ML20217C274
| ML20217C274 | |
| Person / Time | |
|---|---|
| Site: | Palisades  |
| Issue date: | 03/31/1998 |
| From: | Mcelroy W AFFILIATION NOT ASSIGNED |
| To: | |
| Shared Package | |
| ML18068A330 | List: |
| References | |
| CTS-L-R-97-1, NUDOCS 9804230258 | |
| Download: ML20217C274 (89) | |
Text
CTS-L/R-97-1 INDEPENDENT REVIEW IS THE CONSUMERS ENERGY METHOD (WESTINGHOUSE METHOD)
OF DETERMINING THE PALISADES NUCLEAR PLANT'S BEST ESTIMATE FLUENCE BY COMBINING TRANSPORT CALCULATION AND DOSIMETRY MEASUREMENTS TECHNICALLY SOUND AND l DOES IT MEET THE INTENT OF THE PTS RULE ?
l Manuscript Completed: November 1997 Date Published: March 1998 ,
Prepared by: W. N. McElroy '
f Prepared for Consumers Energy Under Purchase Order No. C0025287 by Consultants & Technology Services cts R DO O O 255 P PDR
DISCLAIMER OF WARRANTIES & LIMITATION OF LIABILITIES This Letter / Report (LR) was prepared as an account of consulting work cooponsored by Consumers Energy (CE) under Purchase Order No. C0025287 and Consultants & Technology Services (CTS) under its internally supported ASTM Standards Technology Development, Transfer, & Training (STDTT) Project & Program Work.
Neither CTS nor CE nor any person acting on behalf of them:
(A) Makes any warranty or representation whatsoever, express or implied, (i) with respect to the use of any information, apparatus, method, process, or similar item disclosed in this LR, includirig merchantability and fitness for a particular purpose, or (ii) that such use does not infringe on or interfere with privately owned rights, including any party's intellectual property, or (iii) that this LR is suitable to any particular user's circumstance; or (B) Assumes responsibility for any damages or other liability whatsoever (including any consequential damages, even if CTS or any CTS representative has been advised of the ;
possibility of such damages) resulting from your selection or use of this LR or any information, apparatus, method, process, or similar item disclosed in this LR.
(C) Assumes any legal liability or responsibility for any third party's use, or the results of such use, of any information, apparatus, product or process disclosed in this LR, or represents that its use by such third party would not infringe privately owned rights.
Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacture, or otherwise does not necessarily constitute or imply its
- endorsement, recommendation, or favoring by CTS. The views and opinions of the author do not necessarily state or reflect those of CTS.
Printed in the United States of America Available from Consultants & Technology Services 113 Thayer Drive Richland, Washington 99352 Telephone: 509 943 5953 Fax: 509 946-5979 Price: Optional for Report ; Plus Copying, Shipping and Handling Costs
- Any funds collected will be used for CTS office expenses associated with Standards Technology Development, Transfer & Training (STDTT) for ASTM Subcommittee E10.05 Task Group (TG) E10.05.04 on the E706 Master Matrix of LWR Surveillance Standards; see detailed ;
commentary in the attached March 30,1996 TG Report.
1 Copyright O 1997 Consultants and Technology Services, Richland, WA. All rights reserved. l This material may not be reproduced or copied, in whole or in part, in any printed, mechanical, electronic, film, or other distribution and storage media, without the express written consent of the President of CTS and/or the Manager of Licensing of Consumers Energy. It is not to be reproduced or distributed as quoted, in whole or in part, by any 1 means whatsoever, except with the express written approval of the President of CTS and/or the Manager of Licensing of Consumers Energy.
il
CONTENTS ag ABOUT THE AUTHOR ACRONYMS ix 1.0 OVERVIEW 1 REVIEW PROGRAM AND RESULT 1 REVIEW HIGHLIGHTS AND WORK SCHEDULE 1
2.0 INTRODUCTION
8 3.0 FEDERAL LAW & TEGUL ATIONS (FLR): STANDARDS TECHNOLOGY DEVELOPMENT, TRANSFER & TRAINING (STDTT); AND KEY ISSUES AND SUPPORTING TECHNICAL DOCUMENTATION THAT ARE RELEVANT TO FLR & STDTT AND PAST & FUTURE DISCUSSIONS BETWEEN THE NRC AND PALISADES NUCLEAR PLANT STAFF. 10 FLR & STDTT 10 i
PWR & BWR SURVElLLANCE PROGRAM REGULATORY INSTRUMENTS 12 TABLE 3.1 ATTACHMENT: NRC PHYSICS-DOSIMETRY COMPENDIUM 18 TABLE 3.1 ATTACHMENT: REGULATORY GUIDE 1.99, REVISION 2 25 KEYISSUES 27
)
BENCHMARK FIELD (BF) ISSUE 27 TRANSPORT CALCULATIONAL ISSUE 27 EXPOSURE & RADIATION DAMAGE PARAMETER EXTRAPOLATION ISSUE 28 REGULATORY GUIDE 1.99, REV.2 DATA BASE CONSISTENCY ISSUE 28 FLUENCE ISSUE 29 CHARPY SHIFT ISSUE 29 COPPER AND NICKEL ISSUE 29 NEUTRON FLUX OR DAMAGE RATE ISSUE 30 CONSISTENCY OF METHODOLOGY ISSUE 30 LMFBR-ILRR-FTR PHYSICS-DOSIMETRY CHARACTERIZATION PROGRAM REQUIREMENTS AND RESULTS 31 BIASED FLUENCES IN THE CHARPY EMBRITTLEMENT DATABASE 32 j ill
i i
4.0 REVIEW AND STU6Y OF THE PALISADES NUCLEAR PLANT'S (PNP)
INFORMATION PACKAGE 39 5.0 IS OWNER'S METHOD (WESTINGHOUSE METHOD) OF DETERMINING BEST ESTIMATE FLUENCE BY COMBINING TRANSPORT CALCULATION AND DOSIMETRY MEASURE-MENTS TECHNICALLY SOUND? 39 6.0 IS THE METHOD USED CONSISTENT WITH THE BASIS OF 10 CFR 50.61, THE PTS RULE? 41 7.0 HAS OWNER'S EXPLANATION OF THE BfAS BETWEEN MEASUREMENT AND CALCU-LATIONS PROVIDED SUFFICIENT BASIS TO SUPPORT THE MAGNITUDE OF THIS ;
DIFFERENCE? , 43
, 8.0 HAS OWNER COMMUNICATED ITS POSITION CLEARLY AND IF NOT WHERE WOULD FURTHER EXPLANATION BE USEFUL? 43 )
9.0 FERRET-SAND ll METHODOLOGY 44 REGULATORY REQUIREMENTS AND ASTM STANDARDS 44 WESTINGHOUSE METHODOLOGY & TECHNICAL ISSUES 45 NINTH ASTM-EWGRD SYMPOSIUM ON REACTOR DOSIMETRY 49 PAPER 1: DOSIMETRY IN SUPPORT OF THE EUROPEAN NETWORK AMES 50
! PAPER 2: ORGANIZATION FOR ECONOMIC COOPERATION & DEVELOPMENT, l NUCLEAR ENERGY AGENCY (NEA), NUCLEAR SCIENCE COMMITTEE (NSC) STUDY: ISSUES OF DOSIMETRY FLUENCE COMPUTATIONS 56 PAPER 3: EXPERIMENTAL AND THEORETICAL STUDIES ON VVER-1000 l REACTOR DOSIMETRY 57 i
PAPER 4: NEUTRON DOSIMETRY IN EXTENDED SURVEILLANCE PROGRAM ON THE 4TH UNIT OF NPP DUKOVANY 59 PAPER 5: FAST NEUTRON FLUENCE MONITORING ON NPP DUKOVANY 63 REFERENCES 68 ENCLOSURES 1 - 12 iv
ENCLOSURES
> 1) ATTACHMENT 1: CONSENSUS PROCESS, LAWS (STATUTES), REGULATION, QUALITY ASSURANCE, INSURANCE, & STANDARDS; Section 1.4 of (Mc95].
- 2) ATTACHMENT 2: CONSUMERS POWER CO. RV NEUTRON FLUENCE MEASUREMENT PROGRAM FOR PALISADES NUCLEAR PLANT; CYCLE 1 THRU 11, WCAP-14557, REV.1, MARCH 1996. (SELECTED PAGES FROM SECTION 3: NEUTRON TRANSPORT & DOSIMETRY EVALUATION METHODOLOGIES.)
> 3) DRAFT SECTION 5.0 FOR ANS-19.10 STANDARD; ATTACHMENT 3 of (Lo971
> 4) ASTM SUBCOMMITTEE E10.05 TASK GROUP E10.05.04 REPORT ON E706 MASTER MATRIX LWR SURVEILLANCE STANDARDS; Dated March 30,1996.
J
> 5) LWR PV SDlP: PCA EXPERIMENTS AND BLIND TEST; NUREG/CR-1861, July 1981. (SELECTED PAGES)
> 6) LWR PV SDlP PROGRAM: PCA EXPERIMENTS, BLIND TEST AND PHYSICS- l DOSIMETRY SUPPORT FOR THE PSF EXPERIMENTS; NUREG/CR-3318, I SEPTEMBER 1984. (SELECTED PAGES)
> 7) LWR PV SDlP PROGRAM: PSF PHYSICS-DOSIMETRY PROGRAM; NUREG/CR-3320, VOL. 3, OCTOBER 1987. (SELECTED PAGES)
> 8) LWR PV SDlP PROGRAM: LWR POWER REACTOR SURVEILLANCE PHYSICS-DOSIMETRY DATA BASE COMPENDIUM,1987 UPDATE: NUREG/CR-3319, AUGUST 1985 & APRIL 1987. (SELECTED PAGES)
> 9) PROCEEDINGS OF FIRST ASTM-EURATOM SYMPOSIUM ON REACTOR DOSIMETRY, EUR 5667 elf, PART 1,1977. FIRST PAPER: NEUTRON ENVIRONMENTAL CHARACTERIZATION REQUIREMENTS FOR REACTOR FUELS AND MATERIALS DEVELOPMENT AND SURVEILLANCE PROGRAMS.
> 10) FTR DOSIMETRY HANDBOOK, HEDL MG-166, MARCH 1983 (CONTENTS THRU SECTION 2.0)
> 11) J.R. WORTHAM'S PAPER ON " CONSISTENT VESSEL FLUENCE & PTS EMBRITTLEMENT UNCERTAINTIES," DISTRIBUTED TO ATTENDEES OF THE 9TH ASTM-EUROPEAN SYMPOSIUM ON REACTOR DOSIMETRY, HELD IN PRAGUE, CZECH REPUBLIC, SEPTEMBER 2-6,1996
> 12) J.R. WORSHAM'S PAPER ON " BIASED FLUENCES IN THE CHARPY EMBRITTLEMENT DATABASE," DISTRIBUTED TO ATTENDEES OF THE 9TH ASTM-EUROPEAN SYMPOSIUM ON REACTOR DOSIMETRY, HELD IN PRAGUE, CZECH REPUBLIC, SEPTEMBER 2-6,1996 v
i ABOUT THE AUTHOR William N. Mc Elroy Dr. McElroy has more then 45 years background experience in the nudear physics and engineering, reactor development, standards and operations fields. He is a Fellow of ASTM and present or past member of ANS, American Association for the Advancement of Science, National Academy of Sciencel National Academy of Engineering / National Research Council Evaluation Panel for Center for Radiation Research at NBS, Brookhaven Cross Section Evaluation Working Group (CSEWG), IAEA Working Group on Reactor Radiation Measurements, Phi Lambda Upsilon, Sigma Xi, Sigma Pi Sigrna, and Phi Beta Kappa, and has extensive publications and holds several patents related to his work experience.
He is the Executive Chairman of ASTM Subcommittee E10.05 on Nuclear Radiation Metrology, Past Vice Chairman of ASTM Subcommittee E10.10 on Matrix to Standards for Nuclear Systems Technology, and the Past ASTM Executive Chairman of the ASTM-EURATOM Symposia on Reactor Dosimetry. He has been an ASTM representative on the NUMARC\NUPLEX Subcommittee on Codes and Standards and the ASME Board of Nuclear Codes and Standards Steering Committee on PLEX He is responsible for the coordination, direction, and preparation of the ASTM E706 Master Matrix set of LWR-Pressure Vessel (PV)
Surveillance Dosimetry Improvement Program (SDlP) standards. In this regard, he has provided overall coordination and has contributed to the preparation of a very large series of U.S. Nuclear Regulatory Commission (NRC) LWR PV-SDlP progress and technical reports that provide the physics-dosimetry-metallurgy reference documentation that is needed to support the development, use and application of ASTM standards for PWR & BWR operations and surveillance programs. This overall coordination was by NRC contract with Chuck Serpan, Al Taboada, Neil Randall, and Lambros Lois under the auspices of the Offices of Nuclear Regulatory Research & Nuclear Reactor Regulation of the NRC.
He is the lead instructor and author for the ASTM Committee E10 on Nuclear Technology and Applications One-day Standards Technology Training (STT) Course on " Condition Assessment and Surveillance of Nuclear Reactor Pressure Vessel Steels;" this course has been presented world wide. The other instructors and authors are: E. P. Lippincott (Retired from Westinghouse Electric Corporation), A. L. Lowe, Jr. (Retired from Babcock & Wilcox Company), and the late P. D. Hedgecock (NUTECH Engineers, Inc. and Past Chairman of ASTM Subcommittee E10.02 on Behavior and Use of Nuclear Structural Materials). A Second Edition of the One-day ASTM STT Course " Workbook" was completed and issued in November 1991. The authors state:
"This One-day Workshop provides an overview of the basic principles, techniques, and methodology to be used in the analysis and interpretation of neutron exposure data obtained from light water reactor pressure vessel surveillance programs; and, based on the results of that analysis, establishes a formalism to be used to evaluate the present and future condition of the pressure vessel and its support structures. l vi
(
1
AEA Reactor Services, United Kingdom, in association with ASTM, organized a UK Workshop on " Radiation Damage Correlation Methodology" that was held at the ASTM European Office, Hitchin, Herts, England in December 1991. ASTM presented its One-day STT Course Workshop that provided a broad based introduction to the subject of pressure vessel condition assessment and surveillance which was a centralissue of the AEA-UK Workshop.
This provided an opportunity for the ASTM instructors, UK AEA Technology staff, UK-HM Nuclear Installations Inspectorate (Nil) staff, UK-Nuclear Electric (NE) plc staff, and Scottish Nuclear plc staff to exchange views. Particular emphasis was placed on the exchange and discussion d the most current physics-dosimetry-metallurgy information that has been (or will be ) recommended for use in existing codes, regulatory guides, and standards for the management of age and radiation related degradation of materials and reactor components.
In this regard, in depth consideration was given to the status of state-of-the-art development work on damage correlation methodology and its applicability to NE's and Nil's responsibilities, respectively, to safely manage and regulate the operation of the UK Sizewell "B" PWR and to extend the operating life of UK Magnox gas cooled nuclear power plants.
1 Dr. McElroy is the lead instructor and author for another ASTM Committee E10 on Nuclear i Technology and Applications One-day STT Course on " Radiological Decontamination &
Decommissioning;" this course has only been presented within the U.S in conjunction with ASTM Committee E10 meetings. The other instructors and authors are: P. E. Fuller, (Consulting Services, New Hartford; Past Staff Member of American Nuclear Insurers; Secretary of E10.03), Raymond Gold (Metrology Control Corporation), J. L. Helm (Columbia University), A. S. Kumar (University of Missouri-Rolla), E. P. Lippincott (Consulting Services, Monroeville), R. H. Meservey (EG&G Idaho incorporated, Chairman of E10.03 on D&D &
Extended Life Operation of Nuclear Facilities), P. S. Olson (Rockwell International, Past Chairman of Committee E10), and G. Subbaraman (Rockwellinternational). A Second Edition of the One-day ASTM STT Course " Workbook" was completed and issued in June,1995.
The authors state:
"This One-day Workshop provides an overview of
- Radiological Decontamination and Decommissioning (D&D) and the associated Demolition for nuclear facilities; including research, test, and power reactors, fuel processing facilities, and uranium mills and mining operations.
- Status of development and application of a broad range of ASTM standards and
, supporting documentation that have and are being developed for Nuclear Environmental Restoration and Waste Management (EM) as wellas Environmental Assessment and Risk Management (EARM).
- Building consensus through risk assessment management of DOE EM program: That is, thru 1) systematic, integrated risk assessment with full stakeholder participation,
- 2) progress in science-based risk assessment and management, and 3) President's Commission on Risk Assessment & Management & Consortium for Risk Evaluation &
Stakeholder Participation (CRESP).
vii
- Importance of detailedplanning, guidance andimplementation of welldocumented and standardized procedures and Quality Assurance (QA) practices to minimize risks &
costs and commentary on the political and technical consensus processes.
D&D concerns over environmentalrelated technological, regulatory, codes & standards, political, insurance, underwriting, claims, legal, economic, and health & safety issues (including those associated with the Department of Labor's Occupational Safety &
Health Administration's (OSHA) regulations bearing on Demolition].
- Nuclear site specific issues as they are related to current ASTM standards work. ".
Dr. McElroy's expertise has been significantly enhanced by his direct involvement in the development and presentation of the two ASTM STT One-day Course Workshops. Of particular value has been the interactions and exchange of the most current views and information with course students from nuclear related industries, utilities, universities, and regulatory bodies in the U.S. and other countries.
Dr McElroy's career has included positions as: Senior Research Eng'r at Atomics International (Rockwell International) in the design, development, testing, operation and licensing of research reactors; Manager of Reactor Operations at the llT Research Institute; Westinghouse Hanford Co. (WHC) Fellow Scientist and Manager of the Irradiation Analysis Section at the Hanford Engineering Development Laboratory (HEDL); Manager of the National Dosimetry Center (NDC) at WHC-HEDL. As a part of his nuclear reactor work experience, he has been a reactor operations supervisor, prepared reactor operation and maintenance manuals and provided direction, training and examinations for reactor operations personnel.
He is retired from WHC and Battelle, Pacific Northwest Laboratories (PNL), where he was the Manager of the WHC and PNL National Dosimetry Centers (NDC). He currently is the President of Consultants and Technology Services (CTS), has a B.S. in Chemistry from the University of Southern California and a M.S and Ph.D in Physics from the Illinois Institute of ,
Technology (llT), l l
l 1
viii
l ACRONYMS AMES Ageing Materials Evaluation and Studies AEA Atomic Energy Authority, United Kingdom A M P-ll A Modular Code System for Generating Coupled Multigroup Neutron Gamma y Libraries for ENDF Format, ORNL ANISN One-Dimensional Discrete Ordinates Transport Code, ORNL ANS American Nuclear Society ANSI American National Standards Institute ASME American Society of Mechanical Engineers ASTM American Society for Testing and Materials BPVC Boiler & Pressure Vessel Code q B&W Babcox & Wilcox {
BWR Boiling Water Reactor l
BCL Battelle Columbus Laboratones BMI Battelle Memorial Institute BUGLE-96 A Revised Multi-Group Cross Section Library for LWR Applications Based on ENDF/B VI Release 3 CASMO-3 A Fuel Assembly Burnup Program, Studsvik Energitehnik CE Combustion Engineering i
CE Consumers Energy j CEN/SCK Center for the Study of Nuclear Energy, Mol, Belgium I CF Chemistry Factor CFR Code of Federal Regulation COSA2 Spectrum Adjustment System Based on the STAY'SL Type Least Squares Code, Rossendorf, Germany CSEWG Cross Section Evaluation Working Group, BNL CTS Consultants and Technology Services, Richland Washington, U.S. A DPA Displacement per Atom, an irradiation Exposure Parameter i DOE U.S. Department of Energy DOT-4 One & Two-Dimens'al Neutron / Photon Discrete Ordinates Transport Code, ORNL ,
DORT Two-Dimensional Discrete Ordinates Transport Code, ORNL l DOMPAC Triton Reactor Thermal Shield and Pressure Vessel Mockup, France i DOSCROS84 Dosimetry Cross Section Library in 640 Groups of SAND 11 Type, Netherlands )
DOTSOR A Module in the LEPRICON Computer Code System for Representing the I Neutron Distribution in LWR Cores i DOTSYN A Module in the LEPRICON Computer Code System for Synthesizing Three-Dimensional Fluxes ECC European Community Commission ECN Netherlands Energy Research Foundation EDB Embrittlement Data Base EFPY Effective Full-Power Years ENDF Evaluated Nuclear Data File EOL End-of-Life EPRI Electric Power Research Institute EURLIB 120 Group Coupled Neutron & Gamma Data Library, ECC EURATOM, CCR-Ispra ELXSIR Cross Section Library for LWR Pressure Vessel Irradiation Studies: Part of LEPRICON Computer Code System ix
i l
EURATOM Commission of the European Communities EWGRD European Working Group on Reactor Dosimetry FERRET Least-Squares Adjustment Code System Used by HEDL/PNL-NDC & W
-SANDil FMIPA Freely Migrating interstials Per Atom FMVPA Freely Migrating Vacancies Per Atom FORSS A Sensitivity and Uncertainty Analysis Code System, ORNL FSAR Final Safety Analysis Review GE General Electric HAFM Helium Accumulation Fluence Monitor HEDL Hanford Engineering Development Laboratory HFIR High Flux isotope Reactor at ORNL lAEA international Atomic Energy Agency ILRR Interlaboratory LMFBR Reaction Rate Program IRDF 90 International Reactor Dosimetry File, IAEA JAERI Japanese Atomic Energy Research Institute '
J DF-1.1 JENDL Dosimetry File-1.1 JENDL Japanese Dosimetry File for Material Dosimetry in the JOYO Reactor JOYO Experimental Fast Reactor at Oarai Engineering Center of PNC KORPAS Irradiation Facility at the RBT-6 Reactor in Russia for Reactor Pressure Vessel Steel Specimen irradiations and Dosimetry Benchmarking: See PSF Description LANL Los Alamos National Laboratory LEPRICON Generalized Linear Least-Squares Combination Procedure Code System for PWR Pressure Vessel Surveillance Dosimetry Analysis, ORNL LSL-M2 Statistical, Least Squares, Log Normal A Priori & Posteriori Distrib'n Code, ORNL LMFBR Liquid Metal Fast Breeder Reactor ]
LR-O VVER Type Reactor Benchmark l LWR Light Water Reactor !
LWR-PV-SDlP LWR Pressure Vessel Surveillance Dosimetry improvement Program i MCBEND Monte Carlo Code, UK MCC Metrology Control Corporation, Richland, Washington, U.S.A MCNP A General Monte Carlo N-Particle Transport Code, LANL MDRF Materials Dosimetry Reference Facility at University of Michigan: Developed by NIST and University of Michigan MVP/GMVP General Purpose Monte Carlo Codes for Neutron and Photon Transport Calculations Based on continuous Energy and Multigroup Methods, Japan NDC National Dosimetry Center NEl Nuclear Energy Institute NESDlP NESTOR Shielding and Dosimetry improvement Programme, UK NESTOR Source Reactor at Winfrith NEUPAC(J1) Statistical, Linear Estimation Adjustment Code, Japan NIST National Institute of Standards and Technology, U.S.A NJOY Data Processing System (June 1983 Version) Based on the Bondarenko Self-Shielding Factor Approach for Resonance Absorption.
NMF-90 Neutron Metrology File: Integrated Database for Neutron Adjust. Calculations NPP Nuclear Power Plant NRC Nuclear Regulatory Commission NRI Nuclear Research Institute, Rez plc. Czech Republic X
I 1
t l
l 1
NUMARC Nuclear Management and Resources Council j NUPLEX Nuclear Utility Plant Life Extension !
OECD/NEA Organization for Economic Cooperation & Development / Nuclear Energy Agency l OMB Office of Management & Budget, U.S.A I ORNL Oak Ridge National Laboratory i
- ORR Oak Ridge Research Reactor at ORNL
]
l PCA Poolside Critical Assembly at the Bulk Shielding Reactor at ORNL l PCA-Replica In the ASPIS Shielding Facility of the NESTOR Low-Flux Experimental Reactor, j Winfrith. UK l PLEX Plant Life Extension PNL Pacific Northwest Laboratory, Battelle Memorial Institute PNP Palisades Nuclear Plant J l
! PREVIEW Kernel Based Prog, for Out-of-Core Neut. Fluen'e Est's in VVER-440 Reactors l PSF Pool Side Facility of the Oak Ridge Research Reactor at ORNL
]
l PSU-DOTSOR A Revised Form of DOTSOR, Penn State '
Pressure-Temperature P-T
[ PTS Pressurized Thermal Shock PV Pressure Vessel PWR Pressurized Water Reactor RADAK Statistical, Linear Estimation Experimental Data Processing Program, UK RG NRC Rejulatory Guide RIAR KORPUS was Created at the RBT-6 Reactor (RIAR) for Reactor Pressure Vessel ]
Steel Specimen and Dosimetry Measurement Irradiations: Russian Physics- l Dosimetry-Metallurgy Test Reactor RM Radiometric Monitor
! RNDC Russian Nuclear Data Center, Obninsk l RPV Reactor Pressure Vessel j RT, Reference Nil Ductility Temperature SAILOR A Coupled Self-Shielded 47 Neutron 20 Gamma Ray Cross Section Library for l Light Water Reactors, ORNL SAND 11_ Semi-Iterative Adjustment Code; Interpreted as Application of Least Squares Principle Although Statistical Assumptions are not Spelled Out Explicitly SCK/CEN Center for the Study of Nuclear Energy, Mol, Belgium l
SDO Standards Development Organization SIMULATE-3 Advanced Three-Dimensional Two-Group Reactor Analysis Code, Studsvik of America SENSAK Statistical, Linear Estimation Adjustment Code, UK SINBAD A Shielding Integral Benchmark Archive and Database for PC's
,SBl RMI Dosimetry Cross Section Library, ORNL RSIC Data Library Collection SPECTRA Statistical, Linear Estimation Adjustment Code, SANDIA SRM Standard Reference Material SSC Simulated Surveillance Capsule STAY'SL Statistical, Linear Estirnation Adjustment Code, ORNL STDTT Standards Technology Development, Transfer & Training SSTR Solid State Track Recorder SUSD Cross Section Sensitivity and Uncertainty Package and Extension to 3D Analysis TG Task Group 1
r Xi l
\
l
TWODANT A Code Package for Two Dimensional Diffusion Accelerated Neutral Particle Transport Code, LANL THREEDANT A Code Package for Three Dimensional Diffusion Accelerated Neutral Particle Transport Code, LANL TLD
)
Thermoluminescence Dosimeter TM Temperature Monitor TORT Three-Dimensional Discrete Ordinates Neutron / Photon Transport Code, ORNL TRAMO A Flexible Multigroup Neutron Transport Code Based on the Monte Carlo Method, Germany TRIPOLl-3 Three Dimensional Monte Carlo Code, France TWODANT TWO-Dimensional Diffusion Accelerated Neutral Particle Transport Code, LANL UK United Kingdom U.S. United States U.S.A United States of America '
! USAEC United States Atomic Energy Commission l USE Upper Shelf Energy USNRC U. S. Nuclear Regulatory Commission VB PSF Void Box Experiment at ORNL l
VENUS Pressure Vessel Mockup at Moi, Belgium l VITAMIN-C Fine Group Neutron / Photon Cross Section Libraries Derived from ENDF/B-VI j and Nuclear Data with 171 Energy Groups VITAMIN-B VVER Soviet Designed PWR, same designation as WWER
, W Westinghouse L WGRD VVER Working Group on Reactor Dosimetry for VVER Reactors WHC Westinghouse Hanford Company W-NTD Westinghouse - Nuclear Technology Division j WWER Soviet Designed PWR, same designation as VVER 1
l xii
1 l
1.0 OVERVIEW l I
REVIEW PROGRAM AND RESULT -- Consumers Energy asked Consultants & Technology Services (CTS) to 1) provide the consulting services of Dr. W. N. McElroy to assist the >
Owner in performing a review of the Owner's Palisades Nuclear Plant Pressurized Thermal Shock (PTS) submittals to the NRC dated April 4 and 19,1996 and June 26,1997 and the :
NRC December 20,1996 safety analysis report (SER) and 2) upon completing the review l submit a letterireport to the owner's representative. j l
Based on the extent of the reviewers knowledge, CTS was to provide answers to the following four questions:
QUEST /ON 7: Is the Consumers Energy Method (Westinghouse Method) of determining the palisades nuclear plant's best estimate fluence by combining transport calculation and dosimetry measurements technically sound? .
I OUESTION 2 Is the method used consistent with the basis of 10 CFR 50.61, the PTS I Rule? I OUEST/ON 3: Has Consumers Energy's Explanation of the Bias Between Measurement and Calculations Provided Sufficient Basis to Support the Magnitude of This Difference? 4 OUEST/ON 4- Has the Owner Communicated its Position Clearly and if Not Where Would Further Explanation be Useful?
Further, the consultant was to include any additionalitems that were felt to be relevant to the discussion between the NRC and Palisades Plant staff.
Based on a verv carefulreview and studv of all of the relevant documentation andissues. the answers for Questions 1. 2. 3. and 4 are verv definitelv yest Detailed commentary and supporting documentation related to the above questions & answers as well as the consideration of additionalissues are provided in Sections 3 through 9 and the Enclosures / Attachments and Reference Listing.
REVIEW HIGHLIGHTS AND WORK SCHEDULE -- The April 4,1996 letter to the US Nuclear Regulatory Commission (Sm96] states:
- This letterprovides recent reevaluations of Palisades fluence data. New calculated and best estimate fluence values are presented and discussed. The new best estimate fluence values are derived utilizing a bias factor which is based on Palisades in-vessel and ex-vessel capsule measurements. The previously existing capsule measurements have been reevaluated and updated values are provided. Using the information provided in this letter, a revised PTS screening criteria date, as defined in 10CFR50.61, has been calculated. The axial welds containing heat # W5214 remain the limiting vesselmaterialand are now estimated to reach the screening da te in th e year 201 1. " . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
A ttachment 1 contains the Palisades evaluation of the updated reactor vessel fluence values and capsule fluence measurements. A ttachment 2 contains Westinghouse Report WCAP14557, Revision 1, " Consumers Power Company Reactor Vessel Neutron Fluence Measurement Program for Palisades Nuclear Plant - Cycle 1 through 11. " Attachment 3 contains AEA Report AEA T-0121, " Fluence Calculations for the Palisades Plant. "
NRC staff concurrence with the appropriateness of this updated reactor vessel fluence methodology to project reactor vessellife is requested by November 1,1996. " ,
l A careful review and study of Attachments 1,2, and 3 was completed by November 8,1997.
At the same time an in depth review and study of other relevant and supporting documentation, as identified in Section 2.0, was initiated to address specific issues that had been raised in the three attachments. ,
The June 26,1997 letter to the US Nuclear Regulatory Commission [Bo96) states:
The NRC Interim Safety Evaluation dated December 20,1996 concluded that Consumers Energy had determined the Palisades reactor vessel fluence in accordance with Draft Regulatory Guide DG-1053, "Calculationaland Dosimetry Methods for Determining Pressure VesselFluence. " This draft guide requires thatplants qualify their calculations methods using available measurements and allows a plant to adjust the calculations using a correction factor cetermined from a statistically significant measurement data base. " ....... ... ... .
With the additionalinformation provided by this letter, we believe that Consumers Energy has provided the staff with a sound technicalposition to support the additional 17% reduction in vessel fluence recommended by our April 4,1996 submittal. We believe this technical justification satisfies all Best Estimate analysis requirements and basis of 10 CFR 50.61 and RG 1.99 [Re88). We therefore, request that the staff approve this fluence reduction as proposed. if the staff is unable to concur with our technicaljustification for the fluence reduction, we request that the technicalbasis for the staff position be provided to Consumers Energy as soon as possible. "
The June 26,1997 letter then listed three attachments and six enclosures which provide Consumers Energy's response to issues raised by the NRC staff and information previously placed on the docket are enclosed to facilitate staff review.
A careful review and study of Attachments 1,2, and 3 and Enclosures 1- 6 was completed by November 13,1997. At the same time an in depth review and study of other relevant and supporting documentation, as identified in Section 2.0, was continued to address specific issues that had been raised in the three attachments and six enclosures.
The December 20,1996 NRC letter to Thomas C. Bordine, Manager of Licensing, Palisades Plant [Bo96] states:
By letter dated April 4,1996, you submitted an updated neutron fluence evaluation for the Palisades reactorpressure vesseland requested that it be reviewed and accepted as the basis for the 10 CFR 50.61 PTS vessel embrittlement evaluation. You provided additional information in response to staff questions on June 12, June 21, August 27, September 9, 2
September 19, and October 1,1996. The submittal claimed a 25 % reduction in reactor vesselfluence from the value acceptedin the previous review of the Palisades reactor vessel. "
The staff received contract assistance in this review from BNL. The NIST also provided contract assistance for specific dosimetry questions. Our evaluation indicates that an 8%
reduction in the calculated fluence due to physical changes in the plant is acceptable. The remaining 17%, consisting of a bias resulting from the comparison of calculated to measured fluence values and a reduction due to spectraladjustments can not be approved at this time.
The current Palisades fluence value results in the Palisades reactor vessel reaching the PTS screening criteria in 1999. The 8 % reduction approved by this evaluation results in the vessel reaching the screening criteria in 2003.
This is an interim report. Our review etfort willcontinue at BNL, focusing on the acceptability of the calculation-to-measurement bias and the statistical treatment of the data. Our safety evaluation is provided as Enclosure 1. The BNL technical evaluation report is provided as Enclosure 2. "
A careful review and study of Enclosures 1 and 2 was completed by November 20,1997.
At the same time an in depth review and study of other relevant and supporting documentation, as identified in Section 2.0, was continued to address specific issues that had j been raised in the two enclosures.
The CTS Lette.lReport preparation started on November 20,1997 and was completed on November 26,1997. This effort continued to involve the in depth review and study of all of the relevant documentation identified in Section 2.0 and preparing written technical responses and justification for answering "yes" to all of the four Questions in Section 2.0. The more difficult task was to provide Consumers Energy with sufficient additional technical documentation and justification so that the NRC would accept the results and allow the additional 17% reduction in reactor vessel fluence for the Palisades Nuclear Plant.
This was accomplished by an in depth technical review and study of the technical basis and soundness of the NRC recommended procedures and methodology found in the Code of Federal Regulations 10 CFR Part 50, Appendix G and H, the PTS Rule 10CFR50.61, RG DG-1025, RG DG-1053, RG 1.99 Rev. 2, ASTM E706 Master Matrix Set of Physics-Dosimety-Metallurgy Surveillance Standards and the ANS 19.10 Standard, " Fast Neutron Fluence to PWR Reactor Cavities;" see Section 3.0, Subsection "PWR & BWR Surveillance Program Regulatory Instruments.
More specifically, it was stated above that "The NRC Interim Safety Evaluation dated
' December 20,1996 concluded that Consumers Energy had determined the Pelisades reactor vessel fluence in accordance with Draft Regulatory Guide DG-1053. "
The determination of the Palisades reactor vessel fluence is not based on the RG DG-1053 procedures and methodology, but rather on the application of the Westinghouse Methodology.
This Methodology is based on the requirements of the Code of Federal Regulations 10CFR Part 50, Appendix G and H, the PTS Rule 10CFR50.61, Reg. Guide 1.99, Rev. 2, and the referenced ASTM E706 Master Matrix Set of Physics-Dosimetry-Metallurgy Surveillance Standards (See Table 3.1).
3
As such, the NRC needs to accept the results generated by the use of the Westinghouse methodology since it is based on the procedures, methodology, and data bases specified in the Code of Federal Regulations 10 CFR Part 50, Appendix G and H, the PTS Rule 10CFR50.61, RG 1.99 Rev. 2, and the referenced ASTM Standards.
Based on the present CTS review and study of the submittals by Consumers Energy's staff and the NRC's staff responses, the NRC has not provided an acceptable technical basis for not giving its concurrence with the appropriateness of the PNP updated reactor vessel fluence methodology to project reactor vessel life.
If the NRC's staff wants specific concerns about the validity of the procedures and methodology recommended in the ASTM standards addressed, this can be accomplished through the ASTM voluntary technical consensus process. That is, by working directly with the task groups that are responsible for reviewing technical issues and making changes through the technical consensus balloting process.
More will be said about this in Section 3.0 since it has to do with Consumers Energy's request for additionalitems relevant to discussions between NRC and Palisades Plant's staff. l In [Ca961, Carew and Aronson state in Section VI.1, Best-Estimate Fluence Determination:
". . .. Based on these measurement-to-calculation (M/C) comparisons of the dosimeter reaction rates, a M/C bias of 12( 7)% is determined. This M/C bias is then adjusted using ,
a least-squares adjustment technique to account for uncertainties in the measurement and l calculations. In the case of Palisades this adjustment increases the M/C bias from 12% to }
17%, andimplies the calculations are overt redicting the fluence by 17%. The determination 1 of the M/C bias and the adjustment method are discussed in the following sections. "
In Section VI.2, Fluence Calculation-to-Measurement Blas, they state:
"From Figure 4, it is seen that the in-vesselM/C bias is 1.00 t 0.03 for the dosimeters with thresholds E> 4.0 MeV, and 0.8610.02 for the dosimeters with thresholds E< 4.0 MeV.
In the CPC/Wanalysis this difference is assumed to be due to a spectrum-dependent error in 1 the DORT calculations which results in an exact calculation above E> 4.0 and an over prediction for E< 4.0 MeV. Based on this assumption, a 12 % M/C fluence reduction (i.e., not including the additional FERRET reduction) is applied to the DORT E> 1.0 MeV fluence f.rediction. The application of this M/C spectrum-dependent correction is illustrated in Figure
- 6. While this conclusion may be correct, there are severalotherpossible explanations for the observed 1.00/O.86 difference between the high/ low-energy M/C biases that would not
'requaa this reduction in the DORT calculated fluence. These include: 1) the use of erroneously low dosimeter cross sections for Fe-54 and Ni-58 in the interpretation of the measurements and/or 2) errors in the Fe-54 and Ni-58 measurements. (The number of U-238 and NP-237 dosimeters tnat are included in the in-vessel M/C bias is small and these measurements are subject to relatively large uncertainties.)"
In Section VI.3, Least-Squares Fluence Adjustment, they state:
"A major concern with the application of the FERRETadjustmentis that, while the adjustment 4
i l
l does provide a best-fit of the measured data, the dosimeter cross sections, measured reaction j rates and calculated spectrum adjustments are made without any physical basis. This application of the FERRET adjustment methodology to Palisades is presently being evaluated and the results of this evaluation willbe reported separately when completed. "
l That part of the above statement, "the dosimeter cross sections, measured reaction rates and l calculated spectrum adjustments are made without any physical basis," is not a correct l technical statement and it contradicts what is scientifically known and has been accepted (via j the established world wide ASTM voluntary technical consensus process) in the applicable ASTM E706 Master Matrix Set of LWR Physics-Dosimetry-Metallurgy Surveillance Standards; see Section 9. l l
All of the above three statements have to do with questions and answers about the validity of the procedures and methodology recommended in DG 1053 and the referenced ASTM and l ANS 19.10 standards. Obviously, such questions must be addressed and any needed clarif-ication should be provided in DG 1053 and the referenced ASTM and ANS 19.10 standards to add more certainty (and cost effectiveness) to the process of the regulation and safe operation of nuclear power plants.
How to " Add More Certainty" to the regulatory process through Standards Technology Development, Transfer and Training (STDTT) and the discussion of " Key Issues" are considered in Section 3.0, Subsection FLR & STDTT.
An important aspect of this letter / report is to clarify some of the misunderstandings that Lois, Carew, Aronson and others may have developed as a result of Worsham's two technical papers [Wo96a,Wo96b] that were distributed at the 9th ASTM-EWGRD Symposium on l
Reactor Dosimetry, Prague, Czech Republic, September 2-6,1996; copies of these two papers l are attached. By the incorrect interpretation and use of the Pool Critical Assembly (PCA) I
- reported FERRET-SAND 11 and LSL-MS2 least squares adjustment code results, Worsham
! concluded that [Wo96b]:
f
(
"When FTlbegan evaluation of the calculationalrequirements of the new draft guide, a clear l
implication was that surveillance capsule fluence calculations must be within the uncertainty of previous measurements. The " Margin" term associated with a) RG 1.99, Rev. 2, and b) the generic PTS safety analysis, requires that the uncertainty in the capsule calculational l predictions be less than or equal the uncertainty in the previous measurement predictions.
i Unfortunately, FTlfound that the previous fluence " measurement" predictions in the RG 1.99 embrittlement data base are biased. Investigations of a) the work of Simons, et alia, Lippincott, et alia, and Stallmann, et alia, and b) the reasons for the biased measurements, suggest that the biases are caused by the old FERRET-SAND technology. "
It is noted that Lois attended the 9th ASTM-EWGRD Symposium ana very likely talked to
'PWR-PV Mockup Experiments and Calculational Blind Test.
5
Worsham and received copies of his papers. 2 Worsham concluded that least-squares adjustment codes should not be used for obtaining the best-estimate value of the PV fluence.
This issue is given further consideration in Section 3.0 under the Subsection " Biased Fluences ;
in the Charpy Embrittlement Database." @
His two 9th ASTM EUROPEAN papers [Wo86a,Wo86b] were carefully reviewed and studied to see if there were technical merit for the conclusions presented in his two papers. A technical justification could not be found as reported in Section 3.0, Subsection " Biased Fluence in the Charpy Emrittlement Database."
Further, in the Minutes for the June 2,1997 Orlando ANS-19.10 Meeting [Lo97], it states:
"T. Worsham stated that he was concerned that the fluences in the Charpy embrittlement database may be biased. He noted that the use of LEPRICON is an interesting concept, but is concerned that these codes do not have a physicalbasis for the adjustment. In addition, he suggested that any fluence adjustment be made directly to the transport calculations rather than through FERRET and LEPRICON. '
The International-Interlaboratory PSF Physics-Dosimetry-Metallurgy Experiment yielded the highest quality and accurate physics-dosimetry results reported for a PW3 Pressure Vessel Benchmark Mockup Experiment [Mc87b]. The 5'Fe(n,p), 58Ni(n,p), 5STi(n,p), 53Cu(n,a),
2 '8 U(n,f), and 2 'Np(n,f) and other sensors used in this experiment are those recommended (in the ASTM standards) for use in PWR and BWR surveillance capsules to measure the flux and fluence E> 1 MeV. These are the same sensors used for the PSF physics-dosimetry-metallurgy experiments and in the Palisades surveillance capsules. The technical issues associated with the application of the ASTM Master Matrix Standards and FERRET-SAND ll Adjustment Code Methodology for determining the PNP "Best Estimate" of the fluence and answers to specific technical questions raised by NRC staff are provided in Sections 3 & 9.
Detailed supporting documentation is presented in Section 3.0, Subsection KEY ISSUES, that provides necessary justification to refute the f act that there is a need for having a " physical basis" to justify the use of a least-squares adjustment code for generating the most accurate "Best Estimate" fluence value for the determination of the PTS Rule pressure vessel embrittlement. Section 5.2 of the " ASTM E944 Standard Guide for Application of Neutron Adjustment Methods in Reactor Surveillance" specifies the " Requirements for the Use of Adjustment Codes in Reactor Surveillance." Additional commentary is provided in Table 3.1, PWR & BWR Surveillance Program Regulatory Instruments, and the Table 3.1 Attachment.
In view of FTl continued support of Worsham in assisting in the preparation and revision of
'the draft of the new ANS-19.10 " Fast Neutron Fluence to PWR Reactor Cavities" standard, Consumers Energy should encourage its staff to continue (along with Stan Anderson) its active participation in the standards development work for this new LWR physics-dosimetry j 2
it is notad that the peer review of the papers and subsequent publication of the proceedings of the 9th Symposium is still in progress; as such, only the attendees would have received copies of Worsham's two papers.
6
surveillance standard. Anderson has already prepared a draft of Section 5.0 of this standard on " Determination of Best Estimate Fluence;" see Enclosure 3. This draft is very well written and uses appropriate information extracted from DG 1053, ASTM Standard E706-(llE2),
" Guide for Benchmark Testing of LWR Calculations," and ASTM Standard E944 on " Guide for Application of Neutron Spectrurn Adjustment Methods in Reactor Surveillance." It would be very appropriate to use this write-up for inclusion in a future revision of DG 1053.
1 e
t 7
2.0 INTRODUCTION
Based on the extent of the reviewers knowledge, provide answers to the fel;owing questions:
QUESTION 1: Is the Consumers Energy Method (Westinghouse Method) of determining the palisades nuclear plant's best estimate fluence by combining transport calculation and dosimetry measurements technically sound?
aufST/ON 2: Is the method used consistent with the basis of 10 CFR 50.61, the PTS Rule?
DUESTION 3: Has Consumers Energy's Explanation of the Bias Between Measurement and Calculations Provided Sufficient Basis to Support the Magnitude of This Difference?
DUEST/ON 4: Has the Owner Communicated its Position Clearly and if Not Where Would Further Explanation be Useful?
In order to provide answers for these four questions it was necessary to:
- 1) Review and study the following relevant documentation [See Reference Listing):
- C. Serpan's Prior Papers Related to NRC's Research & Regulatory Positions on the Physics-Dosimetry Derivation of the PTS Rule's Fluence Estimate.
- A. Taboada's Prior Papers Related to NRC's Research & Regulatory Positions on the Physics-Dosimetry Derivation of the PTS Rule's Fluence Estimate.
- N. Randall's Prior Papers Fviated to NRC's Research & Regulatory Positions on the Physics-Dosimetry Derivation of the PTS Rule's Fluence Estimate.
- ~L. Lois's Prior Papers Related to NRC's Research & Regulatory Positions on the Physics-Dosimetry Derivation of the PTS Rule's Fluence Estimate.
- J. Carew's Prior Papers Related to NRC's Research & Regulatory Positions on the Physics-Dosimetry Derivation of the PTS Rule's Fluence Estimate.
- The New 1996 Federal Public Law #104-113 on the " Technology Transfer Improvements Act (TTIA) of 1995," Code of Federal Regulations 10 CFR Part 50, Appendix G & Appendix H, PTS Rule 10CFR50.61, RG DG-1025, RG DG-1053, RG 1.99 Rev. 2, and the ANS 19.10 Standard, " Fast Neutron Fluence to PWR Reactor Cavities." [Lo97)
- Relevant NRC LWR-PV SDlP Physics-Dosimetry-Metallurgy NUREG/CR-XX Program Reports.
- Relevant ASTM Physics-Dosimetry Related Standards and AfiTM Subcommit-tees E10.02 and E10.05 Meeting Minutes and W. N. McElroy's March 30, 1996 ASTM Task Group E10.05.04 Report on the E706 Master Matrix Set of LWR Surveillance Standards: See Attached copy.
- Relevant Prior ASTM EURATOM International Symposia on Reactor Dosimetry 8
Papers, with Emphasis on those Presented at the 9th Symposium, September 2-6, 1996, Prague, Czech Republic. This Symposium was Attended by Lambros Lois as a Result, at least Partially, of Discussions with W. N. McElroy and his Sending Lambros a Copy of the March 30,1996 E10.05.04 TG Report identifying Those World Wide Assisting in Revision, Drafting and/or Review of the ASTM E706 Physics-Dosimetry-Metallurgy Set of LWR Surveillance Standards.
- 2) Review and study the Owner's Palisades Nuclear Plant's (PNP) Information Package on Pressurized Thermal Shock (PTS) submittals to the NRC dated April 4,1996 and June 26,1997 and the NRC's December 20,1996 Interim Safety Evaluation (ISE).
- 3) Provide information on any additional "lSSUES" that are relevant to the discussions between the NRC and the Palisades Plant Staff.
Based on a verv carefulreview and studv of allof the relevant documentation andissues. the answers for Questions 1. 2. 3. and 4 are verv definitelv vest Commentary and supporting documentation related to the above questions & answers as well as the consideration of additional issues are provided in Sections 3 through 9 and the Enclosures / Attachments and Reference Listing.
9 l
t
l 3.0 FEDERAL LAW & REGULATIONS (FLR); STANDARDS TECHNOLOGY DEVELOPMENT, TRANSFER & TRAINING (STDTT); AND KEY ISSUES AND SUPPORTING TECHNICAL DOCUMENTATION THAT ARE RELEVANT TO FLR
& STDTT AND PAST & FUTURE DISCUSSIONS BETWEEN THE NRC AND PALISADES NUCLEAR PLANT (PNP) STAFF l
FLR & STDTT -- In order to conclude that the answer was yes to the four questions in I Section 2, it was necessary to review and study the new 1996 Federal Public Law 104-113 l on the " Technology Transfer Improvements Act (TTIA) of 1995, the Code of Federal Regula-tions 10 CFR Part 50, Appendix G & H, the PTS Rule 10CFR50.61, RG DG-1025, RG DG-1053, RG 1.99 Rev. 2, the referenced ASTM Master Matrix Set of LWR Physics-Dosimetry-Metallurgy Surveillance Standards, and the ANS 19.10 Standard, " Fast Neutron Fluence to PWR Reactor Cavities."
Contained in the new TTIA law is a provision 12(d) that codifies the existing Office of Management and Burfget (OMB) ircular A-119 on Federal Participation in the Development and Use of Voluntary Standards. The law directs that all Federal agencies and departments shall use "J'echnical Standards" that are developed or adopted by voluntary consensus standards uodies, using such technical standards as a means to carry out policy objectives or I
activities determined by the agencies and departments. The National Research Council of the Academy of Science & Engineering report related to the new law says that " voluntary consensus standards are often equally as stringent in the level of protection they require as l mandatory standards would be. It might seem reasonable to expect that private standards developers--industry associations, especially--would seek to set stanoards at the lowest common denominator of safety. .....In fact, however, private standards writers have several incentives to set high standards. Forestalling govemment regulation by developing a private solution to a perceived problem require' a standard stringent enough to satisfy public needs.
(Government participation in standards committees enhances this process from both public '
and private perspectives.).. " [Ko96)
The definition of " Technical Standards" as used in Subsection 12(d) means performance-based or design-specific technical specifications and related management systems practices.
The National Institute of Standards & Technology (NIST) has posted an implementation plan for Public Law 104-113 on the Web [http:llts.nist. gov /ts/htdocs/210lplan.htm]. The law directs NIST to lead a national effort to coordinate standards and conformity assessment activities among federal, state and local government agencies, and the private sector.
.The Private Sector, ASTM, ASME, ANS, NRC, and DOE should consider rneeting with the NIST staff to develop a coordinated national STDTT effort to more effectively specify, manage, and make use of the Code of Federal Regulations 10 CFR Part 50, Appendix G &
Appendix H, the PTS Rule 10CFR50.61, RG DG-1053, RG 1.99 Rev. 2, the ANS 19.10 Standard, and the ASTM E706 Master Matrix Set of Physics-Dosimetry-Metallurgy LWR Surveillance Standards. The use and modification of the FTR Dosimetry Handbook [ lib 3) would provide a good starting point for such a coordinated ef fort; that is, it could be updated t
and modified to serve as an important reference source for research, test, and power reactor
{ applications and surveillance programs. A tremendous amount of LMFBR research and 10
development money was used to develop and maintain the HEDL National Dosimetry Center, which is still being maintained, but as the PNL National Dosimetry Center.
In an 8th ASTM-EURATOM Symposium paper on " Reactor Vessel Dosimetry Assessment:
Perspective of the Materials Engineer" by Steele et al., it is stated:
"Materialand dosimetry specialists, who develop andimplement surveillanceprograms and resolve regulatory concerns for individual reactor vessels primarily utilize three ASTMpublications: ASTMStandardE853-87, ASTMStandardE1005-84 (Reapproved 1991), and ASTM Standards Technology Training Course and Workbook (Mc91]. "
With the assistance of experts and careful use of all referenced standard guides, practices and related references, the materials specialist and other appropriate reactor plant personnel should be capable of determining the effect of neutron exposure in terms of remaining reactor vessel design and safety margins. A case history dealing with the Yankee Rowe reactor vesselissue is presented next. Prior to looking at this case history, the current AS7'M-specifiedpath andpotentialimpediments to resolving outstanding issues are assessed. The pattern of Standard E853-87 sounds straightfor-ward and is summarized as .follows: ..... .. . .... . .. ... .... .
There is predominant emphasis on core physics, operating history, and dosimetry computations. In addition, E853-87 refers to 21 other standards and guidelines, has 70 references, and refers to major NRC Code of Federal Regulations and ASME documents. For materials specialists who are not dosimetry specialists, this approach is highly complex, time consuming and difficult to use. ........ ... .. "
To the three ASTM publications mentioned above, one must add others from those discussed in the March 30,1996 E10.05.04 TG Report; such as ASTM Standards E185, E482, E560, E900, and E944 as well as NRC RG 1.99, Rev. 2, the Draft DG 1053 on Calculational and Dosimetry Methods for Determining PV Neutron Fluence, and the ANS 19.10 Standard on Fast Neutron Fluence to PWR Reactor Cavities.
The complexity and difficulty of having to understand, apply, and use all of these different standards and regulatory instruments is very great and challenging for the U.S. nuclear industry. This must be kept clearly in mind for those accepting the responsibility for the development, drafting, and revision of the ASTM E706 Master Matrix Set of LWR Surveillance and ANS 19,10 Standards.
Additional commentary on Codes, Regulations, Regulatory Guides, and how the ASTM E706
' Set of LWR Surveillance Stande,rds are identified through the ASTM E185 Standard as a part of Appendix H to 10 CFR 50 and as such, by reference, are made a part of CFR will be found on pages 2 thru 8 of the March 30,1996 TG Report, Lowe's commentary on the " Critical Role of Neutron Fluence in Reactor Vessel Integrity" will be found on pages 7-8 of f.he March 30th Task Group report, Enclosure 4. He states:
In the U.S, the situation is enhanced by the wording of 10CFR50, Appendix G. While there is reference to change in materialproperties, nowhere in this appendix is there 11
any reference to neutron fluence, dosimetry, or analysis procedures. To assess the effects of neutron fluence on the material properties as needed for Appendix G analysis,10CFR50, Appendix G references 10CFR50, Appendix H. A review of this appendix has minimum reference to the neutronic requirements of the surveillance program except for an integrated surveillance program: "... there must be an adequate dosimetry program for each reactor. " The remainder of the requirements are buried in ASTM Standard E185. There is little wonder that the role of dosimetry and fluence l analysis in reactor vesselintegrity is not appreciated....
J The publication of 10CFR50.61. .... was the first regulation that highlighted the role 1
of reactor vesselneutrcn fluence on operating integrity. This regulation establishes a l
direct relationship between the reactor vessel screening limitations and the reactor vessel neutron fluence....
l The PTS issue greatly increased the industry's appreciation of the importance of the measurement, accuracy, and precision of fluence calculations.
1 I
In summary, reactor vesselintegrity is based on a number of properties that must be l understood, including nondestructive evaluation of the critical beltline materials, knowledge of effects ofirradiation on materialproperties, and accurate knowledge of neutron fluence. Like a three-legged stool, all three elements are essential to ensure \
reactor vesselintegrity. "
l l With the passage of TTIA, the opportunity now exists for the nuclear industry to bring much !
l more certainty to the process of the federal regulation of the safety of the operation, surveillance programs, and decommissioning programs for nuclear power plants. As examples, additionalinformation is provided in Attachment I on how ASTM voluntary technical standards are bringing much more certaintv to the process of Environmental Assessment and Environmental Risk Management. This information was taken from Reference [Mc95].
PWR & BWR SURVEILLANCE PROGRAM REGULATORY INSTRUMENTS Table 3.1 provides commentary on PWR and BWR surveillance program regulatory instruments. With this information in hand, it appears that the Regulatory Guide DG 1053 methodology does not recommend and support the determination of the "best estimate" of the EOL pressure vessel fluence value by use of the least-squares method and the adjustment l l codes recommended in the E944 Guide; i.e., E944 is never mentioned or referenced in DG i 1053. As such, the DG 1053 procedures and methodology lack the required acceptance (technical consensus) from the industry and the public, and is, therefore, not appropriate for j providing regulatory guidance as it is now written.
l .
That is, this review of the Regulatory Instruments used for the regulation of surveillance !
programs for PWR and BWR pressure vessels clearly demonstrates that the recommended methodology of the new DG 1053 is not consistent with that recommended in the ASTM E706 Master Matrix Set of LWR Physics-Dosimetry-Metallurgy Surveillance Standards that have been developed and adopted over the last - 35 years on a world wide basis by the industry and the public through voluntary technical consensus due process; see Section 9.
Since DG 1053 lacks world wide voluntary technical consensus due process, it needs to be 12
1
[.
revised to be consistent with the procedures and methodology recommended in the ASTM E706 set of standards. Further, the implementation of the use of the DG 1053 as it is now written adds confusion and make it very difficult for licensee's to remain in compliance with Federal Law as now specified in 10 CFR Part 50, Appendix G & H, the PTS Rule 10CFR50.61, .
Beg. Guide 1.99, Rev. 2, and the ASTM E706 Master Matrix Set of LWR Surveillance Standards (that by reference to the ASTM E185 Standard in Appendix H has made the entire l set of E706 standards a part of Federal Law).
l In this regard, on pages 65458 & 65459 of the Federal Register l Vol. 60, No. 243 ITuesday, December 19, 1995 / Rules and Regulations, changes in Appendix H of 10 CFR 50 are discussed. It is stated that:
"The otherprincipalchange to Appendix H clarifies the version of ASTM Standard E185 that aoplies to various portions of the surveillance programs. Appendix H recognizes the need to separate surveillanceprograms into two essentialparts. Specifically the design of the program and the subsequent testing and reporting of results from surveillance capsules. Because the cesign of the surveillance program cannot be changed once the program is in place, the requirements for design of the surveillance program are static for each plant. However, the testing and reporting requirements are updated along with technicalimprovements made to ASTM standard E185. "
i I
13 j i
l l
T i
l TABLE 3.1 l
l PWR & BWR SURVEILLANCE PROGRAM REGULATORY INSTRUMENTS l As discussed in [Mc88], for each LWR nuclear power plant, the physics-dosimetry-i metallurgy surveillance program requirements are identified through the ASTM Standard E185 on " Practice for Conducting Surveillance Tests for Light-Water Cooled Nuclear Power Reactor Vessels, E706(IF)," which in turn is identified as a part of the 10 CFR PART 50, Appendix H.
The requirements for design of the surveillance program are static for each plant.
However, the testing and reporting requirements are updated along with technical improvements made to the ASTM E185 Standard.
Section 8.2.1 of ASTM E185-94 states that: "Theneutron fluence rate, neutron energy spectrum, and neutran fluence of the surveillance specimens and the corresponding maximum values for the reactor vessel shah be determined in accordance with the guidelines in ASTM Standard E482 on " Guide for Application of Neutron Transport Methods for Reactor Vessel SurveiMance" and ASTM Standard E560 on " Practice for l Extrapolating Reactor Vessel SurveiHance Dosimetry Results. "
Section 8.2.2 states that: "The specific method of determination shah be documented. "
Section 8.2.3 states that: "The Neutron fluence rate and fluence values (E> 0.1 and 1 Me V) and dpa rate and dpa values shallbe determined and recorded using a calculated spectrum adjusted or validated by dosimetry measurements. "
There is no existing regulatory requiremont that the prediction of the vessel fluence must be made by an " absolute" fluence calculation in which the transport of the neutrons from the core is calculated out to the vessel cavity, rather than a simple spatial extrapolation of the fluence measurements. This is a new requirement in DG 1053 and it is inconsistent with the existing Federal Law's regulatory requirements of the ASTM E706 Master Matrix Set of Physics-Dosimetry-Metallurgy Standards. Furthermore, this provision has not been considered by ASTM Committee E10 and subjected to ASTM technical
. consensus due process.
l Draft Regulatory Guide DG 1053 also states that:
" Compliance with this guide is not a regulatory requirement of the USNRC. i However, if a licensee elects to use this guide to determine pressure vessel neutron fluence, implementation of the guide would not be satisfied unless the licensee complies with certain specific provisions identified in the Regulatory Position of the guide."
14 l
l
.a
TABLE 3.1 (Cond't)
- Section 3.1.1 of ASTM Guide E482 states that:
"The methodology recommended in this guide specifies criteria for validating computationalmethods and outlines procedures applicable to pressure vessel related neutronic calculations for test and power reactors.
- Section 3.2 Validation states that:
Prior to performing transport calculations for a particular facility, the computational methods must be validated by comparing results with measurements made on a benchmark experiment. Criteria for establishing a benchmark for the purpose of validating neutronic methodology shouldinclude those set forth in ASTM Standard 944 on Guide for Application of Neutron Spectrum Adjustment Methods in Reactor Surveillance" as well as those prescribed in Section 3.2.1. Requirements for Benchmarks.
- Section 3.4.8.2 states that:
Use a spectrum adjustmentprocedure as recommendedin Guide E944 using calculated group fluences and dosimetry data with uncertainty estimates to obtain an adjustment to the calculated group fluences and exposure parameters. Predictedpressure vessel fluences could then incorporate the spectraland normalization data obtained from the adjusted fluences.
- Section 4.2.3 of ASTM Practice E560 states that:
1
" Guide E944 should be used to combine the transport calculation with the dosimeter results. The E944 adjustment procedure should be used to indicate whether the dosimeter measurements and associated uncertainties are consistent with the transport calculation and with uncertainties implied from benchmark tests of the transport code (PCA, VENUS, NESDIP, andappropriate CommercialBWRorPWR). Having established the required consistency, the adjusted transport code results may be used to calculate the neutron field at all points in the pressure vessel wall with the uncertainties estimates derived from the application of the adjustment codes. Direct use of the transport code results with the appropriate (experimentally determined) bias factors and uncertainties is another acceptable approach. "
- Section 3.1 of ASTM Guide E944 states that:
" Adjustment methods provide a means for combining the results of neutron transport calculations with neutron dosimetrymeasurements in order to obtain optimalestimates of neutron damage exposure parameters with assigned uncertainties. The inclusion of measurements reduces the uncertainties for theseparameter values andprovides a test 15 i
I
I TABLE 3.1 (Cond't) l l for the consistency between measurements and calculations and between different measurements. This does not, however, imply that the standards for measurements and calculations of the input data can be lowered; the results of any adjustment procedure can be only'as reliable as are the input data. "
i
- Appendix G of 10 CFR Part 50 specifies the:
j l
" fracture toughness requirements for ferritic materials of pressure-retaining 1
components of the reactor coolant pressure boundary of light-water-cooled nuclear power reactors. "
Appendix H of 10 CFR Part 50, " Reactor Vessel Material Surveillance Program Requirements" "Provides the rules for monitoring the changes in the fracture toughness properties of the RPVbeltline ma terials due to irradiation embrittlemen t using a surveiHanceprogram.
Appendix H references ASTM Standard E185 for many of the detailedrequirements of l surveiHance programs, and permits the use of integrated surveillance programs, l
wherein surveiHance program capsules for one reactor are irradiated in another reactor. "
- PTS Rule 10CFR50.61 requirements state that:
l "For each pressurized water nuclearpower reactor for which an operating license has been issued, the licensee shah have projected values of RT,7s , accepted by the NRC, for each reactor vessel beltline material for the EOL fluence of the material. The assessment of RTers must use the calculation procedures given in paragraph (c)(1) of
\ this section, except as provided in paragraphs (c)(2) and (c)(3) of this section. "
Section (c) states that:
" Calculation of RTers must be calculated for each vessel beltline material using a fluence value, f, which is the EOL fluence for the material. RTers must be evaluated ,
using the same procedures used to calculate RTw7 as indicated in paragraph (c)(1) of i this section, and as provided in paragraphs (c)(2) and (c)(3) of this section. "
Section (c)(1)(iv)(B) states that:
"Iis the best estimate neutron fluence, in units of 10" n/cm (E> 1 MeV) at the clad-base metalinterface on the inside surface of the vessel at the location where that materialin question receives the highest fluence for the period of service in question.
As specifiedin this paragraph, the EOL fluence for the vesselbeltline materialis used in calculating sRTers. "
16 1
l TABLE 3.1 (Cond't)
Regulatory Guide 1.99, Rev. 2: In Section B, Discussion, it is stated:
"The basis for Equotion 2 for 4RTwis contained in publications by G.L Guthrie [Gu841 and G.R. Odette et o'. [Od841.
"The measure of fluet.ne used in this guide is the number of neutrons per squate centimeter having energies greater than 1 million electron volts (E> 1 MeV). The differences in energy spectra at the surveillance capsule and the vesselinner surface locations do not appear to be great enough to warrant the use of a damage function such as displacements per atom (dpa) in the analysis of the surveillance data base
/McB7a/. " (It is noted that Guthrie made use of the reference [Mc87a surveillance capsule derived FERRET SAND ll values of fluence while Odette, primarily, use derived fluence values based on only the measured "Fe reaction rate. The Table 3.1 Attachment provides additional information related to the basis and uncertainties associated with the procedures and methodology of RG 1.99, Rev. 2.)
It is also important to know that RG 1.99, Rev. 2 is essentially the same as the ASTM Standard E900 on " Guide for Predicting Neutron Radiation Damage to Reactor Vessel Materials." RG 1.99, Rev. 2, therefore has been developed on the basis of ASTM voluntary technical consensus due process. It is stated in Section 1.3 of E900 that:
"This guide is Part IIF of the Master Matrix E706 which coordinates severalstandards used for irradiation surveillance of light-water reactor vessel materials. Methods of determining the applicable fluence for use in this guide are addressed in Master Matrix E706, Practices E560 (IC), and E944 (IIA) and E1005 (Illa). The overallapplication of these separate guides and practices is described in Practice E853 (IA). "
l ASTM E900 and RG 1.99, Rev. 2 were developed simultaneously under ASTM voluntary consensus due process, The Regulatory Guide DG 1053's methodology requires that:
l "The prediction of the vessel fluence must-be made by an " absolute" fluence
- calculation in which the transport of the neutrons from the core is calculated out to the vesselcavity, rather than a simple spatialextrapolation of the fluence measurements. "
This is in conflict with the requirements of ASTM E900 and RG 1.99, Rev. 2 and
,, the DG-1053 procedures and methodology hat lot been subjected to ASTM technical consensus due process.
l 1
1 1
17
l I
TABt.E 3.1 ATTACHMENT The following Reference [Mc86) commentary was taken from Section F, Trend Curve Data Development and Testing, prepared by W.N. McElroy, R. Gold, E.P. Lippincott, R.L. Simmons and S. Anderson:
"The status of the development and application of new advancements in L WR-PV-SDlP, such as cavity physics-dosimetry for improving the reliability of current and end-of life (EOL) predictions on the metallurgical conditions of pressure vessels and their support structures, is discussed with appropriate referencing to the currentliterature, Federaland NRC regulations and rules, and the new series of 21 ASTM LWR Surveillance Standards. Application of established ASTM standards is expected to permit the reporting of measured materials property changes and neutron exposures to an accuracy andprecision within bounds of 10 b tn 30%, depending on the measured metallurgical variable and neutron environment.
NRC Physics-Dosimettv Comnendium The NRC physics-dosimetry compendium (Mc87a) is a collation of information and data ceveloped from available research and commerciallight water reactor vessel surveillance program (RVSP) documents and related surveillance capsule reports. The data represents the results of the HEDL (Simons) least-squares FERRET-SAND ll Code re-evaluation of exposure ,
units and values for 47 PWR and BWR surveillance capsules for W, B& W, CE, and GEpower plants (see Figure HEDL-22). Using a consistent set of auxiliary data and dosimetry-adjusted reactor physics results, the revised fluence values (Table HEDL-22) for E > 1 MeV averaged 25% higher than the originally reported values. The range of fluence values (new/old) was from a low of 0.80 to a high of 2.38. l l
These HEDL-derived FERRET-SAND ll exposure parameter values are being used for NRC supported HEDL and other PWR and BWR trend curve data development and testing studies.
These studies are providing results to support Revision 2 of Regulatory Guide 1.99. The information in the compendium is also being made available to the ASTM E10 Committee, to tne Metal Properties Council (MPG) Subcommittee 6 on Materials for Nuclear Reactors, and to others developing improved data bases and trend curves. These curves are used by the utilities and by the NRC to account for neutron radiation damage in setting pressure /
temperature limits, in making fracture analysis, and in predicting neutron induced changes in reactor PV steel fracture toughness and embrittlement during the vessel's service life.
As stated in Subsection " Key Issues" under " Fluence issue".
'."A survey of the fluence results for the A3028 correlation monitor materials made it apparent that the Guthrie data, which employed the NUREG 3319 re-evaluated fluences, provided a better trend curve, indicating the need for, and the potentialbenefit to be gained from, a self consistent basis for expressing capsule fluence. The impact of the reevaluation on individual cata from the Odette et alanalysis are apparent.
Many industry neutronics analysts have argued that such dramatic effects on fluence evaluation are unlikely to exist in more recent surveillance data, which "should" have been 18
l 1
WESTINGHOUSE ~1" x 1" SQUARE CAPSULE LOCATIONS
- ( 2 x 2 ARRAY OF CV SPECIMENS WITH STEEL VOID FILLER)t j f COMBUSTION ENGR. N1" THICK x 2" WIDE CAPSULE LOCATIONS M (1 x 3 ARRAY OF CV SPEC: MENS WITH STEEL VOID FILLERif O (BABCOCK AND 3 x 3 OR 4 x 4 ARRAY OF CV WILCOX SPECIMENS WITH ~21/2" DIAMETER ALUMINUM VOID FILLER)? CAPSULE LOC l g GENERAL ELECTRIC ~ 1/2" THICK x 6" WIDE CAPSULE LOCATIONS (1 x 12 ARRAY OF CV SPECIMENS)t RANGE OF ANTICIPATED THERMAL TO FAST (T/F) FLUENCE RATIOS FOR DIFFERENT RADIAL &
SURVEILLANCE CAPSULE q POSITIONS
, , l
- P Og,,
WATER Y "
O, g
r o.
f/ RIACTOR VESSEL CAVITY
- 7 WATER o g, PRESSURE VESSEL WALL j
O #j. POSSIBLE SURVEILLANCE
, CAPSULE LOCATIONS 6 'SOME CAPSULES ARE EMBEDDED PARTIALLY OR FULLY WITHIN THE OUTER BACK SURFACE OF THE THERMAL SHIELD: THEREFORE, THE T/F RATIO FOR THE FIRST LAYER OF CHARPY SPECIMENS (FACING THE REACTOR CORE) WILL BE VERY LOW (<1), WHILE THE LAST LAYER (FACING THE
, PRESSURE VESSEL) WILL BE MUCH HIGHER (>1),
i SHROUD, BARREL, THERMAL SHIELD, OR THERM 3 PAD (SEGMENTED THERMAL SHIELD) COMclNATION REACTOR CORE REGION
" f(XY) HORIZONAL CROSS-SECTION AL ARF.AY, NOT AXIAL l
- DEPEMPNG ON SURVEILLANCE CAPSULE l DESIGN 8 PLACEMENT HEDL 8400-063 FIGURE HEDL-22. Schematic Representation of In-Vessel Surveillance Capsule Designs and Locations for Operating PWRs and BWRs.
l 19
- - - - - - - - - - - - -----________--_-----------_j
e r %%e %%e em %%e%%%e %%%%%eemet %%% %e%%
e 4 000 ve 3. C. Q. ee + C. O. 3 0+ C 0 0 0. C. C
++++O O C C Q. C. O. C. O. O
+ O. C. C. C + C. C.
gw www www wwwwwwwwww wwwwwwwwww ww wwww e t m-e mee w-e Nem e~ee SONN ee%ew eweNNmNeme ---em%.
O E c% eN= veN-eeMoew e%-cNaN a- e. w.h. %. .C.e.
um N. e. m. e. d. w. C. M. O. N. . C. m. w. e. N. N. m.m. e.c.a.@. @. C. @.
= e%N emN =N-mw-mmm- wweww--w-- mwwe-em e
=4 41 emN mom
-e wCN mehoe-ee-e m-MN=N =*
e-me4%-mNm NNmNNNehme
%-wN-e%
%e%eN-49 w
A e
-OO C
O m
C000000000
m-----
cococococo ocococo
--~~~-- -
0 e e e e e e e e e a e e e e e e e s e a e e e e e e e a e e a e e o N www www wwwwwwwwww wwwwWwwwww wwwwwww g
@ s owe e@N em%eeemece -%eomeoNw. Nome %-w
-U w
y C. N. m. m. Q. e. M. N.e.w.c. o.m. N.=. =. w e.. c. N. e. N. e. e. w. e. . N.N. N.w. .m. -
e== e-m Nw-m=----- NNNNN-NN-N -------
O d M w Z c N
+ --- --- ---------- ---------- -------
NNN NNNNNNNNNN NNNNNNNNNN NNNNNNN w
< 9 e e o NN.N e s 4 a e e e e e e e e e e a e e e 6 e e a e 6 6 e e e e w U q h WwW www wwwwwwwwww WwwwwwwWWw wwwwwww N@N Ceh %emNNehe-e ewmmmmme%m ww--wen W ,
- e. c. @. %. @. @. e. c. @. N. e. c. e. c. e. w. @. %. e. N. N. N. %. e. @. c. c. e. @. @. @. e. @.
=
w y g --- --~ --~~~~- == yo-----e-- n------
g
< e J m --- ~~- ---------- ~~-------- --n----
,J -
e NN%
-NN mmo Nhh
%Nmm% eNm--
N-N=NN----
emwNmmmmNw
-N----
m%eNmme NNN==-N b
e
= www www wwwwwwwwww wwwwwwwwww wwwwwww &
W w e en- 4%Ce 40 -C y H Ne wwe @we demC-ewN-m NNeNw@c-%c wmNedwo e E w N e + AO-m = e Q--O-wONew eew@ w C g m w e N-N- - - --N---g--w coo. O w M
4 a . . O. C. . C. C. C. C. C. . O. . . . O. C.000000000 0000 . C. C D QQC CCO CCCCCCCCCO CCCCCCCCOO C000C00 -o
> w 1 - --- --- ------~~~~ ---------- ------- 3
> mNe MMe adeemeNm*@ ewamwee
- E, 0
. -m-www
~NN www wee-WOam09 -
~c~N-~
wwwwwwwwww N NN~c~-~-~N wwwwwwwwww
-w~~N wwwwwww
~
y a -N wg mee eem
--- --m emeemmemme m-m ---e-m meeeeeeeem-m----m---
emememe---m--- e.
. y +++ +** * +++* **++* + + ++***++* WWWWWwwwww E -=C% C www www wwwwwwwwww wwwwwww w W
O wVw wmm e@e WN-Ca@NwCN -ewAMCeem= @-mANO@
W w e. w. m. C. N. e. M. a. m. w. e. @. e. w. e. N. N. N. e. m. N. N. e. Q. w. N. N. C. =. =. m. e. N.
w NeN NmN -e-emeMMcN e@-m-mN--w mwMMemm W -
~ , a "
N w - e "N Z d mCN e- emcee-w emw-coNw -N-emo- 5
- i - -
. N. c. N. . N. m. e.N.$. ee. w.
. b. b. NN.*-----m-Om e W ~ e. c. . M. M. . . N. o. N. N. e. . e. . e. . a d
C W E w
--- NN- --m me--m- --m-m
- v b L W M Z C 0
= w _C . --- --- ----------mwNe-ee New wo
------- =oowooC NNNCQeN
@ b y --
- e New e-NN NN- N- NN - ----m--= =oo = NNN-- N g
]
W Z > e www www wwwwwwwwww wwwwwwwwww wwwwwww .
J w
U IC mee eee seemmememe meeeeeeeem ememeem UI [
< g - H -..
. +++
...*+++++.
WWWWwwWWWW
+ +++++++
WWWWWWWWWW wwwwWWW
+ ++++++ 4 a
4 g , - www www
-emw- e w . @Nm N b 4 w
H w
. .88 m@N
- m. e. N.
Nec
-w-w2.mNeO.
- e. o. m.
-N-NMNNMN-
- m. e. N. N. m.mN.w N C. . e.m
@@-c@eNNeN
- w. -w.e C.m e NN. N. N. e. N.
NMMMMNN
- c. o. o. E. gm. w O
g he a .
w a -Co Z e mee See ememmememe emmemmemme mememme E e e C 4
m
+ ++ ++*
www www
==......=++
- +
wwwwWWWWWW
. . =+WWWWWWWWWW
+++++++ ..+++++
wwWWWWw em 3 4 w b o 6[*
w T- ..b M eve e 6 :3 N. A. w. ameeE-SNec O O e -
w o. C. N. w e. N. C. . e. . C. w. h. N.
. - e S.S.e.eC.N
. N. w e=
- m. w. o. S.NNONE.@.
- o. e. m. e. = t C
3 Nw= -Ne me-@MTNmN- echecNNgeN -NNmeNN Ot d
< - -- - -- -- - h$LwC.
C .S v y 0 --- --- ---~~mnegn --~g---gge ag--g-- ==S O mee@m@ed N*wMA@@ mee NNee"eNe MO ***NW
=
-w CNN NNg
-Ne NNN NNNNNNNN"e AmNNNNNNNN N%AN eeWN GC =~
w at
-m b gg
-t>
00g EEh OgGO90@ tQ SWEE>Eb&w 9 49 9944G94 Q>>E>>>he>
099tet9 EmQ&>E>
3-G b eC 9
% e www www wwwwwwwwww wwwwwwwwww wwwwwww e . 3
- D e#&
315 0*
a . C' O g 1
%, e*o w Ee*
- x == - -- - -- -
Sl 3l 3 3 3UUUUUM IU! 3 3 3! g o O - . .
WW- w W C
-.*b@
W p # g n 6 g- >>>*23> -y ",
g (w2 (Q6 e>eme>>MD5 DDEDDeMD>M a _3-C de ,
< C .N l
O
> $ mmwwNN-NmN mN-- m-NNN -NmomNm M4 CW b* B 4 l
W 4e l, i e W -d
- yu COw ON b
i E WD w 1 e$-
q w vs T wwwwee && W9 9A$e6 e
g
&mg m.
&@ ggg .C .C .C .C .C-C g TCC 49 AAAAA vuvuQ -*
CC me-*9 W"
w 94444 00 9w" 0000 C
I.
I. bb b www 666621 mg *-@@
.C--
CC .w.v.
ememe W 66 a = =
3 "W
& C hhhh 3 de CE 7"**
e w
m Gwww a em e e mg e -- e e bd C *
- 4m Uw- CC 49000-Cf "f e
WMW e CC ge b b b N gg b b . 600000 . CC***9 *
- t 3 3 33 . . 3 &&MMWh&bbb O-"NNNe
- 3 y y de >>>>E3e Em 20
l l
}
. ~~~e, ~~~~~~ ~~~~gesege,e
- 7 gw
????T
-wwww ??.????
ww w-w .? .? * ? .wwww.Y?.ww...?
Y R e~~ .,. e. m.-
a m m e m. m m m. u ~ ~ ~
E t
~. e ee . ..m . e n.~ ~. ~. . ..+... ~. ~. ~. ~ ~.
. . ~ . . ...
~en--
m
. S. mm u e.~
e.
emee-
. i RO3R$ ***"** S
- , - ~ S*{R8.m~*8.RSS
~~ ~ m~~
w
~
e o o- -- -- oco oooooooo o WWWwW WwWWWW WwWWWwWWWWWW
= .~me-acoa" #-ge#~
$c .~~e ~.e.e~g
~.*9"*.~. gm._. *** -
g
---ee m.mmmm -~-e.m~- m~~ @
- . t
- s. ~.~.~.~.~. ~ . ~ . ~ . ~ . ~ . ~ . ~.~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ . ~ . -
WwWWW WWwwWw WwWWWWwwwWWW t
- 2. . ....
~. ~. e. ~
- m. em.~.
e.m. m -----
o.
m ,
- o. . o. m.
. . m.
. . . m. m... . .
- --- ------ .e, mem
~~-
co--ee wwwwww
~
~~----
wwwwwwwwwwww e~~~~~ -~~
~ - .
g wwwww e M A m
- ~~. M. M - ~ -e w
e- .- - -
e e m. ~m omoe. . o-- . ~. ~. -
,. ~EB~S. . . . . 8.8.8.8.8 B.@.5.8.8.3.8.5.8.8.8.8 ,
ocooo oooooo cococcoooCoo ,o
. -orm-
@.w~N m##
~~~@~~~~.~~~~@~
wwww wwwwww wwwwwwwwwwww g
.N3y eeece mamme~ ocemoocococo
^
mos wwwwW
. ~. . ....*.
wwwwww
~* ~.. N. ~. ~. ~.N* ~+ ~.~.
wwwwwwwwwwww w e
Q ~ * *
.memme g
w y
.$82~.
~. .... 8. .. ce..m.S.- m m -e..~.~.
c o o . ~e. #. o.. . #. ..
~
C w ~mm-N -~ --eeN---~~-- -
o u , ;
v - e.e=~ oc~c~m e - m e #
o s m. . m. e. e. e. c. o. c, o. . e. o. o. ~. -
c E
~
- -oo on -- o - - -
s N ~6~ < U. A i *
- a 8 a -
a gem ~
~ --- -ooo.e
~ --
-~-
~ ~~~~~~.
~
a o
y > g wwwww wwwwww wwwwwwwwwwww a
= l - eweem ~ ene~ eeeeeeeeeece ce E w
- w wwwww wwwwww
. ....- wwwwww wwwww
...... ,8 a
^
a s ~X e m .' eS.o.cee o. ~e. ~m a36
. 8.m.c. u wk 8.m~em~.
. -~
.~. e.
. -m-e
- o. o. . ---.m ~.m
- e. ~..- ~ ~.
u .m==. . '
. e.
>o
. 8 T-emeem ~ ~~=~ ee me. eeeee %.
.o
~, 8
....+ ..+... .......+.... r
- , WWWWW WWWWWW WWWwWWwWWWWW ,, .[
o o 6
- a. o R. E. o ~. S.. m.m e ~o. u ~. 8.3 2.. 3.5.~m.~. .em~ # .:=
5 -w
-es e-e~-~
~-- . m u - .mu~. ..
. a. 8 6*.-g J.
-~..~w,
- ~ ~ ~
~. ~~
-~~wwv~--
.~.~.. a,
- .- .t w.w8
- -e.S. ~1 ~ . ~ . ~ . ~ . ~ . -~~ .w1111 22
.o e
g.
~.
te3>&
wwwww
= ~
wwwwww wwwwww 4 ~111.
wwwwwwwwwwww 3.-~.
c.-
,-g.c
.+a
. sie.S E =5 w.
$ *W* e -m'
- ' a -
Ewk315 ww
- 11 ma 3e.us.d s
3DM3313 2 W A W G Wl *
=n-
..a "
- w -
a-o.u m
. ...~ ~
...~ - ..-o w ..*t g3 we N
3-N3 N
www<ww ...
....M.. e .-..
MmMm *y MM s- em
@ C
- e 2"g.Ng 5 t
.t
--um--
6 w mmmm---~~nnu ..-i*t<o v
. . 6
. 83w-w
_~ ,,
w W
- ttt 5 .A v t- . ......
..... ~ t E " .*
8 . w . -.. e. m . w.
3 .-- g -}. - - , . ,
z jg.jg. wwwwwwww
-=g.- . .=
. ' ,w g g. .
j v
2$ 222 8I 28
.t ., .32x ###di$ s U
4 o
. ., c 1 ,. ,. ,. ,. ,. ,. ,.
A &&&&&&&&&&
Edwmw 21 1 l
1 a
l I1 evaluated on a self-consistent basis. Interestingly, including more recent surveillance data from the PREDB for the correlation monitor material with data in Rev. 2 increases the scatter alarmingly, suggesting that the more modern data are no more consistent than the originalRG 1.99, Rev. 2 surveillance data. It is therefore concluded that an uncriticalstatisticalanalysis i of the raw surveillance data base is unlikely to reduce data scatter and hence the margins. On tne contrary, it is more likely to increase the margins. This would seem to be a strong argument for continuation of the exercise initiated in NUREG 3319. "
Fortunately for the U.S. Nuclear Industry, Lippincott and Anderson have continued the NUREG j 3319 exercise for Westinghouse surveillance capsules and have carefully reported and !
cocumented the results [Li94,Li961. They state [Li961:
"A systematic analysis of dosimetry from Surveillance capsules from Westinghouse plants has shown that excellent consistency of results is obtained. The indicatedprecision of results for both calculations and measurements falls well within the limits required for regulatory compliance.
. ....the measured surveillance capsule results for all the domestic Westinghouseplants were reanalyzed using a consistent methodology to derive the fast neutron fluence (E > 1 MeV).
The studyincluded a totalof 131 capsules which were originally analyzed between 1970 dnd early 1994. The comparison of calculated fluence for these capsules with the measured l results provides a large body of results for establishment of the precision of fluence 1 cetermination. . . .In the reanalysis of capsule fluences from dosimetry measurements, an aoproach slightly different from that typically used for the analysis ofindividualdosimetry sets was adoptedin order to maximize the consistency of the results. The assumption was made that variation of the neutron spectrum for capsules in symmetric locations in the same or similarplants is smallcompared to uncertainties in the measured reaction rates. The analysis procedure was then carried out using the following steps;
- 1) Reevaluate the dosimetry reaction rates for all capsules.
- 2) A verage normalized reaction rates for similar capsules.
- 3) Derive a best estimate spectrum for each capsule location using the average reaction rates.
- 4) Calculate a measured fluence (E > 1 MeV) for each capsule using a weighted average of the fluence indicated by each threshold dosimeter. "
l.ippincott states in Ref. [Li94]:
"Almost allof the capsules with large changes from the original values were included in those analyzed by Simons and reported in NUREG 3319. Comparisons with Simon's results are presen ted in Table .. .. ... .... This table contains the original values for fluence (E > 1 MeV),
the updated values from the work reported in this document, and Simon's values. Most of the updated fluence values are in reasonable agreement with Simons. The exception is the l values for San Onofre Unit I where Simons is substantially higher. The difference probably ;
arises form the spectrum used to derive the fluence from the measurements and to the lack 22
}
l
(
of fission monitor data. The comparisons with the calculated fluences for these capsules indicates that the values in this report are probably correct. On the average, the fluence results from Simons (~ 1985) are 32 % higher than the original values, and 11 % higher than the Westinghouse (~ 1994) updated values. Expected bias between the Simons fluences and the updated values is about 5%. due to the change in fluence location from the center of the capsule to the center of the charpy specimens." i k
Lippincott and Anderson have summarized the results of their updated studies in Table A 7, l t aken here from Ref. [Li94]. In Ref. [Li961, they state: l I
"For all the capsules taken together, the C/M ratio averages 0.88 with a standard deviation of 10W However, the consistency of results for individualplant types is below this 10%
value in all cases. For the newer plant designs (the plants with neutron pads), which are expected to have the most accurate results, the standard deviations is about 5 ?o. This value of 5% can be assumed to represent a reasonable estimate of the precision of fluence analysis far capsules with well designed dosimetry sets.
It is also seen from the results in Table A-7 that some of the capsules from thermal shield plants falloutside the expected bounds based on this precision estimate. Two early capsules \
from the two-loop plants had less dosimetry coverage and were analyzed by a different laboratory from the rest. Exclusion of these two capsules results in the same consistency of results from the remaining 19 capsules as obtained from the laterplants. It is also noted that two four-loop plants have consistent results with larger bias, possibly due to a structural oifference in surveillance capsule location. With these exceptions, the three-loop thermal shield plants exhibit the greatest variation, but the results for individualplants are generally 1 consistent with the 5% precision value observed in the newest plants.
The reanalysis of measured data for a group of capsules using the latest nuclear data libraries based on ENDF/B-VI resulted in no significant change to the measured fluence values. This is expected since changes to the dosimetry cross-sections were smalland any changes in the calculated neutron spectrum are minimized by the FERRET adjustment procedure. The calculated fluence values using the BUGLE-93 transport cross-sections did increase l significantly, however, compared to the SAILOR results. The change in calculated fluence values resultedin an increase in the average C?M bias from 0.88 to 1.07. Thus the new cross sections result in a smaller bias, but seem to overcorrect the previously observed bias. Using i new cross sections, the calculated fluence values may be expected to be conservative for l predictions of radiation effects. The change in bias occurred consistently for the capsules studied, and it is concluded that the standard deviation of the C/M result given in Table A-7 would not be significantly different if all the capsules were evaluated using the newer
' IENDF/8-VI) cross sections. "
In conclusion, Lippincott and Anderson [Li96] state:
"The consistency of the C/M results over the large number of fuelarrangements and types of plants indicates that the calculation of both capsule and vessel fluences can be made with high precision, and after bias is taken into account, high accuracy can be obtained also.
23
I Table A-7 Summary of C/M Results Standard Plant Type Number of Capsules C/M Deviation Fc)
Two Loop Plants 21 0.767 8.9 Three TS Loop Plants:
Beaver V.1. N. Anna 7 1.C ; 7 5.S I
Surry 6 0.903 7.6 Turkey Point 5 0.823 2.7 Robinson 3 - 0.937 4.6 I !
All Three Loop TS Plants 21 0.927 9.9 i l
Three Loop NP Plants 12 0.916 5.1
, Four Loop TS Ptaw $
i
- Sequoyah 1 3 0.748 5.4 1 Zion 1 4 0.780 8.8 Other 4 Loop TS Plants 29 0.924 5.4 l
All Four Loop TS Plants 36 0.893 9.1 Four Loop NP Plants 33 0.868 4.9 Connecticut Yankee 5 0.989 8.8 San Onofre 1 3 0.967 7.4 l
r All Plants 131 0.879 10.3 Notes: C/M is ratio of calculated fluence (E > 1 MeV) to that derived from the dosimetry measurements at the center of the specimens in each capsule. TS indicates thermal shield plants and NP indicates neutron pad plants. Connecticut Yankee and San Onofm I have different geometry and are not included j with the other thermal shield plants.
l l
I 24
I 1
In particular, the consistency of results indicates that the calculations are adequately taking into account the variation due to the change in fuelloading, including high burnup and low leakage fuelmanagement. The calculation of fluence for fuel cycles beyond the time of the latest capsule removal can therefore be relied upon to produce good estimates of relative exposure. The consistency also indicates the high quality of the measurement results. It is concluded that even a single capsule measurement result can be used to significantly reduce f uence uncertainty by defining plant specific bias.
Use of the BUGLE-93 cross section set results in a significant change in the average capsule bias but little change in data scatter. Therefore, all the conclusions regarding the precision of experimental and calculational results drawn on the basis of the older cross section calculations are still valid. In addition, the values of the measured fluence for surveillance capsules derived using the older cross section set are still valid. "
Contiuning the commentary taken from Section F, Trend Curve Data Development & Testing: 5 Regulatory Guide 1.99. Revision 2
{
"In Ref (Ra84), Randall discusses the basis for Revision 2 of Reg. Guide 1.99. As stated, the 1
Guide is being updated to reflect recent studies of the physical basis for neutron radiation damage and efforts to correlate damage to chemical composition and fluence. Revision 2 contains several significant changes. Welds and base metal are treated separately. Nickel l content is added as a variable, and phosphorus is removed. The exponent in the fluence !
factoris reduced, especially at high fluences; and guidance is given for calculating attenuation of damage through the vessel wall.
For PV wallneutron fluence attenuation predictions, the preliminary results of the PSF (Mc85) l comparisons lie within 10% but reaffirm slight deficiencies in the iron cross sections first brought to light by the PCA and PSF startup experiment comparisons (Mc81,WiB3), which show increasing disagreement the further into the PV one goes.
In the planned Revision 2 of Reg. Guide 1.99 (Ra84), the equation used for PV wall fluence 1 attenuation by Randallis Fluence (x) = Fluence / Surface) . e* (1) where x is the depth in the wallin inches, measured from the inside surface. This equation is based on transport calculations by Guthrie et al. (Gu82,GuB2a) for the dpa attenuation through an 8.0-inch vessel wall. These calculations did not account for the deficiencies in the
' iron cross sections mentioned above.
It has been recently noted by Fabry that the Lifn,a) spectrometry data (DeLeeuw, in Mc81) in PCA are consistent with gas proton recoil spectrometry (Rogers, in Mc81) and silicon camage measurements (DeLeeuw, in Mc81), and they indicate largerproportions of neutrons below 1.0 MeV than predicted by ENDF/B-IV; the discrepancy is on the order of 20% in the 1 same direction as nuclear research emulsion (NRE) results reported by Roberts, Gold, and Preston in Ref (Mc85), Section 2.2.1.1, NRE Measurements. This confirmed result does affect the dpa/@> 1 MeV transverse predictions through the reactor PV planned for use in
, 25
I l
Reg. Guide 1.99, Revision 2 (Ra84), and may adversely impinge upon eventual crack-arrest considerations in the safety analysis of ASME-Ill designed vessels. It is recommended, tnerefore, that:
- 1) A new simultaneous evaluation of all experimental data in PCA, the NESDIP replica, and the Molfron Shell Benchmarks should be performed, including the French damage monitor results obtained during the PSF startup program,
- 2) Integralmeasurements using NRE as well as higher threshold-energy sensors (such as
- Nifn,p), "Zn(n,p), or 'Alln,allshouldbe performedin the Molfron ShellBenchmarks, and ,
l
- 3) Continuous gamma-ray spectrometry experiments should be conductedin the NESDIP benchmark, Phase 3, to resolve inelastic gamma-rays produced by fast neutron intr:ractions in iron and there by test the inelastic neutron transport cross section of iron. "
The above confirmed result has been re-confirmed by the LR-O Benchmark studies of Osmera j
et al. see the Section 9, Subsection 9th ASTM-EWGRD Symposium, Paper 3 commentary. {
Using the ENDFIB-VI data base, BUGLE 93 library with the 1D ANISN and 2D Dort transifort codes, Osmera et al. state:
"The spectra measured by both spectrometers were identical in the frame of usual uncertainties known from the measurement in reference fields. The proton recoil socctrometers' experimental (higher upper energy limit) and 2D-DORT, BUGLE-93 Skoda calculation were used for the comparison. The sensitivity of the results to the substitution of SAILOR BY BUGLE-93, P7, was studied with 1D (ANISN) calculation. The differences (improvements) were remarkable. Nevertheless the discrepancy of the 2D-DORT, BUGLE-93, calculation and experiment is substantial for the evaluation of the reliability of the RPV exposure calculation..... .. .The calculation mostly underestimated the fast fluxes.
. . .... The disagreement of calculation and experiment is probably caused by the used group library. In severalstudies the sensitivity of the results to the type of the library are presented (Ha96a,Ch92]. The Monte Carlo calculations of similar problems (Osxx,Woxx] show substantially better consistency with experiment. "
Additional support for these conclusions is given by Hogel et al. in their Prague papers
[Ho96,Ho96al on " Neutron Dosimetry in Extended Surveillance (Capsule) Program on the 4th Unit of NPP Dukovany" and " Fast Neutron Fluence Monitoring (Cavity) on NPP Dukovany;"
see Subsection 9, 9th ASTM-EWGRD Symposium, Papers 4 and 5, respectively.
What this shows is that the results of the application of the procedures, methodology, and j data files recommended in the ASTM E1018 Dosimetry Cross Section File standard are of '
such high quality, that the use of the current ENDFIB-VI, or an earlier version, of the files with one of the codes listed in the ASTM E944 Adjustment Code Standard will provide extremely reliable "best estimate" values of fluence for NPP surveillance and regulatory programs.
26
KEY ISSUES - Detailed commentary is provided on important technicalissues with relevance to the PNP's Physics-Dosimetry Program and the use of the recommended methodology in the ASTM E706 LWR Surveillance Standards; see Attachment ll (of the March 30,1996 ASTM E10.05.04 TG Report) on "TechnicalIssues Relevant to LWR Surveillance Standards" and its
" Reference Listing" and Attachment lli on " ASTM Standards Associated with PWR and BWR Power Plant Licensing, Operation and Surveillance" and its " Reference Listing." issue statements from Attachments ll and ll1 with relevance to the present review and study are:
3 Benchmark Field (BF) Issue The issue of the limitations of BF is addressed by Gold in Refs. [2,81. Lippincott [B] found that because of the large potential impact of the uncertainties in plant specific parameters (reactor geometry etc.) use of only one or two reactors to benchmark plant calculations and determine generic biases in the calculational methodology and cross sections is not adequate.
In Ref. [13], Blaise, de Wouters, and Ait Abderrahim found that further away from the (VENUS) core, the average C/E in the neutron pad from 12 measurements is 1.01 1 0.11 (s.d). In this study, MCBEND provided an essentially unbiased approximation of the
" equivalent fission flux" from the core baffle to the neutron pad, without significant variation of C/E in the range of penetration depth. This conclusion generally confirms previous validation work in slab geometry benchmark fields with a pure U-235 fission source spectrum. This encouraging result, obtained with a realistic core shape and a spectrum of emerging neutrons similar to that in a NPP, has confirmed the choice of MCBEND as a suitable tool for the computation of the fast fluence at the PV and the surveillance capsules of seven Belgian PWRs presently in operation.
Transnort Calculational lasue A number of past and present Symposia papers address the issue of the use of the Discrete Ordinates and Monte Carlo Methods. Of interest here are the conclusions of Gold's study of tne " Limitations of Pressure Vessel Fluence Calculations" [2,B], Helm's review of "Past and Current Methods Used for Fluence Spectrum Estimation" (15,161, Lippincott's review of the
" Assessment of Uncertainties in RPV Fluence Determination" [H], and de Wouters et al.'s study of the " Analysis of PWR PV Surveillance Dosimetry with MCBEND" [12].
Gold concludes that given the current limitations that exist in calculations, the only rational way to determine PV and SS neutron fluence with the necessary accuracy and reliability for
evaluating and predicting steel radiation damage is through , reliance upon experiment. This can be accomplished by application of a least squares adjustment code, as described in ASTM Standard E944 (E706(ll A)], which judiciously combines calculational and experimental results.
Gold uses the existing calculationallimitations to show that the recent draft RG DG 1053 for the determination of pressure vessel neutron fluence is based upon procedures and I'
The numbered references can be found in Attachments ll and Ill.
27 i
i l
l L assumptions that are nat valid. Gold also emphasizes that the limitations possessed by experimental methods must be carefully considered and evaluated. As a consequence, specific and detailed recommendations to improve LWR-PV SS surveillance dosimetry have been advanced elsewhere [S).
In regard to the use of the Monte Carlo method, de Wouters et al. [12] found that the l
MCBEND code provides an accurate approximation of f ast fluence at the level of surveillance l capsules and suggest that it is a suitable method to calculate the fast fluence at the inner l surface of the PWR PV with little or no bias.
1 Exoosure and Radiation Damage Parameter Extracolation issue l_ The complexities and limitations of PV neutron fluence calculati. ns make alternative methods l cf neutron fluence extrapolation desirable. To overcome unce:tainties and systematic biases that can be introduced by calculational methods, an empirict I two step method of extrapola-t:on has been advanced by Gold and McElroy [22,23).
l l Least-squares analyses of the Pool Critical Assembly (PCA) and Pool Side Facility (PSF) ber)ch-l mark dosimetry data have demonstrated that a simple exponential description of either the DPA or fluence is an excellent representation of the radial variation of neutron exposure within l l the PV wall. This simple behavior can be used to advantage in obtaining extrapolated exposure l and damage parameters (such as the Charpy shifts) at points of interest within the PV. The ASTM Standard on Extrapolating PV Surveillance Dosimetry Results, E706(IC), should be I
revised to indicate the availability of this new empirical extrapolation alternative.
l Reaulatorv Guide 1.99. Revision 2 Data Base Consistency issues l l i l
The methodology and data used in RG 1.99, Rev. 2 for predicting Charpy shifts are, 4 l essentially, the same as those used and recommended in the ASTM E900-87 (E706(llF.1)]
Standard. E900-87, however, does not address the issue of ARTm through wall attenuation.
The following critique related to RG 1.99 Rev. 2 consistency issues is based, primarily, on the results of recent studies by R. McElroy. 4 An important part of this work is associated with an in-depth review of the methodology to be considered and recommended for use in future versions of E706(IE), E706(llF.1), E706(llF.2), E706(llF.3) and E706(lllD.2).
The two separate data bases employed to produce RG 1.99, Rev. 2 were quite different. Two l
, Independent trend curves, for both plate and weld, one by Odette et al. and the other by j Guthrie, were combined to provide the most conservative elements of each in a final form embodied in Rev. 2 (51,fi21.
- Symposium Oral Session A presentation by R.J. McElroy on " Embrittlement Trend Curve Development Issues." !
28
)
i Fluence Issue For the two data bases, the fluences differ in the majority of cases, some by more than a factor of two. The reason for this difference is that Odette et al.
most often, employed the fluences quoted in surveillance reports and compiled in the EPRI l data base (53), while Guthrie employed a re-evaluated set of fluences from NUREG-3319 l produced by Simons, McElroy, and Lippincott under the LWR-PV-SDlP [35) as well as some values taken directly from surveillance capsule reports. The 3319 fluences were on the
)
average 25 % higher than the original fluence estimates employed by Odette et al., some being ]
as much as 2.5 times higher. I l
A survey of the fluence results in the PREDB for the A302B correlation monitor materials made it apparent that the Guthrie data, which employed the NUREG 3319 re-evaluated fluences, I
provided a better trend curve, indicating the need for, and the potential benefit to be gained from, a self consistent basis for expressing capsule fluence. The impact of the reevaluation on individual data from the Odette et al analysis are apparent.
]
Many industry neutronics analysts have argued that such dramatic effects on fluence evaluation are unlikely to exist in more recent surveillance data, which "should" have been l l evaluated on a self-consistent basis. Interestingly, including more recent surveillance data from the PREDB for the correlation monitor material with data in Rev. 2 increases the scatter alarm-ingly, suggesting that the more modern data are no more consistent than the original Rev'. 2 surveillance data, it is therefore concluded that an uncritical statistical analysis of the raw I
- surveillance data base is unlikely to reduce data scatter and hence the margins. On the contrary, it is more likely to increase the margins. This would seem to be a strong argument for continuation of the exercise initiated in NUREG 3319. .
)
i Charov Shift issue -- The next inconsistency between the two data bases arises in the Charpy shifts. In this case, Guthrie used the shifts quoted in the surveillance reports, which were largely judged "by eye", or where no 30 f t-Ib result was reported, Randall and Guthrie made their own estimate of the shift from the plotted data. Odette et al, on the other hand, used shifts based on tanh fitted curves to the EPRI data base. Virtually all Charpy shifts differ between the two data bases and in approximately 15% of the cases this difference is substantial.
I l Conner and Nickel lasue -- Differences in Cu and Ni content are also apparent l between the two data bases, though these are less common and therefore of less importance.
It is now clear that bulk Cu content is a poor indicator of the potential for embrittlement. It can be demonstrated with the PREDB that, for welds with Cu content above about 0.23wt%,
the property behavior is effectively independent of Cu content, as might be expected, since
- the Cu solubility at a typical RPV Post-Weld Heat Treatment (PWHT) temperature of 607 14 C is about this level.
It should be noted that widely accepted techniques are now available for the measurement of soluble copper as well as for monitoring the copper precipitation during irradiation. Such techniques would almost certainly demonstrate a strong " embrittlement saturation effect",
particularly at the higher irradiation temperatures where matrix hardening would be reduced by recovery effects.
t 29 l
T
- The role of Ni could be more complex over the range of Cu and Ni contents encountered in the PREDB and it may be necessary to further subdivide materials into low and high Ni groups i
as well as into plate and weld, which is the basis of the current Regulatory Guide. Indeed, further subdivision into low and high Cu might also be warranted due to the non-uniform distribution, especially m welds, of Cu and Ni.
A simple form of the trend curve describing the above behavior would contain a Cu hardening component which increases with Cu up to a level judged to be representative of an upper bound solubility based on the vessel's heat treatment. Above this Cu level, the Cu hardening component would be constant and independent of Cu content.
Neutron Flux or Damaae Rate issue -- Another effect which is of potential importance, particularly in relation to BWR's and accelerated Test Reactor (TR) irradiations. 1 is the effect of neutron flux or damage rate. The copper hardening component increases curing irradiation to a plateau level corresponding to full copper precipitation. The plateau level increases with increasing copper content in solid solution.
1 l There is now considerable evidence that the copper precipitation process, which is irradiation l enhanced, is a function of both fluence and time, such that at low dose rates full precipitation is achieved at lower doses. Neutron flux or damage rate should therefore be included in any reassessment of embrittlement trend curves, since surveillance exposure rates in the order of 8 2 3x10 n/cm s are to be found in the PREDB, and if TREDB irradiations are to be included, they q
can be as high as 10"n/cm 2s. Stallmann et al will be reporting at this symposium on the status of development of the TR Embrittlement Data Base [141.
Consistency of Methodoloav lasue - it is apparent that significant differences existed between the two data base analyses employed in deriving the Rev. 2 trend curve and that neither followed a consistent methodology. A more consistent approach would have been to combine the Odette et al Charpy data with the Guthrie fluences and to ignore data for
)
which copper and nickel values were in doubt or not available. Such an approach appears not to have been attempted, and it is probable that a more self-consistent approach would have reduced the margins and yielded a better trend curve, even given the unrepresentative nature of the copper contribution.
l l An NRC sponsored analysis of the PREDB is in progress and it is to be hoped that the object
! lessons derived from the recognized inconsistencies in the Rev. 2 analysis will be corrected.
- There remains the need to review and revise all fluence data employed in the current analysis using self-consistent dosimetry and neutronics methodologies. Without such a revaluation j there is evidence that margins willincrease significantly.
Additionally, advantage should be taken of mechanistic and microchemical technique developments which should allow more representative estimates of copper and matrix harden-ing effects to be assigned to account for effects of PWHT, capsule temperatures and damage rate.
30 l
Incorporation of all these factors in a strongly phenomenological approach will provide greater ,
confidence in the specification of trend curves, particularly for maximum life attainment of existing NPPs and ALWRs, while maintaining more realistic and definable safety margms.
In regard to the above, the Westinghouse (Li94,Li961 results of the application of the ASTM E 706 Master Matrix, self-consistent dosimetry and neutronics methodologies, were submitted for inclusion in the PREDB physics-dosimetry-metallurgy data base. This data base was used for the development and testing of the recently balloted revised ASTM E900 98 Standard on
" Guide for Predicting Neutron Radiation Damage to Reactor Vessel Materials."
In the " Technical Basis of ASTM E900-98" Attachment to the E900-98 Standard's ballot, it states under item (2), Section 4.0, Data Compilation and Description of Data Base, that:
" Fluence values were updated by ABB-CE, General Electric, Westinghouse, and Framatome Technologies. "
It is my understanding, that except for Westinghouse, the other three vendors did not followed the Table 3.1 recommended ASTM procedures and methodology in updating their surveillance capsule fluence values for the PREDB. That is the methodology required under Sections,1.2 of the Scope Sections 1.2 of E900-98 and Section 1.3 of E900-87. It would appe'ar, therefore, that the more consistent and updated Westinghouse surveillance capsule fluence values were not used in the development of the ASTM E900-98 predictive formulas. That is, since un-determined fluence value biases could be introduced into the data base by the use of the combined ABB-CE, General Electric, Westinghouse, and Framatome Technologie results if the same procedures and methodologies for the determination of the fluence values had not been used by all four service organizations.
LMFBR-ILRR-FTR PHYSICS-DOSIMETRY CHARACTERIZATION PROGRAM REQUIREMENTS AND RESULTS -- The reason that the adjustment code methodology is so well established and validated is because the success of the LWR-PV-SDlP program was dependent on the remarkable success of the LMFBR development program [Mc77,Li83] Relevant excerpts from
" Neutron Environmental Characterization Requirements for Reactor Fuels and Materials Development and Surveillance Programs" [Mc77) and the FTR Dosimetry Handbook [Li831 are l attached to this letter report. In Mc77, Figure 2 represents an estimate of pre and post-1975 state-of-the-art nuclear parameter uncertainties for neutron environmental characterization for U.S. reactor development programs. After 1970, the results and projected estimates are primarily based on EBR ll and the Interlaboratory LMFBR Reaction Rate (ILRR) program dosimetry test results and reactor physics studies. By 1985, estimated uncertainties were in I the range of 5 to 15% (la) for such dosimetry measurement adjusted parameters as neutron
- ' flux and fluence for E> 1 Mev. The basis for the Figure 2 and Table i estimate of 5 to 15%
for dosimetry measurement adjusted values of fast flux and fluence (E> 1 Mev) was the success of the ILRR program [Mc75). In [Mc75, page 180], McElroy and Kellogg state:
The development, design, & operation of nuclear reactors require the accurate prediction of a) fission rates and bumup for fuels and b) neutre;r exposure for neutron induced property changes for fuels and materials.
31
l l
! t 1
l
(
Goal accuracies of as low as 1 % (Ic) have been set for the determination of fission rates, burnup, and neutron fluences for tne fast reactor development program. Based on the ciscussion of the status of fuels and materials fast reactor dosimetry data cevelopment and testing, attainable goalaccuracies presently appear to be in the range of 2 to 5%.
Comparisons are made of CFRMF and SE results with those reported previously for GODi\/A and the "'U fission spectrum. Fission yield results are considered, based on measurements in the high-intensity environment of EBR ll. These results are coupled with those from i CFRMF, 55, and other neutron fields to more clearly define and document exis ting uncertainties associated with reaction-rate, fuelburnup, and flux-spectral-fluence determina-t on for fast reactors. '
1
. l I Information on the application of the SAND ll multiple foil method of neutron-flux spectral characterization, oeveloped for the U.S. A tomic Energy Commission 's fast Reactor Materials Dosimetry Center (FRMDC) at HEDL. is presented in Ref. 3. The SAND ll Monte Carlo error analysis code (Os 76]is used in this paper to assign uncertainties to multiple-foil derived flux l soectra for two reference standard neutron fields, CFRMF and 55. The foil results are compared with spectrometry and calculations to help identify sources of uncertainties in current estimates of flux spectra for these two important neutron fields.
)
The value of absolute total flux at 6 A Wderived by the SAND ll Monte Carlo code for CFRMF
\
is 7.36 x 10'* n/(cm sec) 2.3% (la).
Using the 36 group calculated spectrum as input, the SAND ll Monte Carlo derived value of absolute flux at a reactor power of 1 MW for EE is 7.4 x 10* n/(cm sec) 2.8 % { t al. l Here too, the multiple-foil-derived spectrum is harder than that calculated. "
It is important to understand that in 1975 (and it is still true today), the results of the ILRR
[Mc75], EBR 11 & Fast Test Reactor (FTR) [Li83], and LWR-PV-SDlP [GO89] programs have demonstrated, with great confidence, that without dosimetry measurements, transport calculations can not be used to derive accurate and reliable values of absolute fast flux and j fiuence for benchmark, test, and/or power reactor applications.
BIASED FLUENCES IN THE CHARPY EMBRITTLEMENT DATABASE -- In a 9th ASTM-EUROPEAN Symposium paper by Worsham [Wo96a) he states:
l
~
"Curren tly, calculational results are corrected or adjusted to equal dosimetry measurements and the calculations are used to extrapolate measured vess elfluences. "
In a second Symposium paper [Wo966) he states:
" Corrections or adjustments to equate the calculations and measurements will not provide an acceptable calculational model. The only way to achieve acceptable accuracy and uncertainties in the calculationalmethodology is to correct any errorc and j improve thepoorapproximations. " Measured" vesselfluences thatarepredictedusing '
erroneous orunreliable calculationalmodels willbe erroneous and unreliable. FTlfound
/ 32 l
I that the previous fluence " measurement predictions in the RG 1.99 (Re881 embrittle-ment database, are biased. Investigations of la) the work of Simons, et alla (sib 21, Lippincott, et alia (Mc81], and Stallmann, et alia (St86,Mc84], and (b) the reasons for the biased measurements, suggest that the biases are caused by the old FERRET-SAND ll technology. "
He also states [Wo96b] under " Bias Evaluations - FERRET-SAND and LSL-M2 Comparisons" that:
"If FERRET-SAND produces biased results with fluence values that are too high, as indicated by Equation 5, then the positive bias should be evident when FERRET-SAND results are compared to another least squares adjustment method, such as LSL-M2. Reference [Mc811 provides the comparison of the "PCA Blind Test"results from FERRET-SAND with those from the LSL-M2 predecessor. As wouldbe expected, the FERRET-SAND results showed a positive l bias. These results further confirm that the FERRET-SAND adjustments are the reason that the fluences in the embrittlement database are biased. Reference (Mc81] results for the PCA Blind Test were published in 1981. Three years later, after several modifications to the FERRET-SAND procedure, new FERRET-SAND results for the PCA configuration 8/7 were reported, Reference (Mc84]. These results nearly duplicated those reported from LSL-M2.
If modifications to least squares technique produce a bias, such as the change between'the 1981 and 1984 FERRET-SAND results, the evidence is rather clear the techniques are erroneous. "
Worsham's analysis and conclusions are not valid because he did not perform a review and careful study of all of the available PCAIPSF documentation. This is shown by the comparison of PCA [Mc81,Mc84] and PSF [Mc87bl benchmark adjustment code results presented in j Table 3.2. !
The primary reason for the differences between the 1681 and 3318 results is that different input data and assumptions were used for the 1981 and 1984 studies; for example for the 1681 study, Lippincott included Gold's and Roger's differential neutron measurements (proton-recoil spectrometry results) as reported in Sections 3.1 and 3.2. In the 3318 study, these latter data were neglected and not used. Stallmann did not use any of the differential neutron spectrometry data as input to his LSL-M2 analysis. In Section 7.3 of the 1681 report, its ,
i stated that: l l
"The following causes for the discrepancies between the HEDL (FERRET-SANDill and ORNL (LSL-M2) calculation have been tentatively identified:
i
! . .1) The HEDL calculations include, in addition to reaction rate data, "Li-spectrometry and proton-recoildata including that absolute data as reportedin Section 3.3.4. The large amount ofinput information tends to decrease the uncertainties, but may introduce biases if the data are inconsistent.
- 2) The uncertainties in the foiland fission chamber measurements are also smallerin the HEDL calculations, disregarding the 1 4% core power normalization uncertainty. Photofission corrections were made for the HEDL calculations whereas for the ORNL calculation, the uncertainty for the *U monitor was increased from 6% to 15%.
33 1
TABLE 3.2 COMPARISON OF PCA & PSF BENCHMARK ADJUSTMENT CODE RESULTS i
RATIO (NEUTRDN FLUX E> 1 MEV) FERRET-SANDil I LSL-M2 PCA (a) PCA (b) PSF (c) l PRESSURE CONFIGURATION CONFIGURATION CONFIGURATION I l VESSEL NUREG/CR 1681 NUREGICR 3318 NUREG/CR 3320 V3 i
LOCATION 8/7 12LL3 4/12 _BL7_ 12/13 4/12 4/12 SSC-1 - - - - - -
0.99 l SSC-2 - - - - - -
1.02
- . o.T - - - - - -
1.04
! 1/4-T 1.08 1.05 -
0.98 1.09 1.07 1.01 1/2-T 1.06 1.06 -
0.98 1.09 1.11 1.01 I 3/4-T M M -
M 1JD .LL3 -
l AVERAGE = 1.07 1.06 1.00 1.09 1.10 1.01 l
l RATIO (NEUTRON FLUX E> 1 MEV) SENSAK / LSL-M2 i
PRESSURE CONFIGURATION CONFIGURATION CONFIGURATION I VESSEL NUREG/CR 1681 NUREG/CR 3318 NUREG/CR 3320 V3 LOCATION 8/7 12/13 4/12 8/7 12/13 4/12 4/12 SSC-1 - - - - - -
1.03 !
SSC-2 - - - - - -
0.97 0-T - - - - - -
0.93 1/4-T - - -
1.02 1.07 0.97 0.96 1/2-T - - -
1.03 1.09 0.98 0.96 3/4-T - - -
M 1.09 M -
AYERAGE = 1.04 1.08 1.01 0.97 (a) Used Table 7.3.1 results; (b) Used Tables 7.1.2.1, 7.2.1.1 & 7.2.3.1 results; (c) Used Table C1 & Table 4.4.4 results; (d) Same as (b); (e) Same as (c).
34
- 3) The HEDL calculations assume much smaller uncertainties for the calculationaldata (25 %
than ORNL (50%). Further, the HEDL calculation has a larger number of groups, which are t.ed together via short range correlations. "
In Section 4.2.4 of the 1681 report, its stated that:
"That is, the PCA Experiments and Blind Testprovides necessary, but not sufficient, validation of the analytical tools and dosimetry methods needed for LWR PV in vessel neutronic projections (see Section S.1 for a discussion of both necessary and sufficient conditions). "
See Section 4.2.5 of 1681 on " Recommendations" for additional commentary, if Worsham had carefully reviewed the PCA reports, he would have realized that he could not make the comparisons between the results of the PCA 1681 and 3318 reports because there were valid reasons (stated above) for the Table 3.2 differences in the PCA results. That is, the preliminary PCA results should not have been used as a basis to support his conclusion that the FERRET-SAND ll techniques are erroneous [Wo96bl.
To make such comparisons, Worsham should have use the PSF [Mc87b] results genergted 3 years later. These results had direct applicability to the validation of the accuracy of P0lR and BWR in vessel surveillance capsule and dosimetry methodology. What one finds by comparing the Table 3.2 results of three adjustment codes (FERRET-SANDil, LSL-M2, and SENSAK) is that within the range of the specified uncertainties for the input parameters for the discrete ordinates and Monte Carlo transport calculational and dosimetry measurements, there is absolute /v no bias introduced by the adjustment code methodology. Further, Worsham conclusion that there is a bias in the Reg Guide 1.99, Rev. 2 fluence data base because of the use of FERRET-SANDilderived fluence values in the LWR-PV-SDlP: LWR Power Peactor Surveillance Phvsics-Dosimetrv Data Base Comoendium [Mc87a] is not valid and cannot be justified on a rational scientific basis: see Figure 1, Embrittlement Fluence Factor, in Worsham's paper (Wo97bl.
Worsham is a member of the working group on the ANS-19.10 " Fast Neutron Fluence to PWR Reactor Cavities" and would like to maintain the emphasis in DG 1053, as stated, that:
"The prediction of the vessel fluence must be made by an " absolute" fluence calculation in which the transport of the neutron from the core is calculated out to the vesseland cavity, rather than a simple spatial extrapolation of the fluence measurements. "
Just prior to the above sentence, it is stated in the Guide that:
"The determination of the pressure vessel fluence is based on both calculations and measurements; the fluence prediction is made with a calculation, and the measurements are used to qualify the calculationalmethodology.
Because of the importance and the difficulty of these calculations, the method's qualification by comparison to measurements must be made to ensure a reliable and accurate vessel fluence determination. In this qualification, the calculation-to-measurement comparisons are used to identify biases in calculations and to provide reliable estimates of the fluence 35 l
l i
uncertainties.
LVhen the measurement data are of sufficient quality and quantity that they allow a reliable estimate of the calculationalbias (i.e., they represent a .itaristically significant measurement cata base), the comparisons _to measurement may be ut ed to (1) determine the effect of the yarious modeling approximations and any calculationalt ias and, if appropriate, (2) modify the calculations by applying a correction to account for bias or by modeladjustment or both.
As an additional qualification, the sensitivity of the calculation to the important input and modeling parameters must be determined and combined with the uncertainties of the input and modeling parameters to provide an independent estimate of the overall calculational uncertainty. "
Further, in the Minutes for the June 2,1997 Orlanco ANS-19.10 Meeting [Lo97), it states:
Worsham stated that he was concemed that the fluences in the Charpy embrittlement catabase may be biased. He noted that the use of LEPRICON is an interesting concept, but is concerned that these codes do not have a physicalbasis for the adjustment. In addition, he suggested that any fluence adjustment be made direc../ to the transport calculations rather tnan through FERRET and LEPRICON. " '
It would appear that Worsham supports the NRC and DG 1053 position that the pressure vessel fluence estimates must be based, primarily, on the calculations and not on adjustment code results because:
Framatome Technology Inc. has direct responsibility to the B&W Owner's Group to provide technical support to us members in the consideration of NRC licensing issues and requirements.
Appendix H of 10 CFR Part 50 references the ASTM E185 Standard for conducting surveillance tests for many of the detailed requirements of the surveillance program and permits the use of integrated surveillance programs, wherein surveillance program capsules for one reactor are allowed to be irradiated in another host reactor.
Certainly, it could be beneficial for those utilities with B&W reactors (without any surveillance capsules or cavity dosimetry) to be allowed, primarily, to rely on the plant specific calculations for the determination of the best estimate value of the predicated fluence at the pressure vessel inner surf ace.
l In the review and study of the available world wide supporting technical documentation, most countries with NPP now require the combined use of calculations and different types of dosimetry measurements and benchmarking. That is they are using calculated and measured a) benchmark mockups b) in vessel surveillance capsule dosimetry, c) cavity dosimetry, and a
/
di vessel and other component metal scraping to obtain the dosimetry measurements needed to verify the accuracy of calculations and the "best estimate" PV fluence value using an appropriate " Adjustment Methodology" as recommended in ASTM Standard E944; see the 36
supporting technical commentary provided in Section 9, " Subsection on 9th ASTM-EWGRD Symposium on Reactor Dosimetry" papers. {
s h is interesting to observe that in Worsham's statement in the ANS-19.10 meeting minutes he raises the issue that "The adjustment codes do not have a physical basis for the adjustments." These are the same words that Carew and Aronson used in their November 15, 1996 letter to L. Lois on the " Palisades Cycles 1-11 PV and Cavity Fluence Evaluation"
[Ca96]. This letter report was attached to the Ref. [Ha961 December 20,1996 in the Palisades June 26,1996 and December 20,1996 Information Package.
In Ref. [Ca96], Section IV.3, Carew and Aronson state: i l
A major concem with the application of the FERRETadjustmentis that, while the adjustment coes provide a best-fit of the measured data, the dosimeter cross sections, measured reaction rates and calculated spectrum adjustments are made without any physical basis. This aoplication of the FERRET adjustment methodology to Palisades is presently being evaluated and the results of this evaluation willbe reported separately when completed. "
To better understand the basis for Worsham's conclusion that least-squares adjustment co es should not be used for obtaining the best-estimate value of the PV fluence, his two 9th Symposium papers [Wo86a,Wo86b] were carefully review and studied to see if there were technical merit; none could be found. This was best shown in Table 3.2 by a comparison of the PCA and PSF benchmark adjustment code results for the LSL-M2 (St86], FERRET-SAND b.
ll [Sc79,Mc67,Mc87al, and the SENSAK [Mc79,Mc83) codes. Copies of relevant excerpts from the NUREG/CR 1681, NUREGICR 3318, and NUREGICR 3320, Vol. 3 reports that were used to help reach the conclusion that there wasn't any technical merit are attached as Enclosures 5, 6, and 7, respectively. As stated on page 7.3-4 of the 1681 report, "The information in Table 7.3.2 suggests that the results of the HEDL and ORNL calculations will closely resemble each other as soon as all differences in the input data are eliminated. l This needs to be verified by further studies. " >
"The lack of final values and resolution of discrepancies in some of the measured data, l together with the preliminary nature of the ORNL and HEDL analyses, precludes the oetermination of finalrecommended values of the exposureparameters and theiruncertainties ,
for the PCA. Such values anduncertainties willbe derived once a consensus is reached about i finalinput values and uncertainties. "
^
That is, the final values of the FERRET-SAND lllLSL-M2 ratios will be closer to unity. A listing of the causes for the discrepancies between the HEDL and ORNL calculation have been tentatively identified and are listed below:
As stated on page 7.2-13-4 of the 1681 report by Lippincott, Stallmann, and Thomas,
- Comparisons of derived exposure parameter values in the block show differences between the three laboratories of up to 12W No consistent bias between the results exists, when all 37
tne configurations are considered. These differences will have to be investigated and understood to further increase confidence in the least-squares derived uncertainty. "
Uncertaintiesin the exposureparameters also differbetween the threelaboratories. ORNL has tne lowest uncertainty esticnates which reflect the application of a more sophisticated aoproach and/or tighter tolerances on the input spectrum shape. RR&A has the largest range l _ of uncertainty values; for example, for 4(E> 1), the RR&A uncertainties range from 5% to 16% in the block compared to HEDL values of 6% to 9% and ORNL values of 4% to 7%. "
"A comparison of thepresent results with those previously reportedby HEDL and ORNL (Table
- 7. 3.1 of Mc81), indicates impro ved and closer agreement (previous differences ranged as high l as 22%). Improved methods and different assumptions have enabled ORNL to reduce their error estimates by a factor of 2 or more. HEDL uncertainty estimates are now higher because tne results of each measurement location were handled individually and the proton recoil data were neglected. "
It also must be noted that the foil set [' Rh(n,n'), "5ln(n.n'), esNi(n.p), 27Al(n,a),23sU(n,f) and 2:'Np] used for the PCA [Mc81,Mc841 studies was different than the [5'Fe(n p), "Ni(n,p),
i
! dosimetry programs [Mc87a,Si87). The 5'Fe(n,p),"Ni(n,p),56Ti(n,p),52Cu(n,a),23sU(n,f), and, 2 i7 Np(n,f) sensors are recommended (in the ASTM standards) for use in PWR and BWR surveillance capsules to measure the flux and fluence E > 1 MeV. These are the same sensors used for the PSF physics-dosimetry-metallurgy experiments and in the Palisades surveillance capsules.
1 I
38
i i
4.0 REVIEW AND STUDY OF THE PALISADES NL' t.EAP PLANT'S (PNP)
INFORMATION PACKAGE The review er decided that the review and study of the palisades nuclear plant's information package had to be done while keeping in mind 1) the regulatory aspects and 2) the technical aspects of Consumers Energy's Palisades Nuclear Plant's Surveillance Prcgram.
Table 3.1 on "PWR and BWR Surveillance Program Regulatory instruments" was prepared to address these and other aspects of the CTS review and study program. Commentary on Table 3.1 and its Attachment is provided in Section 3.0, Subsection "PWR and BWR Surveillance Program Regulatory Instruments." The Table 3.1 Attachment is associated with
" Trend Curve Data Development and Testing" and provides relevant background information on FERRET-SAND ll surveillance capsule fluence determination and the development and technical basis for RG 1.99, Rev.2.
In Section 1, commentary on highlights of the CTS review and study effort and the work schedule are presented.
5.0 IS OWNER'S METHOD (WESTINGHOUSE METHOD) OF DETERMINING BEST ESTIMATE FLUENCE BY COMBINING TRANSPORT CALCULATION AND DOSIMETRY MEASUREMENTS TECHNICALLY SOUND This reviewer concluded that this question, designated as " Question 1," had to be answered two ways; from 1) a regulatory and 2) a technical perspective.
Table 3.1 on "PWR and BWR Surveillan::e Program Regulatory Instruments" was prepared to address this question. Commentary on Table 3.1 is provided in Section 3.0, Subsection "PWR and BWR Surveillance Program Regulatory Instruments."
The review and study of the information in Table 3.1 and its Table 3.1 Attachment, the relevant documentation listed in Section 2.0, what has been stated in Sections 3 through 9, as well as other considerations, led this reviewer to conclude that the owner's method (Westinghouse Method) of determining "best estimate" fluence by combining transport calculation and dosimetry measurements is both technically and regulatory sound.
The regulatory guidance provided in DG 1053, however, is not consistent with the existing regulatory guidance provided by Federal Law as now specified in 10 CFR Part 50, Appendix G & H, the PTS Rule 10CFR50.61, RG.1.99, Rev. 2, and the ASTM E706 Master Matrix Set "of LWR Surveillance Standards (that by reference to the ASTM E185 Standard in Appendix H has made the entire set of E706 standards a part of Federal Law).
The main inconsistency here is that in RG DG 1053, it is stated:
"The determination of the pressure vessel fluence is based on both calculations and measurements; the fluence prediction is made with a calculation, and the measurements are used to qualify the calculational methodology.
1 l
39 i
"The prediction of the vessel fluence must be made by an " absolute" fluence calculation in which the transport of the neutron from the core is calculated out to the vessel and cavity, rather than a simple spatial extrapolation of the fluence measurements. "
When the measurement data are of sufficient quality and quantity that they allow a reliable estimate of the calculationalbias (i.e., they represent a statistically significant measurement data base), the comparisons to measurement may be used to (1) determine the effect of the various modeling approximations and any calculationalbias and, if appropriate, (2) modify the calculations by applying a correction to account for bias or by model adjustment or both.
The emphasis in RG DG 1053 for determining the "best estimate" fluence value is on a calculation validated by measurements while in the ASTM standards it is based on using least-squares methodology to generate the "best estimate" fluence value. Direct use of the transport code results with the appropriate bias f actors and uncertainties is another acceptable approach."
Since compliance with DG 1053 is not a regulatory requirement of the USNRC,it is suggested that Consumers Energy just continue to follow the regulatory guidance provided in Federal Law as now specified in 10 CFR Part 50, Appendix G & H, the PTS Rule 10CFR50.61, RG 1.99, Rev. 2, and the referenced ASTM E706 Master Matrix Set of LWR Surveillance Standards. That is, follow the existing regulatory guidance that is incorporated and j recommended in the Westinghouse Methodology and the referenced ASTM Standards.
Stan Anderson has prcpared a draft of Section 5.0 of the new ANS-19.10 " Fast Neutron Fluence to PWR Reactor Cavities" standard. This section is on " Determination of Best Estimate Fluence;" see attached copy, Enclosure 3. This draft is very well written and uses appropriate information extracted from DG 1053, ASTM Standard E706-(llE2), Guide for l Benchmark Testing of LWR Calculations." and ASTM Standard ES44 on " Guide for Application j of Neutron Spectrum Adjustment Methods in Reactor Surveillance."
l l
l 40 1
6.0 IS THE METHOD USED CONSISTENT WITH THE BASIS OF 10 CFR 50.61, THE PTS RULE?
This reviewer concluded that this question, designated as " Question 2," also had to be answered two ways; from 1) a regulatory and 2) a technical perspective.
Table 3.1 on "PWR and BWR Surveillance Program Fiegulatory Instruments" was also used to address this question.
The review and study of the information in Table 3.1 and its Table 3.1 Attachment, the relevant documentation listed in Section 2, what has been stated in Sections 3 through 9, as well as other considerations, led this reviewer to conclude that the owner's method (Westinghouse Method) of determining the "best estimate" value of fluence by combining transport calculation and dosimetry measurements is consistent with the basis of 10 CFR 50.61, the PTS Rule.
The method is also consistent with the basis of RG 1.99, Rev. 2 because of Consumers Energy (Westinghouse Methodology) use of the FERRET-SAND 11 physics-dosimetry least squares adjustment methodology recommended in the ASTM E944 Standard Guide. for Application of Neutron Spectrum Adjustment Methods in Reactor Surveillance. The sa'm e methodology used to develop the NUREG/CR 3319 LWR-PV-SDlP: LWR Power Reactor Surveillance Physics-Dosimetry Data Base Compendium [Mc87a,Si87). 5 That is, it involves the use of the FERRET-SAND Code's adjustment methodology to derive surveillance capsule "best estimate" values of fluence (E> 1 MeV) as reported in the Compendium. This "NRC Physics-Dosimetry Compendium's" best estimate fluence values were used as input for l Guthrie's derivation of Charpy trend curves based on 177 PWR data points [Gu84]. Randall
! [Ra86] in turn, then used Guthrie's results together with those of Odette (Od841 to develop Revision 2 of Regulatory Guide 1.99. [Re88]; see Ron McElroy's commentary in Section 3.0, Subsection " Key issues," on " Regulatory Guide 1.99, Revision 2 Data Base Consistency l Issues" and " Consistency of Methodology issue." Also, see the Table 3.1 Attachment associated with " Trend Curve Data Development and Testing," which includes additional information on the development and technical basis for RG 1.99, Rev.2.
As already stated, the regulatory guidance provided in DG 1053 is inconsistent with the l existing regulatory guidance provided in Federal Law as now specified in 10 CFR Part 50, l Appendix G & H, the PTS Rule 10CFR50.61, RG 1.99, Rev. 2, and the referenced ASTM l E706 Master Matrix Set of LWR Surveillance Standards.
, The main inccasistency here is that if the regulatory guidance in RG 1053 were continued to be used to derive "best estimate" fluence values for subsequent PTS screening criteria analyses; the value of the RTns estimate will be biased in some undefined way; again, see the Table 3.1 Attachment, Section 3, Subsection on " Key issues," and Section 9 commentary.
5 Also designated as the NRC Physics-Dosimetry Compendium: Section 2.2 of Ref. [Mc861 ,
I 41 ]
Such biases have been minimized in the RG 1.99, Rev. 2 best estimate values of fluence (E> 1 MeV) database because of the use of the ASTM 944 recommended adjustment code physics-dosimetry methodology in the development of the NRC Physics-Dosimetry Compendium [Mc87al; See the Table 3.1 Attachment, Subsection on NRC Physics-Dosimetry Compendium, in this regard, PTS Rule 10CFR50.61 requirements state that:
l "For each pressurized water nuclear power reactor for which an operating license has been issued, the licensee shallhave projected values of RT,7s , accepted by the NRC, {
for each reactor vessel beltline material for the EOL fluence of the material. The )
assessment of RTers must use the calculation procedures given in paragraph (c)(1) of this section, except as provided in paragraphs (c)(2) and (c)(3) of tiris section. "
l Section (c) states that: l
" Calculation of RT,7s must be calculated for each vessel beltline material using a fluence value, f, which is the EOL fluence for the material. RT,rg must be evaluated using the same procedures used to calculate RT,or as indicated in paragraph (c)(1) of this section, and as provided in paragraphs (c)(2) and (c)(3) of this section. " '
1 What does this mean? Answer: It means that if Consumers Energy where to follow the guidance in DG 1053 they would not be in full compliance with Federal Law by not using the recommended physics-dosimetry-metallurgy procedures and methodology as specified in the current version of the ASTM E185 and the other referenced ASTM E706 Master Matrix )
Standards. In this regard, it is again repeated that the current Code of Federal Regulations guidance is contained in 10 CFR Part 50, Appendix G & H, the PTS Rule 10CFR50.61, RG 1.99, Rev. 2, and the referenced ASTM E706 Master Matrix Standards.
l
)
l l
l 42 l
l I
l 7.0 HAS OWNER'S EXPLANATION OF THE BIAS BETWEEN MEASUREMENT AND CALCULATIONS PROVIDED SUFFICIENT BASIS TO SUPPORT THE MAGNITUDE OF THIS DIFFERENCE?
This reviewer concluded that this question, designated as " Question 3," also had to be I answered two ways; from 1) a regulatory and 2) a technical perspective.
T able 3.1 on "PWR and BWR Surveillance Program Regulatory Instruments," and its Table 3.1 Attachment were also used to address this question. I Eased on the review and study of all of the material identified in Section 2.0 and what has l
been stated in Sections 3 through 9, the answer to this question is a decided yes! !
8.0 HAS OWNER COMMUNICATED ITS POSITION CLEARLY AND IF NOT WHERE WOULD FURTHER EXPLANATION BE USEFUL?
This reviewer concluded that this question, designated as " Question 4," also had to be answered two ways; from 1) a regulatory and 2) a technical perspective. ,
Table 3.1 on "PWR and BWR Surveillance Program Regulatory Instruments" and its Table 3.1 Attachment were also used to address this question.
I Based on the review and study of all of the materialidentified in Section 2.0 and what has been stated in Sections 3 through 9, the answer to this question is also a decided yes! -
i 1
I l
43
9.0 FERRET-SAND H METHODOLOGY REGULATORY REQUIREMENTS AND ASTM STANDARDS -- As discussed in (Mc88), for each LWR nuclear power plant, the physics-dosimetry metallurgy surveillance program requirements are identified through the ASTM Standard E185 on " Practice for Conducting Surveillance Tests for Light-Water Cooled Nuclear Power Reactor Vessels, E706(IF)," which in turn is identified as a part of the 10 CFR PART 50, Appendix H.
The requirements for design of the surveillance program are static for each plant.
However, the testing and reporting requirements are updated along with technical improvements made to the ASTM E185 Standard.
Section 8.2.1 of ASTM E185-94 states that: "The neutron fluence rate, neutron energy spectrum, and neutron fluence of the surveillance specimens and the corresponding maximum values for the re:sctor vessel shall be determined in accordance with the guidelines in ASTM Standard E482 on " Guide for Application of Neutron Transport Methods for Reactor Vessel Surveillance" and ASTM Standard E560 on " Practice for Extrapolating Reactor Vessel Surveillance Dosimetry Results. "
Section 8.2.2 states that: "The specific method of determination shall be documented. "
Section 8.2.3 states that: "The Neutron fluence rate and fluence values (E> 0.1 and 1 MeV) and dpa rate and dpa values shall be determined and recorded using a calculated spectrum adjusted or validated by dosimetry measurements. "
l
- Section 3.1.1 of ASTM Guide E482 states that:
i "The methodology recommended in this guide specifies criteria for validating computationalmethods and outlines procedures applicable to pressure vessel related neutronic calculations for test and power reactors.
Section 3.2 Validation states that:
Prior to performing transport calculations for a particular facility, the computational methods must be validated by comparing results with measurements made on a benchmark experiment. Criteria for establishing a benchmark for the purpose of validating neutronic methodology shouldinclude those set forth in ASTM Standard 944 l on Guide for AppHcetion of Neutron Spectrum Adjustment Methods in Reactor SurveiWance" as well as those prescribed in Section 3.2.1. Requirements for Benchmarks.
Section 3.4.8.2 states that:
Use a spectrum adjustmentprocedure as recommendedin Guide E944 using calculated group fluences and dosimetry data with uncertainty estimates to obtain an adjustment 44
to the calculated group fluences and exposure parameters. Predicted pressure vessel fluences could then incorporate the spectraland normalization data obtained from the adjusted fluences.
Section 4.2.3 of ASTM Practice E560 states that: {
" Guide E944 should be used to combine the transport calculation with the dosimeter results. The E944 adjustment procedure should be used to indicate whether the dosimeter measurements and associated uncertainties are consistent with the transport calculation and with uncertainties implied from benchmark tests of the transport code (PCA, VENUS, NESDIP, andappropriate CommercialBWR orPWR). Having established the required consistency, the adjusted transport code results may be used to calculate the neutron field at all points in the pressure vessel wall with the uncertainties estimates derived from the application of the adjustment codes. Direct use of the transport code results with the appropriate bias factors and uncertainties is another acceptable approach. "
l
- Section 3.1 of ASTM Guide E944 states that: !
. I
" Adjustment methods provide a means for combining the results of neutron transport calculations with neutron dosimetry measurements in order to obtain optimalestimates of neutron damage exposure parameters with assigned uncertainties. The inclusion of measurements reduces the uncertainties for theseparameter values andprovides a test for the consistency between measurements and calculations and between different measurements. This does not, however, imply that the standards for measurements and calculations of the input data can be lowered; the results of any adjustment procedure can be only as reliable as are the input data. "
WESTINGHOUSE METHODOLOGY & TECHNICAL ISSUES -- The Consumers Energy Method PNestinghouse Method) of determining the palisades nuclear plant's best estimate fluence by combining transport calculation and dosimetry measurements is, as described above, in complete compliance with current regulatory requirements and is based on the application of the methodology and procedures recommended in the ASTM E706 Master Matrix Set of LWR Physics-Dosimetry-Metallurgy Surveillance Standards.
Enclosure 2 to this letter report is Section 3 of Consumers Power Company Reactor Vessel Neutron Fluence Measurement Program for Palisades Nuclear Plant - Cycle 1 through 11 (Attachment 2 of Ref. [Sm96]).
In Section 3, it states:
- As noted in Section 1 of this report, the best estimate exposure of the reactor vessel was oeveloped using a corabination of absolute plant specific neutron transport calculations and plant specific measurements from the reactor cavity andinternalsurveillance capsules. In this section, the neutron transport and dosimetry evaluation methodologies are discussed in some cetail; and the approach used to combine the calculations and measurements to produce the best estimate vessel exposure is presented. "
45
in discussing the Westinghouse Neutron Transport & Dosimetry Evaluation Methodology used for the PNP, reference is made to 12 ASTM standards,6 including the ASTM E706 Master Matrix Standard, the ASTM E853 " Standard Practice for Analysis and Interpretation of LWR Surveillance Results," the ASTM E261 " Standard Practice for Determining Neutron flux.
Fluence, and Spectra by Radioactivation Techniques," and the ASTM E262 " Standard Method for Measuring Thermal Neutron Reactions and Fluence Rates by Radioactivation Techniques. '
The other eight standards are associated with the guidance for the accurate measurement of the disintegration rates (used to derive the reaction rates) for the dosimetry sensors: ASTM l E263 (Fe54), ASTM E264 (NiS8), ASTM E481 (Co & AG), ASTM E523 (Cu63), ASTM E526 i (Ti46), ASTM E704 (U238), and ASTM E705 (Np237). As stated in these standards, general l practice indicates that disintegration rates can be determined with a bias of : 3% (la) and a precision of 1% (la) for the non fission sensors and with a bias of : 5% (1c) and a precision of 1 % (lo) for the fission sensors.
l The specific assignment of uncertainties in the measured reaction rates and the input (trial) spectra used in the FERRET evaluations are specified on page 3-15 of Section 3. For reaction rates, a value of 5% was specified. The values specified for flux normalization and fast flux groups are 30%, which clearly shows the expected rather high uncertainty in the DO.RT absolute calculation of the input values of the shape and magnitude of the neutron flux-
- soectrum.
The establishment and usage of a dosimetry reaction cross-section and uncertainty file for FERRET-SAND ll and other adjustment codes are discussed in the ASTM E1018 " Standard Guide for Application of ASTM Evaluated Cross Section Data File."6 it is stated in Section 7.3 of this Guide on Uncertainty File Usage:
l t
"The cross section uncertainty file shall be used as one input to the determination of the overall uncertainties of processed quantities such as fluences and dpa. It is expected that, using least squares adjustment codes such as FERRET, LSL-M2, STA Y'SL, or LEPRICON, a good statistical evaluation of the uncertainty of processed quantities can be obtained. The ,
use of validated cross section and uncertatnty files willprovide the needed confidence to justify usage of derived exposure parameter values and uncertainties for defining neutron-induced materialproperty change limits for L WR nuclear power plants. " l l
For the FERRET-SAND ll evaluations, the dosimetry reaction cross-section and uncertainty files j were taken from the RSIC Data Library Collection DLC-178, "SNLRML Recommended l Dosimetry Cross-Section Compendium," July 1994 [Gr931. '
'For the Np237, U238, Fe54, NiS8, Ti46, and Cu63, Mannhart [Ma89] has measured the spec-tral average cross sections for each of these reactions in the Cf-252 neutron spectrum and compared the measured values with calculated values using the ENDFIB-V cross section file.
The observed CIM ratios were 0.999,0.970,1.011,0.964,0.935, and 0.955, respectively.
l l
46 i
i
Table 3 of the most recent revision of the ASTM E261-96 Standard provides an updated hsting of these C/M ratios for both the Cf-252 and U-235 fission spectra with the uncertainty in the ratio represented by a sum in quadrature of the experimental and calculated uncertainty.
The cross section and spectrum components of the uncertainty are also hsted. For Cf-252 tqe ratios and la uncertainties are 0.981 (9.43 %), 0.970 (1.76 %), 1.014 (2.65 %), 0.981 (2.83 %), 0.891 (3.24%), 0.984 (3.72%), respectively. For U-235 the ratios and 1c uncertainties are 0.990 (11.0 %), 0.991 (4.98 %), 0.996 (5.91 %), 0.974 (7.16 % . 0.899
( 6.8 6 %), 1.042 (12.9 %), respectively. For Cf-252 and U-235, the calculated fission soectrum shape component of the uncertainty for the six reactions range between 0.23 to 1.38% and 4.21 to 6.05%, respectively. For both spectra, the calculated energy dependent soectral averaged cross-section component of the uncertainty for the last five reactions range l between 0.53 to 2.85%; the value for Np-237 was - 9.3%. These results are based on the '
most recent work of Griffin et al(Gr931 and on other work reported in seven references dating back - 22 years to the November 1976 Vienna IAEA Consultants Meeting on integral Cross-Section Measurements in Standard Neutron Fields [Fa78] and the USAEC's Interlaboratory LMFBR Reaction Rate (ILRR) Program [Mc75,Mc75a,Mc771. The cross section and covariance matrices used for Table 3 evaluations are consistent with the source detailed in Ref. [Gr93).
Only the Np237, U238, Fe54, and NiS8 reactions have a significant effect on the adjus,ted value of the fluence (E) 1 MeV) because of the high energy response (E> - 4 MeV) for the Ti46 and Cu63 sensors. There are not enough neutrons above - 4 MeV to significantly influence the final FERRET-SAND ll adjusted value of the fluence for neutrons above 1 MeV.
This review of uncertainties in the sensor reaction data bases demonstrates that the statements made by Carew and Aronson in Section V1.2 of Ref. [Ca961, below, that there are "several other possible explanations" do not have technical merit! I l
"In the CPC/W analysis this difference is assumed to be due to a spectrum-dependent error in the DORT calculations which results in an exact calculation above E> 4.0 and an over prediction for E< 4.0 MeV.................. Based on this assumption, a 12% M/C fluence reduction (i.e., not including the additionalFERRETreduction)is applied to the DORTE> 1.0 MeV fluence prediction. The application of this M/C spectrum-dependent correction is illustratedin Figure 6. While this conclusion may be correct, there are severalotherpossible explanations for the observed 1.00/0.86 difference between the high/ low-energy M/C biases tnat would not require this reduction in the DORT calculated fluence. These include: 1) the L,se of erroneously low dosimeter cross sections for Fe-54 and Ni-58 in the interpretation of tne measurements and/or 2) errors in the Fe-54 and Ni-58 measurements. "
That is, explanations 1) and 2) above. cannot be justified on a technical basis and are not
'possible explanations for the observed biases between the Fe54 & NiS8 and Ti46 & Cu63 results. The reason for the biases is that some of the input information for the calculation is i incorrect, such as the magnitude of the fission source term and the high energy (E> 4 MeV) {
part of the input form of the composite U235, U238, and Pu239 fission spectrum. The !
FERRET-SAND 11 result for the adjusted form of the DORT calculated input flux-spectrum is 1 correct and cannot be challenged on the basis of errors in the sensor measured reaction rates and ENDF/B-V or VI energy dependent evaluated dosimetry and uncertainty data files!
l 1
47 l
This is further confirmed by a careful study of the Table 3.2 results for the PSF where the average FERRET SAND lllLSL-M2 ratios and SENSAKiLSL-M2 ratios for the six PWR pressure vessel mockup locations for the PSF 4/12 configuration are 1.01 and 0.97, respectively. As stated in Section 3.0, Subsection on Biased Fluences in the Charpy Embrittlement Database, cf this letter report:
"What one finds by comparing the Table 3.2 results of three adjustment codes (FERRET-SAND li. LSL-M2, and SENSAK) is that within the range of the specified uncertainties for the input parameters for the discrete ordinates and Monte Carlo transport calcul5tional and dosimetry measurements, there is absolutelv nobing introduced by the adjustment code methodology.
In Section V1.3 of Ref [Ca96], it is stated that:
"A major concem with the application of the FERRETadjustmentis that, while the adjustment ooes provide a best-fit of the measured data, the dosimeter cross sections, measured reaction r.stes and calculated spectrum adjustments are made without any physical basis. "
In response to this statement, it can be said that the statement is not relevant because the use of a least-squares adjustment code allows the study of biased data. That is, the analyst must not only apply physically based physics and reasoning but must carefully examine'the '
data to determine if it is of adequate quality, which has been done for both the PNP and PSF FERRET-SAND 11 Code analyses.
The FERRET SAND 11 Adjustment Code Methodology is well documented in the Section 3 l (Enclosure 2) report and elsewhere [Li83,Li86,Li94,Mc87a,Wo81]. Further, the methodology is based on the guidelines provided in a number of the applicable ASTM standards; see l Table 3.1. The FERRET-SAND 11, LSL-M2, SENSAK, LEPRICON and other adjustment code l
methodology is well established, has been validated, and has been accepted nationally and internationally for deriving "best estimats" fluence values for test and power reactor l applications; see ASTM Standard E9446and Figure 2 and Table 1 of the next Subsection's Paper 1, " Dosimetry in Support of the European Network AMES."
It is very discouraging, therefore, to have the NRC staff say in their December 20,1996 letter
! to Thomas Bordine, Manager of Licensing, Palisades Nuclear Plant [Bo96), that:
"Our review finds that: 1) the fluence reductions based on physical quantities to be acceptable, because they are based on directly measured parameter, 2) the 12% bias is not acceptable because the measurements are inconsistent and statistically incompatible, and 3) the spectral adjustments of 5% are not acceptable because they represent an intuitive
' averaging and do not evoke any physicalprinciple. The spectral adjustments (5%) are still under evaluation; however, it does not seem likely that the method is adequately justified. "
Based on the ASTM E944 Standard's recommended and validated use of FERRET-SAND 11, LSL-M2, SENSAK, LEPRICON and other adjustment codes referenced in E944 (as well as the commentary and results presented in Papers 1,2,3,4, and 5 of the next Subsection on the 9th ASTM-EUROPEAN Symposium on Reactor Dosimetry), there is no scientific technical merit or justification for making such a statement!
48
l l
er justification for making such a statement! l l
In [Ca961, and as previously quoted, Carew and Aronson state in Section Vl.1, Best-Estimate Fluence Determination: l
.. .. . . .. . Based on these measurement-to-calculation (M/C) comparisons of the dosimeter reaction rates, a M/C bias of 12( 7)% is determined. This M/C bias is then adjusted using a least-squares adjustment technique to account for uncertainties in the measurement and calculations. In the case of Palisades this adjustment increases the M/C bias from 12% to 17%, andimplies the calculations are overpredicting the fluence by 17%. The determination l of the M/C bias and the adjustment method are discussed in the following sections. "
to Section VI.2, Fluence Calculation-to-Measurement Bias, they state:
l "from Figure 4, it is seen that the in-vessel M/C bias is 1.00 0.03 for the dosimeters with tnresholds E> 4.0 MeV, and 0.86 0.02 for the dosimeters with thresholds E< 4.0 MeV.
l In the CPC/Wanalysis this difference is assumed to be due to a spectrum-dependent errorin tne DORT calculations which results in an exact calculation above E> 4.0 and an over l prediction for E< 4.0 MeV. Based on this assumption, a 12 % M/C fluence reduction (i.e.,' hot l including the additional FERRET reduction) is applied to the DORT E> 1.0 MeV fluence prediction. The application of this M/C spectrum-dependent correction is illustratedin Figure
- 6. While this conclusion may be correct, there are severalotherpossible explanations for the l observed 1.00/0.86 difference between the high/ low-energy M/C biases that would not j require this reduction in the OORT calculated fluence. These include: 1) the use of
\
\ erroneously low dosimeter cross sections for Fe-54 and Ni-58 in the interpretation of the l measurements and/or 2) errors in the Fe-54 and Ni-58 measurements. (The number of U-238 l and NP-237 dosimeters that are included in the in-vessel M/C bias is small and these measurements are subject to relatively large uncertainties. "
, in Section VI.3, Least-Squares Fluence Adjustment, they state; 1
"A major concem with the application of the FERRETadjustmentis that, while the adjustment j does provide a best-fit of the measured data, the dosimeter cross sections, measured reaction l l rates and calculated spectrum adjustments are made without any physical basis. This 1 aoplication of the FERRETadjustment methodology to Palisades is presently being evaluated and the results of this evaluation willbe reported separately when completed. "
l NINTH ASTM-EWGRD SYMPOSIUM ON REACTOR DOSIMETRY -- In order to provide i
- additional supporting technical documentation and commentary related to the technicalissues
! raised in the above statements, a preliminary review and study was completed of approximately ~40 of the most relevant papers of the ~80 papers distributed at the 9th ASTM-EWGRD Symposium on Reactor Dosimetry, held in Prague, Czech Republic, September 2-6,1996.
This review and study was only partially completed, but it did provide additional information ,
that strongly supports the use and accuracy of the application of the ASTM and Westinghouse j 49 l t
Methodology and use of the FERRET-SAND 11 or other adjustment codes. 7 Of particular importance here, is the supporting technical commentary (related to selected methodology, standardization and technicalissues) provided in five of the Prague papers. The first paper provides commentary related to the European Network on Ageing Materials Evaluation and Studies (AMES) that was set up in 1993 to bring together the organizations in Europe that have the main capabilities on reactor pressure vessel materials assessment and research. The second paper provides commentary related to the European Organization for Economic Cooperation & Development (OECD), Nuclear Energy Agency (NEA), Nuclear Science Committee (NSC) study on Issues of Dosimetry Fluence Computations.
The European Network on AMES projects and activities and the OECD NEA NSC studies are providing a very important continuation and expansion of:
- 1) The previous NRC supported LWR PV-SDlP related activities at HEDL, ORNL, NBS (NIST), MEA, and UCSB and those supported by CEN/SCK (Mol), by EPRl/ Utilities at B&W, CE, GE & W, by SwRI & BMI, by several UK Laboratories (Harwell: Rolls-Royce
& Associates; Winfrith), and by KFA (Julich, Germany), CEA/CEN (Saclay, France), l GKSS & IKE (Germany), ElR & HSK (Switzerland), the Joint Japanese Special Working Group (UT, JAERI, TC, MHI, TU, Hatachi, JSW and MSPI); and ,
I l
- 2) Present and past ASTM Standards Technology Development, Transfer, and Training l l (STDTT) activities and ASTM's sponsorship and co-sponsorship, respectively, of the j series of international ASTM Symposia on the Effects of Radiation on Materials and the )
ASTM-EWGRD Symposia on Reactor Dosimetry.
1 l
Emper 1: " DOSIMETRY IN SUPPORT OF THE EUROPEAN NETWORK AMES" As stated by Ballesteros, Debarberis, and Voorbraak in their Prague paper [Ba961:
"There is a need to coordinate activities to ensure maintenance of capabilities and facilities, and also to harmonize activities with the objective of common European standards. A European network, AMES (Ageing Materials Evaluation and Studies), was set up in 1993 to bring together the organizations in Europe that have the main capabilities on reactor pressure l vessel (RPV) materials assessment and research (Ba96a,Br95,Es95). !
Several tasks, in the dosimetry field, have been planned for the AMES projects and activities.
In addition, a study is being carried out with the following objectives:
- review of the current situation for different considered reactor types (including WWER reactors)
- summarize the existing reference documentation, methodology and techniques 7
Related to this review and study, this letter / report's Acronyms listing includes some of the acronyms used in - 40 technical papers.
50
- analyze key factors for European harmonization validation and benchmarking of dosimetry practices establish a set of recommendations for the AMES projects post mortem dosimetry validation limitations, uncertainty study and possible improvements A description of the dosimetry activities in support of the AMES European network will be presented. Conclusions and recommendations are also of interest for the dosimetry community, and will contribute to a closer working relationship between specialists in both l materials and dosimetry. "
l The following additional commentary was taken from the Ref. (Ba961 paper:
WORKING METHOD Regarding dosimetry, three options were evaluated as working methods to reach the objectives of this Task Group:
Option 1: Incorporate dosimetry in the materials research proposals.
I i Option 2: Dosimetry proposals are independent and complement materials research proposals.
l Option 3: Combine Option 1 and Option 2. This last option has been considered the most l adequate for the AMES projects.
As an example of application of Option 1, a specific task on retrospective (a posteriori) l oasimetry has been included in a project proposalrelative to the analysis of 20 trepans l available from the core baffle of a Spanish NPP. In this project proposal three different oisciplines are involved, neutron metrology (experimentaland calculations), materialresearch and fracture mechanics.
i Felating to Option 2, a project proposalhas been elaborated and it includes:
l l
- 1. Neutron calculations concerning all different European reactors types (PWR, BWR, Magnox, WWER and materials test reactors).
l
- 2. A benchmark with VENUS data. SCK/CEN will supply all necessary data and will coordinate the benchmark activities.
- 3. Analysis of the impact of uncertainties in fluence calculation procedures, for specific
- reactors, on the brittle behavior of RPVmaterials, according to the embrittlement trend
! curves used in different countries.
- 4. Compile a conversion table between typical reactor locations (surveillance pocition, peak flux positions at inner wall,1/T, 3/4 T, etc.) for different reactor types in as many indices as possible. Differentindexes are used forneutron damage. Normally ~ > 0.1 MeVand ~ > 1.0 MeV for Western materials, and ~ > 0.5 MeV for Russian materials.
51 l
Another indexation used mainly for Magnox reactors is the number of displacements per atom DPA. A direct comparison of the above mentioned quantities is not possible in general. The neutron spectra are different for the different cases as well as the materialnuclearlibraries. Therefore, itis recommendable the generation of a qualified conversion table of all used damage indexes in order to compare available data and studies.
CURRENT ACTIONS A first action, performedbefore the formalconstitution of the Task Group, was to prepare a State-of-the Art report on Dosimetry (Ba96a]. This report, financed by EC-DG XIandavailable for July 1996, shows the international situation on dosimetry and neutron calculations for reactor pressure vessel surveillance. It establishes general recommendations, and identifies technicalareas where more investigation is needed. Its main usefulness la to be a reference document to take decisions on further R&D. Figure 2 depicts the detailing plan for the study.
(Note, to the PCA/ PSF, NESDIP, and VENUS benchmarks, they have added LR-O [Os96,Ho96 Ho96al.) Table 1 and 2 have been includedin the finalreport and show examples of neutron transport methodologies and existing ENDF/B and derived libraries, respectively.
Some other relevant actions of the Task Group are described below. %
- 1. It was no ted the necessity to include a very well calibrated dosimeter, or a set co vering the full range of energies, in all AMES irradiations in order to reduce uncertainties that could affect neutron damage assessment. Some desirable characteristics of this State-of-Art dosimeter are:
- Easily calibrated.
- Stable.
- Relatively insensitive to extremes of environmental conditions.
- Correctable systematic errors.
- Produced in reproducible lots.
- Small dimensions compared to distances over which neutron flux gradients become significant. This is of specialinterest for AMES irradiations in WWER reactors.
It was agreed to develop the specification for this correlation dosimeter.
- 2. Ex-vessel dosimetry and retrospective la posteriori; e.g., vessel wall scrapings.) dosimetry are of great interest for AMES. Several specific tasks are
- , being planned for the materials research project proposals. "
In regard to the Figure 2 LR-O Benchmark, Osmera et al. [Os961 have stated:
"The experimentaland theoreticalinvestigations of the reactor dosimetryproblems startedin NRIand Skoda 20 years ago. Due to the unsatisfactory reliability of the calculations andinput data libraries the programme of the LR-0 Mockup-Ups has been launched. " (See Paper 3 for detailed commentary on the results of the "LR-O Benchmark experimental and theoretical PWR studies on VVER-1000 reactor dosimetry."
52
lwti s ie
' e
- v. m x n
- h I t ev y it r c te e i pm s
b tros el
_ R
_ )
e les ra s e gu s
- w ev an d -
r x min ha e
_ a o DM
( d sr an el t
l e se mse R 1
is v S
_ b n-I i S
l j
Y l
l M-I iT
- A S
_ I
_ s I :
dr l a l l i
d T
- S n
ta 7 r t
sr R
e o O nit F n n
_ uo
- n dM R
a N A
w L t
n P
_ l a
u G
_ s lc a n N C i o I.
t n a l o l d l r n A t
u ' O -
e T
e m
- N l
R L n w
I
/ l < l
_ y l c r e :
t e s R l
i 2
- i m
s N o
l E A i
s k
s V R
- r ll 11 a s G
- e h
m r i I
i tn w D i a
'l r
e S t r
B E e N c l
l in l l -
F S
P X /
O A sN C l P mG s l aA r r e
go M, t
e Pr R E m
a r
eW c P a
g lnW I
a
'. iR, l
- l ms d ed e
vW r
y loo Su B. _
f nC U
l R s W d e
P o C t r
l os pe
~ s sd dr n o I a aC r
T d l n s ta i e
S r a
r b
i.
1 l
t uW
TABLE 1: EXAMPLES OF NEUTRON TRANSPORT METHODOLOGIES REACTOR TYPE NEUTRON NEUTRON DOSIMETRY CROSS- ADJUSTMENT REF, i CROSS-SECTIONS CODE SECTIONS CODE (1) j LOVilSA WWER-440 BUGLE-80 D OT 3.5 IRDF-90 l3p45 l
MAINE YANKE W BUGLE-80 DORT ELSXIR 3 p650 SAILOR (S8P3)
TlHANGE 2 W ELXSIR DOT 4.3 ELXSIR LEPRICON 3065 (58P3) l W VIT AMIN-C DOT 3.5 IRDF-90 3 p155 l
) KOEBERG f DOEL 12 W VITAMIN C DOT 3.5 1 p17 !
, (58P3,17g)
PAKS WWER-40 IRDF-85 1p105 KRSKO W DLC 2D D OT 3.5 IRDF 82 STAY'SL 1p115 IS6P3)
SAILOR DOT 4.2 IRDF-85 LSL M2 2 963 (S8P3, 27g)
LR O WWER 440 VITAMIN-C D OT 3.5 1 p130 mock-up (S8P3)
McGUIRE W XSDRNPM DOT 4.2 ENDF/B V 1p139 ANO-l B&W ANO-il CE B&W Owner Gr. B&W DOT 4 1p155 KRB A (2) BWR ENDF/B IV DOT 4.2 1pl65 1
H.B. ROBINSON W SAILOR DOT 3 ENDF/B V FERRET 1p147 ELXSIR DOT 4.3 ELXSIR LEPRICON 1 p405 Saint Laurent B1 W TRIPOLI STAY'SL 2973 i
(1) Ref =
Reference:
, n pd = reference n, page j
" Ref.1 = 6' ASTM-Euratom Symposium, June 1987, ASTM STP 1001 Ref.2 = 7' ASTM-Euretom Symposium, August 1990. Proceedings. Kluwer Academic Publishers Ref.3 = 8' ASTM-Euratom Symposium, September 1993. ASTM STP 1228 (2) Decommissioned 54 ,
\
TABLE 2: ENDF/B AND DERIVED LIBRARIES YEAR ORIGINAL DERIVED TYPE NUMBER OF REFEF ENCE PUB l LIBRARY USAARY GROUPS
{
1972 ENDF/B 111 DLC 2/100G X 100 DLC 2 1978 ENDF/B IV VITAMIN C X 171 DLC-4 ' '3t ! l
{ .
1984 ENDF/B V VITAMIN-E X 174 DLC 11; j
i
) '
1993 ENDF/BVI VITAMIN-86 X 199 j
]
1980 ENDF:B IV BUGLE 80 X 47 DLC 75 l
1993 ENDFiB IV SAILCR X 47 DLC 76 1984 ENDF/B lV + V ELXSIR X,D 56 tla, 1993 ENDF/B VI BUGLE 93 X 47 l
i k
1982 ENDF/B Ill IRDF-82 0 620 DLC 94 646 1984 ENDF/B lV + V ELXSIR X,D 56 '11 1985 ENDF/B IV IRDF-85 D 620 th l
1990 ENDF/B VI 1RDF 90(6) D 620 lAEA0067 lin (0) X = neutron cross sections; O = dosimetry cross sections III EPRI NP 3654 (1984)
(2) lAEA NDS 141 (OCT/93)
(3) ORNL-RSIC-37 (4)lAEA NDS-41/R (1982)
(5) IAEA-NDS-41 (OCT/93)
(6) IRDF 90/G V.2, version October 1993 Note: The inelastic scattenng cross section of iron was rnodified to a large extent in the ENDF/B V rnod. 3
' (1986). This new cross-section is not included in SAILOR. On the contrary, it is included in the last versions of ELXSIR.
4 55
I Eaper 2: " ORGANIZATION FOR ECONOMIC COOPERATION & DEVELOPMENT, NUCLEAR ENERGY AGENCY (NEA), NUCLEAR SCIENCE COMMITTEE (NSC)
STUDY: ISSUES OF DOSIMETRY FLUENCE COMPUTATIONS" l The following commentary was taken from the Rulko, Kodeli and Sartori Prague paper [Ru961:
j "In order to determine metal damage due to neutron (and gamma if significant) irradiation l'uence and spectra have to be determined at points ofinterest in a reactor from transport calculations and experimentalmeasurements. This fluence is then usedin some damage model to infer the damage of metal. In addition, the coupling between the experimentalresults and transport calculations takes place since fluence is derived from the measurements and/or the calculated neutron spectrum and dosimeter reaction rates are derived from the calculations and measured decay rates using the flux history and transport results.
The measurements are used to validate the calculational results. In addition, the intimate l interdependence of calculations and measurements is often used to adjust the calculated i spectra with measurements vialeast-squares fitting. Usually, uncertaintiesin measured values are much lower than those in calculations and hence adjusting permits to combine the ]
measurements and the calculations in a way to provide the adjusted fluence with the reduced l uncertainties. Typicalrecent results show uncertainties of 15 to 20 percent for unadjusied j methods and results within the measurement uncertainties of 5 to 10 percent for adjusted methods. Opinions vary if the adjustment of calculated spectra using experimental data is necessary ornot. If the object of a given dosimetry study is a validation of transport methods then different unadjusted results of calculations should be compared against measurements to deduce the sources of computationalerrors due to methods used, cross-section data, and modelling assumptions.
COMPUTATIONAL METHODS AND UNCERTAINTIES The computationalschemes presently used fallinto two major categories of discrete ordinates methods and Monte Carlo methods both combined with sensitivity and uncertainty analysis whose application has proved necessary to establish reliable safety margins for target quantities. Sensitivity analysis is the most effective measure of predicting what model and j data improvements are most effective in predicting target quantities. Uncertainty analysis is essentialin establishing the level of confidence one can have in the results of calculations.
There are a number of uncertainties associnted with the computations using transport codes. .
The first set of uncertainties results from numerical approximations such as: the order of l quadrature usedin S ucalculations, the scattering cross-section expansion, choice of meshes in the model, energy group structure, statisticalconvergence criteria, etc. The second set of
- uncertainties results from modelling approximations such as: surveillance capsule placemant, RPV thickness variations, cavity streaming effects, 3-D flux syn thesis, peripheralsubassembly \
source distribution, dimension andmaterialuncertainties. The thirdset of uncertainties results l from uncertainties in nuclear data such as: cross-sections, dosimeter cross-sections, U235 fission spectrum above 6 MeV.
A major effort has been devoted to the analysis of cross section data uncertainties and derived quantities such as displacement per atom (DPA) cross-sections, kinetic energy release in materials (kertria), and gas production data. The full uncertainty analysis requires: }
I 56 l i
l
f
- 1) Least-squares methods for estimating the values of cross-sections and their covariances,
- 2) processing of cross-sections and covariances into a multigoup form,
- 3) sensitivity and uncertainty analysis of target quantities (e.g., fluence and doses), and l 4) a consistency analysis of the experimental and calculated data and improvement through adjustment. "
Paper 3: " EXPERIMENTAL AND THEORETICAL STUDIES ON VVER 1000 REACTOR DOSIMETRY" The following commentary was taken from the Osmera et al. Prague paper [Os961:
" INTRODUCTION - The experimentaland theoreticalinvestigations of the reactor dosimetry problems startedin NRIand Skoda 20 years ago. Due to the unsatisfactory reliability of the calculations and input data libraries the programme of the LR-O mockup-ups has been launched. ~
Several VVER-440 mockups with standard and reduced cores have been investigated. The mockup experimental data have been used in evaluation of the neutron monitors in surveillance specimens programme (H096] and ex-vesselposition measurements (HoS6al.
. ..... . ....The neutron spectra have been measured in the horizonal centralplane along the 1 axis of the mock-up symmetry with the proton recoilspectrometers. The power distribution has been checked by the gamma scanning of the fuel pins. The measuring program emphasizes the RPVcriticallocation and the monitoring ones. The ID and 2D (ANISN, DORT) l transport calculations with the SAILOR and BUGLE-93 libraries have been performed.
In the Skoda design of the surveillance program the specimens are placed in the rectangular boxes fixed at the RPVinner wallin the position of the maximum of the fast flux neutron censity and the acceleration (lead) factor is about 1.9 for the fast fluence above 0.5 MeV.
The specimens boxes are equipped with Fe, Nb, Co, Cu, Np237 and U238 detectors and Cu and Fe wires for space dependent distribution measurement. . ... .... ..... .
EXPERIMENTAL PROGRAMME & EXPERIMENTAL TECHNIQUES - .. . . . . . . Th e L R-O lo w l power limit the measurement with activation detectors, but the differential energy neutron
- .soectrum is a better test of calculation model and methodology than a set of reaction rates usually measured in many benchmarks.
The radial-azimuthal core power distribution use to be checked by means of the gamma scanning of a reasonable set of fuelpins (severalhundreds for the radialand several tents for the azimuthaldistribution). The radialand azimuthaldistributions of the last flux in the vicinity of the core orin the central tubes of the fuelassemblies could be measured with the indium fails, in the positions from the core the proton recoil detectors could be used for the oistribution measurements.
^
57
I-1 l
CALCULA TIONS - The transport codes ANISN, DORT, TORT and the libraries VITAMIN-C, SAILOR, BUGLE-93 have been implemented. The core distnbution was calculated with MOBY DICK, 2D, diffusion, multigroup, pin-to-pin code. Some distributions were calculated using tne " influence functions. " Using the point kernelmethod the contributions of 20 -50 layers over the height of the core to the chosen point are calculated. The effective cross sections are evaluated via 1D transport calculations. The influence function calculation (integral of l
Green functions over a defined area of the core) follows the MOBY DICK results. Using this system it is possible to calculate the time dependent integral (above defined energy) fast ,
l neutron flux density distribution in the position of a neutron monitor.
{
DISCUSSION OFRESUL TS -- The neutron spectra were measured at the outer surface of the barrel, at the PV inner surface, and the FV outer surface. The experimental spectra were I transformedinto the SAILOR (BUGLE) group format, and the broad group approxima tion (O.1, 0.2, 0.5,1.0, 2.0, 3.0, 7.0 and 10,0 MeV) which were used for the comparison with the calculation. Because the absolute powerin LR-O cannot be determined the measurements are l
miative and a monitoring system covering several orders in the neutron flux density is used.
Comparing the shapes of a spectrum the measured and calculated spectra were normalized to integral flux in the energy range 1 to 10 MeV or O.1 to 10 MeV. . . , . ..
The spectra measured by both spectrometers were identicalin the frame of usualuncertainhes i l known from the measurement in reference fields. The proton recoil spectrometers' experi- l
\ mental (higher upper energy limit) and 2D-DORT, BUGLE-93 Skoda calculation were used for the comparison. The sensitivity of the results to the substitution of SAILOR BY BUGLE-93, P7, was studied with ID (ANISN) calculation. The differences (improvements) were remark-able. Nevertheless the discrepancy of the 2D-DORT, BUGLE 93, calculation and experiment i is substantial for the evaluation of the reliability of the RPV exposure calculation.
The calculation rnostly underestimated the fast fluxes.
- Comparing the calculated and measured space energy distributions and space energy indices, broad groups and integral
- f!uxes above defined energy, it could be stated that the tendency in Fig. 3 (Normalized to l integrel flux in the energy interval 1-10 MeV at the barrel outer surface position) is similar as !
, in Fig. 2 (Normalized to integral flux in the energy interval O.1-10 MeV at the barrel outer surface, PVinner surface, and PV outer surface positions) except at the PV outer surface, the oisagreements in attenuations of the integral fluxes over the waterlayer and the RPV (Fig. 4 l Measured and Calculated Indices / Comparison of Integral Fluxes) corresponds to the ones in l l soectra. The results of calculation with the " influence functions" are in Fig. 3 marked as l l 3D-INCAL C. The disagreement of calculation and experiment is probably caused by the used l group library. In several studies the sensitivity of the results to the type of the library are l presented (Ha96s,Ch921. The Monte Carlo calculations of similar problems (Osxx,Woxx]
'show substantiefly better consistency with experiment."
l i
l 8
This is the same trend observed for the PCA Experiments and Blind Test. That is, compared with the measurements, most calculations showed a trend towards under-estimating the fluxes (E > 0.1 and 1.0 MeV) by 5% to 25%, and even more at higher neutron energies and deeper steel penetrations
[Mc81,Mc84).
58
i Eaper 4:
" NEUTRON DOSIMETRY IN EXTENDED SURVEILLANCE PROGRAM ON 4TH UNIT OF NPP DUKOVANY" As stated by Hogel et al. in their Prague paper [Ho96]:
" INTRODUCTION The Standard Surveillance Program for the reactors VVER 440/213 of Dukovany NPPdemonstrateda numberof shortcomings. As to the neutron fluence monitoring tne surveillance containers (capsules) were equipped with only a restricted number of activation monitors - S4Fe, 63Cu, and 93Nb. Moreover, the surveillance container chains contained only severalsets of monitors and thatis why the detailed description of the vertical neutron fluence distribution was not possible..... ... Therefore it was not possible to oetermine reliably the neutron fluence in the individual surveillance specimens. As well the neutron fluence on the reactor vesselcouldbe determined with great uncertainty. Uncertainty in the individualsurveillance specimen neutron fluence led to problems with the application of the mechanical test results to the RPV state and life time assessment.
J l
DETECTORS FOR THE ESP -- The Extended Surveillance Program (ESP) was designed to eliminate all the above described shortcomings.. ..The ESP included two chains of20 ,
surveillance containers each. ....... . . .The chains were irradiated in the NPP Dukovany \
Unit 4, Cycle 7.
Containers were equipped with either Fe and/or Nb wires for neutron fluence distribution f
assessment both inside the container and along the whole chain. Three types of wires were !
used:
- O wire, a containerperimeter neutron fluence distribution monitor, located above the .
surveillance specimens (in several containers).
l-wire, a container height neutron fluence distribution monitor, located along the surveillance specimens (in all containers).
- H wire, a detailed verticalneutron fluence distribution monitor, inserted directly into the specimen notch (in severalcontainers).
Specified surveillance containers contained additional spectrometric sets of neutron fluence monitors Fe, Cu, Ni, Nb, alloy Al-Co and fission detectors 238U and 237Np..... . . . ...
In the second chain the spectrometric sets were equipped with isotopically enriched monitors l 54Fe, 63Cu, 93Nb, and 59Co separated by naturalpellets of Ti. .. ....... . ... . . ..... .
A CTIVITY MEASUREMENTS -- were performed in the two independent laboratories - NRI Rez and SKODA Nuclear Machinery. Both the laboratories used HPGe spectrometer for the 1
- standard measurements and HPGe spectrometer with Be window for the Nb X-ray 1 measurements.
r Before measurement the O-wires were dividedinto 6-8 approximately equalpieces andI-wires l into 4-S pieces depending on their length. By this way the container perimeter and height l 59
I l
neutron fluence distribution was obtained.
1 Local changes of the neutron fluence rate along the core height depending mostly on the fuel i
\ burn up during the operation cycle were also taken into account. Correction to the container orientation was made in order to get the activities in the container axis. Nb activities were corrected for the X-ray self-absorption and for the X ray fluorescence caused by decay of bo th
! 94Nb and 182Ta impurities.
RESUL TS AND DISCUSSION-- Examples of the container height neutron fluence distributions
) (I-wires) are presented...... .. It can be seen that the height distribution changes from 0.8 to
- 1.2 relative to the mean value in the first container placed at the top chain position. In the
? containers from the middle part of the chain .... . the vertical fluence distribution is close to th e con tain er mean value. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
A detailedrelative neutron fluence distribution in the container. . equipped with both H-wires I and 0 wires and spectrometric detector set is shown in .... .. ..The value of the neutron l fluence in the specimen notch varies from 0.83 to 1.16 relative to the spectrometric set position.
l Using the height neutron fluence distribution determined from the O-wires the reaction rates
{
l tar the spectrometric sets were corrected, the results are presented in ... ......
The reaction rates were adjusted to an a priorineutron spectrum which had been measured by Holman (Ho91], Jansky and Marek (Bu86] on the VVER-440 mock-up using differential i proton recoilmethods.
(
The SAND ll adjustment procedure was used with IRDF-90 cross section library. Adjusted neutron fluences after one iteration are given in Tables 3 and 4 for the chain No.1 and 2, respectively. From this process the 238U monitors had to be excluded as they showed inconsistency with the a priorispectrum. Probable reason is the content of 235U was too l high for this type of neutron spectrum. The standard deviation of the measured and calcubstedactivities varies from 1.9 % to 6.1 which shows good consistency between neutron dosimetry data and the a priorineutron spectrum. * ..........
1 \
CONCL USIONS -- From the results presented above the following conclusions can be stated:
l
- 7he perimeter distribution measurement (0-wire) is necessary so that the detailed l description of the neutron fluence field inside the container can be assessed and the vsrticalneutron fluence distribution along the chain can be estimated.
l l
' A ma ior problem with using discrete ordinates transport codes is that the group structure is not fine enough to properly handle the effect of fine structure in the actual neutron spectrum and transport code and dosimetry cross sections. Using the higher resolution measured proton recoil spectrum, the 640 group UAND 11 dosimetry cross sections, and the SAND ll iterative (or FERRET-SANDil least-squares) adjustment code eliminates the need for the use of processed broad group averaged cross sections. This problem is also eliminated or minimized by the use of Monte Carlo transport codes with high group st'uctures up to - 8000.
60 l
- Only in the few bottom and top containers the verticalneutron fluence distribution in the container should be measured using the H wires.
Spectrometric sets should be equipped only with Fe, Cu, Ni, Nb, Ti, and Co. The usage of the fission detector 237Np is not necessary as we are able to measure and correct Nb monitors. The 238 monitors are not suitable for this type of neutron spectrum. The content of 235Uis too high andits influence cannot be eliminated by a Gd cover. To ensure possibility of Nb correction the Nb monitors should be used both bare and covered by Gd. "
With reference to the commentary presented in Paper 3, Osmera et al. have stated:
"The proton recoil spectrometers' experimental (higher upper energy limit) and 2D-DORT, EUGLE-93 Skoda calculation were used for the comparison. The sensitivity of the results to tne substitution of SAILOR BY BUGLE-93, P7, was studied with ID (ANISN) calculation. The cifferences (improvements) were remarkable. Nevertheless the discrepancy of the 2D-DOR T, BUGLE-93, calculation and experimentis substantialfor the evaluation of the reliability of the j RPVexposure calculation. . .. . . . .. . ... . . . .. . .. . . . ... . . .. .. . .. '
The calculation mostly underestimated the fast fluxes. .. .. ..... . . .. . . . ... .
1 i
The disagreement of calculation and experiment is probably caused b y the used group library.
In several studies the sensitivity of the results to the type of the library are presented (Ha96a, Ch92]. The Monte Carlo calculations of similar problems [Osxx,Waxx] show substantially better consistency with experiment."
What the above says is that in the Czeck Republic, by appropriate LR-O Benchmark VVER-440 mockup studies, the NRI Rez and NPP Dukovany staffs have determined that they cannot rely on absolute unadjusted (by dosimetry measurements) transport calculations to determine l reliable surveillance capsule specimen and RPV end-of-life neutron iluence exposure values. ,
i l
That is, and as Hogel et al. have already stated: i "Therefore it was not possible to determine reliably the neutron fluence in the individual surveillance specimens. As well the neutron fluence on the reactor vessel could be determined with great uncertainty. Uncertaintyin theindividualsurveillance specimen neutron fluence led to problems with the application of the mechanical test results to the RPV state and life time assessment. "
61
2 Table 3. Fast neutron fluence (t/cm ) in the container axis at various elevation above core midplane.
chain No.l.
Elevat. SD 0 0 0.398eV 10 kev 0.lMcV 0.5MeV IMeV
[m] % 0. lev 0.398eV 10 kev 0.lMeV 20MeV 20MeV 20MeV 1.217 6.1 8.45E+17 1.01E+ 18 5.67E+18 4.18E+18 1.69E+19 9.38E+ 18 4.84E+ 18 1.119 4.8 2.13E+18 2.56E+18 1.29E+ 19 9.62E+18 3.98E+19 2.19E+19 1.12E+19 0.233 5 6.16E+ 18 7.40E+18 3.72E+ 19 2.78E+19 1.15E+20 6.35E+19 3.28E+19 0.233 3.3 9.68E+19 1.16E+20 1.42E+20 3.88E+19 1.20E+20 6.55E+19 3.35E+19 0.107 2.7 1.24E+18 1.49E+18 1.82E+18 . l.99E+19 1.21E+20 6.57E+ 19 3.29E+19 Table 4. Fast neutron Buence (1/cm ) in the container axis at various elevation above core midplane, chain No.2.
Elevat. SD 0 0 0.398cV 10 kev 0.lMeV 0.5MeV IMeV -
[m] % 0. lev 0.398eV 10 kev 0.!McV 20MeV 20MeV 20MeV 1.23 1.9 5.85E+17 7.02E+17 8.59E+17 1.00E+18 5.1IE+18 27,$E+14 140E+18 1.215 5.9 2.17E+18 2.61E+18 3.19E+18 2.90E+18 1.43E+19 7.88E+18 4.06E+ 18 1.116 3.1 5.67E+18 6.80E+18 8.32E+18 8.41E+18 4.20E+19 2.30E+19 1.17E+ 19 0.99 4.1 4.82E+18 5.79E+18 7.08E+18 1.27E+19 6.85E+19 3.73E+ 19 1.88E+19 0.609 5.4 1.15E+19 1.38E+19 1.69E+19 2.08E+19 1.08E+20 5.88E+19 2.99E+19 0.102 2.9 1.24E+19 1.49E+19 1.83E+19 2.35E+19 1.23E+20 6.72E+19 3.44E+19
-0.41 5.5 1.33E+19 1.59E+19 1.95E+19 2.38E+19 1.23E+20 6.74E+19 3.44E+ 19
-0.79 4.9 1.08E+19 1.30E+19 1.59E+19 2.06E+19 1.07E+20 5.87E+19 2.99E+19
. -0.91 2.4 9.52E+18 1.14E+19 1.40E+19 1.89E+19 9.82E+19 5.34E+19 2.68E+ 19
-1.04 4.7 7.80E+18 9.37E+18 1.15E+19 1.63E+19 . 8.58E+19 4.69E+19 2.38E+19 l
62
Paper 5: " FAST NEUTRON FLUENCE MONITORING ON NPP DUKOVANY" As stated by Hogel et al. in their Prague paper [Ho96al:
The results and experiences gained by the ex vessel fast neutron fluence measurements performed at the nuclear power plant (NPP) Dukovany are summarized in the present paper.
The continuous monitoring is obligatory for all four units of NPP. Insertion of the activation cetectors into the reactor cavity and their withdrawal are performed during the refueling outage. The time corrections for the decay during the irradiation were performed with respect to the time course of reactor power . The neutron spectra measured by differential methods on mock up experiments in the LR-O experimental reactor were used to determine the fast neutron flux density and fluence in the reactor cavity, on the inner surface, and 1/4 T of the FPV thickness. Discussion of results and uncertainty propagation is performed.
INTRODUCTION -- The knowledge of fast neutron spectrum and fluence impinging on the reactor pressure vessel IRPV) of nuclear power reactors is one of the basic criteria for evaluation of its lifetime. The monitoring of the neutron fluence on the RPVinner surface in VVER 440 reactors can be performed only indirectly, eitherin the surveillance containers, or on the RPV outer surface, in the reactor cavity. In both cases we have to know the conversion factors for determination of neutron fluence on the inner surface and on the '1/4 of the pressure vessel thickness, respectively. These factors can he obtained experimentally in mock-up experiments or from calculations The standard surveillance program on all four units of NPP Dukovany was finished after five cycles. But it was not too suitable for determination of the neutron fluence on the RPVinner surface. In essence, it was not even designed for this purpose. First of all, the detector choice was also considerablylimited. Secondly, the detectors wereplaced eccentricallyin the surveillance containers which can cause, together with unknown orientation of the container in relation to the core, a systematic error up to 20% in neutron fluence in surveillance soecimen positions.
Ex-vessel fluence monitoring in the reactor cavity has many indisputable advantages.
Detectors can be located at anyplace behind the & V and therefore it is possible to obtain the complete azimuthal and vertical neutron fluence distribution. They are replaced after each cycle which also enables to use nuclear reactions having a shorterlife-time (nickel, titanium, f p.), and activation detectors can be readily replaced or added into the detector set. The measurement and the evaluation can also be performedin a relatively short time period after tne end of cycle (~ one month). The fact that the conversion factors for criticalpoints of the RPV can be obtained more accurately and reliably from this measurements then from the
. detectors located in the surveillance containers is considered to be a big advantage as well.
Regular fast neutron fluence monitoring in the reactor cavity startedin 1992 and one canjust pity thatit was notperformed since the beginning of the operation as a logicalsupplement to the surveillance program.
FLUENCEDETERMINA TIONINRPV CRITICAL POINTS-- There are two criticalpoints of VVER 440 RPV. The azimuthal and verticalmaximum of neutron fluence, and the weld No4 lying in the lower part of the reactor core. The spatialneutron fluence distribution measurements enable us to find the location of the fluence maximum and also to determine the conversion 63 l
l
f.sctor for the weld position.
Two methods are used to obtain the neutron fluences from the measured reaction rates of the activation detectors. In !mth of them, the spectra measured by Holman (Ho91], Jansky and Marek (bub 6] on the VV. , 440 mock-up using the differentialproton recoilmethod, play the cecisive role. These expenments were carried out on the LR-O Benchmark Reactorin NRI-Rez.
In the first method, the so-called effective cross sections a,, are calculated for single detectors using the definition....... . . ... .... .... . .... . .. ... .. .So we can write
@,, = Reaction Rate / a,,
where @,,is the neutron flux density.
Using the evaluated reaction rates and the effective cross sections calculated in spectra measured on the mock-up, we can simply obtain the neutron flux density above the chosen energy for allused activation monitors. Itis obvious that this method can be used only when the neutron spectrum is well known.
The second way of experimental data handling is the spectrum unfolding from evaluited reaction rates using the SAND ll code, where the spectra measured on mock-ups by cifferentialmethods are used as the guess spectra. The neutron flux densities and fluences are calculated from the resulting spectra. In all cases the dosimetry sensor cifferential cross sections from IRDF 90 library are used for the evaluation.
The values of fast neutron flux densities over 1 MeVin the reactor cavity from two different measurements gained by both methods are presentedin Table 2. Good agreement of results obtained from nuclear reactions having considerable different energy responses confirms the aoplicability of spectra measured on the mock-up for the activation data evaluation.
Finally, the fluences on the inner surface of the pressure vesselandin 1/4 of RPV thickness are obtained by multiplying the values evaluated for the outer surface by attenuation factors obtained from the mock-up experiment. The factors were measured using both differential and activation methods.
DISCUSSION OFRESUL TS -- The evaluated maximal fluences above 1 MeV on the RPVinner surface in measured cycles are in the range 3.96-4.82xlO ' m, depending on the power cistribution in the core and the cycle length.
..The following main components contribute to the total uncertainty of the evaluated fast neutron fluence on the inner surface of RPV:
- The uncertainty of the measured activities and evaluated reaction rates, I
- the evaluation of :he neutron flux density and fluence from reaction rates, and
- the uncertainty of the attenuation factor.
As to the first item, we can say that the determination of the uncertainty of reaction rates is 64
satisfactorily solved. In Table 3 there are the variances and the covariance matrix of the evaluated reaction rates from the measurements performed on Unit 1. Cycle 8. The list of partial components is presented in Table 4. Some of these values itime course of reactor power, recalculation to the maximum) are fully correlated. The correlation coefficients of gamma detector efficiencies are given by the program ETA, and the remaining values are supposed to be uncorrelated.'
The missing covariance matrices of differential cross sections and ones of the spectra measured on the mock-ups doesn 't allow us to determine statistically correctly the remaining two components.
In Table 5 there are the ratios of fluences obtained from single detectors using the effective cross section method to fluences obtained from SAND ll. The rather small dispersion of the presented values in the upperpart of table confirms the applicability of spectra measured on tne mock-up for our evaluation of the activation data. Nb, Np and U detectors were not included in the SAND ll unfolding, as we have performed only a few measurements and we are not sure the measuring and evaluation process is fully correct.
7he uncertainty of the attenuation factorgiven by Holman is 15% for the threshold 0.5 MeV and 10% for 1.0 MeV. Our conservative estimate of the totaluncertainty of the fast neufron fluence on the inner surface of the RPVis 15% for 0.5 MeV and 10% for 1.0 MeV. "
I 65
Tab.2: Measured reaction rates and evaluated fast neutron flux densities. The comparison of results obtained using effective cross section method and from SAND 11 code Reaction Unit 1. cycle 9 Unit 4. cycle 7 tr lsl @i ,, lm'2 s l rr lsl @, o lm' s 'l "Fe(n.p) 1.18E 15 2.07E+ 14 1.03E 15 1.81 E+ 14
Ni(n. p) 1.58E-15 1.98E+ 14 1.42E-15 1.77E+ 14 "Ti(n.p) 1.96E- 16 2.23 E+ 14 1.68E-16 1.92 E+ 14
Cu(n.u) 1.23E-17 2.15 E+ 14 1.07E 17 1.87E+ 14 "Mn(n.2n) 1.28E 17 1.82E+14 1.19E- 17 1.69E+ 14 "Y(n.2 n) 1.07E 17 1.95E+14 Mean value 2.03E+14 I 82E+14 SAND 11 2.04E+ 14 1.79E+14 Tab.3. Reaction rates, their variances and correlation matrix from the measurement in bie unit 1, cycle 8.
Reaction rr isl 61%l Correlation matrix (x100)
"Fe(n.p) 1.163E-15 3.2 100
Ni(n.p) 1.599E 15 2.8 62 100 "Ti(n.p) 1.856E-16 2.6 67 78 100
Cu(n.a) 1.184E-17 2.6 67 78 84 l(x)
,, l 66
1 ab 4 Partial components for calculation of variances and correlation matris of esaluated reaction rates.
Reacuon Vanance [%l cla eps g N K M N. , ssg 6
"Fe(n.p)"Mn 0 77 - 01 I l' 2' l.7 -
b
".N (n.p)"Co 0.77 O.3 0.1 I l' 2' - -
6 "Ti(n.p)"'Sc 0 76 - -
I t* 2' - 02 6
"'Cu(n.a)*'Co 0.77 - -
I l* 2* -
02 eta detection efficiency eps gamma branching ratio g foil mass N counting statistics K time course ofirradiation M recalculation to the maximum No number of target nuclei in the foil ssg gamma self absorption in the foil
- a fully correlated b corr (etal, eta 2)= corr (etal, eta 3)= 1.00 corr (eta 2, eta 3)= corr (eta 3, eta 4)= 1.00 corr (etal, eta 4)= 0.94 corr (eta 2, eta 4)= 0.93 -
Tab.5: The ratio of fluences obtained from single monitors to fluences evaluated by SANDil (without Nb, Np and U)
Unit 1 2 3 4 C)cle 8 9 10 8 9 10 7 8 9 6 7 8 Fe 1.01 1.04 0.93 0.94 0.98 0.96 0.98 U.96 1.02 1.04 0.98 0.97 Ni 0.97 1.00 0.95 0.98 0.97 1.06 1.01 0.99 1.00 1.02 0.95 0 94 Ti 1.02 1.06 1.05 1.06 1.05 1.07 1.06 1.04 1.% l.06 1.03 1.02 Cu 1.00 1.07 0% l.00 1.02 1.03 1.01 0.99 1.04 1.00 1 02 1.01 Mn 0.86 0.88 0.90 0.96 0.92 0.86 0.85 0.91 0.88 0.94 0.93 Y 0.93 0.97 Np 0.88 09.1 U 0.90 1.04 Nb 1.27 1.23 1.12 1.03 1.08 1.18 1.08 1.14 67 I
4
REFERENCES
[As97] " ASTM E706 Master Matrix of LWR Surveillance Standards", Annual Book 01 ASTM Standards, Vol.12.02, ASTM, Philadelphia, Current Edition.
[Ba96] A. Ballesteros, L. Debarberis, and W. Voorbraak " Proc. 9th ASTM-EURATOM Svmoosium on Reactor Dosimetrv. Prague, Czech Republic, Sept. 2 6, 1996.
[Ba96a] A. Ballesteros, J. Bros, C. Cueto Felgueroso, AMES Report on " Dosimetry and Neutron Transport Methods for Reactor Vessel Surveillance,"
COSU-CT94-0073 ES, EC-DGXI, January 1996.
[Bo96) T. C. Bordine, " Docket 50-255-License DPR-20-Palisades Plant: Reactor Vessel-Fluence Measurement Process Best Estimate Clarification." Consumer Energy Letter to U.S. NRC, Document Control Desk, June 26,1997.
[Br95] M. Brumovsky, A. Kryukov, F. Guillemot and V. Levin, "Results from the Phase ill of the IAEA Coordinated Research Programme: Optimizing of Reactor Pressure Vessel Surveillance Programmes and Their Analysis," Proceedings of the Specialists Meeting on irradiation Embrittlement and Mitigation, Finlaild, Espoo, October 1995. {
l
[Bu86] J. Burian, B. Jansky, Z. Turzik, M. Marek, and J. Racek, " Spectrum Measurements on VVER-440 Mock-Up," UJV 7883 R,T,1986, in Czech.
[Ca90] J. F. Carew et al., " Status of a New Regulatory Guide on Methods and
' Assumptions for Determining Pressure Vessel Fluence", Proc. of the 7th ASTM-EURATOM Svmoosium on Reactor Dosimetrv, Strasbourg, France, I August 27-31, 1990.
[Ca96] J. F. Carew and A. Aronson, " Palisades Cycles 1-11 Pressure Vessel and Cavity Fluence Evaluation," BNL Letter to L. Lois, November 15,1996; attached to the ;
NRC Letter from J. N. Hannon to T. C. Bordine, Manager, Licensing, Palisades Plant [Ha96). J
[Ch92] B. J. Chang, Z.-W. H. Lin, Ch. Ch. Wun, " Calculations of EURACOS Iron i Benchmark Exp. Using HYBRID Method, Nucl. Sc. Eng.,112, p. 54,1992.
[Es95] U. von Estorff, S. Crutzen, L.M. Davies, C. English, "AMES Ageing Materials Evaluation and Studies: A progress Report," Workshop on Ageing on NPP Component Materials, St. Petersburg, 28.02-02.03, 1995.
[Fa78] A. Fabry, W. McElroy, D. Gilliam, E. Lippincott, J. Grundl, and D. Hanson,
" Review of Microscopic Integral Cross Section Data in Fundamental Reactor Dosimetry Benchmark Neutron Fields," Consultants Meeting on Integral Cross-Section Measurements in Standard Neutron Fields, International Atomic Energy Agency, Vienna, November 1519,1976, IAEA 208, Neutron Cross Sections for Reactor Dosimetrv.1978.
68
[Fe93) J. D. Feit, " Voluntary Standards -- A Common Sense Approach: The Depart-ment of Energy and Nongovernment Standards Bodies: An Equal Partnership to Assure Success," ASTM Standardization News, p. 36, September 1993.
[Go89] R. Gold and W. N. McElroy, "The LWR PV-SDlP: Past Accomplishments, Recent Developments, and Future Directions", Reactor Dosimetrv: Methods. Aeolica.
tions. and Standarotzation. 6th ASTM-EURATOM Svmoosium on Reactor Dosimetrv, Jackson hole, Wyoming, May 31 to June 6,1987, STP 1001, p.
44, ASTM, May 1989.
[Go94] R. Gold, " Neutron Fluence Determination for LWR Pressure Vessels," Reactor Dosimetrv. ASTM STP 1228, Harry Farrar IV, E. Parvin Lippincott, John G.
Williams and David W. Vehar Eds., ASTM, Philadelphia, p. 104 1994.
[Gr931 P. J. Griffin, J. G. Kelly, T. F. Luera, and J VanDenburg, SNLRML Recommended Dosimetrv Cross Section Comoendium, Sandia National i Laboratories, Albuquerque, NM, Report SAND 92-0094, November 1993.
(Gu82) G. L. Guthrie, W. N. McElroy and S. L. Anderson, "A Preliminary Study of the Use of Fuel Management Techniques for Slowing Pressure Vesel s Embrittlement," Proc. of the 4th ASTM EURATOM Svmoosium on Reactor Dosimetrv. Gaithersburg, MD, March 22-26,1982, NUREG/CP-0029, Vol.1, NRC, Washington, DC, pp. 111-120, July 1982.
(Gu82a) G. L. Guthrie, W. N. McElroy and S. L. Anderson, " Investigations of Effects of Reactor Core Loadings on Pressure Vessel Neutron Exposure," LWR-PV-SDlP:
Quarterly Progress Reoort. October 1981 - December 1981. NUREG/CR-2345, Vol. 4, HEDL-TME 81-36, NRC, Washington, DC, Section E and Appendix A, pp. HEDL HEDL-36 & HEDL A1 - HEDL A46, October 1982.
- [Gu841 G.L.Guthrie, "Charpy Trend Curve Based on 177 PWR Data Points,"
l LWR-PV-SDlP: LWR Pressure Vessel Surveillance Dosimetrv Imorovement Program, NUREG/CR-3391,Vol. 2, HEDL TME 83-22, Hanford Engineering Development Laboratory, Richland, WA, April 1984.
[Hi94] A. L. Hiser, Jr, "USNRC Plans for Evaluating and improving RPV Integrity and irradiation Embrittlement Rules and Regulatory Guide," Reactor Dosimetrv.
ASTM STP 1228. Harry Farrar IV, E. Parvin Lippincott, John G. Williams and David W. Vehar Eds., ASTM, Philadelphia,1994.
[Ha96) J. N. Hannon, Palisades: Evaluation of Updated Reactor Pressure Vessel Fluence l Values (TAC NO. M95134), NRC Letter to T. C. Bordine, Manager, Licensing, Palisades Plant, December 20,1996.
[Ha96al H. L. Hansaw, A. Hagighat, and J. C. Wagner, "Multigroup Cross-Section Generation with Spatial and Angular Adjoint Weighing," Transactions of the American Nuclear Society, p. 175,1996.
69
l l
l [Ho91] M. Holman, Exoerimental Research of Soatial and Energetical Distribution of l Neutrons in Inner Shielding of Light Water Reactors, Doctoral Thesis. VSSE l Plzen,1991, in Czech.
[Ho961 J. Hogel, M. Hort, M. Marek, B. Osmera and F. Tomasek, " Neutron Dosimetry in Extended Surveillance Program on the 4th Unit of NPP Dukovany," Proc. 9th ASTM-EURATOM Svmoosium on Reactor Dosimetrv. Prague, Czech Republic, Sept. 2-6,1996.
[Ho96al J. Hogel and M. Hort, " Fast Neutron Fluence Monitoring on NPP Dukovany,"
Proc. 9th ASTM-EURATOM Svmoosium on Reactor Dosimetrv. Prague, Czech Republic, Sept. 2-6,1996.
[Ko961 K. Kono, "OMB A-119 Becomes L'6w: A Win-Win Situation for Both the Public and Private Sectors," ASTM Standardization News, p. 40, May 1996.
[Li83) E. P. Lippincott and W. N. McElroy, Ed., FTR Dosimetrv Handbook, HEDL MG-166, Hanford Engr. Devel. Lab., Richland, WA, March 1983.
[Li86] E. P. Lippincott, Evaluation of Surveillance Caosule and Reactor Cav'itv Dosimetrv from H.B. Robinson Unit 2. Cvele 9. Prepared for USNRC and Carolina Power & Light Company, Westinghouse Report WCAP-11104 and NRC Report NUREG/CR-4576, December 1996.
[Li941 E. P. Lippincott et al., Westinohouse Surveillance Caosule Neutron Fluence Reevaluation. Westinghouse Report WCAP-14044, April 1994.
[Li961 E. P. Lippincott and S. L. Anderson, " Systematic Evaluation of Surveillance Capsule Data," Proc. 9th ASTM-EURATOM Svmoosium on Reactor Dosimetrv.
Prague, Czech Republic, Sept. 2-6, 1996.
[Lo89) L. Leis, " Reactor Dosimetry and Nuclear Reactor Reg 4 tion," Proc. 6th ASTM-EURATOM Svmoosium on Reactor Dosimetrv. Jack an Hole, Wyoming, May 31-June 6,1987, STP 1001, p.12, ASTM, May 1989.
[Lo91] A. L. Lowe, Jr.., " Role of Neutron Fluence in Reactor Vessel Integrity", AliS Transactions. Vol. 64, p. 491, San Francisco, Calif., Nov. 10-14,1991.
[Lo97] L. Lois and J. Carew, " Minutes from the Orlando, FL, ANS-19.10 Meeting, June 2,1997 and Confirmation of the November 16,1997 Meeting in Albuquerque, NM," Memorandum, BNL, September 20,1997.
[Ma89) W. Mannhart, " Status of Cf-252 Neutron Spectrum as a Standard," Reactor ,
Dosimetrv. Methods. Acolications. and Standardization. 6th ASTM-EURATOM l Svmoosium on Reactor Dosimetrv. Jackson Hole, Wycming, May 31 to June j 6,1987, STP 1001, p. 340, ASTM, May 1989. 1 I
l i
j 70
)
- \
[Mc67] 'N N. McElroy, S. Berg and T. Crocket, SAND 11: A Comouter- Autom_atf.4 Iterative Method of Neutron Flux Scectra Determined bv Foil Activation, AFWL-TR 67-41, Vol. l-IV, Air Force Weapons Laboratory. Kirkland AFB, NM.
July 1967.
[Mc75] W. N. McElroy et al., " Materials Dosimetrv and Technical Pacers: Interlaboratorv LMFBR Reaction Rate (ILRR) Program." Nuclear Technology, Complete Volume Vol. 25, No. 2, February 1975.
[Mc75al W. N. McElroy and L. S. Kellogg, " Fuels and Materials Fast-Reactor Dosimetry Data Development and Testing," Nuclear Technology, Vol. 25, No. 2, pp.180-223, February 1975.
[Mc77] W. N. McElroy, R. A. Bennett and D. L. Johnson, " Neutron Environmental Characterization Requirements for Reactor Fuels and Materials Development and Surveillance Programs," HEDL-SA-911, and Proc. of the 1 st ASTM EURATOM Svmoosium on Reactor Dosimetrv. Petten, The Netherlands, September 22-26, 1975, EUR 5667, Vol. I, Commission of the European Communities, pp.1-26, 1977.
[Mc79] A. K. McCracken, "Few Channel Unfolding in Shielding - The SENSAK Code."
Proc. of the 3rd ASTM-EURATOM Svmoosium on Reactor Dosimetrv. Ispra, Italy, October 1-5,1979, EUR 6813, Vol. II, Commission of the European Communities, p. 732,1980.
[Mc81] W. N. McElroy, Ed., LWR-PV-SDlP: PCA Exoeriments and Blind Test.
NUREG/CR-1861, HEDL-TME 80-87, NRC, Washington, DC, July 1981.
[Mc82] W. N. McElroy et al., " Surveillance Dosimetry of Operating Power Plants", Proc.
of the 4th ASTM-EURATOM Svmoosium on Reactor Dosimetrv, Gaithersburg, MD, March 22-26,1982, NUREG/CP-0029, NRC, Washington, DC, Vol.1, p.
3, July 1982.
[Mc83] A. K. McCracken and A. Packwood, The Soectral Unfolding Program SENSAK, DPT/SC/P(83)13, UKAEA, Winfrith, UK,1983.
[Mc84] W. N. McElroy, LWR-PV-SDlP: PCA Exoeriments. Blind Test. and Phvsics-Dosimetrv Succort for the PSF Exoeriments, NUREG/CR-3318, HEDL TME 84-1, NRC, Washington, DC, September 1984.
[Mc85] W. N. McElroy, F.B.K. Kam, J. A. Grundl, E.D. McGarry, and A. Fabry, Eds, LWR-PV-SDlP: 1984 Annual Reoort. NUREG/CR 3746, Vol. 3, HEDL-TME 84-31, NRC, Washington, DC, April 19985.
[Mc86] W. N. McElroy and E. P. Lippincott, Eds, LWR-PV-SDlP: 1985 Annual Reoort, NUREG/CR-4307, HEDL-TME 85-14, NRC, Washington, DC, January 1996.
71
(Mc87a] W. N. McElroy, Ed., LWR-PV-SDlP: LWR Power Reactor Surveillance Phvsics-Dosimetrv Data Base Comoendium. NUREGiCR-3319, HEDL- TME 85-3.
Hanford Engr. Devel. Lab., Richland, WA, March 1987 Update.
[Mc87b] W.N. McElroy. and R. Gold, Eds., LWR-PV-SDlP: PSF Physics-Dosimetrv Prooram, NUREG/CR-3320, Vol. 3, HEDL-TME 87-3, NRC, August 1987.
[Mc881 W. N. McElroy and L. S. Kellogg, " Dosimetry Adjusted Reactor Physics Parameters for PV Neutron Exposure Assessment", Transactions of ANS, Vol.
57, p. 223, Nov.1988.
[Mc91] W. N. McElroy, E. P. Lippincott, A. L. Lowe, Jr., and P. D. Hedgecock, C.ondi:
tion Assessment and Surveillance of Nuclear Reactor Pressure Vessel Steels, CASNRPVS/91-1, ASTM Standards Technology Training Course Workbook for a One-Day Workshop Sponsored by ASTM Committee E-10 on Nuclear Technology and Applications, ASTM,1916 Race St., Philadelphia, PA,1st Edition, November 1991.
[Mc93] W. N. McElroy, E. P. Lippincott, A. L. Lowe, Jr., and P. D. Hedgeco.ck, Condition Assessment and Surveillance of Nuclear Reactor Pressure Ves'sel l Steels, CASNRPVS/91-1, ASTM Standards Technology Training Course Workbook for a One-Day Workshop Sponsored by ASTM Committee E-10 on Nuclear Technology and Applications, ASTM,1916 Race St, Philadelphia, PA 19103, 2nd Edition, August 1993.
[Mc94] W. N. McElroy, 8th ASTM-EURATOM International Svmoosium on Reactor Dosimetrv Kevnote Session & Consensus Standards: Trends in Enerov &
Nuclear Poliev " CTS-RP-94-1, Consultants & Technology Services,113 Thayer Drive, Richland, WA 99352, February 1994.
[Mc94a] W. N. McElroy, R. J. McElroy and R. Gold, "TechnicalIssues Relevant to ASTM LWR Surveillance Standards," Beactor Dosimetrv. ASTM STP 1228, Harry Farrar IV, E. Parvin Lippincott, John G. Williams and David W. Vehar Ec.a.,
ASTM, Philadelphia, p. 9,1994.
[Mc94b] W. N. McElroy, R. J. McElroy, R. Gold, E. P. Lippincott, and A. L. Lowe, Jr,
" ASTM Standards Associated with PWR & BWR Power Plant Licensing, Opera-tion & Surveillance," Reactor Dosimetrv. ASTM STP 1228, Harry Farrar IV, E.
Parvin Lippincott, John G. Williams and David W. Vehar Eds., ASTM, Philadel-phia, p.19,1994.
[McS5] W. N. McElroy, P. E. Fuller, R. Gold, J. L. Helm, A. S. Kumar, E. P. Lippincott, R. H. Meservey, P. S. Olson, and G. Subbaraman, " Standards Technology Development, Transfer, and Training Manual on Radiological Decontamination and Decommissioning," CTS-M-951, Consultants & Technology Services,113 Thayer Drive, Richland, WA, 2nd Edition, June 1995. !
72
I
[0d84] G. R. Odette et al., " Physically Based Regression Correlations of Embnttlement Data from RPV Surveillance Programs,* Electric Power Research institute, NP-3319, January 1984.
10s76) C. A. Oster, W'JN. McElroy, R. L. Simons, E. P. Lippincott and G. R. Odette, a Modified Monte Carlo Program for SAND-Il with Solution Weichtino and Error ;
Analvsis. HEDL-TME 76-60, Hanford Engineering Development Laboratory, Richland, WA, August 1976.
1
[0s961 B. Osmera et al., " Experimental and Theoretical Studies on VVER 1000 Reactor 1 Dosimetry" Proc. 9th ASTM-EURATOM Svmoosium on Reactor Dosimetrv, I Prague, Czech Republic, Sept. 2-6, 1996. )
i
[0sxx) 8. Osmera, J. Kynel et al., "The Benchmark for MCNP Gamma and Neutron Spectra Calculation in VVER Geometry," Undated.
[Pr97] Proceedinos of the 9th ASTM-EUROPEAN Svmoosium on Reactor Dosimetrv.
Prague, Czech Republic, Septemrir 2-6, 1996.
[Ra04] P. N. Randall, " Basis for Revision 2 of the US NRC Regulatory Guide 1.99,"
Proc.10th MPA Seminar, Stuttgart, FRG, October 10,1984.
[Ra861 P. N. Randall, " Basis for Revision 2 of the U.S. Nuclear Regulatory Commission's Regulatory Guide 1.99," Radiation Embrittlement of Nuclear I Reactor PV Steels: An International Review, (2nd Vol.), L. E. Steele, Ed., ASTM STP 909, ASTM, Philadelphia,1986.
[Re88) Regulatorv Guide 1.99. Revision 2: Radiation Embrittlement of Reactor Vessel Materials. U.S. Nucl. Reg. Com., Washington, DC, May 1988.
[Ru96] R. P. Rulko, l. Kodeli and E. Sartori, " Review of the Status of Reactor Vessel Embrittlement Prediction," Proc. 9th ASTM-EURATOM Svmoosium on Reactor !
Dosimetrv. Prague, Czech Republic, Sept. 2-6, 1996.
[Sc79] F. A. Schmittroth, FERRET Data Analvsis Code. HEDL-TME 79-40, Hanford Engineering Development Laboratory, Richland, WA, September 1979.
[Se87) C. Z. Serpan, Jr , "NRC Research in Aging and Extended Life for LWR Primary
. . Systems", Transactions of ANS. Vol. 55, p. 266, November 1987.
[Se891 C. Z. Serpan, Jr., " impact of USA-Euratom Cooperative Dosimetry Research on NRC Regulation of Light Water Reactors", Reactor Dosimetrv: Methods.
Aoolications. and Standardization. 6th ASTM EURATOM Svmoosium on Reactor l Dosimetrv, Jackson Hole, Wyoming, May 31 - June 6,1987, STP 1001, p. 7, j ASTM, May 1989.
I i
73
[Se941 C. Z. Serpan, Jr., " Impact of U.S. NRC Reactor Vessel Embrittlement Research on Regulation of Nuclear Power Plants: Summary of Keynote Session Address.
Reactor Dosimetrv. ASTM STP 1228, Harry Farrar IV. E. P. Lippincott, J. G.
Williams and D. W. Vehar Eds., ASTM, Phil., p. 3,1994
[Si821 R. L Simons, et al., "Re-Evaluation of the Dosimetry for RPV Surveillance j Capsules," NUREG/CP-0029, Vol. 2, Proceedinos of the 4th ASTM-EURATOM j Svmoosium on Reactor Dosimetrv." NBS, Gaithersburg Maryland, March 22-26,1982. l
[Si87] R. L. Simons, Sections 4.1.2, 4.1.3, 4.2 4.5; W. N. McElroy, Ed . ,
LWR-PV-SDlP: LWR Power Reactor Surveillance Physics-Dosimetrv Data Base Comnendium, NUREG/CR-3319, HEDL-TME 85-3, Hanford Engineering Development Laboratory, Richland, WA, March 1987 Update.
[Sm96] R. W. Smedley, " Docket 50-255-License DPR-20-Palisades Plant: Updated Reactor Vessel Fluence Values," Consumers Energy Letter to U.S. NRC, Document Control Desk, April 4,1996.
[St861 F. W. Stallmann, "LSL-M2: A Computer Program for Least-Square Logarithrhic Adjustment of Neutron Spectra," NUREG/CR-4349 (ORNL/TM-9933, (ORNL/TM-9933), ORNL, March 1986.
1
[St941 L. E. Steele, B. F. Beaudoin, E. C. Biemiller, R. A. Van Konynenburg, and W. L.
Server, " Reactor Vessel Dosimetry Assessment: Perspectives of Materials Engineer," Reactor Dosimetrv. ASTM STP 1228, p. 490, H. Farrar, E. P.
Lippincott, J. G. Williams and D. W. Vehar Eds., ASTM, Philadelphia,1994.
[Ta85) A. Taboada, P. N. Randall, and C. Z. Serpan, Jr., " Status of Regulatory Issues and Research in Light Water Reactor Surveillance Dosimetry in the U. S.",
Proc. of the 5th ASTM-EURATOM Svmoosium on Reactor Dosimettv. !
Geesthacht, Federal Republic of Germany, Sept 24-28,1984, EUR 9869, Vol.
1, p.185, Commission of the European Communities,1985.
[Ta891 A. Taboada, P. N. Randall, and C. Z. Serpan, Jr., " Overview of U.S. Research and Regulatory Activities on Neutron Radiation Embrittlement of Pressure Vessel Steel", Radiation Embrittlement and Surveillance of Nuclear Reactor Pressure Vessel Steels: An international Review. Third Volumn, ASTM STP 1011, L. E.
Steele, Ed., p. 27, February,1989.
[W 331 M. L. Williams et al., " Validation of Neutron Transport Calculations in Benchmark Facilities for Improved Damage Fluence Predictions," Proc. of the 11th WRSR information Meetina. Gaithersburg, MD, October 24-28, 1983, NUREG/CP-0048, Vols.1-6, NRC, Washington, DC,1983.
[Wo81] D. W. Wootan and F. Schmittroth, Comoarison of SAND ll and FERRET.
Hanford Engr. Devel. Lab. Report HETL-TC 1589, January 1981.
74
i l'No96al J. R. Worsham Ill, " Consistent '!essel Fluence and PTS Embrittlement Uncertainties," hoc. 9th ASTM-EURATOM Svmoosium on Reactor Dosimetrv, Prague, Czech Republic, September 2 6,1996.
('No96bl. J. R. Worsham'lil, Biased Fluences in the Charpy Embrittlement Database,"
Proc. 9th ASTM-EURATOM Svmoosium on Reactor Dosimetrv. Prague. Czech .
Republic, September 2 6,1996. '
[Woxx] R. de Wouters and B. Osmera, " Calculations of a VVER-1000 Mock-up with MCBEND, Undated.
l 1
l i
'l 75 .
r
i ENCLOSURE 5 CONSUMERS ENERGY COMPANY PALISADES PLANT DOCKET 50-255 Palisades Reactor Vessel Fluence Resume & Independent Review Completed by Eckhard Polke Siemens AG
'. 1 l
l i
l 1
Curiculum Vitae
- 1. Family name: Polke
- 2. First name: Eckhard
- 3. Date of birth: 26.01.1947
- 4. Nationality: Gerrnan
- 5. Civil status: married
- 6. Education: 10/1969 - 4/1972 University of Dusseldorf 5/1972 -1/1981 University of Heidelberg Academic degrees: Dipl. Phys. and Dr. rer, nat.
- 7. Key Qualifications: calculations of the reator core surrounding gamma- and neutron-radiation for PWR, BWR and WWER responsible for the fluence calculations for rpv
-surveillance programs
- 8. Professional Experience Record:
3/1981 -1998 SEIMENS AG (Power Generation Group)
Nuclear Power Generation Division, Erlangen Department: Radiation Protection Design Responsible for calculation of the reactor core surrounding gamma- and neutron-radiation
- 9. Publications: Comparison of Fast Fluences Determined via Fe , Nb , and ThO2-Detectors with Theoretical Values Proceedings of the 7* ASTM-EURATOM Symposium on Reactor Dosimetry 1992 Fluence Surveillance by Scraping Samples from the Inner Surface of the Thermal Shield in the Nuclear Power Plant E Obrigheim in Germany (KWO)
Reactor Dosimetry, ASTM STP 1228,1994 SEIMENS-KWU Experience in Evaluating Fluence ~ Detectors inside and Outside the RPV in German Light Water Reactor Plants, Ninth International Symposium on Reactor Dosimetry Prague 1996
. n uw2 SF.lMENS-KWU NDS3/Po
_-