ML20206G622
| ML20206G622 | |
| Person / Time | |
|---|---|
| Site: | Millstone  |
| Issue date: | 05/04/1999 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML20206G617 | List: |
| References | |
| NUDOCS 9905100087 | |
| Download: ML20206G622 (24) | |
Text
"
p ut y
t UNITED STATES 1
j NUCLEAR REGULATORY COMMISSION
~
%...../g WASHINGTON, D.C. 20006-0001 SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION REQUEST TO APPLY LEAK-BEFORE-BREAK STATUS TO THE PRESSURIZER SURGE LINE PIPING l
MILLSTONE NUCLEAR POWER STATION. UNIT 2 NORTHEAST NUCLEAR ENERGY COMPANY DOCKET NO. 50-336
1.0 INTRODUCTION
By letter dated November 9,1998, Northeast Nuclear Energy Company (NNECO)
(Reference 1), the licensee for the Millstone Nuclear Power Station, Unit 2 (Millstone 2),
requested that the NRC review and approve its application to remove consideration of the dynamic effects of postulated ruptures of the Millstone 2 pressurizer surge line (PSL) piping from the facility's licensing basis. NNECO's submittal was based on an application of Title 10 of the Code of Federal Regulations, Part 50 (10 CFR 50), Appendix A, Ger,sral Design Criteria 4, which states:
However, dynamic effects associated with postulated pipe ruptures in nuclear power units may be excluded from the design basis when analyses reviewed and approved by the Commission demonstrate that the piobability of fluid system piping rupture is extremely low under conditions consistent with the design basis for the piping.
For the purpose of this demonstration, NNECO submitted leak-before-break (LBB) analyses prepared by Structural Integrity Associates (SIA) for the PSL piping (Reference 1). LBB evaluations developed using the general criteria contained in NUREG-1061, Volume 3, " Report of the U.S. Nuclear Regulatory Commission Piping Review Committee, Evaluation of Potential
- for Pipe Breaks," (Reference 2) and/or Draft Standard Review Plan (DSRP) Section 3.6.3 have been previously approved by the Commission as demonstration of an extremely low probability of piping system rupture.
2.0 REGULATORY REQUIREMENTS AND STAFF POSITIONS Nuclear power plant licensees have, in general, been required to consider the dynamic effects which could result from the rupture of sections of high energy piping (fluid systems that during normal plant operations are at a maximum operating temperature in excess of 200 *F or at a 9905100087 990504 PDR ADOCK 05000336 P
PDR jy
p,
- maximum operating pressure in excess of 275 psig). This requirement has been formally included in 10 CFR 50, Appendix A, General Design Criteria 4 which states, " structures, l
systems, and components important to safety...shall be appropriately protected against
. dynamic effects, including the effects of missiles, pipe whipping, and discharging fluids, that may result from equipment failures and from events and conditions outside the nuclear power unit." For facilities such as Millstone 2, which were licensed prior to the advent of the General i
Design Criteria, these requirements were included as part of plant-specific licensing reviews.
NNECO recently identified a condition at Millstone 2 in which sections of closed-loop piping i
near the unit's American Society for Mechanical Engineers (ASME) Code Class 1 PSL piping ~
j would not be adequately protected from a failure of the PSL. The licensee identified this
- condition to the NRC staff in Licensee Event Report 98-005-00 and per the preceding discussion has addressed the problem by performing an LBB evaluation for the PSL piping.
i The philosophy of " leak-before-break" behavior for high energy piping systems, developed by the NRC in the early 1980s, was used in certain evaluations stemming from Unresolved Safety Issue A 2, " Asymmetric Blowdown Loads on PWR Primary Systems," and was subsequently expanded for applications resolving issues regarding defined dynamic effects from high-energy
. piping system ruptures. The general criteria developed by the NRC for performing LBB analyses were thoroughly detailed in NUREG-1061, Volume 3 and summarized in Draft Standard Review Plan Section 3.6.3," Leak-Before Break Evaluation Procedures,"which was published for public comment in August 1987.
3.0 LICENSEE'S DETERMINATION The following discussion contains information supplied by NNECO in its November 9,1998, letter to the NRC and the attachments to that letter. These attachments included a report prepared by SIA for NNECO: SIR-98-096, Rev. O, " Pressurizer Surge Line, Leak-Before-Break
- Evaluation, Millstone Nuclear Power Station, Unit 2." In addition, another report prepared by SIA for NNECO: SIR-99-021, Rev. O, was submitted to provide additional information in the licensee's February 26,1999, response to the staff's request for additional information (RAI).
The licensee also provided additional information in a letter dated March 5,1999, in response to a second staff RAI.
3.1 Identification of Analyzed Piping and Piping Material Properties NNECO's submittal identified and analyzed the PSL from its connection to the reactor coolant system hot leg to the pressurizer. This piping is shown in Figure 1.
The PSL was identified as having the following material components. The nozzle material connecting the PSL to the reactor coolant system hot leg was a low-alloy cart >on steel forging, American Society for Testing and Materials Specification (ASTM) A-105, Grade 11. The nozzle at the pressurizer was fabricated from an ASTM A-508 Class 2 forging. Centrifugally-cast stainless steel (SS) safe ends manufactured from ASTM A-351, Grade CF-8M materials were attached to each nozzle forging by bimetallic welds fabricated by the shielded metal arc welding (SMAW) process using Inconel 182 filler metal. The remainder of the piping was made from ASTM A-351. Grade CF-8M centrifugally-cast SS piping spool pieces and the piping welds were fabricated using SS filler materials and SMAW processes.
4 i I For the material properties used in the PSL LBB evaluations, NNECO/SIA used consistent sets of stress-strain and J-resistance (J-R) curve information based on the material being evaluated
- at a particular location (carbon steel, cast SS, or SS weld metal). For the carbon steel nozzles, the licensee referenced J R and stress strain curve data from the generic characterizations in 1
the EPRI Ductile Fracture Handbook (Reference 3). Since several generic carbon steel characterizations are provided in Reference 2, NNECO/SIA examined available certified material test report (CMTR) information (Charpy test data, chemical composition) to support the generic characterizations selected. NNECO/SIA also developed a method to adjust these properties for temperature since the generic data were given for 550 *F, while the conditions in the surge line during normal operation were between 600 *F and 650 *F. This adjustment was based on the consideration of the change in the ASME Code minimum values for these material properties between 550 *F and 650 *F.
For the centrifugally-cast SS safe end and piping spool pieces, NNECO/SIA proposed a more detailed analysis of the material properties. Cast SS materials are subject to thermal aging (embrittlement) at reactor operating temperatures. This embrittlement degrades the fracture toughness and increases the yield and ultimate strength (stress-strain properties) of the j
materials. The rate and severity of these changes are expected to be a function of the g
chemical composition of the cast SS (amount of 5-ferrite, molybdenum content), the casting l
process, and the time and temperature of service. The NNECO/SIA analysis utilized the work I
s in References 4 and 5 to estimate the fracture toughness and tensile properties of the surge
]
' line cast SS materials. The NNECO/SIA methodology examined the chemical composition
)
given for each cast SS spool piece in the CMTRs and developed a corresponding J R curve i
and stress-strain property representation for each spool piece. NNECO/SIA took the
)
Ramberg-Osgood parameters for the stress strain curve in the aged condition from Reference 5 and obtained the yield and ultimate strengths from the CMTRs for each spool piece. In addition, in response to the staff's request for the licensee to analyze severe thermal stratification transients (which may occur with the PSL at other than normal,100 percent power operating temperature), NNECO/SIA developed a methodology for performing temperature adjustments to the cited tensile properties. When analyzing a nodallocation (generally associated with a specific weld location, see Figure 1), NNECO/SIA used most conservative properties from the spool pieces to either side of the weld in the fracture mechanics analyses, except at nodes 14 and 15, where the licensee considered explicitly material on either side of the weld.
Like the cast SS materials, SS weld material is also known to be subject to thermal aging. The
'NNECO/SIA analysis used stress-strain properties at 550 *F from Reference 3 for the SMAW weldments and would adjust those properties based on the variation of the ASME Code minimums with temperature, if necessary For the fracture toughness, the lower bound curve provided in Reference 6 for thermally aged SMAW materials was used. These stress strain and fracture toughness properties were also conservatively assumed to apply to the inconel i
182 welds at either nozzle-to-safe end location.
l l
1
~
t 4
4 3.2 General Aspects of the Licensee's LBB Analysis The ana4ses provided by NNECO sought to address four principal areas which were consistent with the criteria established for LBB analysis acceptability in NUREG-1061, Vol. 3 or DSRP Section 3.6.3 and extended the analysis to account for conditions of particular interest when evaluating the surge line analysis. These conditions are the thermal stratification loads in the PSL during normal operation (NOP) which may affect the leakage size flaw, and the severe thermal stratification loads generated in the PSL during heatup and cooldown transients which may affect the critical flaw size. It should be noted that the licensee's original submittal only addressed the explicit loading conditions addressed in NUREG-1061, Vol. 3 and that consideration of the effect of the thermal stratification loads on the LBB analysis was incorporated in their February 26,1999, response to the staff's RAl. The original submittal did consider the impact of the thermal stratification loads on the fatigue analysis by citing the results of CEOG Topical Report CEN-387-P, Rev.1-P-A (Reference 7), which was developed in response to NRC Generic Letter 88-11 (Reference 8). However, the licensee's assessment of the loads due to thermal stratification for the fracture mechanics evaluations were based on a linearization of the top-to-bottom thermal gradient in the piping cross-section, with a correction factor of approximately 4 percent on the calculated loads to account for nonlinear effects. This approach will be addressed further and contrasted with a more conservative staff position nn the determination of thermal stratification loads, the approach approved in the staff's review of CEN-387-P, Rev 1 P-A, when the NRC staff analysis is discussed in Section 4.0.
The NNECO/SIA approach was as follows. First, demonstrate that the subject piping is a candidate for LBB analysis by showing that the piping is not particularly susceptible to active degradation mechanisms or atypicalloading events. Second, establish the critical through-wall flaw size under which analyzed locations would be expected to fait under NOP plus safe-shutdown earthquake (SSE) loading conditions or under extreme thermal stratification transients during heatup and cooldown operations (which as a group will be referred to herein as " critical flaw loading conditions"), whichever is most limiting. Third, establish the leakage size flaw under NOP loads for each location. For this application the leakage size flaw was that flaw which provided 10 gallons per minute (gpm) of reactor coolant system leakage based on the Millstone 2 leakage detection system capability of 1 gpm and a required margin of 10.
Fourth, evaluate the margin between the critical flaw size and the appropriate leakage flaw size and evaluate the stability of the leakage flaw under loading conditions of v'2 times the limiting critical flaw loading conditions.
3.3 Evaluation of Pressurizer Surae Line Pipino SIA prepared the analysis of the PSL piping that was submitted to the staff as Attachment 3 to the November 9,1998, letten was prepared for the licensee by SIA as report number SIR 096, Rev. 0.. The NNECO/SIA assessment of the impact of thermal stratification concems was submitted to the staff as Attachment 2 (SIA report number SIR 99-021, Rev. 0) to the licensee's February 26,1999, RAI response. This section summarizes the results of the NNECO/SIA results for the four subject areas noted in Section 3.2 above and incorporates information from both SIA reports.
t
. Initially, NNECO's submittal addressed the issue of potential piping degradation mechanisms and stypicalloading conditions. Per the discussion of the limitations of LBB analyses in NUREG-1061, Volume 3, the LBB approach should not be considered when operating experience has indicated particular susceptibility to failure from the effects of corrosion, water hammer, or fatigue. NNECO's submittal concluded that the PSL piping at Millstone 2 has not been shown to be particularly susceptible to the effects of water hammer, intergranular stress corrosion cracking, or flow-assisted corrosion. NNECO included a fatigue analysis which concluded that the PSL is not subject to sufficient loading cycles to cause significant growth of postulated cracks.
The licensing basis fatigue analysis for the Millstone 2 surge line was performed as part of the response by the CEOG to NRC Bulletin 88-11 (Reference 8). In response to this bulletin, the CEOG performed an ASME Section 111 bounding fatigue analysis to demonstrate that the design of the surge lines in CEOG plants complied with the licensing design basis for CEOG plants, considering the additional loading due to stratification of the water in the surge line during various plant operating conditions. The results of this analysis were described in Reference 7. The CEOG calculated a bounding cumulative usage factor (CUF) of 0.938 for all CE plants, including Millstone 2. The staff considers a CUF less than the ASME Code design limit of 1.0 as acceptable to demonstrate that piping is an appropriate candidate for LBB evaluation.
Next, the NNECO/SIA analysis evalu:.ted the PSL piping by developing the applied loads under NOP plus SSE loading and the loads which occurred under specific bounding, heatup, and cooldown transients. The loads submitted by the licensee for NOP (which include piping deadweight, thermal expansion loads, and the contribution of the thermal stratification during NOP, but exclude the intemal pressure) are given in Table 1. The Operating Basis Earthquake (OBE) loads (which must be multiplied by a factor of two to establish the SSE loads) are given in Table 2. The loads generated for five bounding thermal stratification conditions are given in Tables 3 through 7. Note that although the licensee submitted the information in Tables 1 through 7 for all nodes, only a limited number (including bounding nodes 2 and 3) are included in the tables for comparison. In addition, not all of the thermal stratification cases for which information was submitted were explicitly evaluated since the three which were evaluated were sufficient to ensure the performance of a bounding analysis. For the summations, the loads were added algebraically. The stratification loading conditions chosen for the snalysis were based on the licensee's review of facility operational experience and procedures. These conditions have also been used by the licensee to delineate an operational curve, which limits the pressurizer-to-hot leg temperature gradient. The licensee committed in their February 26, 1999, letter to incorporate this operational limit curve (shown in Figure 2) into facility operating procedures to support LBB approval.
For the purposes of LBB analyses, the critical flaw size can be defined as the longest preexisting through-wall flaw which could exist without growing unstable to double-ended pipe rupture under the limiting critical flaw loading conditions. The analysis performed by SIA was based on the J-integral / Tearing Modulus (elastic-plastic fracture mechanics) approach to flaw stability which is applicable for the materials of most interest in this analysis. Formally, piping I
r-w 6-failure is predicted when the applied J exceeds J,e (the mateial property value at which crack growth initiates) and the rate of increase of the applied J with crack extension (dJ/da) exceeds the rate of increase of the piping material's J-R curve with crack extension (d(J R)/da).
The analysis in SIR-98-096, Rev. O and SIR-99-021, Rev. O calculated the critical flaw size by using SIA's pc-CRACK code. To do this, SIA first determined the tensile and bending stresses from the applied loads (including pressure) at nodallocations. The analysis then assumed that the stresses applied at an analyzed location were all tensile stresses and determined a entical flaw size (a) under these conditions. Then it was assumed that the i
stresses applied were all bending stresses and determined a critical flaw size (a,) under such conditions. A linear interpolation was then performed between the two results as:
a, = a, * (o, / (o, + oi)) + a, * (o, / (o, + o ))
i where o, and a were the bending and tensile components, respectively, of the overall stress i
and a, was the combined critical flaw size. It should be noted that while the licensee's original analysis for the critical flaw sizes addressed the NOP plus SSE critical flaw loading condition for all nodal locations, the information submitted later regarding critical flaw sizes under stratification conditions only examined the nodallocations determined by the staff and the licensee to be potentially controlling based on the prior information. Table 8 summarizes the licensee's results for the two nodal locations (2 and 3) that were subsequently determined to be bounding. Node 2 was, however, identified as being associated with the entire hot-leg nozzle safe-end location and was further subdivided into locations 2a,2b, and 2c, and the material regions present at each of these locations, as shown in Figure 3. The column headings of " Case 1", " Case 2", and " Case 3" refer to the stratification loading cases shown in Tables 3,4, and 5, respectively. The results of the original NNECO/SIA critical flaw size 1
analysis based on the NOP plus SSE loads is provided in the final column.
NUREG-1061, Vol. 3 also requires that the acceptable leakage flaw size be demonstrated to be stable under loads which are equivalent to /2 times the limiting critical flaw loading condition. Using the methodology outlined above, the analyses then determined what size flaw would be stable under such conditions and compared it to the leakage flaw size discussed below. Table 9 summarizes this information for node 2 and provides the flaw size which was stable at each materiallocation under /2 times the loads from the loading cases noted in the paragraph above. The licensee noted that this demonstration for limiting aged cast SS material at node 2 showed that the criteria for a margin of two between the critical-to-leakage flaw size would be the controlling criteria for the node / material combinations of interest and that additional calculations regarding the stability criteria for other nodes or materials were not necessary.
The NNECO/SIA analysis then determined the leakage flaw size under NOP conditions. Again in this case, NOP loads also included the loads due to thermal stratification at normal,100 percent power operation. The loads provided by the licensee for these conditions were given
E 4,
in Table 1. The leakage flaw analysis then performed by SIA was based on the use of the Pipe Crack Evaluation Program (PICEP, Revision 1) computer code developed by the Electric Power Research Institute (EPRI) for calculating two-phase flow through cracks in light-water reactor piping. By entering the piping cross-section description, material property characteristics for the SS SMAW material, the cast SS piping, or the carbon steel nozzle and the loads for the appropriate node into the program, the NNECO/SIA analysis determined the flaw size which generated 10 gpm of leakage. As stated in Section 3.2, the 10 gpm value was selected based upon the Millstone 2 containment leakage detection system capability of detecting 1 gpm of leakage in the course of 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> (consistent with NRC Regulatory Guide 1.45, " Reactor Coolant Pressure Boundary Leakage Detection Systems," guidance) and a i
margin on leakage of 10 as required by the methodology outlined in NUREG-1061, Vol. 3.
The leakage flaw sizes calculated by NNECO/SIA for each of the materiallocations at bounding nodes 2 and 3 are given in Table 10.
The conclusions of the licensee's analysis are summarized in Table 11 for bounding nodes 2 and 3. To reiterate, to establish that the PSL qualifies for LBB approval, ths licensee sought to demonstrate that the critical flaw size in column 2 was a factor of 2 greater than the leakage flaw size in column 3, and that the flaw size (given in column 4) which was stable under loads
/2 times the limiting critical flaw loading condition, was equal to or greater than the leakage flaw size in column 3. Again, the " stability" flaw size in column 4 was not provided for all node / material combinations f.ince the result given for the limiting cast SS material demonstrate j
adequate margin between the leakage and " stability" flaw sizes. It should be noted that although in three cases the critical-to-leakage flaw size ratio was just less (1.97,1.98,1.99, highlighted in bold in Table 11) than the nominal factor of two, the licensee concluded that this was sufficient to establish the PSL for LBB approval.
4.0 STAFF EVALUATION Based on the information provided by the licensee regarding the materials comprising the Millstone 2 PSL piping and its loads under NOP, SSE, and severe thermal stratification.
conditions, the staff independently assessed the compliance of the PSL piping with the LBB criteria established in NUREG-1061 Vol. 3. While the staff has determined that the conclusions of the licensee's analyses were correct in that LBB behavior would be expected for the PSL piping, the following sections will focus on the differences between the details of the staff's analysis, conducted per NUREG-1061, Vol. 3, and the licensee's.
4.1 Identification of Analyzed Piping and Piping Material Properties The staff examined the list of materials identified for the PSL and concluded that the materials of primary interest for the LBB analysis would be the cast SS piping and safe ends or the SMAW materials because all were susceptible to thermal aging. Based on the staff's decision (discussed below) to use generic, bounding aged material properties for each material type (cast SS or SMAW metal), initial calculations demonstrated that for a given loading set, the assessment of the cast SS material at any node would bound the assessment of the SMAW material. Therefore, the staff analyzed the behavior of the cast SS materials at the locations of interest and only those calculations will be included in this discussion.
!=
i l
. NUREG-1061, Vol. 3 specifies particular aspects which should be considered when developing materials property data for LBB analyses. First, data from the testing of the plant-specific piping materials is preferred. However, in the absence of such data, more generic data from the testing of samples that possess the same material specification may be used. More specifically, it was noted in Appendix A of the NUREG that "[m]aterial resistance to ductile crack extension should be based on a reasonable lower-bound estimate of the material's J-resistance curve," while Section 5.2 of the NUREG stated that the materials data should include, " appropriate toughness and tensile data, long-term effects such as thermal aging and i
other limitations."
Appendix A to the SIA report SIR-98-096, Rev. O did provide chemical composition and tensile data for the cast SS piping spool pieces and safe ends in the unaged condition. It is the staff's position that the use of yield and ultimate tensile strengths based on unaged CMTR data is not, in general, acceptable given the requirement to account for thermal aging effects. This decision is supported by the data presented in Reference 5, which shows significant changes in the
' tensile behavior of cast SS after aging. Therefore, the staff used the tensile properties shown in Table 12, which have been used in previous staff analyses and were developed from Reference
- 5. The Ramberg-Osgood parameters (a and n) cited by the licensee for their analysis also came from Reference 5 and were not significantly different than the values used in prior staff analyses. Therefore, the NRC staff continued to use Ramberg-Osgood parameters from prior staff analyses for cast SS (shown in Table 13 for 554 *F). The staff, however, did use the temperature adjustment methodology given in the licensee's February 26,1999, submittal to modify the cast SS tensile properties in Table 12 when analyzing off-normal thermal stratification transients.
The NRC staff also continued to use generic, bounding, J-R curve data (in this case, for centrifugally-cast,5-ferrite >15 percent, CF-8M cast SS) from Reference 3 in lieu of the spool piece-by-spoo! piece representations given by the licensee. The basis for this decision was that recent discussions with the author of NUREG/CR-4513 (Reference 4) during the assessment of cast SS aging issues for the Calvert Cliffs license renewal have indicated that the use of the correlations in NUREG/CR-4513 may be limited to material pieces with not greater than 25 percent 5-ferrite (the limit of the database used to develop the correlations). Since the critical j
90 degree elbow spool piece (denoted as C-4620-4, see Figure 1) between nodes 2 and 3 was reported by the licensee to have a calculated 5-ferrite content between 26.4 and 28.4 percent, the staff sought additional data to verify the use of an appropriate J-R curve. Reference 9 provided this additional data (specimen EK, tested at 608 'F after aging at 617 *F for 30,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br />) for a material with a calculated value of 28.0 percent ferrite content using Hull's equivalent factors. The staff determined that use of the bounding generic representations at room temperature and 554 *F for centrifugally-cast CF-8M SS from Reference 4 was consistent with, or conservative when compared to, the data from Reference 9. Note, however, that this review has not determined the adequacy or inadequacy of the methodology or spool piece-by-spool piece J-R curve representations submitted by NNECO/SIA since they were not required to perform the NRC staff's independent assessment.
1
.g.
4.2 General Aspects of the Staffs LBB Analysis The staff's analysis was performed in accordance with the guidance provided in NUREG-1061, Vol. 3.- Based on the information submitted by the licensee, the staff determined the critical flaw size at the potential bounding PSL locations using the codes compiled in the NRC's Pipe Fracture Encyclopedia Reference 10. For the purposes of the staffs evaluation, the critical location was defined by those locations at which materials with low postulated fracture toughness existed in combination with high ratios of stratification or SSE-to-NOP stresses.
This is because high stratification or SSE stresses tend to reduce the allowable critical flaw size while low NOP stresses increase the size of the leakage flaw required to produce 10 gpm of leakage. In particular, when evaluating the critical flaw in the thermally aged CSS base materials, the staff used the LBB.ENG2 code developed by Brust and Gilles (Reference 11).
The staff also used the LBB.ENG3 code developed by Battelle (Reference 11) to evaluate the SS SMAW material when the initial scoping calculations were performed to show that the SMAW material would not be limiting when compared to the aged cast SS materials. The LBB.ENG3 methodology is_significantly different from the other codes in Reference 10 and from the licensee's analysis in that LBB.ENG3 explicitly accounts for the differences in the stress-strain properties'of the weld and adjoining base material when determining the effective energy release from the structure with crack extension. The same criteria as discussed in
. Sections 3.3 and 3.4 with regard to the applied J exceeding the material J,c and the applied dJ/da exceeding the material's d(J-R)/da were used to identify the critical crack size.
The staff then compared the critical flaw at the bounding location to the leakage flaw which provided 10 gpm of leakage under NOP conditions to determine whether the margin of two defined in NUREG-1061, Vol. 3 was achieved. The leakage flaw size calculation was carried out using the same PICEP analytic code used by the licensee, with a difference in the assumed crack surface roughness. The 10 gpm value was defined by noting that the i
compliance of the Millstone 2 containment leakage detection system with the positions in Regulatory Guide 1.45 indicates that this system would be able to detect a 1 gpm leak in the course of 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> and a factor of 10 is applied to this 1 gpm detection capability to account for thermohydraulic uncertainties in calculating the leakage through small cracks. The stability of the leakage flaw under loadings a factor of v'2 greater than the limiting " critical flaw loading conditions" was subsequently evaluated to check the final acceptance criteria of NUREG-1061, Vol. 3.
4.3 NRC Staff Evaluation of the Millstone Unit 2 Pressurizer Surge Line The staff initially reviewed the licensee's information for addressing the acceptability of the PSL piping based on the absence of active oe.1radation mechanisms or atypicalloading conditions. The staff concurred with the licens te's conclusion regarding the absence of operating experience which would indicate a particular susceptibility of the PSL to corrosion damage or water hammer events.
e
_ The staff also concurred with the licensee's evaluation of the susceptibility of the PSL to thermal fatigue damage, based on its review of CEN-387-P, Rev.1-P-A. The staff and its consultant, the Brookhaven National Laboratory, reviewed this report and found in its safety evaluation (References 7 and 12) that the CEOG analysis adequately demonstrated that the bounding surge line and nozzles meet ASME Code Section lll stress and fatigue requirements for the 40-year design life of CEOG plants, including the effects of thermal stratification and thermal striping phenomena.' Fatigue is therefore not considered a significant contributor to the potential for pipe rupture of the surge line.
Appendix B of Reference 1 provided the intamal forces and moments at various sections or i
nodes of the PSL. These moments were determined from a finite element analysis of the PSL
- using the program ANSYS, subjected to dead weight, surge line thermal expansion, RCS thermal expansion movement, thermal stratification, and OBE seismic loading. The thermal
. intamal moments were combined in various loading combinations with dead weight to provide resultant square-root-of-sum-of-squares (SRSS) resultant momants at the selected sections.
The licensee provided the resultant moments for various load cases in its letter of February 26, 1999.
Based on the SRSS moments provided by the licensee, the staff concluded that the cast SS piping at nodes 2,3,13, or 16 of the piping model would be limiting for the PSL piping evaluation. However, the NRC staff did not fully concur with the licensee's methodology for calculating the bending moments due to thermal stratification. The staff concluded that the methodology used by the licensee, which was based on a linear top-to-bottom temperature gradient with about a 4 percent correction to account for non-linearity, may be non-conservative. Altematively, the NRC staff chose to use a more conservative step-wise temperature gradient to calculate the stratification temperature moments, consistent with the methodology in CEN-387-P, Rev.1 P A that the staff had approved for the resolution of Bulletin 88-11. As a result, the NRC staff determined that the thermal stratification contribution to the SRSS moments calculated by the licensee should be increased by approximately 22 percent, for the casar showa in Tables 1,3,4,5,6, and 7. The earthquake loads in Table 2 are unaffected. The arresponding loads used in the NRC staff analysis are given in Tables 13 through 19.
Similar to the NNECO/SIA analysis, the staff first calculated the minimum critical flaw size for each critical flaw loading case at the potentially controlling nodes (2, 3,13,16) by using the
- staff's material property data (temperature-corrected as necessary) and load data. The results for stratification cases 1 through 3 and the NOP plus SSE loads are given in Table 20.
The staff then calculated the flaw size which would be stable at loads equal to /2 times the loads from each critical flaw loading condition. These " stable flaw sizes" are given for each critical flaw loading case / node combination in Table 21. When comparing the licensee's and staff's results (data from Table 8 and 9 to data from Table 20 and 21, respectively), the smaller (more conservative) critical and " stable" flaw sizes resulting from the staff calculations are expected due to the staff's use of greater loads for the thermal stratification loading cases, particularly the limiting Case 1. Again, these " stable flaw sizes" must be shown to be equivalent to the calculated leakage flaw size under NOP conditions (see below) and the critical flaw sizes in Table 20 must be shown to be twice as large as the leakage flaw sizes.
I
~
4
. The staff then used the PICEP code to determine the leakage flaw size for nodes 2,3,13, and
- 16. Using the surface roughness value that the staff has used in previous LBB evaluations of E = 0.003 inch, the staff determined that the 10 gpm of leakage flaws for each node would be as shown in Table 22. It should be noted that the similarity between the NNECO/SIA results in Table 10 and the staff's results in Table 22 must be attributed to a number of factors. The material tensile properties used in each analysis were somewhat different, as was the surface roughness, and these differences would have been expected to result in the leakage flaw calculated by the staff to be larger than that calculated by the licensee. However, the increase in the NOP thermal stratification stresses used by the staff (because of the step-wise top to-bottom temperature gradient) offset these material property differences (higher NOP stresses resulting in a larger cracls opening area and a correspondingly small flaw size for 10 gpm of leakage) and the final results were, therefore, comparable.
The results of the staff's analysis are then shown in Table 23 for the limiting cast SS at nodes 2,3,13, and 16, and these may be compared to the licensee's results in Table 11. Whereas the licensee's analysis resulted in a minimum critical-to-leakage flaw size ratio of 1.97 (at the node 2b location when using the stratification case 1 loads) and acceptable margins on leakage flaw size stability for all cases, the staff's results indicate a minimum critical-to-leakage flaw size ratio of 1.84 (at the node 3 location when using the stratification case 1 loads) and
- acceptable margins on leakage flaw size stability for all cases. However, in previous LBB evaluations, the staff has concluded that margins of slightly less than two on the critical-to-leakage flaw size are acceptable provided that a full margin of 10 is maintained on the leakage uncertainty. It is the staff's position that relaxation from the guidance written in 1984 on this point is acceptable based on the work which has been completed in the areas of piping fracture (e.g. the Intemational Piping Integrity Research Group (IPIRG) work) and the evaluation of minimum material properties to more appropriately bound the behavior of primary system piping materials. Therefore, the staff has concluded that the above evaluation confirms the licensee's conclusion that the analyzed portions of the Millstone 2 PSL piping will exhibit LBB behavior.
5.0 CONCLUSION
Based on the information and analysis supplied by the licensee, the staff was able to independently assess the LBB status of the analyzed portions of the Millstone 2 PSL piping.
The staff has concluded that: (1) for this analysis, the loading conditions which determined the critical flaw size were based on the initial heatup stratification loads (case 1); (2) that because of thermal aging effects, the cast SS piping would be the limiting material at each nodal location, and; (3) based on thesc considerations, it has been demonstrated that the Millstone 2 PSL will exhibit LBB behavior. Furthermore, the licensee should be permitted to credit this conclusion for eliminating the dynamic effects associated with the postulated rupture of these sections of piping from the Millstone 2 facility licensing basis, consistent with the provisions of 10 CFR 50, Appendix A, General Design Criteria 4.
l Principal Contributors: M. Mitchell M. Hartzman Date:
May 4,_1999 l
4,.
6.0 REFERENCES
- 1. Letter of November 9,1998, from M. L. Bowling, Jr., Northeast Nuclear Energy Company, to the NRC Document Control Desk, with attachments.
- 2. NUREG-1061, Volume 3," Report of the U.S. Nuclear Regulatory Commission Piping Review Committee, Evaluation of Potential for Pipe Breaks," November 1984.
- 3. EPRI Report No. NP-6301 D, " Ductile Fracture Handbook," June 1989.
- 4. Chopra, O.K.," Estimation of Fracture Toughness of Cast Stainless Steels during Thermal Aging in LWR Systems," NUREG/CR-4513, ANL-93/22, Rev.1.
5.' Michaud, W.F., et al., " Tensile-Property Characterization of Thermally Aged Cast Stainless Steels," NUREG/CR-6142, ANL-93/35.
- 6. Gavenda, D.J., et al.," Effects of Thermal Aging on Fracture Toughness and Charpy-impact Strength of Stainless Steel Pipe Welds," NUREG/CR-6428, ANL-95/47.
- 7. CEN-387 P, Rev 1-P-A," Pressurizer Surge Line Flow Stratification Evaluation,"
May 1994.
8.-
USNRC Bulletin 88-11, " Pressurizer Surge Line Thermal Stratification",
December 20,1988.
- 9. Jayet-Gendrot, S., Ould, P., Meylogan, T., " Fracture Toughness Assessment of in-Service Aged Primary Circuit Elbows Using Mini C(T) Specimens Taken From Outer Skin," PVP-Vol. 304, Fatigue and Fracture Mechanics in Pressure Vessels and Piping, ASME 1996.
- 10. Pipe Fracture Encyclopedia, produced on CD-ROM by Battelle Columbus Laboratory for the U.S. Nuclear Regulatory Commission,1997.
- 11. Brust, F.W., et. al., " Assessment of Short Through-Wall Circumferential Cracks in Pipes,"
NUREG/CR-6235, BMi-2179.
- 12. Letter of July 14,1993, from J. W. Shea, NRC, to R. F. Burski, Combustion Engineering Owners Group, with enclosed Safety Evaluation.
l
r 1
t
. 1 Table 1:
Loads Calculated by NNECO/SIA for Normal Operational Conditions (including Deadweight, Thermal Expansion, and Thermal Stratification Loads) l Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-3.377 536.6 3
-20.126 496.8 16
-16.055 918.2 13
-16.711 1378.8
- Sign convention is that negative axial forces are compressive. Intemal pressure is not included.
Table 2:
Loads Calculated by NNECO/SIA due to Operating Basis Earthquake Seismic Conditions (Safe Shutdown Earthquaka Loads Defined to be Two Times the OBE Loads)
Node Axial Force (kips)
Total Bending Moment (in-kips) 2 4.0 277.6 3
4.0 277.6 l
16 32 125.7 13 7.2 192.7 Table 3:
Initial Heatup Stratification Loads Calculated by NNECO/SIA (Case 1) i (Pressurizer Temp.= 410*F, Hot Leg Temp.= 70*F, System Pressure = 262 psig) l Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-0.556 2445.6 3
-6.800 2330.8 l
16
-6.373 2370.9 l
1 13
-6.825 2645.1
- Sign convention is that negative axial forces are compressive. Intemal pressure is not included.
l l
1
\\
. Table 4:
Intermediate Pressure Stratification Loads Calculated by NNECO/SIA (Case 2)
(Pressurizer Temp.= 576*F, Hot Leg Temp.= 334*F, System Pressure = 1270 psig)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-1.806 1811.5 3
-14.310 1596.1 16
-12.020 1833.7 13
-12.634 2150.0
- Sign convention is that negative axial forces are compressive. Intemal pressure is not included.
Table 5:
Initial Cooldown Stratification Loads Calculated by NNECO/SIA (Case 3)
(Pressurizer Temp = 653*F, Hot Leg Temp.= 483*F, System Pressure = 2235 psig)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-2.564 1338.8 3
-18.211 1100.6 16
-14.894 1469.9 13
-15.577 1847,8
- Sign convention is that negative axial forces are compressive. Internal pressure is not included.
Table 6:
Intermediate Temperature Stratification Loads Calculated by NNECO/SIA (Case 4)
(Pressurizer Temp.= 610*F, Hot Leg Temp.= 400*F, System Pressure = 1645 psig)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-2.142 1600.2 3
-16.036 1370.3 16
-13.291 1664.5 13
-13.935 2002.1
- Sign convention is that negative axial forces are compressive. Internal pressure is not f
included.
Table 7:
' Intermediate Temperature Stratification Loads Calculated by NNECO/SIA (Case 5) 1
i
- )
(Pressurizer Temp. = 495'F, Hot Leg Temp. = 200*F, System Pressure = 635 psig)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-1.161 2156.0 3
-10.567 1986.6 l
16 9.215 2119.0 13
-9.751 2405.6
- Sign convention is that negative axial forces are compressive. Intemal pressure is not included.
Table 8:
Licensee Determination of Critical Flaw Sizes, Node 2 and 3 Materials Node Materiel Critical Flaw Critical Flaw Critical Flaw Critical Flaw Size No.
Stratification Stratification Stratification NOP + SSE Case 1 Case 2 Case 3 Loads 2a A105 Grade il 17.44 17.90 17.76 19.68 2a SMAW 18.56 18.36 18.22 20.16 2b SMAW 16.00 15.82 15.76 17.44 j
2b Cast SS 13.90 14.00 14.28 16.08 2c Cast SS Piece 1 12p 12.90 13.10 14.96*
2c SMAW 15.26 15.00 14.84 16.56 2c Cast SS Piece 2 13.60 13.54 13.66 14.96*
3 Cast SS Piece 1 14.20 14.80 15.14 14.10*
3 Cast SS Piece 2 13.40 14.18 14.66 14.10*
- Only one value given for the critical flaw size under NOP + SSE loads since licensee's original analysis for this case in SIR-98-096, Rev. O was based on using a bounding assessment for the cast SS at each unique location. SMAW material at Node 3 was not reanalyzed under stratification conditions since it had been demonstrated previously (based on the node 2 data) that the cast SS material would be bounding.
Table 9:
Licensee Determination of Stable Flaw Sizes under (/2
- Critical Flaw Loading Conditions), Node 2 Cast SS Material
. Table 9:
Licensee Determination of Stable Flaw Sizes under (/2
- Critical Flaw Loading Conditions), Node 2 Cast SS Material Node Material Stable Flaw Stable Flaw Stable Flaw Stable Flaw No.
Size Size Size Size Stratification Stratification Stratification NOP + SSE Case 1 Case 2 Case 3 Loads 2b Limiting
'10.318 10.560 10.936 12.90 j
Cast SS 1
1 2c Limiting 9.229 9.354 9.631 11.62 Cast SS 1
)
Table 10:
Licensee Determaination of Leakage Flaw Sizes Under Normal Operating Conditions, Nodes 2 and 3 Node No, Material 10 GPM Leakage Flaw Size (inches) 2a A105 Grade ll 8.26 2a SMAW 8.33 2b SMAW 6.97 2b Cast SS 7.06 2c Cast SS Piece 1 6.48 2c SMAW 6.40 2c Cast SS Piece 2 6.49 3
Cast SS Piece 1 6.74 3
Cast SS Piece 2 6.74
- SMAW material at Node 3 was not included since it had been demonstrated previously that the cast SS material would be bounding.
. Table 11:
Comparison of Licensee Analysis of Leakage and Critical Flaw Sizes, Nodes 2 and 3
Node Material Critical Flaw from Leakage Flaw Flaw Size Stable No.
Limiting Load Case from Normal under(2 times Operation Loading Limiting Critical Flaw Conditions Load Case 2a A105 Grade il 17.44 8.26 N. A.*
2a SMAW 18.22 8.33 N. A.*
2b SMAW 15.76 6.97 N.A.*
2b Cast SS 13.90 7.06 10.318 2c Cast SS Piece 1 12.82 6.48 9.229 2c SMAW 14.84 6.40 N. A.*
2c Cast SS Piece 2 13.54 6.49 9.229 j
3 Cast SS Piece 1 14.10 6.74 N. A.*
3 Cast SS Piece 2 13.40 6.74 N. A.*
- Again, the " stability" flaw size in column 4 was not provided for all nodo/ material combinations since the result given for the limiting cast SS material demonstrate adequate margin between the leakage and " stability" flaw sizes.
Table 12:
Parameters used in Staff Evaluation of Millstone 2 Aged Centrifugally-Cast SS Parameter Value at Room Temperature Value at Operatina Temperatures (70*F)
(550'F to 650*F)
Young's Modulus 28500 ksi 25500 ksi Yield Strength 52.6 ksi 32.8 ksi Ultimate Tensile Strength 79.8 ksi 72.8 ksi Sigma-zero 56.6 ksi 32.8 ksi Epsilon-zero 0.00185 0.00129 Ramberg-Osgood Alpha 2.10 1.276 Ramberg-Osgood n 5.50 6.60 2
2 C
2474 in-lbs. / in 3035 in lbs. / in n
0.33 0.31 Note: J = C(aa)".
- Table 13:
Loads at Calculated by the NRC Staff for Normal Operational Conditions (including Deadweight, Thermal Expansion, and Thermal Stratification Loads)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-3.377 604.4 3
-20.126 520.5 16
-16.055 960.8 13
-16.711 1420.5
- Sign convention is that negative axial forces are compressive. Internal pressure is not included.
Table 14:
Loads Calculated by the NRC Staff due to Operating Basis Earthquake Seismic Conditions (Safe Shutdown Earthquake Loads Defined to be Two Times the OBE Loads)
Node Axial Force (kips)
Total Bending Moment (in-kips) 2 4.0 277.6 3
4.0 277.6 16 3.2 125.7 13 7.2 192.7 Table 15:
Initial Heatup Stratification Loads Calculated by the NRC Staff (Case 1)
(Pressurizer Temp.= 410*F, Hot Leg Temp.= 70*F, System Pressure = 262 psig)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-0.556 2990.9 3
-6.808 2864.5 16
-6.373 2887.9 13
-6.825 3276.5
- Sign convention is that negative axial forces are compressive. Internal pressure is not included.
1
, Table 16:
Intermediate Pressure Stratification Loads Calculated by the NRC Staff (Case 2)
(Pressurizer Temp.= 576*F, Hot Leg Temp.= 334'F, System Pressure = 1270 psig)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-1.806 2197.0 3
-14.310 1967.9 16
-12.020 2189.6 13 12.634 2542.9
- Sign convention is that negative axial forces are compressive. Intemal pressure is not included.
Table 17:
Initial Cooldown Stratification Loads Calculated by the NRC Staff (Case 3)
(Pressurizer Temp = 653*F, Hot Leg Temp.= 483*F, System Pressure = 2235 psig)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-2.564 1606.2 3
18.211 1347.8 16
-14.894 1703.3 13
-15.577 2096.0
- Sign convention is that negative axial forces are compressive. Internal pressure is not included.
Table 18:
Intermediate Temperature Stratification Loads Calculated by NRC Staff (Case 4)
(Pressurizer Temp.= 610'F, Hot Leg Temp.= 400*F, System Pressure = 1645 psig)
Node Axial Force (kips)*
Total Bending Moment (in-kips) 2
-2.142 1933.4 3
-16.036 1687.9 16
-13.291 1966.6 13
-13.935 2331.2
- Sign convention is that negative axial forces are compressive. Intemal pressure is not included.
l 1 Table 19:
Intermediate Temperature Stratification Loads Calculated by NRC Staff (Case 5)
(Pressurizer Temp. = 495'F, Hot Leg Temp. = 200*F, System Pressure = 635 psig)
Node Axial Force (kipt)*
Total Bending Moment (in-kips) 2
--1.161 2628.0 3
-10.567 2446.7 16
-9.215 2563.1 13
-9.751 2902.7
- Sign convention is that negative axial forces are compressive. Internal pressure is not included.
Table 20:
NRC Staff Determination of Critical Flaw Sizes for Cast SS, Nodes 2,3,13, and 16 Node Material
- Critical Flaw Critical Flaw Critical Flaw Critical Flaw Size No.
Stratification Stratification Stratification NOP + SSE Loads Case 1 Case 2 Case 3 2b Cast SS 13.1 13.7 14.3 16.0 2c Cast SS 12.2 12.8 13.4 15.1 3
Cast SS 12.6 13.7 14.6 15.7 16 Cast SS 12.5 13.0 13.3 15.0 13 Cast SS 11.3 11.8 12.0 13.0
- It was previously determined that the aged cast SS would be limiting when compared to either the carbon steel forging material or the aged SS SMAW material at any given location. Based on a comparison of the loads at the various nodes, only those shown above could be limiting in the NRC staff's analysis.
i
Table 21:
NRC Staff Determination of Stable Flaw Sizes under (/2
- Critical Flaw Loading Conditions), Node 2,3,13, and 16, Cast SS Material Node Material Stable Flaw Size Stable Flaw Size Stable Flaw Size Stable Flaw Size No.
Stratification Stratification Stratification NOP + SSE Case 1 Case 2 Case 3 Loads 2b Cast SS 9.4 10.6 11.6 13.5 2c Cast SS 7.7 8.7 10.2 12.6 3
Cast SS 8.2 10.4 11.9 13.0 16 Cast SS 8.1 9.1 10.0 12.4 13 Cast SS 6.8 7.4 7.8 9.1 Table 22:
NRC Staff Determination of Leakage Flaw Sizes Under Normal Operating Conditions, Cast SS, Nodes 2,313, and 16 Node No.
Material 10 GPM Leakage Flaw Size (inches) 2b Cast SS 6.97 2c Cast SS 6.32 3
Cast SS 6.84 16 Cast SS 6.00 13 Cast SS 5.19
- SMAW materials were not included since it had been demonstrated previously that the cast SS material would be bounding.
Table 23:
Result and Comparison of NRC Staff Analysis of Leakage and Critical Flaw Sizes, Nodes 2,3,13, and 16 Node Material Critical Flaw from Leakage Flaw Flaw Size Stable No.
Limiting Load Case from Normal under(2 times Operation Loading Limiting Critical Flow Conditions Load Case 2b Cast SS 13.1 6.97 9.4 2c Cast SS 12.2 6.32 7.7 3
Cast SS 12.6 6.84 8.2 16 Cast SS 12.5 6.00 8.1 13 Cast SS 11.3 5.19 6.8
~
1
-E-l
$0615-(60$ 02)
C-4620 3 (50W2)
($0s.03)
Ic 4a2o.41 Ic 4e25 il 3
(605-13)
IC-461611 (505 06)
EC 4410 3l Met g
4
_ (40443)
[C-32231RI (505 04 1)
(Sos 04)
[Cas20 21 (C-4618.tl v
V (50504*
(50$-08) r IC 4416 ti 10 4ets.4l%
O* g O wao. w.nn.r.
t i spow r c.u r
Ic se2 i O u.i.n. iia.ni,r r
==.
Figure 1 Surge Line Piping Materials and Node Identification h.
1 l
l
\\
l l
~~<- --..
_y_
h R
e
~
p s
\\
h
' E) y 8a o
jT 3
8
}
u
\\
a u.
a g
e, n
h
,E y
N E
Ht s-g 6-F w
a u
e z c
8 nz R
E n.
E k
.E
=
3
=2 s
\\
e
\\
E 8
b a
l d 'aanneJadwalJazpnsssJd
0 o
.?
l HOT LEG N0ZZLE 00 = 13.5" SMAW Wald 00 = 12.75" C-46181 SA-105 Cast Stainless 2a 2b 2C O
i
- 1.6875"J
'L.- t = 1.3125" t=
48" suosn Figure 3 Locations for Evaluatic.i at Hot Leg Nozzio 4
e
- r**=*n.
e =......,,,,,,, _,
,,,