ML18010B171
| ML18010B171 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 08/06/1991 |
| From: | James Shea Office of Nuclear Reactor Regulation |
| To: | Walsh L PUBLIC SERVICE CO. OF NEW HAMPSHIRE |
| References | |
| IEB-88-011, IEB-88-11, NUDOCS 9108220209 | |
| Download: ML18010B171 (29) | |
Text
UNITED STATES NUCLEAR REGULATORY COMMISSION WASHINGTON. O.C. 3I555 August 6, 1991 Hr. Lawrence A. Walsh, Chairman Westinghouse Owners Group New Hampshire Yankee P.O.
Box 300 Seabrook Statiorr
- Seabrook, New Hampshire C3874
Dear Yr. Walsh:
SUBJECT:
AUDIT OF WESTINGHOUSE ON GENERIC DETAILED ANALYSIS OF PRESSURIZER SURGE LItlE THERMAL STRATIFICATION (NRC BULLETIN 88-11)
The NRC staff and its consultant, Brookhaven National Laboratory, audited Westinghouse on November 8 and 9, 1990.
The purpose was to review the Westinghouse Owners Group (WOG) report "Pressurizer Surge Line Thermal
,Stratification Generic Detailed Analysis" (WCAP-12369).
The MOG report provided ASt1E stress and fatigue evaluations for participating Mestinghouse plants to address NRC Bulletin 88-11 Action Item 1.d, which requires licensees to update their stress analyses to ensure code compliance.
During the audit, representatives of Westinghouse and the WOG gave a presenta-tion on the MOG program methods and results, and provided responses to the audit team's comments and questions.
The audit team selected and reviewed a number of calculations and documents.
A copy'f the audit report is forwarded as an Enclosure.
Based on the results of the audit, the staff's questions and concerns on the WOG report were resolved.
The methodology used to analyze and evaluate, the stress and fatigue effects due to thermal striping and thermal stratification was found to be acceptable.
However, for the forty-three plants included in the WOG generic anaTysis, only fifteen plants demonstrated acceptability in stress and fatigue usage in the surge 'line.
The remaining twenty-eight Westinghouse plants will require plant-specific analyses.
In addition, all licensees must evaluate pipe supports and combined piping loads other than just the thermal stratification and striping:loads to verify that their resultant stresses are in compliance with the ASIDE Section III Code.
Hr. Lawrence A. Walsh August 6, 1991 A copy of the staff's Safety Evaluation on MOG Report WCAP-12369 is forwarded as Attachment 3 to the Enclosure for your information.
Sine
- ely,
Enclosure:
Audit Trip Report cc w/enclosur e:
. Yir. Steven Tritch Manager of Engineering Technologies Westinghouse Electric Cor poration P.O.
Box 355 Pittsburgh, Pennsylvania 15230
'eph W. Shea, Lead Project Manager PA X-all Project Directorate I-2 Division of Reactor Projects I/II Office of Nuclear Reactor Regulation
ENCLOSURE PURpOSE:
Audit of Westinghouse Owner's Group (WOG)
Pressurizer Surge Line Thermal Stratification Generic Detailed Analysis to Address NRC Bulletin 88 11 Issues LOCATION: Westinghouse Electric Corporation, Pittsburgh, pA DATES:
November 8
& 9, 1990 NRC PERSONNEL:
WOG PERSONNEL:
S.
Hou (NRC),
G. DeGrassi (BNL)
G. Kammerdeiner, D. Roarty, T.H. Liu, and others (See Attachment 1)
The staff of the Mechanical Engineering Branch and its consultant from Brookhaven National Laboratory conducted an audit of Westinghouse.
The purpose of this audit was to complete the review and evaluation of the Westinghouse Owner's Group (WOG) generic detailed analysis of pressurizer surge line thermal stratification.
The WOG program for generic detailed analysis was implemented in June, 1989.
The program provided ASNE stress and fatigue evaluations for participating Westinghouse plants to address NRC Bulletin 88-11 Action 1.d, which requires updating their stress and fatigue analyses to ensure code compliance.
The methods and results of this program were summarized in WCAP-12639 dated
- June, 1990.
This report was reviewed by the staff and its consultant and a list of comments and questions was prepared and submitted to WOG in advance for further discussion (see Attachment
- 2).
During the audit, representatives of WOG and Westinghouse gave a presentation on the WOG program methods and xesults and provided responses to all our questions on the report.
In addition, the audit team selected and reviewed a
number of calculations and reports.
A list of audit participants is included in Attachment 1.
Highlights of the audit are summarized below.
A safety evaluation report on the WOG generic analysis was prepared and is included in Attachment 3.
1 ~
WO D RESU The WOG generic analysis was applicable to 43 of the 55 Westinghousq'lants.
The remaining 12 plants had performed plant specific analysis.
In order to reduce the analytical efforts, the 43 plants were grouped based on their similarity of'urge line response to thermal stratification.
The most important parameters considered in the grouping process were piping layout and support
design configuration, axial temperature profile distribution and thermal transients from plant operation.
As a result of this effort, the 43 plants were reduced to 17 analysis groups.
The major elements of the WOG program included the following activities.
o collection and interpretation of surge line monitoring data from 21 plants.
o Update of existing design transients to account for thermal stratification based on review of plant monitoring
- data, operating records, and operating procedures.
o Development of thermal striping transients.
o Global piping analysis of the 17 plant groups for thermal stratification.
o Local stress analysis to account for the nonlinear stratification temperature gradients.
o ASME Code stress and fatigue analysis to evaluate the effects of stratification and striping for the 17 plant groups.
For the 43 plants analyzed by the WOG report, the results of the analysis demonstrated acceptable ASME equation 12 stress and fatigue usage for 15 plants.
However, Licensees of these 15 plants must also perform additional work to verify the applicability of the analysis and perform additional evaluations which were not part of the WOG program, such as evaluation of support
- loads, pipe displacements, integral welded attachments, pressurizer
- nozzle, etc.
Plant specific analysis of the remaining 28 plants will be needed to demonstrate acceptability.
It is anticipated that some of these plants will require modification.
The WOG generic analysis results also support the technical basis for the existing JCO (justification for continued operation) which indicated acceptability of continued operation for an additional ten heatup and cooldown cycles for all but 4 of the 43 plants in the evaluation.
For those 4 plants, additional technical basis of JCO was prepared.
II.
WOG DO V
W STAFF U
Westinghouse provided written responses to the NRC questions on WCAP-12639 (Attachment 2).
Since they were provided as a draft proprietary
- document, they are not included in this report.
Westinghouse will formally transmit proprietary and non-proprietary versions of the responses to NRC shortly.
The audit team reviewed the responses and found them acceptable.
Some of the
responses were discussed in depth and Westinghouse was asked to provide documentation for detailed review.
The results of these discussions and, document reviews are summarized below:
IX.1 0
0 e Westinghouse was asked to provide documentation describing the technical basis for the JCO's for the 4 plants whose original JCO was not supported by the generic analysis.
In the new JCO's Westinghouse identified conservatisms in the generic analysis such as system bT, earthquake
- cycles, and profile definition.
Based on plant specific
- data, Westinghouse performed an assessment and concluded that all four plants should be acceptable for at least 10 additional heatup/cooldown cycles.
The audit team found this approach acceptable to justify continued operation.
II.2 F t' u
s:
Westinghouse was asked to provide sample fatigue results for review.
Westinghouse uses a computer program which automatically computes the fatigue usage for all pairs of load sets corresponding to the heatup/cooldown, normal and upset transients.
The sample outputs gave the audit team information on the transient load sets which provide the greatest contribution to the cumulative fatigue usage factor.
II.3 S ste s
bu o
As part of the process to update design transients, Westinghouse used historical operating records from 10 plants to determine a distribution of maximum system 4T for the design life.
The audit team questioned the adequacy of this approach since it is based on averaged versus bounded data.
Westinghouse provided the system aT data from the 10 plants and the system nT distribution used in the calculation of the design transients.
The audit team reviewed this information and found that the distribution was actually more conservative than the distribution shown in the table on page 4-5 of WCAP-12639.
The distribution bounded most of the 10 plants in the group.
A few plants had more severe distributions but data for these plants were limited.
Taking into consideration the various other conservatisms,
- the, methodology was judged to be acceptable.
II.4 t
o St at'n order to perform local thermal stress analysis of surge line piping, Westinghouse first performed heat transfer analyses to determine'emperature distributions.
The audit team reviewed the calculations to define the heat transfer film coefficient used in the finite element analysis.
The calculation used classical formulas for forced and free convection.
Appropriate ranges of
q ~ ~
~
temperatures and fluid velocities were considered.
The audit team found the approach acceptable.
ZZ.S Westinghouse used the transfer function method to calculate thermal stress.
The method was described in WCAP-12315, The author of that document provided a description of the theoretical basis and discussed the applicability of the method with the audit team.
Benchmarking problems to verify the accuracy of the method were included in the report.
The audit team found the methodology acceptable but asked Westinghouse to submit the WCAP report to the staff for further review.
ZZ ~ 6 t
a s
Various elements of the thermal striping analysis method were reviewed by the audit team.
Westinghouse explained how both the striping finite element model and the STRFAT2 program vere used to calculate the striping fatigue usage. factor.
Westinghouse also explained hov the heat transfer film coefficient was calculated and pointed out the conservatisms in the calculation.
The technical basis for Figure 9-2 of WCAP-12639, "Attenuation of Thermal Striping Potential by Molecular Conduction" was revieved in detail.
The curve was calculated from a classical conduction heat transfer formula.
The assumptions and model were judged to be conservative for all anticipated flow conditions in the surge line.
- Overall, the audit team found the methodology used to evaluate fatigue usage due to striping to be acceptable.
ZZ.7 W
e The audit team reviewed a copy of the WOG pressurizer surge line operator interview questionnaire.
The purpose of the questionnaire vas to obtain information that vould assist Westinghouse in defining the operational characteristics of PWR plants.
It was intended to determine such characteristics as maximum system hT allowed by administrative procedures, average system aT during normal heatups/cooldowns, frequency of plant operations such as RCP starts/stops, pressurizer operations and boron concentration equalizations.
The audit team found the questionnaire to be comprehensive and detailed.
Data collected from this effort provides additional confidence in the design transients used in the analysis.
ZZZ.
0 FO Zn order to satisfy the requirements of NRC Bulletin 88-11 action l.d, licensees of'he 15 plants shown acceptable by the WOG generic analysis must individually perform a
plant specific applicability program to update stress reports and verify
~
II V
4
~',
V I
)
applicability of the generic analysis to the specific plant.
Such additional evaluations were not included as part of the wOG program.
Specific activities to be performed include the following:
Review past operating records to verify that system zT limits assumed in the analysis were not exceeded.
Verify operational methods to ensure that they are consistent with the methods assumed in the analysis.
Perform a plant specific global piping analysis to determine and evaluate the adequacy of support loads and pipe displacements.
Verify the applicability of the assumed seismic OBE bending moments and check the allowable resultant moments for combined deadweight and OBE at the hot leg nozzle safe end weld.
IV.
Evaluate the effects of stratification on stress and fatigue at integral welded attachments (lugs, pi,ates, etch')
~
Evaluate the effects of stratification on stress and fatigue of the pressurizer nozzle.
As a result of this audit, all questions and concerns on WCAP-12639 were resolved.
The methodology used to ana],yze.and, evaluate the stress and fatigue effects due to thermal stratification and thermal striping was found acceptable.
For the 43 plants included in the WOG analysis, the results demonstrated acceptable ASME Section ZIZ"equation '12 stress and fatigue usage in the surge line and reactor coolant loop nozzle for 15 plants.
.However., additional analysis for these 15, plants are needed,.to.,verify applgcability of the 'generic WOG. analysis and update stress reports for each individual plant (see ZII above).
The remaining 28 Westinghouse plants. that could not'be shown acceptable by the generic detailed analysis..will require plant specific analysis to demonstrate acceptability.
Zt is anticipated that some of the plants can be shown acceptable by removing some of the conservatisms inherent in the generic approach.
. Zt is likely, however, that some plants will require, mcipifieations..
'our of the 28 plants which could not be shown, acgoptable by the generic detailed analysis were also unacceptable for continued operation for ten additional heatup/cooldown cycles.
For those four plants, additional plant-specific information.for justifying continued operation was presented by Westinghouse and was found acceptable by the staff and its consultant.
E
~
~
ATTACHMENT 1 Audit Meeting Attendees
- pl' P~
WOG/NRC MEETING - 11/8/90 Sam S.
Palusamy T H. Tiu Brad Maurer Jenifer Santander Riyad Qashu Rajni Patel Giuliano DeGrassi Shou-nien Hou David Roarty Greg Kammerdeiner Mike Davis Hovard Sandner SCE SCE BNL-NRC NRC DLC/WOG Duke Power
~Q f, g~ 4 s'il Il
~
g
+0
~
AQ E
e
~ ~ gal I.V
~
~
~ \\4 gf
WOG/NRC Meeting - 11/09/90 1 ~
2 ~
David H. Roarty Shou-nien Hou NRC 3.
Giuliano DeGrassi BNL/NRC 4.
5.
6.
7.
8.
9.
Zenifer Santander Mike Davis T.H. Liu Sam S.
Palusamy Rajni Patel Brad Maurer SCE Duke Power 10.
C.Y. Yang ll.
R.L. Brice-Nash 12.
E;L. Crawford 13.
R.
Qashu 14.
Howard Sandner SCE
Attachment 2
NRC/BNL Questions on WCAP-12639
~
~
stoso W
1 ~
2 ~
3 ~
4 ~
The generic detailed analysis demonstrated acceptable ASME Section III Equation (12) stress and fatigue usage for 15 out of 43 plants.
Please identify the 15 plants which were shown acceptable, the 28 plants which have not yet been shown acceptable, and the 12 plants which were qualified by plant specific analysis.
For each plant, provide the calculated equation (12) stress and the fatigue usage factor based on the most current analysis.
Explain why the previous justification for continued operation still applies to those plants which were not qualified by the generic analysis.
Provide a
description and schedule for completion of the plant specific analyses to be performed.
The generic detailed analysis doe's not support the conclusions of the existing JCO for four plants.
Identify these plants and provide additional )ustification for continued operation.
What specif ic instructions (in addition to WCAP-12639) are being provided to individual Licensees to demonstrate applicability of the generic analysis to their plant, update their analysis and perform additional evaluations if needed.
Provide examples.
Will all Licensees be required to update their analysis of record for the surge line?
Hov vill differences in the Code of Record be reconciled?
3 0
0 1 ~
Provide additional information on the correlation of measured pipe OD temperature 'to fluid temperature distribution.
How closely does the measured aT at the pipe OD match the fluid aT inside the pipe?
To what degree of accuracy can the measurements predict the vertical fluid temperature distri-bution including the hot-to-cold interface depth.
How are the uncertainties accounted for in the stress analysis?
Provide examples'
~
2 ~
Describe the basis for selection of the five hot-to-cold interface levels shown in Figure 3-4 to define axial stratifi-cation profiles along the length of a particular surge line.
Were the selection criteria confirmed by measurements?
4.0 2 ~
To what extent was Plant monitoring data used to confirm the normal and upset stratification transient data Presented in Table 4-1?
considering the relatively low dTs for the normal and upset transients listed in Table 4-1 (compared to the heatup/
cooldown transients in Table 4-2),
did any of the events significantly contribute to fatigue usage?
Could a dT cutoff be defined below which the thermal stresses are less than the endurance limit?
3 ~
4 ~
5.
The distribution of system 4T ranges presented in Section 4.5 was based on a review of historical records from 10 plants.
While the data may be representative for the sample of ten plants, it may not be representative for a single plant within the group.
For example, certain plants within the sample may have had consistently higher aT ranges than others because of differences in operating practices.
Provide additional justification to demonstrate that the system aT distribution is representative and conservative for any plant in the MCIG program.
Was the detailed data reduction described in Section 4.6 and summarized in Tables 4-3 through 4-6 performed for each of the ten plants?
Did the bounding distribution use this type of information from all ten plants for each mode of operation?
Please explain how data from different modes 'of operation was factored into the development of Table 4-2 data.
Were different hT values used for each mode?
6.
7 ~
8.
Section 4.6 states that a cooldown contains less than half of the cycles of a heatup and therefore the number of cycles for heatups were multiplied by 1.5 to reflect both heatup and cooldown.
Were the temperature ranges of the'ooldown"cycles shown to be bounded by the temperature ranges of the heatup cycles?
h Identify the plant which indicated significantly " higher stratification cycles at the nozzle as stated in Section 4.6.
What geometric effect was,judged to cause this'?
Identify the plants with significantly higher cycles associated with performing venting operations duri6cf"heatup as stated in Section 4.6.
F
~
~
Table 4-2 shows fewer total nozzle transients in the nozzle than in the pipe.
This is attributed to turbulent mixing which occurs at the nozzle when the reactor coolant pump is operating.
- However, even when the pump is operating, stratification does occur in the pipe and the global bending will induce nozzle'stresses.
How are these stresses accounted for?
10..
Are striping transients associated only with heatup and cooldown? If so, explain why striping does not occur during normal or upset transients.
5 0
1 ~
Please identify each plant associated with the plant. numbers in. Table 5-1.
2 ~
Describe the criteria used to define the enveloping support/
restraint configuration within a subgroup.
Provide examples to illustrate.
6 0
G
~
~
1 ~
How willthe potential for exceeding snubber and spring hanger travel ranges be checked?
What specific information and instructions will WOG provide to the individual licensees?
The analysis of a representative surge line with enveloped supports willnot provide displacements and support: loads that can be used for design purposes.
How will the individual plants be able to verify support adequacy and potential interferences with whip restraints or other adjacent structures?
3 ~
4 ~
7.
Do the temperatures presented in Table 6-. 1 represent fluid or metal temperatures?
Are fluid and metal temperatuies assumed to be equal in,this 'analysis?
Identify the plants listed in Table 6-2.
d 1 ~
Provide a,brief description of the heat transfer analysis performed to determine local thermal stresses in the piping and hot leg nozzles.
Were only steady state conditions considered'?
Considering the variations in fluid'elocities and temperatures, how were conservative values of film coefficients arrived at?
~
~
0
~
4,
9.
3 ~
4 ~
The ASME Code, Section
- ZZZ, 1986 Edition was used in the analysis.
Since all surge lines were originally designed to earlier Code editions or to other piping codes, will a code reconciliation be performed for each plant?
Provide a description of the "transfer function method,"
an example of its application,. and a copy of Reference 3.
How will the assumed envelope of OBE moments be verified?
If the thermal striping stress intensity and peak stress range was calculated from a 2-D finite element analysis using the model shown in Figure 9-1, please clarify why and how 1-D heat transfer analysis stresses from the computer program "STRFAT2" vere used.
The paper by Fugimoto, et al., "Experimental Study of Striping at the Interface of Thermal Stratification" suggests that the surface film coefficient in the interface region may be as much as seven times the nominal value.
What impact would this have on the results?
6.
Describe the methodology and significant assumptions used in developing Figure 9-2.
Was a flow rate of 90 gpm assumed?
Would the curve change significantly at different flow rates?
Provide the basis for the assumed OBE moments summarized in the table in Section 9.3.1.
8 ~
9 ~
Please clarify the requirements for equation 13 qualifi-cations.
Xt appears that the 15-plants which were shown acceptable for equation 12 and maximum usage factor must still be checked for meeting equation 13 as part of the plant specific evaluation.
Moments of the hot leg nozzle must be compared to'the allowable moments in Table 9-5.
Are all other components of the surge line qualified to equation 13?
Are the additional plant specific evaluations for the 28 plants that have not been qualified to equation 12 or fatigue usage being performed as part of the WOG program?
How will these results be reported and what is the schedule for completionP b'
I
~ ~
A Please explain how the permanent plant temperature
'sensor mounted in the surge line (TE 450) can be used to determine maximum fluid i0.
)
~
~
~
~
~
2 ~
3 ~
4.
Is each plant reQuired to perform a plant specific global piping analysis?
What specific instructions will WOG provide to each utility?
Section 10.4 refers to moments tabulated in section g.2.1.
Shouldn't it refer to the table in section 9.3.1?
The pressurizer nozzle evaluation is outside of the scope of the HOG program.
Have any preliminary evaluations been performed to ensure that the pressurizer nozzle is not a
concern?
5 ~
Will the results of the plant specific detailed analyses for those plants not shown acceptable under the generic analysis be reported in a future WOG report?
>> ~
~
r>>
~ >>
"~
~
k>>
I
>> w'
~
~
ATTACHNENT 3 Safety Evaluation Report
+ 'lq ~
I p
s ~
)
~
~
SAFETY EVALUATION REPORT ON THE WESTINGHOUSE OWNERS GROUP PRESSURIZER SURGE LINE THE~ STRATIFICATION GENERIC DETAILED ANALYSIS WCAP-12639 1.0 iNTRODUCTION NRC Bulletin No.
88-11
. requested all PWR licensees to establish and implement a program to confirm pressurizer surge line integrity in view of the occurrence of thermal stratification and inform the staff of the actions taken to resolve this issue.
Licensees of operating PWR's were requested to take the following actions:
Action l.a-Action 1.b-Action 1.c-Action 1.d-Perform a visual inspection walkdown (ASME Section XI, VT-3) at the first available cold shutdown which exceeds seven days.
Perform a
plant speci fic or generic bounding analysis to demonstrate that the surge line meets applicable design codes and other FSAR and regulatory commitments for the design life of the plant.
The analysis is requested within four months for plants in operation over ten years and within one year for plants in operation less than ten years.
If the analysis does not demonstrate compliance with these requirements, submit a
justification for continued operation (JCO) and implement actions 1.c and 1.d below.
Obtain data on thermal stratification, thermal stx'iping, and line deflections
. either by plant unspecific monitoring or through collective. efforts
--among plants with a similar surge line design.
If through collective efforts, demonstrate'similarity in geometry and operation.
k Perform detailed stress and fatigue analyses of the surge line to ensure compliance with applicable code requirements incorporating any observations from 1.a.
The analysis should be based on the applicable
$18nt specific or referenced data and should be completed within two yeaxs.
If the detailed analysis is unable to show compliance, submit a JCO and a
description of corrective :actions.
for effecting long term resolution.
~ If/ II
'I
~
~
~
Although not required by the
- Bulletin, licensees were encouraged to work collectively to address the technical concerns associated with this issue.
In response, the Westinghouse owners Group (WoG) implemented two programs to address the issue of surge line stratification in Westinghouse plants.
In the first program, a generic bounding evaluation was performed to satisfy Bulletin Action 1.b.
Since the results of this evaluation provided less than full design life verification, this report was used later by each of the woG plants as technical basis for justifying continued operation (JCO).
Based on the evaluation, it was deemed acceptable for all WOG 'plants to continue power operation for at least ten
- additional heatup/cooldown cycles.
The bounding analysis methods and results were summarized in a Westinghouse topical report, WCAP-
- 12277, which was submitted to the staff in June 1989.
The staff reviewed the report and concluded that the bounding evaluation provided a sound technical basis for justifying continued operation until completion of the Bulletin requested action l.d by the end of 1990.
This report, combined with acceptable plant specific visual inspection results, satisfied Bulletin Actions 1.a and 1.b for all Westinghouse plants.
The second WOG program was implemented to develop a detailed analysis of the surge line to demonstrate Code compliance for the design life of plants and satisfy Bulletin Actions 1.c and 1.d.
The program provided ASME stress and fatigue evaluations based on individual detailed analyses of groups of plants.
The methodology and results of the generic detailed analysis were summarized in WCAP-12639 which was submitted to NRC in June 1990.
The staff reviewed the WOG report and conducted an audit at'estinghouse offices in November 1990.
The following section summarizes the staff evaluation of the program.
2 ~ 0 STAFF EVALUATION The WOG generic detailed analysis was applicable to 43 of the 55 Westinghouse plants.
Plant specific analyses had been previously performed for the other 12 -plants.
Zn. order'o'.reduce the analytical efforts, the 43 plants were'ategorized'" into 17 analysis groups based on their similarity of response to thermal stratification.'he WOG program involved the development of updated design transients to account for 'tratification and striping, and the analysis of each structural group for global and local stresses 'fee'erifying ASME Code compliance in stress and fatigue.
The individual licensees are responsible for demonstrating applicability of the WOG generic analysis to their specific plant.
The major areas of staff review and evaluation are summarized below.
2.1 Update of Design Transients Westinghouse updated their design transients for pressurizer surge line to reflect stratification effects.
Two major categories of transients were considered:
heatup and cooldown transients and normal and upset transients.
For each design transient, the original uniform temperature distribution was modified to a stratified distribution with a corresponding number of cycles.
Xn addition, Westinghouse developed a
new set of transients for thermal striping.
The development of updated transients used information from an operational study and results of several plant monitoring programs.
In the operational
- study, Westinghouse reviewed heatup and cooldown procedures as well as historical records from several plants.
Heatup and cooldown 'operation was of primary concern because the maximum nTs between the pressurizer and hot leg occur during these modes.
In addition, Westinghouse conducted interviews with reactor operators and shift supervisors at a representative sample of plants to gain additional insight into the variation of operating methods.
Based on a
review of WOG plant geometries and earlier monitoring experience gained in plant specific
- programs, Westinghouse provided recommendations regarding the need for additional monitoring data to cover all variations of WOG plants.
Westinghouse received surge line monitoring-data from a total of 21 domestic plants.
Typical monitoring programs involved the installation of temporary sensors on the surge line piping.
Externally mounted RTD's or thermocouples were attached to the outside surface of the pipe at various circumferential and axial locations.
These sensors provided data on the top to bottom temperature distribution along the longitudinal axis of the pipe.
Several plants also installed sensors to detect vertical and horizontal movements at, locations along the pipe axis.
Data was typically collected at frequent intervals during heatups and cooldowns when system hT was high. Data was also collected during steady state operation but at a reduced frequency.
In addition, existing plant instrumentation was used to record various system parameters for correlation of plant operation actions with stratification in the surge line.
The data was typically provided to Westinghouse, in tabular form or in time history plots.
In updating the heatup and cooldown transients, the total number of heatup-cooldown cycles remained..-- unchanged (200)..-
However, sub-events and the associated number of. occurrences were redefined, based on historical records and.
monitoring data.
Westinghouse reviewed operating records from 10 plants to.determine a conservative distribution of maximum system nT ranges to be used over the design life.
Monitoring data from 10 plants was analyzed to develop a bounding distribution of cycles at various ranges of
relative strength of stratification (ratio of pipe aT to system sT).
The information was used to develop a table of numbers of cycles at various maximum stratification hT values for heatup/cooldown for the design life of the plant.
Zn updating the normal and upset transients, Westinghouse redefined the thermal fluid conditions based on the existing design transient system parameters and the knowledge gained from the monitoring programs.
The redef ined thermal fluid conditions conservatively accounted for thermal stratification.
The result of this effort was a table of maximum stratification aT values and corresponding cycles for all normal and upset transients.
Westinghouse reviewed the monitoring data to verify that all recorded normal or upset transient data was enveloped by the updated design transients.
Westinghouse developed a
new set of transients for thermal striping.
The frequency of fluid oscillation was conservatively derived from various experimental studies referenced in WCAP-12639.
Westinghouse assumed that each stratification transient would initiate striping oscillations.
The differential temperature was assumed to be the full aT which would decrease with time because of conduction between the hot and cold layers of fluid.
The attenuation of thermal striping was factored into the amount of time that each level of aT was assumed.
The end result was a table of striping transients shown as numbers of initiation cycles for several aT levels.
The staff reviewed the methodology and raised several questions which were discussed during the November 1990 audit at Westinghouse offices.
During the audit, the staff also reviewed some of the monitoring data and other related documentation.
Based on the review of the information provided during the audit, all questions were adequately resolved.
The staff found the methodology used by Westinghouse to update design transients acceptable.
Westinghouse used conservative methods and assumptions to incorporate stratification into their normal and upset transients.
To the extent possible, monitoring data was used to confirm the
" conservatism of the revised normal and upset transients.
The definition of thermal striping transients was based on 'onservative application of experimental data and anticipated stratification conditions in the surge line.
The development of, updated heatup and cooldown transients relied heavily on plant monitoring data, procedural limits and historical data.
By considering distributions of maximum system nT and relative strengths of stratification observed in several plants, Westinghouse developed a reasonably conservative table of numbers of cycles at corresponding maximum stratification dT values for heatups and cooldowns during the life of a plant.
To provide additional confidence, each licensee will be required to review their operating records and procedures to verify that the input and assumptions of the generic analysis are applicable to their specific plant.
2 '
Pipe Stress Analysis In order to minimize the number oi analyses, Westinghouse divided the 43 plants into 17 analysis groups based'n similarity of design and response to stratification.
The three ma)or parameters considered in establishing the groups were structural layout and support design configuration, axial temperature profile distribution, and thermal transients from plant operation.
During the
- audit, the staff reviewed the guidelines for defining enveloping design configurations and found them to be conservative.
For thermal stratification loading, the piping analysis was divided into two parts.
The global piping system analysis addressed the restraint. effects from pipe supports on the piping system.
The local stress analysis considered the effects of the non-linear temperature gradient in the pipe at several locations including structural discontinuities.
In the global analysis, a
piping model which typically included pipe, elbow, linear and non-linear support elements was prepared for each of the 17 analysis groups.
Bounding thermal stratification loadings were defined and applied.
The results provided maximum pipe loads at critical.
locations for each group.
The local stress analysis determined the local axial stresses which result from the step change in temperature that occurs at the hot-to-cold interface along the pipe including the structural discontinuity stresses in the nozzle transition region.
Westinghouse developed detailed finite element models of the surge line piping and hot leg nozzles to calculate these local thermal stresses.
A number of stratification load cases were defined and analyzed to determine temperature and stress distribution.
The results of the global and local stress analyses were combined as needed to perform the ASME Code evaluation.
Stresses and fatigue usage due to thermal striping were evaluated separately.
The fluid dT and correspondi.ng number of cycles of striping initiation transients were developed from design transients and plant monitoring data.
Initially, the striping differential temperature was assumed to be the full aT but decreasing with time because of conduction between the hot and cold.
layers of fluid.
A striping attenuation curve was developed, and for each striping initiation cycle, the 4T was assumed to follow this curve in five degree 'temperature steps.
The total numbers of striping cycles at each five degree step were determined by multiplying the number of striping initiation cycles which have a temperature step at that
- level, by the frequency of striping oscillation and amount of time that the aT was determined to be at that step..'he frequency of oscillation was based on values observed in water model flow tests performed for the Liquid Metal Fast Breeder Reactor and in experimental studies of thermal striping which were performed in Japan by Mitsubishi Heavy Industries.
Thermal striping stresses were determined by finite
element analysis and by a Westinghouse computer program STRFAT2.
Stresses were intensified by the appropriate ASME Code stress indices for peak stress.
V During the westinghouse
- audit, the staff reviewed sample calculations covering key analysis methods and assumptions.
This included the methodology for heat, transfer and stress
- analysis, calculations for determining film coefficients for stratification and
- striping, models used in the striping analysis and the technical basis for the attenuation curve used in the striping analysi's.
The staff found the methodology and assumptions to be reasonably conservative and acceptable.
2.3 ASME Code Evaluation and Results
'\\
The stress and fatigue evaluation was based on ASME Code, Section ZIl, 1986 Edition.
Westinghouse generally applied the methods of NB-3200 to evaluate the surge line components and reported the results in terms of the NB-3650 piping stress equations.
Stresses were classified in accordance with Code guidelines using stress indices from NB-3680 where appropriate.
Stresses due to pressure,
- moments, and thermal loads were combined for checking against the limits of Code Equations 12 and 13, cumulative fatigue usage and thermal stress ratchet requirements.
The results of the evaluation were presented in WCAP>>12639.
Westinghouse concluded that all analysis groups met the thermal stress ratchet requirements.
- However, only 15 plants in five groups met the Equation 12 and cumulative fatigue usage factor limits.
For Equation 13 qualification, Westinghouse determined limiting values for the resultant moments due to deadweight and OBE for the plant groups that met the other limits.
Therefore, each of the 15 plants must demonstrate that their plant specific OBE and deadweight moments in the surge line are enveloped by values used by Westinghouse.
I'he staff reviewed the Code evaluation results and agreed that they provide an acceptable basis for qualification of the surge lines for the 15 plants subject to plant specific. verification of applicability of the generic analysis and completion of additional plant specific evaluations needed to address items not covered by the generic analysis.
2.4 Applicability Demonstration For the 15 plants which were found acceptable by the WOG generic degailed
- analysis, Westinghouse provided
. guidelines regarding additional work that licensees should perform to-verify applicability of the generic analysis for their specific plant and to complete additional evaluations which were outside of the scope
of the WOG generic analysis.
These guidelines are summarized below.
Licensees should review past operating records to verify that system nT values have not exceeded the maximum system aT used in the analysis.
They should verify that the operational method used in the WOG generic analysis is applicable to their plant.
It is expected that system aT will be controlled in the future by plant operations procedures to avoid the possibility of having to reconcile the effects of exceeding these values.
WCAp-12639 provided a list of operational recommendations to minimize the chances for exceeding these values in the future.
Each plant must demonstrate adequacy of pipe supports and acceptability of piping displacements.
Additional plant specific piping analysis may be needed to determine these loads and displacements.
Licensees must verify that seismic OBE moments assumed in the fatigue analysis are applicable or conservative.
Allowable resultant moments values for combined deadweight and OBE at the hot leg nozzle safe end weld must be checked.
The WOG generic detailed analysis did not address the effects of thermal stratification on stress and fatigue at integral welded attachments (lugs,
- plates, etc.)
or on the pressurizer nozzle.
Plant specific evaluations of these areas are needed to complete the surge line qualification.
Plant specific detailed analysis must be performed for all plants that were not shown acceptable under the generic a'nalysis.
For some plants, modifications may be necessary.
The staff reviewed the above plant specific applicability requirements and agreed with the Westinghouse recommendations.
3.0 CONCLUSION
S Based on the review of the WCAP-12639 and additional information provided by Westinghouse during the November 1990 audit, the staff concludes that the WOG methodology for evaluation of stress and fatigue effects on the surge line due to thermal stratification and thermal striping is acceptable.
The WOG generic detailed analysis demonstrated acceptable ASME Section IIINB-3650 Equation 12 stress and fatigue usage in the surge line and reactor coolant loop nozzle for the following 15 plants:
Zion 1
& 2 Haddam Neck, Salem 1
& 2 Millstone 3 McGuire 1
& 2 Ginna Catawba 1
& 2 San Onofre 1
Prairie Island 2
Wolf Creek Callaway
For these plants, the WOG generic detailed analysis program results can be used to satisfy the requirements of NRC Bulletin 88-11, Actions 1.c and 1-d, provided that plant specific applicability is demonstrated and additional evaluations which were not included as part of the WOG program are performed.
Applicability requirements include:
o Review of operating records to ensure that system aT limits assumed in the analysis were not exceeded o
Verification of operational methods to ensure that they are consistent with the methods assumed in the analysis.
Limits on system aT for future operation are recommended.
o Verification of applicability of seismic OBE bending moments used in the fatigue analysis and combined deadweight and OBE moments at the hot leg nozzle.
Additional plant specific evaluations to be performed include:
o Evaluation of adequacy of pipe support loads and pipe displacements.
o Evaluation of effects of stratification on stress and fatigue at integral welded attachments (lugs, plates, etc.)
o Evaluation of effects of strati,fication on stress and fatigue of the pressurizer nozzle.
A total of 28 Westinghouse plants could not be shown acceptable by the WOG generic analysis.
They include the following plants:
D.C. Cook 1
& 2 Farley 1
& 2 H.B. Robinson 2
Shearon Harris Byron 1
& 2 Braidwood 1
2 Watts Bar 1
& 2 Surry 1
& 2 Pr'airie Island 1
4 This group, of plants will require plant specific analysis to demonstrate Code compliance.
It is anticipated that some of these plants can.'e shown acceptable by removing somci of the conservatisms inherent in the generic approach.
It is likely, however, that some of these plants will require modifications.
DISTRI BUTION NRC LPDR LPDR PD2-1 Reading SVarga 14 E4 GLainas 14 Hl BNozafari OGC PAnderson EJordan NNBB 3701 ACRS (10)
- LReyes, RII JRaleigh 14 E21 SHou 7 E23 JNorberg 7
D2
i