchor Bolts.
- 4. ASE B&PV Code,Section XI, 1977 Edition, with 1978 Addenda.
- 5. G.E. Report NED0-24583-1, " Mark 1 Containment Program Structural Accept-ance Criteria Plant Unique Analysis Application Guide", dated October, 1979.
'6. Structural Mechanics Report SMA-12101.05-R001, " Design Approach for FSTF Data for Combining Harmonic Amplitudes for Mark 1 Post-Chug Response Calculations", dated May,1982.
- 7. General Electric Coniputer Program RVFOR-04, A Program to Compute ~ SRV Line Clearing Forces, General Electric Company, Scn Jose, Calif.
- 8. Welding Research Council Bulletin No.107, " Local Stresses in Spherical and Cylindrical Shells due to External Loadings", dated March,1979.
J. General Electric Report No. MPR-751 " Mark 1 Containment Program, Aug-mented Class 2/3 Fatigue Evaluation Methud and Results for Typical Torus Attached and SRV Piping Analysis", dated November,1982.
- 10. G.E. Report NED0-218?S, Rev. 2, " Mark 1 Containment Program Load Defini-tion Report", dated November,1981.
r T a Report WM ggg Revision 1 REFERENCES (CONTINUED)
- 11. NRC " Safety Evaluation Report, Mark 1 Containment Long-Term Program",
NUREG-0661, dated July, 1980.
- 12. Structural Mechanics Associaties Report SMA-12101.04-R0020 " Response Factors Appropriate for use with C0 Harmonic Response Combination Design Rules", dated March 1982.
- 13. NYPA Letter No. JPN-83-46, Transmittal to NRC of Vacuum Breaker Analysis Method, dated May 20, 1983.
- 14. General Electric Computer' Program RVR1Z, A Program to Compute SRV Water Reflood, General Electric Company, San Jose, California.
- 15. STAR 0YNE, A general purpose computer program for Structural Analysis, System Developement Corporation, Santa Monica, California.
- 16. STAAD, A Computer Program for Frame-Structure Analysis, Research Engineers, Cherry Hill, New Jersey.
- 17. ASME Power Piping Code, ANSI B31.1-1977 edition.
- 18. USNRC IE Bulletin 79-14, " Seismic Analysis of As-Built Safety-Related Piping Systems", dated July 2,1979.
)
_ ; -- -- w w m w
': 1 Table x m Ma e 50,9-3 STRUCTURAL ACCEPTANCE CRITERIA FOR CLASS 2 AND 3 PIPING SYSTEMS m** wa N +
' O WO o e sp f
+
- Na e N
=
(D I.
SB A + SRV 58A + $RV + EQ "O SRV SRW SRA $8A + EO i EVENT rnMBIM A11(*5 + 18A IBA + EQ ISA + $2V ISA + SRV + EQ DBA DBA + EQ DBA + SRv DBA + EO + SRV [
O CO, CO. PS CO, CO, i* CH CO.CH (1) 04 PS CO.CH P$ CM PS CO.CH CH CO.CH O 5 0 S 0 5 0 5 0 5 0 5 0 $ 0 $ 0 S l TYPf' 0F EARTHQl'AKE 4 6 8 9 10 11 12 13 14 15 16 17 IP 19 20 21 22 23 24 25 26 27 COMP th AT IfW NL'P'P t R 1 2 3 5 7 LDADS X X X X X X X X X X X X E I 1 Normal (2) N X X X X X X X X X X X X
,. I I N X X X X X X X I I Earthquake IQ X X X X X X X X X X X X I I 2 I I I SRV Discharge SRV X X X X X X X X X X X X X X X X X X I I I thermal 7, X 1 X X X X I I I '
X X X X X 1 I 1 I I I I I I t I I X X X X X I X X X X P8Pe Pressure P K I I R X X IDCA Fool Swell P yg I y
ie Lot. A Condensett6n p E X X X X X X X X m gf Oscillation CO E X X X X X X X X X q LOCA Chugging X X I, P,C X
ST91'CTL'RAL ELDtDIT ROW Essential Piptog Systees B B B B B B B B B S S S S with IBA/DBA 10 e a B B 8 Y 8 8 B B B B B B t
(4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4) (4)
S (3) (3) (4) (4) * * * * *
- - - - ~ -
with SRA 11 a a s e e a e e B e s e -
.i (3) (3) (4) (4) (4) (4) (3) ()) (4) (4) (4) (4) i*J EtDf t2f"_1]SL
,e . Pepina Systems D D D D D D D D D D D D D D D D D D With tth/DBA 12 8 C D D D D D D D (5) (5) (5) (5) (5) (5) (5) (5) (5) (5) (5) (5) (5) (5) (5) (5)
(5) (5) (5) (5) (5) (5) (5) (5) (5) (5) ,.
i - - - - - * * *
- With $PA 13 C C D D D D D D D D D D - - -
(5) (5) (5) (5) (5) (5) (5) (5) (5) (5) (5) (5)
S*
1 ' f
' .)
li
)- .
4 I-I
?-
f-2 W Revision 1 EN SENICES NOTES TO TABLE _1_
- 1. Where drywell-to-wetwell pressure differential is normally utilized as a load mitigator, an additional evaluation shall be perfonned without SRV loadings, but assuming the loss of the pressure differential. Service Level D Limits shall apply for all structural elements of the piping system for this evaluation. The analysis need only be accomplished to the extent that integrity up to and including the first pressure boundary isolation valve is demonstrated, including operability of that valve.
If the normal plant operating condition does not employ a drywell-to-wetwell pressure differential, the listed Service Level assignments shall be applicable. '
- 2. Normal loads (N) consist of dead loads (D).
- 3. As an alternative, the 1.2 S h limit in Equation 9 of NC-3652.2 may be replaced by Level C (1.8 S h
) pr vided that all other limits are satis-fied. Fatigue requirements are applicable to all columns with the excep-tion of 16, 18, 19, 22, 24 and 25.
- 4. Footnote 3 applies, except that instead of using Level C (1.8 Sh) I" Equation 9 of NC-3652.2, level D (2.4 S h) may be used.
- 5. Equation 10 of NC or ND-3650 shall be satisfied, except that f atigue requirements are not applicable to columns 16, 18, 19, 22, 24 and 25, since pool swell loadings occur only once. In addition, if operability of an active component is required to ensure containment integrity, operability of that component must be demonstrated.
a Report 97 gg Revision 1 -Al ENGNEERING SERVICES APPENDIX 1 USE OF C0 LOAD FOR St%LL BORE PIPING Experience with large bore piping analysis showed that DBA condensation oscillation was usually the most severe Mark 1 load for torus attached piping. This is consistent with the continuous nature of the C0 load (as opposed to the transient nature of some other Mark 1 loads) and the frequency content of CO, which is in a range of typically high piping response.
Experience on large bore piping for the first four plants completed by TES follows:
No. of Large Bore Systems Available for No. Controlled Evaluation by C0 or Seismic
- FitzPatrick 15 14 Pilgrim 14 11 Millstone 11 9 Vermont Yankee 13 11 53 45 i
Of the eight cases not controlled by C0, C0 loads were very close to the maximum, as follows:
l Ratio of Controlling Stress Case to C0 Case Pilgrim .999, .953, .958 Millstone .fj9, .65(1)
Vermont Yankee .960, .53(2)
FitzPatrick .71(3)
- Evaluation did not include drag loads on internal piping - small bore systems do not have internal piping.
Technical Report Y TR-5321-2 -Al Revision 1 In five of these eight cases, C0 stresses are practically equal to the controlling cases. Of the other three cases (1) and (2) are special cases that do not apply to small bore piping; (3) is also a special case as discussed below.
Case (1) is a atmospheric control (vacuum breaker) line that connects at three points at the top of the torus. The multiple connections and the penetration location make this line particularly susceptible to pool swell impact on the upper shell. There is no comparable small bore system.
Case (2) is an RCIC return line which has a long internal section which is responding at a high level to shell motion. The maximum stress in this line is inside the torus. Small bore systems do not have internal piping, so this does not apply.
Case (3) showed very high seismic stress and was re-supported before Mark 1 loads were applied. Reanalysis produced low stresses for all loads (maximum combined stress was 41% of the allowable). Based on this, we conclude that the evaluation of any similar small bore line would be controlled by seismic, and therefore would be covered by our small bore analysis method.
The decision to limit analysis.of small bore piping to DBA C0 as the only Mark 1 load was based on the foregoing. Seismic, thermal and weight were also
'consi'dered, in addition to DBA C0.
__ __ _ _ . _ ._ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ =
s Technical Report TN TR-5321-2 -A2 ENGNEBUNG SERVICES Revision 1 APPENDIX 2 32 Hz Cutoff for Condensation Oscillation Analysis All condensation oscillation response of TAP systems due to torus shell motion used an input frequency cutoff of 32 Hz.
This practice began early in the TAP analysis work and was the result of a decision to cut off shell response frequencies at 32 Hz during the contain-ment analysis. The 32 Hz cutoff for containment analysis is discussed in Appendix 2 of Reference 1, and was based on the fact that both high input energy and high modal responses occurred below that frequency. Use of the 32 Hz cutoff was shown to produce only a small error that was considered negli-gible. On this same basis, the 32 Hz cutoff was applied to C0 analysis for TAP.
Later in the TAP analysis work, it became evident that the 32 Hz cutoff would not be realistic for post chug; input frequencies to 50 Hz were used for post chug. At this time, the decision to cut off C0 frequencies at 32 Hz was reviewed. Spectra were generated for several penetrations showing the C0 shell motion up to 50 Hz. Figures A4-1, A4-2, A4-3 and A4-4 illustrate typical spectra for rotation and displacement at TAP penetration points for a similar torus, analyzed by TES. These show clearly that shell response above 32 Hz is negligible for C0, and support the initial position.
h
)
k5C M E' E %
yy{ ,D A *
($E O
0 5
4 A, 5 ,0 4
5
, 3 L '
A )
I D 7 A 3 R
- E , 03 E D )
1 S 0 Z N N l i
2- O - (
P L A S L Y E E l . 5 C E R i 2 N R S E U L U G LED E Q
I E F i D l R
S A F O 0 0 L l
. 2 C l U
A (
B D
5
-,1
,0 1
- ,5
, 0 0 0 0 0 0 0 0 4 0 6 2 8 4 2 2 1 1 0 0
- E',_ 55 W t = a N
9 4 l
J
$a3OE $8" l NsM*>k hv Es.E O[
,0 5
5
,4
,0 4
=
,5 t
f 3
0 I
T )
A 7 T 3 '
0 ,0 R E
- D 3 )
E 0 N Z
2 S l -
ll
- t (
2 0 L L A PS E Y
,5 C E E li l R R S 2 E l
F EHDA [ E R
F S OL l
,0 0 t 2 C U (
A B
D
,5 1
,0 1
,5
- - - - - - 0 0 0 0 0 0 0 0 4 0 6 2 8 4 2 2 1 1 0 0 T ? eb' aS ~ E f
j1 ,!l ! ll :l1 l l
9 1
FIGURE A2-3 E 1 a$ a T DBA C0 SHELL RESPONSE-RADIAL yM~n S.
(UNLOADED SHELL-NODE 23) $dd 5'
E, r
3.00 -
2.50 - l
~
O l ,
7 1 2.00 -
=
6
~
& 8 y 1.50 - p d
a c
h 1.00-I
~
0.50 -
A m 0 . . . . . , , .- . .
0 5 10 15 20 25 30 35 40 45 50 FREQUENCY (HZ) l
=
m W
h FIGURE A2-4 5,. ln 9 v wo DBA .C0 SilELL RESP 0fiSE-ROTATION {"{ ~~
(UNLOADED SHELL-fl0DE 23) x e
E!
n 2.40 -
2.00 -
m O
7 1.60 -
= r p 1.20 -
S as
" 0.80 -
g 0.40 -
0 0 5 10 15 20 25 30 35 40 45 FREQUENCY (!!Z) w