NRC Inspector General Report on San Onofre failure and CFR50.59
Nuclear Regulatory Commission (2014-10-02) Office Inspector General
This Page:
https://copswiki.org/Common/M1510Media Link:
https://copswiki.org/w/pub/Common/MediaTopicHeader/NRC-InspectorGeneralReportOnSongs-ML14276A478.pdfMore Info:
Nuclear Energy,
Shut San Onofre
NRC-InspectorGeneralReportOnSongs-ML14276A478.pdf: NRC Inspector General Report on SONGS and 10CFR-50.59
Searchable Text
Page 1-----------------------
OFFICE OF THE INSPECTOR GENERAL
U.S. NUCLEAR REGULATORY COMMISSION
NRC Oversight of Licensee’s Use of 10 CFR 50.59 Process To Replace SONGS’ Steam Generators
Case No. 13-006
EVENT INQUIRY
Page 2-----------------------
UNITED STATES NUCLEAR REGULATORY COMMISSION
WASHINGTON, D.C. 20555-0001
OFFICE OF THE INSPECTOR GENERAL
October 2, 2014
MEMORANDUM TO: Chairman Macfarlane
FROM: Hubert T. Bell
Inspector General
SUBJECT: NRC OVERSIGHT OF LICENSEE’S USE OF 10 CFR
50.59 PROCESS TO REPLACE SONGS’ STEAM
GENERATORS (OIG CASE NO. 13-006)
This accompanies the results of an Office of the Inspector General (OIG), U.S. Nuclear
Regulatory Commission (NRC), event inquiry into concerns pertaining to NRC’s
oversight of Southern California Edison’s application of the 10 CFR 50.59 process for
the steam generator replacements in San Onofre Nuclear Generating Station (SONGS)
Units 2 and 3. In addition, public interest groups and Congress specifically questioned
SONGS’ use of the 10 CFR 50.59 rule to replace the steam generators without first
obtaining NRC prior approval through a license amendment. Therefore, OIG also
sought to ascertain from NRC officials whether SONGS required a license amendment
for the steam generator replacements and whether the problems at SONGS could have
been identified through NRC’s license amendment review process.
We have also provided this event inquiry report to the appropriate Majority and Ranking
Members of Congress with oversight responsibilities for the NRC.
If you have any questions, please contact me, at 301-415-5930, or Joseph A.
Mc Millan,
Assistant Inspector General for Investigations, at 301-415-5929.
Attachment: As stated
cc: Commissioner Svinicki
Commissioner Ostendorff
Page 3-----------------------
Office of the Inspector General
EVENT INQUIRY
NRC Oversight of Licensee’s Use
of 10 CFR 50.59 Process To Replace
SONGS’ Steam Generators
Case No. 13-006
Page 4-----------------------
TABLE OF CONTENTS
SUMMARY iii
I. BASIS AND SCOPE 1
II. BACKGROUND AND CHRONOLOGY. 2
III. DETAILS 11
ISSUES:
1. Missed Opportunities During NRC Region IV
2009 Inspection 11
Findings 29
2. AIT Review of SCE’s 10 CFR 50.59 Evaluation 30
Findings 46
3. NRC Oversight of SONGS UFSAR 47
Findings 54
i
Page 5-----------------------
SUMMARY
Basis and Scope
The Office of the Inspector General (OIG), U.S. Nuclear Regulatory Commission (NRC),
initiated this event inquiry in response to concerns pertaining to NRC’s oversight of
replacement steam generators installed at San Onofre Nuclear Generating Station
(SONGS) Units 2 and 3 in 2010 and 2011, respectively. Southern California Edison
(SCE), the license holder for SONGS, replaced the steam generators subsequent to its
application of the regulatory process described in 10 Code of Federal Regulations
(CFR) 50.59, “Changes, Tests and Experiments.” 10 CFR 50.59 establishes the
conditions under which licensees may make changes to their facility or procedures and
conduct tests or experiments without prior NRC approval (i.e., without an amendment to
their NRC license).
In January 2012, approximately 1 year after SONGS replaced its Unit 3 steam
generators, control room operators identified a leak in one of Unit 3’s two steam
generators, and the plant was shut down in accordance with plant procedures. Initial
inspection confirmed one small leak in one tube in one of the two steam generators.
Continuing inspections of all of the steam generator tubes in both Unit 3 steam
generators discovered unexpected wear, including tubes rubbing against each other as
well as against retainer bars. At the time the Unit 3 leak was identified, Unit 2 was shut
down for a routine refueling outage. Subsequent inspections of all Unit 2 steam
generator tubes also discovered unexpected wear.
Over the next approximate year and a half, SCE pursued evaluation of Unit 3 and
restart of Unit 2; however, on June 7, 2013, SCE announced its decision to permanently
cease operations of SONGS Units 2 and 3. SCE’s June 12, 2013, letter to NRC
conveying this decision did not provide the reason for the permanent shutdown.
OIG’s event inquiry examined NRC’s oversight of SCE’s application of the 10 CFR
50.59 process for the replacement steam generators in SONGS Units 2 and 3. OIG
also sought to ascertain from NRC officials whether SONGS required a license
amendment for the steam generator replacements and whether the problems at
SONGS could have been identified through NRC’s license amendment review process.
Background
Nuclear power reactors are licensed based on a given set of requirements, depending
primarily on the type of plant. This set of requirements is called the plant’s “licensing
basis.” A principal licensing basis document is the plant’s final safety analysis report
(FSAR). The FSAR and the plant’s NRC license and associated technical specifications
are the principal regulatory documents describing how the plant is designed,
constructed, and operated. The FSAR is also a key reference document used by NRC
inspectors during both plant construction and operation, and it must be sufficiently
iii
Page 6-----------------------
detailed to permit the staff to determine whether the plant can be built and operated
without undue risk to public health and safety.
Because a plant’s design and operation are not static, certain changes are necessary
over the course of a facility’s operating life. Reactor licensees must follow NRC
regulations to justify and implement changes in the design basis and licensing basis for
their facilities, and they are required to document such changes in the FSAR. 10 CFR
50.71(e) requires the FSAR to be periodically updated. The objectives of 10 CFR
50.71(e) are to ensure that licensees maintain the information in the updated FSAR
(UFSAR) to reflect the current status of the facility and address new issues as they arise
so that the UFSAR can be used as a reference document in safety analysis.
NRC has defined the changes that a licensee may make to a licensed facility without
prior NRC approval. Pursuant to 10 CFR 50.59 (c)(1), the holder of a license may,
without obtaining a license amendment, (1) make changes in the facility as described in
the FSAR (as updated), or (2) make changes in the procedures as described in the
FSAR (as updated), and conduct tests or experiments not described in the FSAR (as
updated) as long as a change to the technical specifications incorporated in the license
is not required, and the change, test, or experiment does not meet any of the eight
10 CFR 50.59 (c)(2) criteria. If any of the criteria in 10 CFR 50.59 are not met (i.e., the
change involves modification to the technical specifications or involves one of the eight
criteria), the license holder must apply to NRC for a license amendment and obtain
NRC’s approval before implementing the change. NRC staff document their safety
analysis of a license amendment request in a safety evaluation providing the technical,
safety, and legal basis for NRC’s disposition of the license amendment request.
Licensee Implementation of 10 CFR 50.59 Process
The Nuclear Energy Institute’s (NEI) November 2000 Guidelines for 10 CFR 50.59
Implementation (NEI 96-07)1 identifies the three following steps in the 10 CFR 50.59
process:
- Applicability and Screening. Determine if a 10 CFR 50.59 evaluation is required. First licensee determines if an evaluation is applicable to the proposed activity and, if so, performs screening to determine if the activity should be evaluated against the 10 CFR 50.59 evaluation criteria.
- Evaluation. If it is determined that a given activity requires a 10 CFR 50.59 evaluation, the licensee applies the eight 10 CFR 50.59 evaluation criteria (10 CFR 50.59(c)(2) (i-viii)) to determine if a license amendment must be obtained from NRC. This is a written evaluation.
1
In its November 2000 Regulatory Guide 1.187, Guidance for Implementation of 10 CFR 50.59, Changes, Tests, and
Experiments, NRC states that NEI 96-07 provides methods that are acceptable to the NRC staff for complying with
the provisions of 10 CFR 50.59.
iv
Page 7-----------------------
- Documentation and Reporting. Document and report to NRC the activities implemented under 10 CFR 50.59. Records maintained must include a written evaluation that provides the basis for the determination that the change, test, or experiment does not require a license amendment.
Frequency of Use
Nuclear reactor licensees have used the 10 CFR 50.59 process thousands of times to
make changes without NRC preapproval. Licensees conduct about 475 10 CFR 50.59
screenings per unit per year, and about five 10 CFR 50.59 evaluations per unit per year
for a nationwide total of about 49,000 screenings and evaluations per year.
Since 1989, 53 of the 65 plants that utilize steam generators have replaced their steam
generators under 10 CFR 50.59, while 6 replacements were made subsequent to a
license amendment.
NRC Oversight of Licensees and Their Application of the 10 CFR 50.59 Process
NRC inspects licensees’ application of the 10 CFR 50.59 process through an NRC
Reactor Oversight Process (ROP) baseline inspection procedure (IP), IP 71111.17,
“Evaluations of Changes, Tests, or Experiments and Permanent Plant Modifications.”
This triennial inspection is intended to provide assurance that required license
amendments have been obtained.
Findings
Issue 1. Missed Opportunities During NRC Region IV 2009 Inspection
OIG found that NRC missed an opportunity during a 2009 triennial baseline inspection
of SONGS’ implementation of the 10 CFR 50.59 process to identify weaknesses in the
SONGS steam generator 50.59 screening and evaluation package. While a Region IV
inspection team selected the SONGS Unit 2 steam generator 10 CFR 50.59 screening
and evaluation package as one of 35 items sampled during a 2009 triennial baseline
ROP inspection at SONGS, the inspection team did not identify various shortcomings
noted more recently by NRC subject matter experts who reviewed the steam generator
screening and evaluation package subsequent to SONGS’ shutdown due to problems
with steam generator design.
The 2009 inspection team concluded from its review of the 35 items sampled that
SONGS had correctly determined that the changes SONGS made could be made
without a license amendment. However, the NRC subject matter experts who reviewed
the Unit 2 steam generator screening and evaluation package following SONGS’
shutdown identified questions pertaining to the Unit 2 steam generator 10 CFR 50.59
screening and evaluation, some of which NRC says cannot now be answered based on
available information. The questions raised by the subject matter experts pertain to (1)
insufficient support for 10 CFR 50.59 evaluation conclusions that contributed to the
v
Page 8-----------------------
decision that a license amendment was not needed and (2) methodology changes that
should have been considered for screening but were not listed in the screening
documentation. OIG found that (1) without knowing whether everything that should
have been screened was screened, and the outcomes of these screenings, and (2)
without reviewing additional information concerning the evaluation conclusions, there is
no assurance that NRC reached the correct conclusion in its 2009 inspection that
SONGS did not need a license amendment for its steam generator replacement.
OIG found that the primary inspector who reviewed the SONGS Unit 2 steam generator
10 CFR 50.59 screening and evaluation package during the 2009 baseline inspection
(at approximately the same time installation of the Unit 2 steam generators
commenced) described conducting a review that aligned with inspection guidance, but
said that in hindsight, with the experience he now has, he might have probed further into
certain aspects of the screening and evaluation package. This inspector, and others
interviewed during the investigation, identified a need for improvement in training and
guidance to inspectors for the 50.59 inspection. Although several senior managers
acknowledged some of the shortcomings in the SONGS screening and evaluation
package, they supported NRC’s inspection approach, which relies on sampling and
judgments made by inspectors with different backgrounds and experience levels. One
senior manager expressed confidence in the 50.59 inspection process, and noted that
the purpose of NRC’s 50.59 inspection is not to identify design flaws, but rather to
determine whether licensees are correctly implementing the 50.59 rule and reaching the
correct conclusions as to the need for NRC preapproval. At the same time, senior
managers, subject matter experts, and inspectors expressed general agreement that
NRC needs to improve its 10 CFR 50.59 inspection training and guidance.
Issue 2. AIT Review of SCE’s 10 CFR 50.59 Evaluation
OIG found that although an NRC Region IV2 Augmented Inspection Team (AIT),
established to assess the circumstances surrounding the tube leak and unexpected
wear of tubes in the Unit 3 steam generators, included a review of the SONGS 50.59
steam generator package to determine whether SONGS needed a license amendment
prior to installing the new steam generators, the AIT did not document an answer to this
question. In its initial July 18, 2012, inspection report, the AIT communicated that the
Office of Nuclear Reactor Regulation (NRR) Project Manager assigned to perform the
review identified one unresolved item (URI number 10, “Change of methodologies
associated with 10 CFR 50.59 review”) for which additional information was needed to
determine if performance deficiencies exist or if the issues constituted violations of NRC
requirements. The URI described two instances that failed to adequately address
whether the change involved a departure of the method of evaluation described in the
UFSAR. Although NRC’s November 9, 2012, AIT followup report documented the
2 NRC’s Region IV regional office in Arlington, Texas, oversees NRC regulatory activities in the western and southern
mid-western United States.
vi
Page 9-----------------------
closure of this URI, and stated that neither change would have required a license
amendment, it did not answer the overall question of whether a license amendment was
required.
The AIT Team Leader and the current Region IV Deputy Regional Administrator told
OIG that based on what NRC reviewed during its inspections, the conclusion was that a
license amendment was not needed, although each allowed that the sampling approach
used to perform this assessment could have missed something. The Acting NRR
Director said he could not determine if an amendment was needed or not due to the
gaps that may exist regarding items that may require screening and/or evaluation. The
current Region IV Deputy Regional Administrator said additional inspection would be
required to answer whether a license amendment was required, and questioned
whether it would be a prudent use of resources to go back and accomplish that. The
former Region IV Deputy Regional Administrator said that in hindsight, he believes that
SONGS should have requested a license amendment from NRC prior to making the
change. He also believes the steam generator design was fundamentally flawed and
would not have been approved as designed. He said the AIT discussed a potential
50.59 criteria violation because of the design issues; however, the AIT ultimately
identified a design control violation.
OIG found that NRC’s justification for closing out URI number 10 does not align with
specific language in 10 CFR 50.59 concerning NRC approval for a change in
methodology, but was based instead on Region IV’s interpretation (in consultation with
NRR) of the rule. 10 CFR 50.59 (a)(2)(ii) reflects that changes from a method described
in the UFSAR to another method are permissible without NRC preapproval if that
method has already been approved by the NRC for the “intended” application. In
closing out the URI, however, the AIT followup report determined the change of
methods would not have required a license amendment based on NRC’s approval for
the use of the method at other nuclear power plants in “similar” applications. OIG notes
that while the AIT characterized the issue as a change in methodology, it justified
closing the matter based on approval for a “similar” application rather than the
“intended” application as stated by the rule.
OIG also notes that while the AIT inspection report identified an unresolved issue
pertaining to the SONGS 10 CFR 50.59 screen and evaluation package, the NRR
technical specialist who reviewed the package used a sampling approach and did not
identify many of the shortcomings described under issue 1 of this report.
Issue 3. NRC Oversight of SONGS UFSAR
OIG found that NRC does not consistently use one of its primary oversight methods to
assess whether licensees are keeping their power plant licensing basis documentation
up to date as required by 10 CFR 50.71(e). Although licensees are required, per 10
CFR 50.71(e), to biannually submit UFSAR updates reflecting the current status of the
facility so that the document can be used as a reference document in safety analysis,
the NRR project managers tasked to review these submittals do not always conduct the
vii
Page 10-----------------------
reviews within the required 90-day timeframe. Moreover, although licensees also must
biannually submit, per 10 CFR 50.59(d)(2), information concerning changes made
under 10 CFR 50.59 without NRC prior approval, NRR project managers -who are
instructed to consider this information during their review of 10 CFR 50.71(e) submittals
-do not always take the 10 CFR 50.59(d)(2) information into consideration during their
reviews. OIG found that while NRC expects a plant’s UFSAR to accurately reflect a
plant’s licensing basis, the former Region IV Deputy Regional Administrator said that
during the SONGS AIT, Region IV staff noted the licensee had made many changes to
the steam generators over a 25-year period that were not reflected in the UFSAR or
consistent with the original Safety Analysis Report (SAR.)
OIG reviewed documentation of project manager reviews in two NRR branches and
found project managers reviewed only 5 of the 21 most recently received licensee
UFSAR submittals within the 90-day timeframe, while 7 were reviewed between 90 days
and a year after receipt, and 9 reports more than a year after receipt. Moreover, only
two of the project manager reviews contained a reference to review of 10 CFR 50.59
documentation submitted by licensees even though project manager guidance directs
that this occurs. OIG also found that over a 10-year period, NRC staff documented two
reviews of changes to SONGS’ UFSAR, although the licensee submitted six UFSAR
updates during this period as required, and neither NRC review mentioned
consideration of 10 CFR 50.59 changes.
Although senior NRC managers expect the project managers to conduct the reviews
within the required timeframe, and to consider changes made under 10 CFR 50.59 as
part of that review, two NRR project managers interviewed said the reviews are
considered a low priority. Neither of the project managers included the 10 CFR 50.59
information in their reviews of 50.71(e) submittals; one thought this review was
conducted by a different NRR group and the other thought the 10 CFR 50.59
information was used by regional inspectors for a different purpose.
In contrast, the Deputy Executive Director for Reactor Preparedness Programs
considers NRC’s oversight of 10 CFR 50.71(e) to be critical for enabling NRC to know
whether a plant is in compliance with its licensing basis, and considers the project
manager review of 50.71(e) submittals to be a priority. While the former NRR Director
also expected project managers to conduct the required reviews to assess whether
changes made by the licensees have generally been updated into the FSAR, he viewed
the project manager’s review as a bookkeeping exercise that is based on the
experience of the project manager. He noted that the FSAR review is a self-imposed
requirement and if NRC is not meeting its own internal guidance, then it should either
meet the requirement or change the guidance based on safety significance.
viii
Page 11-----------------------
I. BASIS AND SCOPE
The Office of the Inspector General (OIG), U.S. Nuclear Regulatory Commission (NRC),
initiated this event inquiry in response to concerns pertaining to NRC’s oversight of
replacement steam generators installed at San Onofre Nuclear Generating Station
(SONGS) Units 2 and 3 in 2010 and 2011, respectively. Southern California Edison
(SCE), the license holder for SONGS, replaced the steam generators subsequent to its
application of the regulatory process described in 10 Code of Federal Regulations
(CFR) 50.59, “Changes, Tests and Experiments.” 10 CFR 50.59 establishes the
conditions under which licensees may make changes to their facility or procedures and
conduct tests or experiments without prior NRC approval (i.e., without an amendment to
their NRC license).
In January 2012, approximately 1 year after SONGS replaced its Unit 3 steam
generators, control room operators identified a leak in one of Unit 3’s two steam
generators, and the plant was shut down in accordance with plant procedures. Initial
inspection confirmed one small leak in one tube in one of the two steam generators.
Continuing inspections of all of the steam generator tubes in both Unit 3 steam
generators discovered unexpected wear, including tubes rubbing against each other as
well as against retainer bars. At the time the Unit 3 leak was identified, Unit 2 was shut
down for a routine refueling outage. Subsequent inspections of all Unit 2 steam
generator tubes also discovered unexpected wear.
Over the next approximate year and a half, SCE pursued evaluation of Unit 3 and
restart of Unit 2; however, on June 7, 2013, SCE announced its decision to permanently
cease operations of SONGS Units 2 and 3. SCE’s June 12, 2013, letter to NRC
conveying this decision did not provide the reason for the permanent shutdown.
OIG’s event inquiry examined NRC’s oversight of SCE’s application of the 10 CFR
50.59 process for the replacement steam generators in SONGS Units 2 and 3. In
addition, public interest groups and Congress specifically questioned SONGS’ use of
the 10 CFR 50.59 rule to replace the steam generators without first obtaining NRC prior
approval through a license amendment. Therefore, OIG also sought to ascertain from
NRC officials whether SONGS required a license amendment for the steam generator
replacements and whether the problems at SONGS could have been identified through
NRC’s license amendment review process.
A total of 30 NRC employees and 4 SCE employees (including one former employee)
were interviewed for this event inquiry.
Page 1
Page 12-----------------------
II. BACKGROUND AND CHRONOLOGY
Plant Licensing Basis and 10 CFR 50.59 Change Process
Nuclear power reactors are licensed based on a given set of requirements, depending
primarily on the type of plant. This set of requirements is called the plant’s “licensing
basis," defined in 10 CFR 54.3 as “the set of NRC requirements applicable to a specific
plant and a licensee’s written commitments for ensuring compliance with and operation
within applicable NRC requirements and the plant-specific design basis (including all
modifications and additions to such commitments over the life of the license) that are
docketed and in effect.”
A principal licensing basis document is the plant’s final safety analysis report (FSAR).3
The FSAR and the plant’s NRC license and associated technical specifications are the
principal regulatory documents describing how the plant is designed, constructed, and
operated. The FSAR is also a key reference document used by NRC inspectors during
both plant construction and operation, and it must be sufficiently detailed to permit the
staff to determine whether the plant can be built and operated without undue risk to
public health and safety.
Because a plant’s design and operation are not static, certain changes are necessary
over the course of a facility’s operating life. Reactor licensees must follow NRC
regulations to justify and implement changes in the design basis4 and licensing basis for
their facilities, and they are required to document such changes in the FSAR. 10 CFR
50.71(e) requires the FSAR to be periodically updated. The objectives of 10 CFR
50.71(e) are to ensure that licensees maintain the information in the updated FSAR
(UFSAR) to reflect the current status of the facility and address new issues as they arise
so that the UFSAR can be used as a reference document in safety analysis.
NRC has defined the changes that a licensee may make to a licensed facility without
prior NRC approval. Pursuant to 10 CFR 50.59s(c)(1), the holder of a license may,
without obtaining a license amendment, (1) make changes in the facility as described in
the FSAR (as updated), or (2) make changes in the procedures as described in the
FSAR (as updated), and conduct tests or experiments not described in the FSAR (as
updated) as long as:
3 The principal application document for a reactor construction permit is a preliminary safety analysis report, which is
submitted at the time an operating license is sought, and is subsequently updated to become the FSAR for the
facility. 10 CFR Part 50 defines, in general terms, the information that must be supplied in a safety analysis report for
a nuclear power plant.
4 Per 10 CFR 50.2, design bases means that information which identifies the specific functions to be performed by a
structure, system, or component of a facility, and the specific values or ranges of values chosen for controlling
parameters as reference bounds for design.
6 10 CFR 50.59 was promulgated in 1962, revised in 1968, and revised again in 1999. According to the October 4,
1999, Federal Register Notice announcing the final 1999 rule, the 1999 revision clarified the specific types of
changes, tests, and experiments conducted at a licensed facility or a certificate holder that require evaluation, and
revised the criteria that licensees and certificate holders must use to determine when NRC approval is needed before
such changes, tests, or experiments can be implemented. The final rule also added definitions for terms that had
been subject to differing interpretations and reorganized the rule language for clarity.
Page 2
Page 13-----------------------
- A change to the technical specifications incorporated in the license is not required, and
- The change, test, or experiment does not meet any of the following 10 CFR 50.59 (c)(2) criteria:
i. Result in more than a minimal increase in the frequency of occurrence of
an accident previously evaluated in the FSAR (as updated).
ii. Result in more than a minimal increase in the likelihood of the occurrence
of a malfunction of a structure, system, or component (SSC) important to
safety previously evaluated in the FSAR (as updated).
iii. Result in more than a minimal increase in the consequences of an
accident previously evaluated in the FSAR (as updated).
iv. Result in more than a minimal increase in the consequences of a
malfunction of an SSC important to safety previously evaluated in the
FSAR (as updated).
v. Create a possibility for an accident of a different type than any previously
evaluated in the FSAR (as updated).
vi. Create the possibility for a malfunction of an SSC important to safety with
a different result than any previously evaluated in the FSAR (as updated).
vii. Result in a design basis limit for a fission product barrier as described in
the FSAR (as updated) being exceeded or altered.
viii. Result in a departure from a method of evaluation described in the FSAR
(as updated) used in establishing the design bases or in the safety
analyses.
10 CFR 50.59 also stipulates that the FSAR (as updated) is expected to include FSAR
changes resulting from evaluations performed pursuant to the regulation (10 CFR 50.59
(c)(3)) and directs licensees to maintain records of changes in the facility (10 CFR 50.59
(d)(1)) and to submit a report containing a brief description of any changes, tests, and
experiments made under this regulation, including a summary of the evaluation of each
(10 CFR 50.59 (d)(2)). This report must be submitted to NRC at intervals not to exceed
24 months.
If any of the criteria in 10 CFR 50.59 are not met (i.e., the change involves modification
to the technical specifications or involves one of the eight criteria listed above), the
license holder must apply to NRC for a license amendment and obtain NRC’s approval
before implementing the change. NRC staff document their safety analysis of a license
amendment request in a safety evaluation providing the technical, safety, and legal
basis for NRC’s disposition of the license amendment request.
Page 3
Page 14-----------------------
Implementation of 10 CFR 50.59 Process
The Nuclear Energy Institute’s (NEI)6 November 2000 Guidelines for 10 CFR 50.59
Implementation (NEI 96-07)7 explains that 10 CFR 50.59 is the process that identifies
when a license amendment is required prior to implementing changes to the facility or
procedures described in the updated FSAR or tests and experiments not described in
the updated FSAR, and it notes the importance of maintaining and updating the FSAR
in accordance with 10 CFR 50.71(e). It states that 10 CFR 50.59 provides a threshold
for regulatory review -not the final determination of safety -for proposed changes,
tests, and experiments (referred to in the document as “activities”).
NEI 96-07 also provides a short background on NRC’s 1999 revision to 10 CFR 50.59
(“the first changes to the regulation in more than 30 years”), stating that the changes
were prompted by the need to resolve differences in interpretation of the rule’s
requirements by the industry and the NRC that came into clear focus in 1996. The NEI
document states that the 1999 changes made 10 CFR 50.59 more focused and efficient
by:
- Providing greater flexibility to licensees, primarily by allowing changes that have minimal safety impact to be made without prior NRC approval.
- Clarifying the threshold for “screening out” changes that do not require full evaluation under 10 CFR 50.59 by adoption of key definitions.
NEI 96-07 identifies the three following steps in the 10 CFR 50.59 process:
- Applicability and Screening. Determine if a 10 CFR 50.59 evaluation is required. First licensee determines if an evaluation is applicable to the proposed activity and, if so, performs screening to determine if the activity should be evaluated against the 10 CFR 50.59 evaluation criteria.
- Evaluation. If it is determined that a given activity requires a 10 CFR 50.59 evaluation, the licensee applies the eight 10 CFR 50.59 evaluation criteria (10 CFR 50.59 (c)(2)(i-viii)) to determine if a license amendment must be obtained from NRC. This is a written evaluation.
- Documentation and Reporting. Document and report to NRC the activities implemented under 10 CFR 50.59. Records maintained must include a written evaluation that provides the basis for the determination that the change, test, or experiment does not require a license amendment. The licensee must submit, at least every 24 months, a report with a brief description of any changes, tests,
6 NEI, with member participation, develops policy on key legislative and regulatory issues affecting the nuclear
industry and services as a unified industry voice before Congress, Executive Branch agencies, and Federal
regulators, as well as international organizations and venues.
7 In its November 2000 Regulatory Guide 1.187, Guidance for Implementation of 10 CFR 50.59, Changes, Tests, and
Experiments, NRC states that NEI 96-07 provides methods that are acceptable to the NRC staff for complying with
the provisions of 10 CFR 50.59.
Page 4
Page 15-----------------------
and experiments, including a summary of the evaluation of each. NEI 96-07
notes that this reporting requirement is identical to that for UFSAR updates
such that licensees may provide these reports to NRC on the same schedule.
NRC reactor licensees typically develop internal procedures to apply 10 CFR 50.59,
based on guidance from NEI such as NEI 96-07, and the Utilities Service Alliance’s8
USA 50.59 Resource Manual. The USA 50.59 Resource Manual’s stated purpose is to
provide guidance for implementing the 10 CFR 50.59 process and the document is
based on, and incorporates the implementation guidance provided in NEI 96-07 (which,
as noted in footnote 1, is endorsed by NRC in Regulatory Guide 1.187).
OIG learned that SONGS’ internal procedure incorporates the implementation guidance
provided in NEI 96-07, and references the USA 50.59 Resource Manual to implement
the 10 CFR 50.59 process. As required by the site Quality Assurance program, the
SONGS 50.59 procedure describes a plant Nuclear Oversight Board, which reports to
plant management and is responsible for reviewing the evaluations for adequacy.
Frequency of Use
OIG learned that nuclear reactor licensees have used the 10 CFR 50.59 process
thousands of times to make changes without NRC preapproval. Licensees conduct
about 475 10 CFR 50.59 screenings per unit per year, and about five 10 CFR 50.59
evaluations per unit per year for a nationwide total of about 49,000 screenings and
evaluations per year.
Since 1989, 53 of the 65 plants that utilize steam generators have replaced their steam
generators under 10 CFR 50.59, while 6 replacements were made subsequent to a
license amendment.
NRC Oversight of Licensees and Their Application of the 10 CFR 50.59 Process
Reactor Oversight Process
NRC’s mission is to license and regulate the Nation’s civilian use of radioactive
materials to protect public health and safety, promote the common defense and
security, and protect the environment. The agency does not operate the plants, but
establishes requirements for the design, construction, operation, and security of
commercial nuclear power plants in the United States. Since 2000, NRC has used the
Reactor Oversight Process (ROP) to verify that U.S. reactors are operating in
accordance with NRC rules, regulations, and license requirements. The ROP is NRC’s
program to inspect, measure, and assess the safety and security performance of
operating commercial nuclear power plants, and to respond to any decline in their
performance. The ROP’s inspection component includes three major elements:
8 Utilities Service Alliance (USA) is a not-for-profit cooperative designed to facilitate collaboration among its member
utilities.
Page 5
Page 16-----------------------
- Baseline inspections -the minimum required at all plants.
- Plant-specific supplemental inspections - performed at those plants with performance below established thresholds.
- Generic safety issue, special, and infrequent inspections - performed to address specific safety significant issues.
As noted in NRC Inspection Manual Chapter 2515, “Light-Water Reactor9,10 Inspection
Program -Operations Phase," the NRC inspection program covers only small samples
of licensee activities in any particular area. The sample sizes specified in the inspection
procedures are based on the relative importance of the area covered by the procedures
to the other areas inspected by the program. They are also based on the inspectors
choosing a “smart” sample instead of a statistically based random sample because the
risk-informed nature of the inspection program requires the inspections to be focused on
those aspects of plant operations and licensee activities that could pose the greatest
risk to public health and safety. Per the Inspection Manual, because the NRC does not
have objective criteria for evaluating positive findings (i.e., no findings), they are not
documented in baseline inspection reports.
Inspection Procedure (IP) 71111.17
NRC inspects licensees’ application of the 10 CFR 50.59 process through an ROP
baseline inspection, IP 71111.17, “Evaluations of Changes, Tests, or Experiments and
Permanent Plant Modifications.”11 This triennial inspection monitors the effectiveness of
the licensee’s implementation of changes to facility SSCs, risk significant normal and
emergency operating procedures, test programs, and the UFSAR in accordance with
the requirements of 10 CFR 50.59. The inspection is intended to provide assurance
that required license amendments have been obtained.
Chronology of SONGS Steam Generator Replacement and Failure
Located in northwest San Diego County, near San Clemente, California, and licensed to
SCE, SONGS Unit 2 began commercial operation in 1983 and Unit 3 in 1984.12 Units 2
and 3 used two steam generators per unit; the steam generators are large heat
exchangers that convert heat from the reactor into steam to drive the turbine generators
and produce 1,150 megawatts of electric power per unit.
9 The term light water reactor is used to describe reactors using ordinary water as coolant, including boiling water
reactors (BWRs) and pressurized water reactors (PWRs), the most common types used in the United States.
10 Units 2 and 3 at SONGS are light water reactors (PWRs) that use ordinary water as the coolant.
11 IP 71111.17 (issued January 31, 2008) is the reference number for this baseline inspection procedure during the
timeframe under examination by this event inquiry. In March 2013, the procedure was revised to add clarity to
terminology, enhance sample selection, and guidance), renumbered, and renamed IP 71111.17T, “Evaluations of
Changes, Tests, and Experiments and Permanent Plant Modifications." Unless noted otherwise, this report
references language in the 2008 procedure because it was in effect at the time of the events, addressed in this
report.
12 Unit 1 went into service January 1, 1968, and was retired in 1992.
Page 6
Page 17-----------------------
In October 2001, SONGS created a team to explore the purchase of four replacement
steam generators and in February 2004, SCE filed an application with the California
Public Utilities Commission for the SONGS steam generator replacement project, which
was expected to cost about $680 million. While SONGS Units 2 and 3 were licensed to
operate until 2022, the existing steam generators had exhibited degradation and were
predicted to reach the end of their operating life within the next several years.
In September 2004, Mitsubishi Heavy Industries in Kobe, Japan, was awarded the
contract to fabricate the new steam generators.
In December 2005, Bechtel was awarded the installation contract, and the California Public Utilities Commission
approved the steam generator replacement project.
On June 7, 2006, SCE notified NRC of its intent and timeline to replace Units 2 and 3
steam generators under 10 CFR 50.59. The SCE briefing document indicated there
would be no associated power uprate and that associated technical specification13
changes were scheduled to be identified in 2007. The briefing document identified the
following key design improvements: larger surface area, alloy 690 thermally treated
tubing, improved anti-vibration bar design, integral steam nozzle, improved material for
tube supports, and a forged shell. The briefing document also identified that while both
the original and replacement steam generators were identical in height and upper
diameter (65’6” and 22’, respectively), the replacement steam generator would weigh
643.6 tons, which was 23.6 tons heavier than the original and the replacement would
have more tubes than the original (9,727 versus 9,350).
SCE documented its 10 CFR 50.59 screen and evaluation for the Unit 2 and 3 steam
generator replacement in engineering change packages NECP 800071702 and NECP
800071703, respectively. SCE concluded that the steam generator activities could be
implemented per plant procedures without obtaining a license amendment.
Installation of Unit 2 replacement steam generators began in September 2009 and was
completed in April 2010.
NRC Region IV inspection activity (NRC Region IV provides
oversight of SONGS) included a review of selected portions of modifications for the
replacement steam generators to determine if changes were in accordance with 10 CFR
50.59; no issues were identified.
Installation of Unit 3 replacement steam generators began in October 2010 and was
completed in February 2011. NRC Region IV inspection activity included a review of
key design aspects and modifications associated with the steam generator
replacements; no issues were identified.
On January 31, 2012, SONGS Unit 3 shut down due to indications of a steam generator
tube leak. Steam generator tube inspections confirmed one small leak on one tube in
one of the two steam generators. Continuing inspections of all of the steam generator
tubes in both Unit 3 steam generators discovered unexpected wear, including tube to
13 Technical specifications set forth the limits, operating conditions, and other requirements imposed upon facility
operation for the protection of public health and safety.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 18-----------------------
tube as well as tube to tube support structural wear. In addition, pre-planned testing of
the SONGS Unit 2 steam generator tubes was in progress as part of a regularly
scheduled refueling outage when the event occurred in Unit 3. Testing results from Unit
2 also revealed unexpected tube wear at the retainer bars. The integrity of steam
generator tubes is important because the tubes provide an additional barrier inside the
containment building to prevent release of radioactive steam.
On March 14, 2012, three tubes in Unit 3 failed a pressure test indicating they would be
more likely to rupture during certain plant events that affect the pressure inside the
steam generator.
On March 15, 2012, the NRC commenced an onsite Augmented Inspection Team (AIT)
assessment of the SONGS Unit 3 steam generator tube degradation. AITs are used by
NRC to review more significant events or issues at NRC-licensed facilities. The
inspection team included inspectors from the NRC’s Region IV office, Region II office,
NRC headquarters in Rockville, MD, and the SONGS Resident Inspector.
On March 27, 2012, NRC Region IV issued a Confirmatory Action Letter (CAL) to
SONGS, Units 2 and 3, for commitments to address steam generator tube degradation.
This CAL was to remain in effect until the NRC had (1) reviewed SCE’s response to the
actions listed in the letter, including responses to staff’s questions and the results of
evaluations, and (2) the staff communicated to SCE in written correspondence that it
had concluded that SONGS Units 2 and 3 could be operated without undue risk to
public health and safety, and the environment. The CAL noted that during evaluations
and steam generator pressure testing for Unit 3, eight tubes failed pressure testing
indicating that these tubes could have failed under some accident conditions. For Unit
2, six tubes required plugging, and 186 additional tubes were plugged as a
precautionary measure.
On July 18, 2012, NRC Region IV issued an NRC AIT Report 05000361/2012007 and
05000362/2012007. The team concluded that plant operators responded to the
January 31, 2012, steam generator tube leak in accordance with procedures and in a
manner that protected public health and safety. Plant safety systems worked as
expected during the event. The NRC team identified 10 unresolved items requiring
additional review for regulatory action. One of the items, “Change of methodologies
associated with 10 CFR 50.59 review,” pertained to the SONGS 50.59 replacement
steam generator evaluation.
On October 3, 2012, SCE submitted a response to the March 27, 2012, CAL. The
response included detailed information demonstrating completion of Unit 2 CAL actions.
The response also included SCE’s Unit 2, Return to Service Report and a list of new
commitments identified.
On November 8, 2012, the Commission issued an Order to the Atomic Safety and
Licensing Board Panel and the NRC staff to address portions of a June 18, 2012,
petition submitted by Friends of the Earth (an environmental advocacy group). The
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 19-----------------------
petition requested that NRC order SCE to submit a license amendment application for
the design and installation of the SONGS Units 2 and 3 replacement steam generators
and to suspend SCE’s licenses until they were amended. The petition stated the
licensee was required to obtain a license amendment when it replaced the original
steam generators at SONGS. The Commission Order (CLI-12-20) directed the staff to
examine this portion under the 10 CFR 2.206 process. 10 CFR 2.206 allows any
member of the public to raise potential health and safety issues in a petition to the NRC.
On November 9, 2012, NRC Region IV issued NRC AIT Followup Report
05000361/2012010 and 05000362/2012010. The team closed 8 of the 10 unresolved
items; this included the item pertaining to the 10 CFR 50.59 evaluation. Although
inspectors determined that the change in method of evaluation did not require a license
amendment prior to implementing the change, a minor violation14 of 10 CFR 50.59(d)(1)
was identified because the evaluation did not provide a correct basis for the licensee’s
determination that the change did not require a license amendment prior to
implementing the change.
On June 7, 2013, SCE announced its decision to permanently shut down SONGS Units
2 and 3. Upon learning this, the NRC terminated its review of SCE’s CAL Response for
Unit 2, dated October 3, 2012.
On June 12, 2013, SCE submitted a Certification of Permanent Cessation of Power
Operations letter to the NRC, certifying that Units 2 and 3 had permanently ceased
power operations.
On June 28 and July 22, 2013, SCE certified all fuel had been permanently removed from Units 3 and 2 reactors, respectively.
On September 20, 2013, NRC Region IV issued a NRC CAL Response Inspection
Report 05000361/2012009 and 05000362/2012009 that closed out the remaining two
unresolved issues (URI) from the AIT inspection. The report documented that one
unresolved item was of very low safety significance (Green) for Unit 2, and one finding
that was preliminarily determined to have low to moderate safety significance (White) for
Unit 3.
On December 23, 2013, the NRC issued a Final Significance Determination of White
Finding and Notice of Violation regarding NRC Inspection Reports 05000361/2012009
and 05000362/2012009. This design control finding involved the failure to verify the
adequacy of the thermal-hydraulic and flow-induced vibration design of the Unit 3
replacement steam generators, which resulted in significant and unexpected steam
generator tube wear and loss of tube integrity on Unit 3 Steam Generator 3EO-88 after
11 months of operation.
14 Violations of minor safety concern are those below Severity Level IV and Green ROP’s findings.
9
Page 20-----------------------
On March 14, 2014, the NRC Office of the Executive Director for Operations (EDO)
initially approved an extension to August 29, 2014, to issue a proposed decision on the
Friends of the Earth, 10 CFR 2.206 petition. The extension was approved “owing to the
complexity of issues raised in the petition.” An additional extension was requested and
approved for Feburary 27, 2015.
In a March 20, 2014, memorandum, NRC’s EDO provided direction to the Region IV
Regional Administrator, the Director of the Office of Nuclear Reactor Regulation (NRR),
and the Director of the Office of New Reactors for staff to evaluate lessons from the
SONGS steam generator tube degradation event and identify and implement
appropriate actions. A charter for the lessons learned review assigned the NRR
Director to oversee the review process, and to present a lessons learned report to the
EDO by December 22, 2014. The charter included the 10 CFR 50.59 process among
several topics to be addressed by the lessons learned review. With regard to 10 CFR
50.59, the charter included the following items to consider:
- Does the 10 CFR 50.59 rule continue to be adequate for major or complex component replacements?
- Does the agency need to provide additional 10 CFR 50.59 guidance and information to (a) licensees for large or complex component replacements, (b) inspectors for their review of 10 CFR 50.59 evaluations of large or complex component replacements, and (c) project managers for their review of 10 CFR 50.59 evaluations?
10
Page 21-----------------------
III. DETAILS
ISSUE 1: Missed Opportunities During NRC Region IV 2009 Inspection
Background
As noted in the background section of this report, NRC inspects licensees’ application of
the 10 CFR 50.59 process through a triennial (every 3 years) baseline inspection IP
71111.17, “Evaluations of Changes, Tests, or Experiments and Permanent Plant
Modifications.” Although the procedure was most recently updated in March 2013, a
January 31, 2008, version of the procedure was in effect in September 2009, when a
Region IV inspection team selected the SONGS unit 2 steam generator replacement
50.59 evaluation as part of a sample to assess for its triennial conduct of IP 71111.17 at
SONGS.
The four-page 2008 inspection procedure directed inspectors to (a) triennially review 6
to 12 licensee evaluations required by 10 CFR 50.59 and 12 to 25 changes, tests, or
experiments that were screened out by the licensee and (b) triennially review 5 to 15
permanent plant modifications. The overall resource estimate was 172 to 212 hours for
the entire inspection, which “should be performed by engineering specialists
knowledgeable in the affected subject areas.” (NRC Inspection Manual, Chapter 2515,
states that IP 71111.17 will normally be performed by regional specialists who have
achieved at least basic certification in accordance with IMC 1245, “Qualification
Program for the Office of Nuclear Reactor Regulation Programs.” NRC’s inspector
qualification process does not include a formal training course in 10 CFR 50.59
inspections (i.e. IP71111.17); however, there is an individual study activity and an
on-the-job activity for reactor engineer inspectors.)
OIG learned that the SONGS 50.59 Unit 2 steam generator screening and evaluation
was selected for review during 2009 relative to requirement (a) above (i.e., triennially
review 6 to 12 licensee evaluations required by 10 CFR 50.59 and 12 to 25 changes,
tests, or experiments15 that were screened out by the licensee). IP 71111.17 contained
the following four specific inspection steps relative to this effort:
1. Verify that when changes, tests, or experiments were made, evaluations were
performed in accordance with 10 CFR 50.59. Verify that the licensee has
appropriately concluded that the change, test or experiment can be
accomplished without obtaining a license amendment.
1510 CFR 50.59 defines change as a modification or addition to, or removal from, the facility or procedures that
affects a design function, method of performing or controlling the function, or an evaluation that demonstrates that
intended functions will be accomplished.
11
Page 22-----------------------
2. Verify that safety issues related to the changes, tests, or experiments have been
resolved.
3. For the changes, tests, or experiments that the licensee determined that
evaluations were not required, verify that the licensee’s conclusions were correct
and consistent with 10 CFR 50.59.
4. Verify, as appropriate, that design and license basis documentation used to
support the changes, and procedures and design and license basis
documentation affected by the changes, reflect the design and license basis of
the facility after the change has been made.
OIG notes that with regard to “evaluations,” IP 71111.17 directs inspectors to verify that
“when changes, tests, or experiments were made [emphasis added], evaluations were
performed in accordance with 10 CFR 50.59.” This language suggests that inspectors
will be assessing modifications that have already been implemented.
NRC’s 2009 IP 71111.17 Inspection Includes Review of SONGS 50.59 Unit 2 Steam Generator Replacement Evaluation
OIG learned that from August 24, 2009, through September 4, 2009, a three-member
inspection team from Region IV conducted a triennial IP 71111.17 inspection at SONGS
that included the SONGS Unit 2 replacement steam generator within the inspection
sample. The Region IV inspection team was composed of a team leader and two team
members; the team leader and one team member had achieved full-qualification as an
NRC inspector approximately 14 months prior to the SONGS inspection and the other
team member was undergoing the qualification process at the time of the inspection.
(Although Region IV conducted a Plant Modification Inspection IP 71111.18 of SONGS
Unit 3 on December 31, 2010, and a steam generator installation inspection on May 10,
2011, neither inspection team selected the Unit 3 steam generator as part of the
inspection sample.)
The Region IV team conducted the IP 71111.17 inspection at SONGS and reported its
inspection results within the body of an NRC integrated inspection report dated
(16) November 5, 2009. The integrated inspection report described the results of a
3-month period of inspection by resident inspectors and region-based inspectors.
Section 1R17, “Evaluations of Changes, Tests, or Experiments and Permanent Plant
Modifications (71111.17),” of the integrated inspection report described the three-
member team’s inspection scope and findings. Section 1R17 stated that the inspectors
reviewed 8 evaluations required by 10 CFR 50.59; 15 changes, tests, or experiments
16 “San Onofre Nuclear Generating Station - NRC Integrated Inspection Report 0500361/2009004 and
05000362/2009004, and Notice of Violation,” describes the results of the 3-month period (June 24 through
September 23, 2009.) The report identified a cited violation of “very low safety significance” and four other findings of
very low safety significance, which NRC opted to treat as noncited violations because of their low safety significance
and because they were entered in the licensee’s corrective action program. OIG noted that none of the violations
were outcomes of the IP 71111.17 inspection.
12
Page 23-----------------------
that were screened out by the licensee; and 12 permanent plant modifications. An
attachment to the inspection report listed, by number, the 15 screens, 8 evaluations,
and 12 plant modifications the inspectors reviewed. Included within the list of eight
evaluations reviewed was number 800071702, which OIG learned was the number
SONGS assigned to its 10 CFR 50.59 screening and evaluation pertaining to its Unit 2
replacement steam generators.17
Section 1R17 stated:
The inspectors verified that when changes, tests, or experiments were
made, that evaluations were performed in accordance with 10 CFR
50.59 and that licensee personnel had appropriately concluded that the
change, test, or experiment can be accomplished without obtaining a
license amendment. The inspectors also verified that safety issues
related to the changes, test, or experiments were resolved. The
inspectors reviewed changes, tests, and experiments that licensee
personnel determined did not require evaluations and verified that the
licensee personnel’s conclusions were correct and consistent with 10
CFR 50.59. The inspectors also verified that procedures, design, and
licensing basis documentation used to support the changes were
accurate after the changes had been made.
In addition, Section 1R17 described the work performed to inspect plant modifications
and stated inspectors verified that supporting design and license basis documentation
had been updated accordingly and was still consistent with the new design, and that
procedures, training plans, and other design basis features had been adequately
accounted for and updated. In summary, Section 1R17 stated that the activities
performed “constitute completion of one sample” and that no findings of significance
were identified.18
OIG also learned that a March 4, 2010, inspection report (San Onofre Nuclear
Generating Station -Unit 2 Steam Generator Replacement Project Inspection Report
05000361/2009007) relied, in part, on section 1R17 from the November 2009 integrated
inspection report, to conclude that no findings of significance were identified with regard
to a portion of the inspection addressing “Design and Planning Inspections.” The March
4, 2010, report documented the results of NRC’s conduct of IP 50001 (Steam Generator
Replacement Inspection) over a 10-month period of inspection by resident and regional
inspectors. Section 50001.02 (Inspection Requirements) directs inspectors to conduct
selective inspections that will (1) verify that selected design changes and modifications
to SSCs described in the FSAR are reviewed in accordance with 10 CFR 50.59 using
17 SONGS used the following identifier: NECP 800071702, Steam Generator Replacement Unit 2, ASC D0018051, in
its documentation.
18 Appendix A to Inspection Manual Chapter 2515, “Light-Water Reactor Inspection Program -Operations Phase,"
states, “The purpose of reporting the results of baseline inspections is to document the scope of inspections and any
substantial negative findings in support of the assessment process. . . . The NRC does not have objective criteria for
evaluating positive findings. Therefore, the assessment process does not incorporate positive findings and they will
not be documented in baseline inspection reports."
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 24-----------------------
procedure IP 71111.17 as guidance, and (2) review key design aspects and
modifications for the replacement steam generators and other modifications associated
with steam generator replacement utilizing IP 71111.17 as guidance.
Review of NECP 800071702
OIG reviewed SONGS’ Unit 2 Steam Generator Screening and Evaluation (identified by
SONGS as NECP 8000071702) package, which states it was completed on September
18, 2009.19 The screening portion of the document is divided into five parts:
• Section I, “General Information," which describes the proposed activity and
identifies the portion of the activity covered by the screen. For example, this
section identifies specific values of major design basis parameters for the
replacement steam generators that were not consistent with those of the original
steam generators, physical changes between the old steam generators and the
new, and methods of evaluation for the steam generators as described in the
updated FSAR and their respective sections in the updated FSAR.
• Section II, Screen Concerns, which provides the licensee’s answers to five
screening questions used to determine whether the parameter changes, physical
changes, and methods of evaluation identified in Section I needed to be
evaluated under 10 CFR 50.59 to determine whether NRC approval was needed
prior to making the change.
• Section III, Screen Conclusion, which reflected the overall result of the screen as
adverse20 based on the responses in Section II. Section III listed three issues
that would need to undergo a 10 CFR 50.59 evaluation to determine whether
NRC preapproval was needed. Each of the three issues pertained to methods of
evaluation.
• Section IV, References, which lists supporting documents used to conduct the
screen analysis such as topical reports containing analysis details. It also lists
sections of the UFSAR reviewed by the licensee as part of the screening and
evaluation.
• Section V, Summary, which notes that the September 18, 2009, 10 CFR 50.59
screen operation replaced an earlier version with two corrections but that the
conclusion of the original screen remains valid.
19 OIG learned that SONGS provided the IP 71111.17 inspection team with an earlier version of the
10 CFR 50.59 replacement steam generator screening and evaluation package prior to their August 4, 2009, initiation
of the inspection. However, the September 18, 2009, revision reviewed by OIG is the official docketed version that
SONGS submitted to the NRC, on February 14, 2013, in response to a request from NRC associated with SONGS’
Unit 2 CAL response (see page 8 of this report for information on the CAL). The September 18, 2009, revision refers
to an earlier version of the document, but states the only items that changed were two licensee controlled
specifications and the overall conclusion of the original screen remains valid.
26 NEI 96-07 describes adverse in the context of the impact on the performance of SSCs and the bases for the
acceptability of their design and operation.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 25-----------------------
The evaluation portion of the document provides the licensee’s description of the three
method of evaluation the licensee identified, based on the screen, that needed to be
evaluated under 10 CFR 50.59. The activities identified as requiring evaluation were:
• The original steam generator stress analyses described in the UFSAR utilized
the ANSYS computer program, whereas the replacement steam generator
analyses used the ABAQUS computer program.
• The seismic analysis of the reactor vessel internals for the replacement steam
generators was performed in accordance with the methodology described in one
topical report, while the corresponding original steam generator analyses
described in the UFSAR referenced a different topical report.
• The tube wall thinning evaluation for the replacement steam generators changed
the methods of evaluation from those described for the original steam generators
in the UFSAR as follows:
o The original steam generator analysis used the CEFLASH computer
programs for main steam line break (MSLB) mass-energy blowdown
analysis; the replacement steam generator analyses used manual
calculation methods using the maximum differential pressure across the
tube wall during the MSLB.
o The original steam generator loss-of-coolant-accident (LOCA) analysis
contained a two-step process using the STRUDL and ANSYS computer
programs to calculate displacement histories and tube stresses,
respectively. The corresponding replacement steam generator analysis
determined tube stresses from blowdown forces using only the ANSYS
computer program.
o The original steam generator analyses considered primary loop branch
line pipe break plus design basis event (DBE) and MSLB plus DBE as
separate events for determining tube stress. For the replacement steam
generators, the LOCA, DBE, and MSLB events were combined into one
“limiting event” and the stresses for this combined event were calculated.
The evaluation documents the licensee’s answers to the eight 10 CFR 50.59 criteria
(see page 3 of this report for a listing of the eight criteria) relative to the three changes
to method of evaluation. The licensee answered the first seven criteria (which apply to
physical changes) as “N/A” and answered “no” to criterion viii (which applies to changes
in methods of evaluation), followed by an explanation for the answer.
The evaluation document concluded that based on the results of the evaluation, the
steam generator replacement could be implemented without a license amendment.
15
Page 26-----------------------
OIG Observations Concerning SONGS 10 CFR 50.59 Screening and Evaluation
OIG examined the SONGS’ 10 CFR 50.59 screening and evaluation in light of NEI 96-
07 guidance and identified examples where information in the screening and evaluation
did not appear to meet NEI 96-07 guidance pertaining to the need to document results
of 10 CFR 50.59 screens and evaluations.21
First, OIG compared the SONGS’ Unit 2 steam generator 10 CFR 50.59 screening and
evaluation against the UFSAR that would have been available to the 2009 inspection
team and identified at least 14 changes in methods of evaluation used to test the new
design in the UFSAR that were not listed in the SONGS screening. These were listed in
Section 3.9 of the SONGS UFSAR, “Mechanical Systems and Components.”
Second, OIG identified input parameters22 and methods of evaluation where additional
information would be required to support the licensee’s conclusions. For example, in
the 50.59 screen document, there are references to UFSAR sections where revised
values of the replacement steam generator major design parameters are different than
the values of the corresponding operating steam generator parameters. However, the
screening document does not describe if these new values are input parameters, which
are classified as changes to the facility and need to be evaluated under 10 CFR 50.59
criteria i through vii, or if they are elements of a method that need to be evaluated
against criteria viii.23 Other examples are the evaluations for tube wall thinning and
seismic analysis where the licensee performed new analyses but did not provide
sufficient information to support their conclusion that a license amendment was not
needed.
Interviews of SONGS 2009 IP 71111.17 Inspection Team Members
Team Leader
The Team Leader of the SONGS 2009 IP 71111.17 inspection, a reactor inspector in
Region IV’s Division of Reactor Safety (DRS), did not recall any specifics of the 2009
inspection, including whether the SONGS replacement steam generator was selected
21 NEI 96-07 states that 10 CFR 50.59 recordkeeping requirements apply to 10 CFR 50.59 evaluations performed for
activities that screened in, not to screening records for activities that screened out. However, it instructs licensees to
maintain documentation of items that screen out in accordance with plant procedures and states the basis for the
conclusion should be documented to a degree commensurate with the safety significance of the change. NEI 96-07
states that typically, the screening documentation is retained as part of the change package.
22 NEI 96-07 describes input parameters as those values derived directly from the physical characteristics of SSC or
processes in the plant, including flow rates, temperatures, pressures, dimensions or measurements (e.g., volume,
weight, size, etc.), and system response times.
23 NEI 96-07 distinguishes methods of evaluation from evaluation input parameters. Changes to methods of
evaluation (i.e., where an input parameter is considered to be an element of the methodology) described in the
UFSAR are evaluated under criterion 10 CFR 50.59(c)(2)(viii), whereas changes to input parameters described in the
UFSAR are considered changes to the facility that would be evaluated under the other seven criteria of 10 CFR
50.59(c)(2), but not criterion (c)(2)(viii).
16
Page 27-----------------------
for review or whether she personally reviewed any aspects of the screen and
evaluation. She recalled there were two ongoing inspections at the time pertaining to
the steam generator. She did not recall any discussions with the team members about
any aspects of the 2009 SONGS 10 CFR 50.59 inspection.
The Team Leader described her role as the administrative coordinator of the inspection
but with some technical oversight. For example, she conducts a pre-inspection review
of licensees’ 50.59 screens and evaluations to determine safety significance and inform
sample selection; assists team members during the inspection if questions arise; and
post-inspection, works to prepare the inspection report, which is based on the “feeders”
prepared by the inspectors for inclusion in the resident inspector’s integrated inspection
report. If an inspector found an issue during the inspection, she will question the
inspector to ensure the issue was not characterized incorrectly. If the inspector has no
findings, the Team Leader asks a few questions to feel comfortable that the inspector
looked at everything. She did not view herself as an authority on the team, and said
team members were “pretty much even.”
The Team Leader thought existing 10 CFR 50.59 guidance could be improved. She
said she attended a November 2011 counterpart meeting for alignment between the
regions on implementation of the 50.59 inspection procedure. She recalled that each
region interpreted the inspection procedures differently. Additionally, the Team Leader
said there was no specific training for 50.59. She thought NRC has a good 50.59
inspection program, but it needs to be revamped to eliminate these discrepancies.
Team Member 1
OIG learned that the team member who was assigned to review the replacement steam
generator evaluation as part of the 2009 50.59 inspection was a Region IV DRS
Reactor Inspector, and the 2009 SONGS 50.59 inspection was approximately his
second 50.59 inspection. He did not remember details about his specific observations
concerning his review of the screen and evaluation, but he recalled using IP 71111.17
and NEI 96-07 as his primary guidance for the inspection. He also recalled spending
about 2 to 3 days on the steam generator screen and evaluation (out of about 2 weeks
spent onsite for the entire inspection) and recalled that the other two members of the
inspection team also participated in reviewing the screen and evaluation. He said the
team “pulled the screen, which was very large, for the 50.59 for the Unit 2 steam
generators and went line by line to ensure that there was no adverse impact pertaining
to installation of the new steam generators at SONGS Unit 2.”
Team Member 1 remembered talking with SONGS engineers and requesting additional
information about the items that screened in for evaluation because “we need to
independently verify that what they wrote [in the document] is backed up by what
they’ve got out in the field. So, we would have looked at calcs, we would have pulled
engineers in to show us the paperwork behind how they came up with what they wrote”
in the evaluation.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 28-----------------------
While Team Member 1 could not recall exactly what he or other team members asked
the SONGS engineers relative to the change in code from ANSYS to ABAQUS, he
remembered discussing a statement in the evaluation (“The results of these sample
analyses demonstrated that in all cases the ANSYS and ABAQUS results varied from
theoretical solutions by no more than 1 percent, and ABAQUS and ANSYS results
themselves were also within 1 percent of each other”) with SONGS engineers. He said
while neither he nor any of the other team members was an ABAQUS or ANSYS “guru,”
they would have asked to see evidence that this statement was supported.
Team Member 1 recalled looking at the UFSAR for the 2009 inspection but did not
recall his observations at the time. However, he said his approach would be to go
through each UFSAR chapter and “Google search steam generator. Every time steam
generator comes up, I’m going to read the pertinent information.” He said the licensee
needs to identify all sections affected by the design change and he will “balance what
they found versus what I found through the entire FSAR.” If he identifies a difference,
he questions it with the licensee staff. He said that each team member has their area of
expertise and, as a mechanical inspector, his expertise was with mechanical issues. He
said he personally probably reviewed two of the items that screened in for evaluation,
based on his expertise. He said he would not have been involved with any “seismic-
type” questions.
Team Member 1 explained the team documented its inspection by writing a “feeder,”
which contains information that is incorporated into a quarterly inspection report. He
said the feeder contains the number of items inspected; it does not provide details of
what was examined, or how or who examined the screen or evaluation. The feeder will
list the document number of the screen and evaluation in the reference section of the
report.
Team Member 1 did not recall looking at section 3.9 of the UFSAR, titled “Mechanical
Systems and Components,” where methodology changes were listed as part of going
through the steam generator screen, but that he likely had access to those pages. OIG
pointed out several methodology changes included in Section 3.9 that were not included
in the screening analysis, and the inspector said he would have expected those items to
be in the screening. He said if he were doing the inspection now and came across
UFSAR items indicating a methodology change that was not captured in the screen, he
would question the licensee. He would provide the licensee with the procedure that
said it had to be in the screen and question why it was not. He said, “if they didn’t have
a justification and it was in their updated FSAR, I’d give them a violation immediately."
Team Member 1 told OIG that training for 50.59 inspections could be improved. He
went into this inspection right out of college and said it would have been helpful to have
had more training, although it was his understanding that NRC is doing more mentoring
now with new inspectors. He said NRC needs to insure inspectors are fully trained and
versed in the 50.59 process. At the time of this inspection, regional inspectors were not
involving the headquarters technical experts, like they do now. Team Member 1 also
thought the 50.59 guidance available to inspectors is too vague. For example, he said
18
Page 29-----------------------
“more than a minimal increase” should be defined by a specific value in 10 CFR 50.59.
(OIG notes that “more than a minimal increase” is the terminology used in several of the
10 CFR 50.59(c)(2) screening criteria.)
Furthermore, he noted that an inspector’s knowledge and background influences the
conclusion of an inspection. He said, “I could read it one way, [you] could read it a
completely different way,” and subsequently come up with a different conclusion. He
added that inspector “skill sets are to determine and ask appropriate questions that
could lead us to potential issues.”
Team Member 2
Inspection Team Member 2 was a metallurgical and materials engineer and he was
going through the inspector qualification process at the time of the inspection. Team
Member 2 recalled having been involved in the Unit 2 integrated inspection; however,
he did not recall his involvement with the 50.59 inspection. (The Team Leader
confirmed Team Member 2 was involved with the 50.59 inspection.)
Team Branch Chief
A Region IV DRS Branch Chief told OIG that during the 2009 integrated inspection at
SONGS that included the 10 CFR 50.59 review of the SONGS replacement steam
generator, he was in a 3-month rotational position acting as the Branch Chief
responsible for the Region IV DRS Engineering Branch that did the SONGS 71111.17
inspection. It was during this period that NRC issued its SONGS NRC Integrated
Inspection Report 05000361/2009004 and 05000362/2009004 (containing the results of
the 2009 10 CFR 50.59 inspection). The Branch Chief did not work with the inspectors
on the scope of their work or provide any oversight of their onsite inspection activity, but
he said he would have looked at the results they brought back and would have
approved the “feeder” report that they prepared for inclusion in the integrated inspection
report. The Branch Chief did not recall participating in a regional debriefing where the
IP 71111.17 inspection team would have briefed senior regional officials on their
findings, but he said this would have occurred prior to the writing of the feeder report.
He also did not specifically recall approving the feeder report that the team prepared or
anything about its contents.
Interviews of NRC Subject Matter Experts
OIG interviewed four NRC staff members who are knowledgeable about the 50.59
process to ascertain their perspectives regarding SONGS 10 CFR 50.59 replacement
steam generator screen and evaluation; three of these staff members are recognized
within NRC as subject matter experts concerning 10 CFR 50.59 and the fourth is an
experienced senior reactor inspector. OIG asked two of these subject matter experts
(an NRR Branch Chief and regional Branch Chief) to review the SONGS Unit 2, 10 CFR
50.59 screening and evaluation to ascertain their perspective on the adequacy of the
document. OIG also interviewed the third subject matter expert (an NRR program
19
Page 30-----------------------
manager), who contributed to the AIT report, and one member of the SONGS AIT (a
Senior Reactor Inspector). Each employee was questioned concerning their
perspective on the 2009 10 CFR 50.59 inspection.
Both Branch Chiefs whom OIG asked input from provided a written summary of their
responses, and both identified various shortcomings in SONGS’ screening and
evaluation and thought that an NRC 10 CFR 50.59 inspection should have identified at
least some of the issues. The issues they identified collectively included the following:
• Although the purpose of any 50.59 process should be to evaluate changes to the
facility as described in the FSAR, the SONGS evaluation rarely discusses the
actual changes to the FSAR. You cannot tell from the evaluation what exact
statements were changed.
• For a document that is supposed to evaluate adverse changes to “license and
design basis” functional requirements, there is little mention of what the actual
design and license basis requirements are. This displays a significant lack of
understanding of the 50.59 process and requirements at the plant.
• There are a number of general statements in the screen that are never
supported. Just stating these types of conclusions does not make it so; this is a
large, involved screen that, after dissected, lacks substance. Examples cited
regarding the comparison of the original steam generators (OSG) against the
replacements include the following:
o “These differences represent a vast improvement over the OSG materials
in terms of corrosion, erosion-corrosion and wear resistance.”
o “These differences also represent functional improvements over the OSG
components.”
• There are items in the evaluation that need further explanation such as the
following:
o Use of ABAQUS instead of ANSYS -this is an entirely new methodology
and the licensee never discusses whether ABAQUS has been approved
by NRC for this application. This is rudimentary for evaluating a change to
methodology.
o Tube wall thinning analyses changing from CEFLASH, STRUDL, and
ANSYS to manual calculations and ANSYS -there is no justification in the
document for the conclusion that use of ANSYS is consistent with its
intended application, constraints, and limitations. The change from
CEFLASH to “manual calculations” is clearly a new methodology (i.e.,
without providing the documentation to support it meets NEI 96-07
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 31-----------------------
guidance). Also, combining the tube stress analysis into one limiting event
appears to be a new methodology.
• There are some instances where the licensee and their contractor appear to
have deviated from NRC-approved guidance and, as a result, failed to perform
10 CFR 50.59 evaluations related to portions of the steam generator
replacements. The most prevalent deviation from guidance involves areas where
reanalysis was completed to demonstrate that all required safety functions and
design requirements were met. NRC-approved guidance states that in these
circumstances, the change is considered to be adverse and a 10 CFR 50.59
evaluation is required.
• Another area of deviation from guidance involves the licensee’s understanding
and use of guidance related to changes to one or more elements of evaluation
and changes from one method of evaluation to another. Each discussion related
to methods of evaluation is incomplete or flawed. In general, these discussions
lack adequate detail to support the stated conclusions.
• Review of this document would take substantial effort by an inspector. There
really are not enough hours allotted within the baseline 50.59 inspection to
perform a good review. It would take a tremendous amount of time just to get the
required documents to support the licensee’s conclusions, if such documents
exist.
One of the Branch Chiefs noted in his review that 10 CFR 50.59 is an administrative
process to determine if facility changes can be implemented without prior NRC staff
approval by way of the license amendment process of 10 CFR 50.90. He wrote, “Errors
in executing the 10 CFR 50.59 process do not directly impugn the ultimate acceptability
of the design and analysis associated with the proposed change.”
The Branch Chief also told OIG that training was an area that needed improvement and
that the quality of a 10 CFR 50.59 inspection is dependent on the inspector’s
knowledge, experience, and background. He said the guidance is complex and there is
a lot of judgment that is applied in using it. The Branch Chief said the only way to
provide effective oversight is to make sure the inspectors have the tools and training to
effectively execute the inspection procedures. He said that currently, the only training
people get on conducting 50.59 inspections is through the inspector qualification
process and that it would be much better to have some kind of recurrent refresher
training or lessons learned. The Branch Chief said some regions do more than others
in that regard.
The Branch Chief also noted that NRC 50.59 inspections generally occur after the fact
and it is the licensee’s responsibility under the license to complete this process properly,
using their procedures, and our inspection activity is reviewing that and aimed at holding
them accountable on a sampling basis for the quality of the projects they produce and
adherence to their procedures. He said while there may be an opportunity - if an
21
Page 32-----------------------
inspector reviews something while it is being worked on -to identify something that can
change the course of the licensee’s path, but typically the activities are already done in
the field, or on their way to being done, before NRC starts looking.
The other Branch Chief told OIG that the 50.59 regulation is complex and NRC
inspectors need clear guidance, specific training on the 10 CFR 50.59 inspection
process, and increased hours to perform the inspections. In his opinion, the NEI 96-07
guidance is too vague, allows for too many judgment calls, and needs solidifying of
definitions. From his experience, the licensee and NRC routinely get into
disagreements because of interpretation of the guidance.
This Branch Chief told OIG that had he reviewed the SONGS 50.59 in 2009, he would
have come to the conclusion that, without additional documents, he would have
absolutely no reasonable assurance that SONGS could pass a 50.59 inspection. He
would have told the licensee that their documentation was inadequate and not
documented properly, and the licensee may have had time before the install to produce
adequate documentation. The Branch Chief said he would have expected the 2009
inspectors to ask questions and follow up on the unsupported general statements made
by the licensee in the 50.59 documents; however, he could not determine from the
documentation he reviewed whether a license amendment was needed.
The Senior Reactor Inspector who was on the AIT told OIG he noticed issues with
regard to changes in methodology and, therefore, suggested to the AIT team lead that
someone take a second look at the 50.59 to determine how to treat the methodology
changes described in the evaluation section. OIG asked this inspector, who performs
50.59 inspections, to review section 3.9 of the UFSAR and provide his opinion as to
whether any of the methods of evaluation listed should have been screened based on
applicable guidance for implementation of 10 CFR 50.59. The Senior Reactor Inspector
concluded that at least 15 of the 41 codes listed in section 3.9 should have been
screened and evaluated against the criteria in 10 CFR 50.59.
The NRR Program Manager who contributed to the AIT report was not an AIT member,
but was selected for his technical expertise in 50.59s to review the licensee’s 50.59
evaluation performed as part of the replacement steam generator modification for the
AIT. The NRR Program Manager thought the region did its job in 2009, but he identified
two issues for the AIT where SONGS 50.59 documentation included inadequate support
for methodology changes. He did not know why the 2009 inspection did not catch the
issue and said it is possible that the inspector raised these questions and the licensee
offered an explanation that the inspector found acceptable. The NRR Program
Manager also commented that inspectors can watch an activity every month -for years
even -and then a different inspector will come in and find something the others never
identified.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 33-----------------------
NRR Response to OIG Questions About Methodology Changes
In an August 25, 2014, memorandum to OIG, the Acting NRR Director reported that
based on a review of all docketed information, including NRC inspection reports, NRC
staff determined there was insufficient information to answer specific questions
regarding whether certain methodology changes utilized with SONGS replacement
steam generators that were not described in the SONGS 50.59 screen should have
been described. The memorandum responded to OIG questions concerning whether
14 methodology changes should have been described in the screen. Specifically, NRC
could not determine if computer codes used with the original steam generators were
replaced by different codes; and if replaced, NRC could not determine if the codes had
been benchmarked for use in the replacement steam generators. According to the
memorandum, it was not unexpected that NRC did not have on hand all the information
needed to answer OIG’s questions “based on NRC Inspection Manual 2515, ‘Light-
Water Reactor Inspection Program,’ which states, ‘[t]he NRC inspection program covers
only small samples of licensee activities in any particular area,’ and ‘individual
inspectors are expected to exercise initiative in conducting inspections, based on their
expertise and experience.’”
The memorandum reported to OIG that it is plausible some of the methodology changes
identified may have “screened out” and therefore would not have required an
evaluation. The memorandum noted that screening documentation is not required by
10 CFR 50.59 and not subject to 10 CFR 50.59 documentation and reporting
requirements.
In response to OIG questions where SONGS indicated codes and analytical
calculations were used for the original steam generators, but were not used for the
replacement steam generators, NRC staff reported they do not know why SCE did not
mention these codes or any replacement codes in their 50.59 screening. The
memorandum advised that implementation guidance in NEI 96-07, with regard to
“Applicability,” provides many possible reasons an activity may require information in
the UFSAR to be updated that would not require documentation under 10 CFR 50.59.
Interviews of NRC Region IV Management
Former Deputy Regional Administrator
The former Region IV Deputy Regional Administrator24 said that his expectations for
50.59 inspections are no different from any other inspection. He expects a thorough
assessment (inspection) consistent with the guidance and the program requirements.
He also expects that to the extent that problems are identified by the staff, a path is
identified to ensure resolution, especially if it is a safety issue. Regarding the 50.59
inspection process, the fundamental question is to understand whether or not the
24 The former Region IV Deputy Regional Administrator held this position from December 2010 to March 2013. From
March 2013 until his retirement in November 2013, he was the Regional Administrator. This report refers to him as
the former Deputy Regional Administrator.
23
Page 34-----------------------
licensing staff, NRR, needs to conduct a review before a change is permitted to the
licensing basis of the plant. He said that samples are taken to see if the licensee
comported with criteria of the regulation.
The former Deputy Regional Administrator told OIG that he was not directly involved
and did not have any firsthand knowledge of the specific inspection activities planned or
conducted during the 2009 Region IV integrated inspection at SONGS. While he read
the inspection report, he had not discussed the 50.59 review with the inspector who
conducted the review to gain an understanding of what the inspector reviewed and what
direction he was given by his supervisor. He believed that in hindsight, SONGS needed
a license amendment based on what was known today (2013), however, he could not
speak to what was known in 2009 or 2010. The former Deputy Regional Administrator
explained that the design, as built, was fundamentally flawed and would not have been
approved under any conditions. The overall design was unacceptable because of the
adverse thermal-hydraulic conditions and the upper tube structure support being
inadequate.
Regarding inspection guidance procedures relevant to the 50.59 process, the former
Deputy Regional Administrator stated that the inspection guidance appears to be
focused on the updated FSAR, and provides some practical guidance. However, it
does not address the issue of why a change might need to have pre-approval. The
inspection guidance does not cover the details of the rule, why each of the 50.59 criteria
exists, and how to interpret them.
According to the former Deputy Regional Administrator, the inspection guidance can be
improved. He noted that he has heard from staff members that they are dissatisfied
with NEI guidance. The challenge is that there are so many different types of
components, structures, and systems and it is hard to write a procedure that captures
all those different circumstances. Having a detailed inspection plan that allows one to
probe into the important areas is part of inspection preparation and for an effort like the
steam generator replacement, which is like inspecting a system, there would need to be
more individuals involved, more resources, more time, and more preparation. He
added that the steam generator replacement inspection guidance is focused on a
number of activities including opening up containment, removing the old generators,
placing new generators, patching the containment and verifying containment integrity.
The guidance allocates 350 hours for this activity and only approximately 60 hours are
devoted to design, including the 50.59 review, which is not very much time to actually
delve into a complex component such as steam generators.
The former Deputy Regional Administrator said that the agency has to decide if it is
important enough to inspect every one of these (steam generators) as they come along,
and that inspections are funded and adequately supported. He recognized that there
are only 12 more units that may change their steam generators. The program guidance
needs to be reviewed and could be enhanced and the resource allocation needs to be
reviewed. The former Deputy Regional Administrator added that guidance for
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 35-----------------------
inspecting in-service steam generators experiencing tube degradation provides a lot of
insight, but has not been updated in 18 years. Some of the guidance is out of date and
some needs to be strengthened.
The former Deputy Regional Administrator said that there are not many findings that
result from [steam generators or 50.59] inspections. When the staff initially started
conducting steam generator replacement inspections they were a huge deal and the
inspection plan was discussed at the branch, division, and regional administrator’s level
and there was a lot of communication with the other regions. However, in his view, over
time these inspections have become routine.
Former Regional Administrator
The then-NRC Region IV Regional Administrator25 told OIG he recalled being briefed by
the Resident Inspector assigned to SONGS on the steam generator replacement project
inspection, but he did not recall the aspects of preparation or reporting from the 2009
Integrated Inspection conducted by Region IV. However, he said that all inspections get
briefed verbally in Region IV in a setting where supervisors or managers are present;
thus, he presumed an outbrief did occur for the 2009 Integrated Inspection.
The former Regional Administrator stated inspectors are certified to inspect the
evaluation packages per agency guidance. Region IV follows program office guidance
on 50.59 and the inspection procedure. He expects inspectors to look at calculations
and the engineering behind the design changes. Although these reviews are never 100
percent because they are done through sampling, his expectations are that the
inspectors look hard and that they challenge. The former Regional Administrator wants
them to be as thorough as they can be, but their time is limited. So they can never look at
everything. He expects them to follow guidance and do a good job of fact-gathering, and
then do good analysis and draw their inspection conclusions. This is true with all
inspections, not just 50.59 inspections.
The former Regional Administrator told OIG that inspectors review the important functions
of the important systems and components that are being changed and the evaluation
behind the change to see if it is justifiable that it is not increasing the risk -not triggering
those criteria from the 50.59 rule. He said the licensee’s evaluation inputs are those
design inputs, the design objectives, “and if they say the right things, it will pass
50.59. And if it's wrong, then something is going to get built and put in the plant that
doesn't function correctly. And that's what happened at San Onofre.” He said that
knowing what they know now, “the steam generators as designed were basically
unlicensable. We wouldn't approve them.”
The former Regional Administrator advised 50.59 could be strengthened with explicit
instruction to the inspector in preparation for their inspections. An overt effort should be
placed to assemble the FSAR information on the subject of the change, compile all
25 The former Region IV Regional Administrator served in this position from the Fall of 2007 until his retirement in
March 2013.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 36-----------------------
information relating to the subject, wherever it is mentioned, to facilitate a comparison to
the screens and evaluations on that subject. He stated questions can then be asked of a
change: Where is the screen? Where is the evaluation? How was it done? He
mentioned a limiting factor with 50.59 is that a lot of information contained in the licensing
basis is not contained in the FSAR and may not be reviewed under 50.59.
The former Regional Administrator felt the NRC should consider excluding some design
changes from 50.59. He said it is worth the time and effort for the NRC to do a license
amendment on a major reactor coolant system component. He said a license amendment
would not accomplish design validation, but it will get to a certain set of criteria and, in the
case at SONGS, it would have caused reviewers to ask good questions to determine
“What’s behind it?” For example, he said that based on guidance in the Standard Review
Plan,26 one item that certainly would have been questioned by reviewers was the
acceptance criteria of 95 percent (void fraction27), because no other plant in the industry is
over 90 percent. Some reviewer would have said that this is an outlier and we need to
understand that.
The former Regional Administrator stated to leave the review of such a large and broad
component with the scope and depth of engineering of the steam generator with the
inspection oversight program may be risky. The inspection program will miss things
because it is not encompassing enough to review to the level of detail or scrutiny as does
the Standard Review Plan.
Current Region IV Deputy Regional Administrator
When asked to interpret the differences of opinion from the 2009 inspection to the
subject matter expert reviews during this inquiry, the current Region IV Deputy Regional
Administrator said he did not know the amount of time taken in 2009 as compared to the
AIT and AIT followup and any additional reviews. He said the different outcomes could
reflect the level of experience of inspectors. Additionally, the 2009 inspectors and
AIT/AIT followup team members would have had the benefit to ask for clarification from
the licensee, where the subject matter experts interviewed by OIG may have not. He
could not assess the adequacy or draw a conclusion about the 2009 inspection based
on the comments of the subject matter experts and OIG review; he would need more
information to determine if the 50.59 evaluation was accomplished incorrectly. He said,
“There’s no compelling reason at this point based on the questions to go do additional
inspection on a plant that’s been shut down and is in the decommissioning process.”
However, he added the fact that the AIT cited SONGS for a couple of minor violations,
indicates the 2009 inspection team did not conduct a proper 50.59 evaluation.
26 NRC prepares Standard Review Plans to establish criteria that NRC staff responsible for the review of applications
to construct and operate nuclear power plants use in evaluating whether an applicant/licensee meets the NRC's
regulations.
27
Void fraction is the calculated value (i.e., parameter) for the volume of gas in a given gas-liquid flow area. It is an
important parameter used to characterize two phase flow of circulating water through the steam generator; and
also, a key parameter used in computer models to predict the flow pattern transitions, heat transfer, and pressure
drop.
26
Page 37-----------------------
The RIV Deputy Regional Administrator commented that if two 50.59 inspectors have
the same information and come to different conclusions, we need to take a look at our
guidance and make sure that it is clear and more objective. He advised based on what
he knows about the SONGS issues, the agency may benefit from a different inspection
approach to do some kind of screening and determine if a more detailed design-type
review for major component replacement inspections is necessary. This issue is being
addressed with the lessons learned review. Additionally, he believes there could be
improvements in training and guidance, as evidenced by the subject matter experts
having different observations in reviewing the SONGS replacement steam generator
evaluation. And if experts have different views, inspectors with less experience may
have more variability in their conclusions.
Interviews of NRC Headquarters Managers
The former NRR Director28 told OIG he was not familiar with the 2009 Region IV
integrated inspection pertaining to the SONGS 50.59 and he could not speak to the fact
that there were no inspection findings related to the 50.59 screening and evaluation.
However, he said that inspectors do not look at everything and are trained to sample.
NRC does not have the resources, including time or manpower, to review everything
and so inspectors sample. Inspectors are given guidance and are entrusted to use
judgment about what they should or should not review. Although he was unfamiliar with
the details of the AIT inspection, he was aware that the AIT identified issues pertaining
to the 50.59 evaluation, one of which was a minor violation.
The former NRR Director said that the problem with the SONGS steam generators was
a design issue. The 50.59 process would not have prevented the steam generators
from leaking and this is not the purpose of the 50.59 process. The 50.59 process is not
going to stop a licensee from buying bad equipment, or from operating a plant
incorrectly. It is not NRC’s job to make sure that a licensee buys good generators. The
NRC’s job is to protect public health and safety.
The former NRR Director told OIG that the 50.59 process is a threshold as to whether a
licensee can take a particular action without NRC approval. The NRC staff recently
completed a review of 50.59s that were conducted over a 12-year period, which had
identified 138 findings of low safety significance. According to the former NRR Director,
this information indicated that NRC was reviewing the 50.59s and that mistakes were
being made; however, none of these findings were particularly safety-significant.
Consequently, based on this data, he does not believe that he has a lot of problems
with the generic use of 50.59 by licensees or a problem with NRC oversight of the 50.59
process.
According to the former NRR Director, if there were problems with the 50.59 process, it
would have manifested itself in many more issues than just the steam generator issue.
He noted that a steam generator replacement involves a major component, which
requires doing a number of things that are significant with regard to a nuclear power
28 The former NRR Director served in this position from May 2009 to June 2014.
27
Page 38-----------------------
plant, including breaking into the primary system. The former NRR Director added that
NRC has not had an issue with the approximately 53 nuclear power plants that have
changed their steam generators under the 10 CFR 50.59 process.
According to the former NRR Director, based on the information provided by OIG
pertaining to methodology changes, it appeared that NRC may have done a bad job of
reviewing the SONGS 50.59 during the 2009 inspection; however, one should be
careful before concluding that this was a broader problem than SONGS. The inspection
program is working and has worked very well in the past. He did not have data to
conclude that NRC needs to do a larger inspection sample. What happened at SONGS
was not particularly safety-significant. The plant shut down as intended in accordance
with the plant’s technical specification. The licensee found that there was a leakage
and took the right actions to shut down the plant.
The former NRR Director said that NRC should assess what happened at SONGS and
whether there are any lessons to learn. It was a good question to ask whether the NRC
was doing enough of a review of the 50.59s being conducted. Nevertheless, as the
NRR Director responsible for the operational safety of 100 nuclear power plants and
research and test reactors, he has limited resources. He and the NRC staff need to
remain focused on safety. He is not concerned about SONGS anymore because the
plant would never be in operation again. What is of interest to him is that the mistake is
never repeated and that what happened at SONGS is communicated to other licensees
so they do not repeat the mistake.
Deputy Executive Director for Reactor and Preparedness Programs
NRC’s Deputy Executive Director for Reactor and Preparedness Programs told OIG that
the licensee’s 50.59 review is intended to control changes in the plant. Safety of the
plant is the licensee’s primary responsibility, and NRC relies on licensees to implement
technical specifications and the conditions of their license. A licensee may make
changes in their facility without obtaining a license amendment only if they meet the
eight criteria in 10 CFR 50.59. NRC makes sure the licensee is applying 50.59 correctly
and coming to the agency for review prior to their implementation by implementing IP
71 1 He said the 50.59 review (i.e., IP71111.17) is not a safety review by the
- .17.
NRC staff and is not intended to pick up problems with design flaws like 10 CFR 50.90
(i.e., license amendment requests) reviews; however, it could. Also, he said, “We’re
only going to be able to sample, and you always want to make sure that you’re sampling
the items with the highest likely safety significance input.”
OIG discussed with the Deputy Executive Director the assessments provided by the
four agency staff knowledgeable about 10 CFR 50.59; the Deputy Executive Director
said he was not bothered that agency technical experts could have questions and end
up disagreeing on some technical issues, although he said, "it certainly raises some
questions.” His main concern is that the process is not overly driven by subjectivity and
judgment. He envisioned a process where a team interacts and collaboratively work
through issues and concludes with a decision that everyone understands, even if they
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 39-----------------------
do not necessarily totally align with it. The Deputy Executive Director said that the
50.59 process has to be repeatable and predictable by inspectors whether they have 10
or 30 years of experience. There are regulations and guidance that endorse an
acceptable way to conduct 50.59 inspections.
The Deputy Executive Director commented that the high frequency with which licensees
use the 50.59 process coupled with the relatively low frequency of issues identified by
NRC suggests to him that training could be a factor. He believes that if we look at
training in a broad sense - qualification training, on-the-job-training, and experience - we
may be able to understand how training influenced inspectors’ decisions made or
should have been made differently. He did not know what improvements will result from
the agency’s ongoing lessons learned review for the SONGS steam generator failure,
but said there will be some. For example, communications with external stakeholders
will be an area that NRC improves and possibly areas with the 50.59 process and how it
gets implemented.
The Deputy Executive Director was aware that stakeholders are concerned with who
was at fault with respect to the steam generator problems; however, he said that from a
safety perspective, he focused on determining the cause of the failure and making sure
the NRC had the information it needed to make a decision about restart. “They screwed
it up and we didn’t pick it up. Turns out we didn’t look at it in detail because they did a
50.59 review -didn’t believe they needed to get us to look at [a license amendment
request].’’
Findings
OIG found that NRC missed an opportunity during a 2009 triennial baseline inspection
of SONGS’ implementation of the 10 CFR 50.59 process to identify weaknesses in the
SONGS steam generator 50.59 screening and evaluation package. While a Region IV
inspection team selected the SONGS Unit 2 steam generator 10 CFR 50.59 screening
and evaluation package as one of 35 items sampled during a 2009 triennial baseline
ROP inspection at SONGS, the inspection team did not identify various shortcomings
noted more recently by NRC subject matter experts who reviewed the steam generator
screening and evaluation package subsequent to SONGS’ shutdown due to problems
with steam generator design. The purpose of the triennial baseline inspection (IP
71111.17, “Evaluations of Changes, Tests, or Experiments and Permanent Plant
Modifications") is to provide assurance that required license amendments have been
obtained.
The 2009 inspection team concluded from its review of the 35 items sampled that
SONGS had correctly determined that the changes SONGS made could be made
without a license amendment. However, the NRC subject matter experts who reviewed
the Unit 2 steam generator screening and evaluation package following SONGS’
shutdown identified questions pertaining to the Unit 2 steam generator 10 CFR 50.59
screening and evaluation, some of which NRC says cannot now be answered based on
available information. The questions raised by the subject matter experts pertain to (1)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 40-----------------------
insufficient support for 10 CFR 50.59 evaluation conclusions that contributed to the
decision that a license amendment was not needed and (2) methodology changes that
should have been considered for screening but were not listed in the screening
documentation. OIG found that (1) without knowing whether everything that should
have been screened was screened, and the outcomes of these screenings, and (2)
without reviewing additional information concerning the evaluation conclusions, there is
no assurance that NRC reached the correct conclusion in its 2009 inspection that
SONGS did not need a license amendment for its steam generator replacement.
OIG found that the primary inspector who reviewed the SONGS Unit 2 steam generator
10 CFR 50.59 screening and evaluation package during the 2009 baseline inspection
(at approximately the same time installation of the Unit 2 steam generators
commenced) described conducting a review that aligned with inspection guidance, but
said that in hindsight, with the experience he now has, he might have probed further into
certain aspects of the screening and evaluation package. This inspector, and others
interviewed during the investigation, identified a need for improvement in training and
guidance to inspectors for the 50.59 inspection. Although several senior managers
acknowledged some of the shortcomings in the SONGS screening and evaluation
package, they supported NRC’s inspection approach, which relies on sampling and
judgments made by inspectors with different backgrounds and experience levels. One
senior manager expressed confidence in the 50.59 inspection process, and noted that
the purpose of NRC’s 50.59 inspection is not to identify design flaws, but rather to
determine whether licensees are correctly implementing the 50.59 rule and reaching the
correct conclusions as to the need for NRC preapproval. At the same time, senior
managers, subject matter experts, and inspectors expressed general agreement that
NRC needs to improve its 10 CFR 50.59 inspection training and guidance.
ISSUE 2: AIT Review of SCE’s 10 CFR 50.59 Evaluation
Background
In mid-March 2012, Region IV established an Augmented Inspection Team (AIT) to
assess the circumstances surrounding the tube leak and unexpected wear of tubes in
the Unit 3 steam generators.29 A Region IV Division of Reactor Safety Branch Chief
was assigned to lead the team, which also included a Region IV Resident Inspector, a
Region II Senior Reactor Inspector, and four headquarters-based engineers (two from
NRR, one from the Office of New Reactors, and one from the Office of Nuclear
Regulatory Research). According to the March 16, 2012, AIT charter memorandum
29 Management Directive 8.3, NRC Incident Investigation Program, states it is NRC’s policy “to ensure that significant
operational events involving reactor and material facilities licensed by the NRC are investigated in a timely, objective,
systematic, and technically sound manner; that the factual information pertaining to each event is documented; and
that the cause or causes of each event are ascertained.” MD 8.3 explains that events may involve responses by an
incident investigation team or less formal responses by an AIT or a special inspection team, depending on the level of
response required. NRC Inspection Procedure 93800, Augmented Inspection Team, states the AIT is responsible for
identifying generic safety concerns in a timely manner and emphasizing fact finding, i.e., fully understanding the
circumstances surrounding the event and probable causes, and not an AIT’s responsibility to examine the regulatory
process to determine whether the process contributed to the cause or course of the event or determine whether NRC
rules or requirements were violated, or recommend enforcement actions.
30
Page 41-----------------------
from the then Region IV Regional Administrator to the AIT Team Leader, the inspection
was chartered to:
• Identify the circumstances surrounding the tube degradation.
• Review the licensee’s actions following discovery of the conditions.
• Evaluate the licensee’s review of potential causes of the unusual steam
generator tube wear.
• Assess adequacy of licensee's actions to prevent recurrence.30
Included within the AIT’s inspection scope was a specific item (number 4), identified to
OIG by the AIT inspection team leader, which read as follows:
“Collect and assess differences in steam generator design and
manufacturing between Units 2 and 3. Review all design and
manufacturing changes to ensure they were properly reviewed and
approved in accordance with procedures.”
The AIT Team Leader told OIG that although this paragraph did not state 50.59
specifically, the paragraph’s intent was to review the licensee’s 50.59 activities
associated with components that may have contributed to the tube leak. The
former Director, Division of Reactor Safety, NRC Region IV, also confirmed the
50.59 review was intended from the beginning under item 4 of the charter.
Review of AIT Inspection Reports
The AIT conducted its inspection of SONGS from March 19 to June 18, 2012, and on
July 18, 2012, NRC issued San Onofre Nuclear Generating Station - NRC Augmented
Inspection Team Report 05000361/2012007 and 05000362/2012007, describing the
AIT’s results. The AIT determined that the plant operators responded to the January
31, 2012, steam generator tube leak in a manner that protected public health and safety
and all safety systems performed their functions to support the safe shutdown and
cooldown of the plant, but that “the loss of steam generator tube integrity is a serious
safety issue that must be resolved prior to further power operation."
With regard to AIT scope item number 4 pertaining to review of all design and
manufacturing changes, the AIT concluded that no significant differences existed in the
design requirements of Units 2 and 3 replacement steam generators, and that based on
the UFSAR description of the original steam generators, the steam generators’ major
design changes were reviewed in accordance with the 10 CFR 50.59 requirements.
The AIT identified 10 “unresolved” items that warranted additional NRC followup
inspection or review. Item number 10 pertains to the licensee’s 10 CFR 50.59 steam
generator review.
30 The AIT charter was revised on May 16, 2012, in part to also evaluate unexpected wear in the Unit 2 steam
generators.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 42-----------------------
(1) Adequacy of the post trip/transient procedure.
(2) Evaluation and disposition of the Unit 3 loose parts monitor alarms.
(3) Design of the retainer bar.
(4) Control of original design dimensions.
(5) Evaluation of and controls for divider plate repair.
(6) Atmospheric controls of Unit 3 steam generators during shipment.
(7) No tube bundle support used during shipping.
(8) Evaluation and disposition of accelerometer readings during shipping.
(9) Adequacy of Mitsubishi’s thermal-hydraulic model.
(10) Change of methodologies associated with 10 CFR 50.59 review.
Section 13 (“Office of Nuclear Reactor Regulation (NRR) Review of SONGS 50.59
Evaluation”) of the overall AIT inspection report, summarizes the scope of the inspection
effort that resulted in unresolved item 10, which pertains to the 10 CFR 50.59 review.
Section 13 states that the “NRR technical specialist”:31
• Reviewed all of the design changes associated with the replacement steam
generators to determine whether the changes to the facility or procedures, as
described in the UFSAR, had been reviewed and documented in accordance
with 10 CFR 50.59 requirements.
• Reviewed the various information used by SCE to review the changes being
made to the replacement steam generators, including calculations, analyses,
design change documentation, procedures, the updated final safety analysis
report, the technical specifications, and plant drawings.
• Determined if the design changes to the replacement steam generators were a
change to the facility or procedures as described in the updated final safety
analysis report or a test or experiment not described in the updated final safety
analysis report.
• Verified that safety issues related to the changes were resolved.
• Compared the safety evaluations and supporting documents to the guidance and
methods provided in NEI 96-07, “Guidelines for 10 CFR 50.59 Implementation,”
Revision 1, as endorsed by NRC Regulatory Guide 1.187, “Guidance for
Implementation of 10 CFR 50.59, Changes, Tests, and Experiments,” to
determine the adequacy of the 10 CFR 50.59 evaluations.
The AIT report stated that in reviewing SCE's 10 CFR 50.59 evaluation, the NRR
technical specialist found two instances that failed to adequately address whether the
change involved a departure of the method of evaluation described in the updated final
safety analysis report:
(a) Use of ABAQUS instead of ANSYS: Updated Final Safety Analysis Report
Sections 3.9.1.2.2.1.11 and 3.9.1,2.2.2.3 were revised to reflect that the
SONGS Unit 2 and 3 original steam generators stress analyses for reactor
31 OIG learned through interviews that the NRR technical specialist referred to in the AIT report was not one of the
AIT members, but an NRR Project Manager who was asked to perform work pertaining to the SONGS 10 CFR 50.59
review in support of the AIT effort.
32
Page 43-----------------------
coolant system structural integrity utilized the ANSYS computer program,
whereas the replacement steam generators analyses utilized the
ABAQUS computer program. The SCE’s 50.59 evaluation incorrectly
determined that using the ABAQUS instead of ANSYS was a change to an
element of the method described in the updated final safety analysis
report did not constitute changing from a method described in the updated
final safety analysis report to another method, and as such, did not
mention whether ABAQUS has been approved by the NRC for this
application.
(b) Use of ANSYS instead of STRUDL and ANSYS: Updated Final Safety
Analysis Report Section 5.4.2.3.1.3 was revised to reflect that the SONGS
Units 2 and 3 evaluation of tube stress under loss of coolant accident
conditions for the original steam generators consisted of a two-step
process utilizing the STRUDL and ANSYS computer programs to calculate
displacement histories and tube stresses, respectively, while the
corresponding replacement steam generators analysis determined tube
stresses from blowdown forces using only the ANSYS computer program.
While SCE’s 50.59 evaluation correctly considered this a change from a
method described in the FSAR to another method, the 50.59 evaluation
did not mention whether the method has been approved by NRC for this
application.
These two instances were identified as unresolved issue URI 05000362/2012007-10,
“Evaluation of Departure of Method of Evaluation for 10 CFR 50.59 Processes."
Section 13.0 concluded that additional review and followup would be required to review
the departure of the method of evaluation used during the stress analysis calculations
associated with the replacement steam generators.
On November 9, 2012, NRC issued San Onofre Nuclear Generating Station - NRC
Augmented Inspection Team Follow-Up Report 05000361/2012010 and
05000362/2012010, reflecting the results of NRC’s followup inspection of 9 of the 10
unresolved items identified by the AIT in its July 18, 2012, report. The AIT followup
team was composed of four members: the same Team Leader, Region IV Resident
Inspector, and Region II Senior Reactor Engineer who served on the initial inspection,
plus a Region IV Engineer who was not on the original team.
The November 9, 2012, AIT followup report stated that NRC had closed 8 of the 10
unresolved items. Included among the eight closed items was URI 05000362/2012007-
ID, “Evaluation of Departure of Method of Evaluation for 10 CFR 50.59 Processes.”
With regard to the use of ABAQUS instead of ANSYS to conduct a stress analysis for
reactor coolant system structural integrity, the followup inspectors determined that the
change of methods “would not have required a license amendment based on NRC’s
approval for the use of ABAQUS at other nuclear power plants in similar applications.”
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 44-----------------------
To reach this conclusion, the inspectors reviewed the Comanche Peak Updated Safety
Analysis Report, and several other reports32 describing uses of ABAQUS for
applications similar to RCS structural integrity. However, the inspectors determined that
the licensee’s decision to use ABAQUS instead of ANSYS for reactor coolant system
structural integrity analyses constituted changing a method of evaluation described in
the SONGS updated safety analysis report and subsequently identified a minor violation
of 10 CFR 50.59(d)(1), which requires the licensee to maintain records of changes in
the facility for changes that do not require license amendments.
With regard to the use of ANSYS instead of STRUDL and ANSYS to evaluate tube
stress, the inspectors did not identify a violation of 10 CFR 50.59. They noted that the
licensee had used ANSYS to calculate tube stresses for both the original and
replacement steam generators, so the licensee’s use of ANSYS for this purpose did not
constitute a change from the method described in the UFSAR. The inspectors noted
that for the original steam generators, the licensee had analyzed a combination of
events using STRUDL to calculate displacement histories for those events, which
provided additional margin for analysis done by ANSYS. However,
the inspection report stated that for the replacement steam generators, the licensee had
analyzed for the most limiting event, and had sufficient margin, so STRUDL was not
needed.
OIG Observations Concerning AIT Reports
OIG noted that while the July 18, 2012, AIT report identified a URI related to change of
methodologies associated with SONGS’ 10 CFR 50.59 review, it did not identify (1) the
14-plus changes in methods of evaluation used to test the new design in the updated
FSAR, (2) the numerous input parameters and methods of evaluation where additional
information would be required to support the licensee’s conclusions, and (3) the general
and unsupported statements and conclusions in the screen and evaluation that would
require additional information to determine if these statements were supported and
conclusions valid.33
In addition, OIG noted that the language NRC used in the followup AIT inspection report
to explain why the use of ABAQUS instead of ANSYS (a “change of methods”) would
not have required a license amendment does not align with the language in 10 CFR
50.59. The AIT followup inspection report states that the determination that “the change
of methods would not have required a license amendment was based on the NRC
approval for the use of ABAQUS at other nuclear power plants in “similar applications."
32 These reports included ORNL/NRC/LTR-04/15, “Probabilistic Structural Mechanics Analysis of the
Degraded Davis-Besse RPV Head,” September 2004 (ML 042600455); D. Rudland, D.J. Shim, H. Xu, and G.
Wilkowski, “Summary Report on Evaluation of Circumferential Indications in Pressurizer Nozzle Dissimilar Metal
Welds at the Wolf Creek Power Plant to Nuclear Regulatory Commission Washington DC,” April 2007 (ML
071560398); NUREG/CR-6854, “Fracture Analysis of Vessels - Oak Ridge FAVOR, v04.1, Computer Code: Theory
and Implementation of Algorithms, Methods, and Correlations,” September 2004 (ML061580369); NUREG/CR-6765,
“Development of Technical Basis for Leak-Before-Break Evaluation Procedures," May 2002 (ML 021720594);
NUREG/CR-6774, “Validation of Failure and Leak-Rate Correlations for Stress Corrosion Cracks in Steam Generator
Tubes,” May 2002 (ML 021510286).
33 These items were described under the Issue 1 section of this report.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 45-----------------------
However, 10 CFR 50.59 (a)(2)(ii) describes that changes in methods of evaluation
described in the UFSAR to another method requires that the latter method has been
approved by NRC for the “intended application.”
Interviews of AIT Participants
Former Division of Reactor Safety Director
The former Director,34 Division of Reactor Safety (DRS), Region IV told OIG that he was
responsible for deciding if an AIT was warranted and that he put the inspection team
together. He stated the 50.59 review was intended from the beginning; even if the term
“50.59” was not stated verbatim in the charter, 50.59 was certainly part of what Region
IV expected and understood would be looked at, and further recognizing that NRC had
done an inspection for that 50.59 modification (i.e., the 2009 Integrated Inspection,
which included the 50.59 replacement steam generator package).
The former DRS Director also served as the technical manager for the AIT followup
report. During that period, he had weekly phone calls with his counterparts in NRR for
the technical issues. He remembered closing the 50.59 URI and that they had strong
dialogue back and forth with headquarters. He recalled that in the end, the licensee
was cited for not identifying the methodology change; however, the change performed
the same analysis and was as effective as the original methodology and did not negate
the entire 50.59.
AIT Team Leader
The AIT Team Leader told OIG that he tasked the AIT Senior Reactor Inspector from
Region II to conduct the team’s evaluation of SONGS’ 10 CFR 50.59 evaluation as part
of the AIT effort. He oversaw and held daily debriefs concerning the team’s activities,
but was not a 50.59 expert, and the 50.59 review was the responsibility of the Region II
Senior Reactor Inspector who served on the AIT. The Team Leader also recalled that a
headquarters specialist from NRR who was not part of the AIT was also involved in the
10 CFR 50.59 review effort. He recalled that the Region II Senior Reactor Inspector
used NEI’s NRC-endorsed guidance (NEI 96-07, Guidelines for 10 CFR 50.59
Implementation) to review the 50.59 screens and evaluations and determine whether a
license amendment was needed. The Branch Chief said he thought the AIT did a
“pretty thorough scrub” before “essentially” concluding that a license amendment was
not required. He said the AIT did find a couple of minor violations, but said, “The AIT
could have missed something too. We had a lot of stuff to look at. ..We didn’t look at
everything."
The AIT Team Leader informed OIG the AIT charter was developed to identify the
circumstances surrounding the tube degradation, review the licensee’s actions following
the discovery of the conditions, evaluate the licensee’s review of potential causes of the
unusual steam generator tube wear, and assess the adequacy of licensee’s actions.
34 The former DRS Director served in this position from March 2013 to January 2014.
35
Page 46-----------------------
The AIT was focused on the tube degradation. He stated the AIT was not meant to look
at all the design changes -only the design changes that could have an impact on the
steam generator tube leak. Furthermore, the AIT inspection is an incident based
inspection; he explained it is not a 50.59 inspection but questions arose regarding the
adequacy of 50.59 and the AIT did more work to address those questions.
The AIT Team Leader explained the AIT cited SCE with a thermal hydraulic modeling
design violation of (10 CFR Part 50) Criterion III, (Design Control), which resulted in
Region IV issuing SCE an apparent white violation. He said, Region IV concluded that
they did not believe the velocity information was available at the time (during the 50.59
process), like it is now. Had the information been available, SCE would have had to do
something differently from a 50.59 perspective. He advised that, collectively, the team
thought if SONGS had done a more thorough evaluation, they potentially could have
identified the problem if they had checked the adequacy of some of the information that
they had questions on, specifically questions regarding velocities and questions
concerning design control. The Team Leader recalled this was the reason Region IV
focused on design control, and that NRC headquarters was consulted and concurred.
Region II Senior Reactor Inspector
The Region II Senior Reactor Inspector on the AIT told OIG he was initially tasked by
the AIT team leader to evaluate the differences between the Units 2 and 3 steam
generators, and later asked to look into the 50.59 evaluation, although this was not in
the original charter. The Senior Reactor Inspector said the decision to review the
licensee’s 50.59 evaluation arose after the AIT member from the Office of Nuclear
Regulatory Research ran an independent thermal-hydraulic model of the replacement
steam generators and identified some inconsistencies between his model and the
licensee’s model. The Senior Reactor Inspector said the AIT Team Leader
subsequently instructed him to look at the 50.59 to determine what the FSAR included
about modeling thermal-hydraulic conditions and if methodologies had changed.
The Region II Senior Reactor Inspector told OIG that based on his review of the FSAR,
he concluded that the methodology used on the original steam generators for thermal-
hydraulic modeling was not described. He said that 10 CFR 50.59 specifies that if the
licensee departs from the methodology as described in the FSAR, then a license
amendment is needed. However, because the FSAR did not contain what was used for
the original steam generators, there was no basis to conclude a departure from
methodology had occurred. He said, “if the methodology is not in the FSAR, they didn’t
depart from it. So legally, by 50.59, they don’t meet that criteria.”
In addition, the Senior Reactor Inspector said they (i.e., AIT team members) looked at
other information in the FSAR pertaining to the steam generators that could have
required a license amendment based on 50.59. To do this, they looked at how the
FSAR described the steam generator design and its functions and compared it with the
new generators to assess how the change impacted the design functions or the
methods of performing and controlling the functions as described in the FSAR. He said
36
Page 47-----------------------
while they used Inspection Procedure 71111.17 as a guide for their review, they did not
implement it line by line as would be done for the triennial inspection. He said the AIT’s
determination was - based on review of the FSAR, the engineering change package
describing the new design, the 50.59 screen and evaluation, and other items -there
was no indication that the licensee needed a license amendment.
Office of Nuclear Reactor Regulation Project Manager
The Office of Nuclear Reactor Regulation (NRR) headquarters specialist to whom the
AIT Team Leader referred was an NRR Project Manager assigned as the 10 CFR 50.59
Program Manager since 2009. The NRR Project Manager told OIG he was asked by
the AIT Team Leader to conduct an independent review of the SONGS replacement
steam generator 50.59 to determine if the licensee made the correct determination with
respect to the need for prior NRC review and approval. He was not told how to conduct
the review and he was not an official member of the AIT. The Project Manager did not
work onsite at SONGS, but the AIT provided him the 50.59 screen and evaluation to
review relative to SONGS’ 10 CFR 50.59 review of the replacement steam generators
plus he had on-line access to the licensee’s documents. He began his review on or
about May 2, 2012, and provided a written response to the AIT Team Leader on June 7,
2012, documenting the results of his review for inclusion in the AIT Inspection Report.
The Project Manager said he reviewed the entire 50.59 screening, the entire 50.59
evaluation, and a “smart sampling” of associated reference documents. He did not
specifically follow IP 71111.17 in reviewing the SONGS 10 CFR 50.59 documentation,
but conducted his review based on his knowledge of NEI 96-07. He did not compare
every single statement against its design change package to confirm the accuracy. If
he had a question about a change to the facility as described in the FSAR, he would
look at it. Any questions he had with respect to the design, he would review the design
information that he had available to him. He said that reviewing the 50.59 entails
reviewing a sampling and based on his years of experience as an inspector, he said,
“you don’t expect 100 percent of everything, but you review it. . . and you dig deeper
into things that don’t sound right.” He said that all inspections are done by sampling.
He described the 50.59 document as largely a stand-alone document that reiterates
relevant information from the FSAR and what the design change said to ultimately
determine if it reached the criteria35 threshold for NRC review.
The Project Manager said that an unresolved issue (URI), with a couple of concern
areas, came from his review and are described in the AIT report, in particular, the
section “regarding methods of evaluation and a change from one computer program to
another computer program.” He said both of these areas, related to methods, were
captured by his URL
The Project Manager was involved in discussions and email exchanges with the AIT
followup team and manager concerning the dispositioning of his URI. Regarding his
first methodology concern area, in which the licensee replaced the computer code
35
The criteria are described in 10 CFR 50.59 and page 3 of this report.
37
Page 48-----------------------
ANSYS with ABAQUS, he said this was a methodology change and the regulation is
clear: If you use a new method, you have to justify it based on whether the NRC has
previously approved that method for the intended application. He communicated to
Region IV AIT followup team members that if ABAQUS had not been previously
approved by the NRC for the specific application, a license amendment would be
needed.
The Project Manager recalled this was an item the licensee objected to because they
said it was a revision to an existing method, and not a new methodology, whereas he
believed it was a new methodology. He was not sure what information the licensee
provided to the AIT followup inspectors, or to what extent the NRC inspectors looked for
where NRC had previously approved this method, but that the region closed the item
out. He said he reviewed it and went with the region’s judgment.
Regarding his second methodology concern, in which the licensee replaced the
computer codes of STRUDL and ANSYS with just using ANSYS for calculating a safety
event, the Project Manager said he closed this concern on a new technical
understanding of the calculation and not based on 50.59 guidance.
The Project Manager acknowledged the methodology changes discussed in Issue I of
this report (i.e., steam generator related methodology changes reflected in the UFSAR
that were not mentioned in the licensee’s 10 CFR 50.59 screen). He said he did not
notice this during his review (he checked the UFSAR to see that changes mentioned in
the evaluation were reflected, but did not do a reverse comparison to see if all the
changes to the UFSAR were reflected in the 50.59), but the licensee should have
included them in the description of changes. However, he said, “just because they
should have included that, and therefore they did not evaluate . . . does not mean it
would have resulted in the need for prior NRC approval. It just means they didn’t
completely document it.”
The Project Manager attributed the problems at SONGS to “design flaws” and said that
while the 10 CFR 50.59 process can pick up design type problems, it is not intended to.
The purpose of 10 CFR 50.59 is not to identify design deficiencies but to determine
whether prior NRC review and approval is required. He said design deficiencies should
be picked up through licensee quality assurance programs and potentially through NRC
oversight of licensees’ design control process through inspection, which is
accomplished through a sampling process.
AIT Followup Team Member
A Region IV Senior Project Engineer who was not on the initial AIT told OIG that he was
the inspector responsible for reviewing URI 10, “Evaluation of Departure of Method of
Evaluation for 10 CFR 50.59" for the AIT followup inspection and he prepared the
writeup for the followup inspection report. He said he consulted primarily with the NRR
Project Manager who identified the URI and the AIT Team Leader to reach a conclusion
and that each concurred with the result. With regard to the closure of UR1 10, the
38
Page 49-----------------------
Senior Project Engineer said that during the followup inspection, the licensee provided
some examples where ABAQUS was used for similar tube stress calculations. He said
that no one had performed the exact tube stress calculations that SONGS had
performed, so they looked for similar examples where a finite element model was used
to calculate tube stresses. He said, “We felt the similarity was the finite element model
because tubes are tubes and metal is metal and . . . people had used ABAQUS in
similar geometries to do similar-type calculations.” He said the examples were similar
enough to say that they had documentation that ABAQUS was an appropriate
environment for doing those kinds of calculations.
The Senior Project Engineer said the team did not go back and find evidence where the
NRC had written safety evaluation reports for each of these. He said all they needed
were examples of cases where ABAQUS had been used for applications similar to
reactor coolant system (RCS) structural integrity analyses. He said, “Our standard was
use of ABAQUS for applications similar to RCS structural analysis, not methods that
had been approved by the NRC for the intended application. The intended application
is tube stress analyses."
After discussion with OIG about the 10 CFR 50.59 provision concerning the need for
NRC approval for the “intended application” when changing from a method of evaluation
concerning methodology changes described in the FSAR to another method (10 CFR
50.59 (1)(2)(ii)), the Senior Project Engineer said the inspection plan was “probably
flawed" because they did not look for explicit approval by the NRC for the intended
application.
With regard to the second area of concern within this URI, the Senior Project Engineer
said that it was not a change in methodology because they went from using STRUDL
and ANSYS to using just ANSYS to analyze a limiting event. However, when OIG
discussed the use of a manual calculation to replace STRUDL, he said the regulatory
basis for closing this area was “probably not” adequate, based on the rule.
Review of Closeout Justification by Subject Matter Experts
The two Branch Chiefs who reviewed the SONGS 10 CFR 50.59 screening package at
OIG’s request also reviewed the AIT closeout justification; both felt NRC should not
have closed out the URI. Both stated that use of ABAQUS instead of ANSYS was a
new methodology and the licensee’s 50.59 documents do not discuss whether
ABAQUS has been approved by the NRC for the application of RCS stress analysis.
One of the Branch Chiefs told OIG that it appears the inspectors relied on the licensee’s
statement that the methodology that ABAQUS had been approved, but he could find no
evidence to that effect in the documents that were referenced in the inspection report.
The other Branch Chief told OIG that NEI 96-07 guidance states that licensees who cite
approved methodology from another licensee need to document their review of the
method, approved application, safety evaluation report, and related documentation and
verify that applicable terms, conditions, and limitations are met and to ensure the
method is applicable to their type of plant.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 50-----------------------
With regard to the second methodology for tube wall thinning analysis, one of the
Branch Chiefs stated the change from CEFLASH, STRUDL, and ANSYS to manual
calculations and ANSYS is clearly a new methodology. As just discussed, 10 CFR
50.59 requires that a change in methodology requires NRC approval for the intended
application.
Region IV Response to OIG Questions
In an August 14, 2014, memorandum to OIG, the Region IV Regional Administrator
provided the region’s analysis supporting the closure of UR1 10 in the AIT followup
report. The memorandum described that NEI 96-07, Section 4.3.8.2, “Guidance for
Changing from One Method of Evaluation to Another,” communicates two paths for
NRC approval. One path is where a vendor submits a topical report and NRC issues a
safety evaluation report documenting generic NRC approval for the use of a specific
analysis methodology by a given class of power plants. The second path consists of
NRC approval of a specific analysis for a given plant via a license amendment.
The Region IV memorandum acknowledges that the AIT followup inspection report lists
several documents that did not involve either of those two paths, but rather described
the use of ABAQUS in research activities previously completed for NRC. However,
following the list of research documents, the memorandum states that the AIT followup
inspection report notes that the inspectors reviewed “the Comanche Peak Updated
Safety Analysis Report, Section 3.6B.2.2.2 (“High-Energy Piping Other Than RCS Main
Loop’) that described using ABAQUS for piping dynamic responses resulting from a
postulated pipe rupture.” The memorandum notes that NRC approved the use of
ABAQUS when it granted licenses to operate the Comanche Peak Steam Electric
Station, Units 1 and 2, and that the use of ABAQUS is described in NUREG-0797,
Supplement 17, “Safety Evaluation Report Related to the Operation of Comanche Peak
Steam Electric Station, Units 1 and 2.”
The memorandum states that Region IV technical staff, in consultation with NRR staff
knowledgeable and experienced in 10 CFR Part 50.59 reviews and associated
inspection processes, determined that the AIT followup inspection report discussion of
Comanche Peak provided an acceptable example consistent with NRI 96-07, Section
4.3.8.2, where ABAQUS has been approved by NRC for the intended application.
Interview of NRC Office of the General Counsel Attorney
An NRC Staff Attorney from the Office of the General Counsel, Office of the Assistant
General Counsel for Operating Reactors, told OIG that at Region IV’s request, he
reviewed the draft URI closeout language, not for directing the staff to close or not to
close the URI, but instead for the clarity of the information: “how does this make sense,
can you write this better in plain language?” The Staff Attorney said he first became
involved with the replacement steam generators at SONGS in approximately May 2012
with the petition filing by Friends of the Earth.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 51-----------------------
The Staff Attorney explained he did not review the supporting examples SCE provided
to RIV regarding SONGS methodology change from ANSYS and ABAQUS; as such, he
could not determine if the examples cited by SCE either supported or did not support
the staffs findings. However, he said that NEI guidance 96-07 provides a variety of
examples of what constitutes methodology change approval by the NRC for an intended
application. He recalled that NRC approval could be in the form of specific approval
somewhere within a licensing document, such as a UFSAR, or approval could be of a
topical report where the NRC has generically approved that topical report in a safety
evaluation report. With respect to the Comanche Peak example provided by Region IV
for approval of ABAQUS, he concluded the staff approved an safety evaluation report
through the licensing process for Comanche Peak, and it was the staffs call to
determine the relevance of the Comanche Peak approval to the application of ABAQUS
at SONGS.
The Staff Attorney advised in the end, Region IV cited a recordkeeping violation and
that was within the staffs legal authority to do. He related from a technical standpoint,
he does not have the technical expertise and would rely that the “tech review” got it
right, and from a legal standpoint Region IV could justifiably close this matter as a minor
violation. Based on the fact that Region IV closed this URI, he would infer they had
adequate support to do so.
Interviews of Region IV Managers
SONGS Special Project Branch Chief
A Region IV Branch Chief with oversight responsibility for SONGS since 2009 who also
served as Branch Chief of the SONGS Special Project Branch told OIG, in hindsight, if
NRC were to go through the 50.59 questions now, some of those screening questions
would have to be answered yes and would require NRC approval. But back then [2007-
2009], NRC did not know and SONGS did not know what NRC knows now [regarding the
FIT-III thermal-hydraulic model], and an [NRC Office of Investigations] investigation was
ongoing [as of February 2013] to determine what information was available. He advised
all indicators are that SONGS was not aware of the failure of the code error (used with the
FIT-III) and SONGS could not have predicted it.
Former Deputy Regional Administrator
The former Deputy Regional Administrator told OIG that the AIT reported to the Region
IV Regional Administrator. The former Deputy Regional Administrator was not directly
involved in the AIT report and he became involved in approximately September 2012,
after the AIT held its exit conference with the licensee, but before the AIT (followup)
report was issued [November 2012]. During the exit conference, Region IV learned that
SONGS had identified an error in a software program pertaining to the void fraction.
According to the former Deputy Regional Administrator, SONGS’ 50.59 did not meet the
criteria 50.59 (2)(c)(ii) because design changes resulted “in more than a minimal
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 52-----------------------
increase in the likelihood of occurrence of a malfunction of a structure, system or
component important to safety previously evaluated in the Final Safety Analysis Report
as updated.” This conclusion is based on a combination of factors, the adverse
thermal-hydraulic conditions coupled with inadequate upper tube bundle support, which
caused the steam generator failure. SONGS did not believe they violated this criteria
even though SONGS was aware that certain calculations pertaining to stability ratios
flow velocity looked wrong. The former Deputy Regional Administrator said that
SONGS exercised engineering judgment and decided that the values were acceptable.
He noted that there is no requirement that defines the value of the void fraction.
The former Deputy Regional Administrator added that the AIT follow-up report
addressed the flow velocity and thermal-hydraulic conditions as unresolved items, which
were to be dispositioned through the escalated enforcement process. While the AIT
report addresses these issues, the report did not reach a conclusion on the 50.59
violation even though he and some staff members concluded that the licensee violated
50.59 (2)(c)(ii). He noted that this issue was not pursued because of the 2.206 petition
submitted to the NRC Petition Review Board by the Friends of the Earth. The petition
posed the very question regarding whether the licensee violated 50.59 and the Petition
Review Board has the responsibility to make a recommendation to the NRR Director
who can then reach a decision on the issue.
The former Deputy Regional Administrator said that while he and some of the staff
members involved in the AIT believed that SONGS violated 50.59 (2)(c)(ii) during its’
implementation of the 50.59 process, other staff members did not share this view.
Some staff felt the criteria was violated but that the licensee did not have control over
this, therefore, the licensee should get discretion because of what they knew or did not
know at the time. According to the former Deputy Regional Administrator, the staff’s
focus and emphasis was not so much on whether one could point to a single criteria,
but whether or not it (the violation) was fair and reasonable and met the standards for
discretion.
As noted under Issue 1, the former Deputy Regional Administrator said in hindsight, he
believes that SONGS should have requested a license amendment from NRC prior to
making the change. He also believes the steam generator design was fundamentally
flawed and would not have been approved as designed.
The former Deputy Regional Administrator also stated that, ultimately, Region IV could
not reach a conclusion on whether or not the licensee violated 50.59 because the Office
of Investigation had an open investigation into whether the licensee had willfully violated
50.59, and because of the 2.206 petition. Region IV could not “get out in front” of the
agency. The region decided that the Petition Review Board had the responsibility to
disposition the matter. He said that in the end, the AIT identified a design control
violation.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 53-----------------------
Former Region IV Regional Administrator
The former Region IV Regional Administrator told OIG that because the event
surrounding the SONGS steam generator tube degradation was unprecedented in the
industry, Region IV immediately established an AIT. Also, there was Congressional
criticism that SONGS should have obtained a license amendment to replace the steam
generators and so it was important to review the issue as early as possible so that
Region IV could draw a conclusion as to whether there was something amiss with
respect to the 50.59 process and how the licensee had implemented the steam
generator design change. His intent was for the AIT to conduct a thorough review and
reach solid conclusions. He recalled being briefed by the AIT after the team completed
its inspection and he concurred on the AIT report.
According to the former Regional Administrator, the AIT conducted a thorough review,
identified significant issues, and reached defensible conclusions. The AIT had a large
area to cover and looked at a number of items, including fabrication, transportation, and
installation of the steam generators. The AIT may have conducted a focused 50.59
because NRC rarely, if ever, conducts a 100-percent review. Consequently, the AIT
could have very easily selected items they viewed as important aspects to review to
determine if the licensee reached the right conclusion. In his view, the AIT focused on
design issues and the 50.59 process is not intended to identify design issues.
The former Regional Administrator was not aware that the AIT did not review all the
methodology changes in the SONGS 50.59 steam generator replacement process. He
was aware that the AIT has raised methodology-type questions because some of it
surfaced in the AIT report and that the team considered methodology changes.
However, if the region missed some of the methodology changes it was because
inspections have always been no more than a sampling. If the AIT did not focus on the
methodology changes, it was because there was something about their approach, plan,
and scope of review, coupled with guidance they received that took them in the direction
they took. It was not malevolent or intent, but a part of the mentality and process of how
inspectors pick their sample and selection process.
Current Region IV Deputy Regional Administrator
The current Region IV Deputy Regional Administrator said that based on what the NRC
looked at during its inspections, the agency made definitive statements that the licensee
did not require a license amendment. However, he acknowledged there could be other
aspects of the 50.59 that were done incorrectly that would require or would have
required a license amendment. He said that without additional inspection, NRC does
not know the answer. He said to make a definitive decision on whether a license
amendment request was required, the agency would have to talk about the resources
needed to accomplish that. He said, “It comes down to a prudent use of resources to
go back and accomplish that.”
The Deputy Regional Administrator stated with SONGS it was an error with code, a
design issue -like an error carried forward-type of issue. The issue was a model error
43
Page 54-----------------------
that never got caught. He does not believe the licensee’s 50.59 evaluation would have
picked it up. He also doubts the NRC license amendment review process would have
picked this up. If the licensee explains the model used and reports the value reached,
NRC may accept the output. The NRC review process is not going to identify an error
that the licensee made in either the modeling or the inputs to the modeling or the
assumptions that went into the modeling.
Interviews of Headquarters Managers
Former NRR Director
The former NRR Director said he was not familiar with the AIT inspection and whether a
member of his staff assisted the AIT (the NRR project manager). He was not aware of
the focus of the AIT or direction given to the AIT. He said it appeared that the project
manager also sampled and identified two issues for AIT follow up, one of which was a
minor violation.
The former NRR Director noted that whether the NRC should have conducted a much
more thorough review because of external stakeholder interest is a different question
and should have been a management decision. However, from a safety significant
standpoint, the answer would have been that it did not warrant a more thorough review.
He noted that there was no release offsite that was consequential to anyone.
Interview of NRR Acting Director
The Acting NRR Director told OIG the region does not have latitude to deviate from the
50.59 rule [similar versus intended use]. But, in this case, it is NRC’s job to have the
licensee provide the justification, as the licensee is responsible for doing the 50.59
evaluation.
The Acting NRR Director informed OIG that the examples cited by the licensee in the
AIT followup report (as described in footnote 32), which included three NUREG
contractor reports, are reports of research that have been done for the NRC Office of
Nuclear Regulatory Research and do not constitute NRC approval. They were
examples of how the NRC used the code for a vessel head, and dissimilar welds, which
are good uses of ABAQUS, but they are not steam generator tube bundle interactions
with tube support plates. His inference in reading the next paragraph of the AIT
followup report was that inspectors looked at Comanche Peak’s FSAR and recognized
SCE’s examples were not NRC approvals and they looked for another analog and
presented the Comanche Peak analysis of ABAQUS. He stated it is debatable if
Region IV found the best application to cite in closing the open URI 10 issue.
With specific regard to Region IV’s closeout of URI No. 10 and whether the basis used
was consistent with 10 CFR 50.59, the Acting NRR Director advised there could
probably be a diversity of views from the staff on this because ABAQUS and ANSYS
are both generic, widely-used finite element analysis codes. He was not certain he
agrees with the region’s position that the use of ABAQUS was a change of
44
Page 55-----------------------
methodology; his read of the 50.59 rule and that understood by his staff members is that
the change to use ABAQUS constitutes a change in element of methodology.
The Acting NRR Director explained that if a modification change is reflected in the post¬
modification FSAR, there should be somewhere in the paper trail whether a screen out
or in, or an evaluation was conducted. There should be some indication of how the
change was assessed against 50.59 criteria, and the results of that assessment
indicating if the change met or did not meet the 50.59 criteria. He stated it is possible
the items that were not addressed by either the 2009 or AIT inspections were not
reviewed, or not selected as a sample inspection. Based on the gaps that may exist
regarding items that may require screening and/or evaluation, he could not make an
overall determination that a license amendment was not required for the replacement
steam generators at SONGS. Likewise, he could not conclude that an amendment was
required. He said there are reasonable questions that should be evaluated as part of
50.59 and we do not know if the questions were addressed. Until those items,
evaluations, are completed, he cannot conclude either way if an amendment was
required. He can conclude for the evaluations inspected, “generally yes,” an
amendment was not required.
The Acting NRR Director did not believe the NRC license amendment review process
would have caught the problem at SONGS, especially given what was found regarding
the fluid elastic instability in the secondary side of the steam generator. He said the
trigger to require a license amendment is intended to focus on safety and not
necessarily to avoid a bad investment by the licensee. The safety outcomes of this
event were well within the licensing basis of the facility and were a low-consequence
event in terms of public exposure impacts. The result was not a significant safety
impact. In the end, SONGS had a tube leak that was well within the design basis
accident analysis. And it was not clear to him that the analysis that supported the
design would have driven the NRC to a conclusion that would have identified and
anticipated the wear identified in the secondary side.
The Acting NRR Director said that industry has replaced the vast majority of steam
generators on pressurized water reactors under 50.59 from the late 80’s and early 90’s
forward, and up until SONGS, these replacements have occurred without issue.
Deputy Executive Director for Reactor and Preparedness Programs
The Deputy Executive Director told OIG that the complex issues with the SONGS
replacement steam generators are problems the NRC has not previously
witnessed. Had SCE submitted an amendment for review, the NRC would have
touched those issues that resulted in the flaws in that generator (which experienced the
tube leak). He further explained that a 50.59 review (inspection) is not a safety review
and that the design problem should have been found by the licensee. Hypothetically,
he stated the NRC could have potentially found the problem in the licensing review
process; but the 50.59 would not be the process to find that. The Deputy Executive
Director told OIG the 50.59 process is utilized to determine if the NRC is required to
45
Page 56-----------------------
review a change proposed by the licensee or if the NRC relies on the licensee’s review
of the change.
With regard to the closeout of UR1 10 concerning the use of ABAQUS, the Deputy
Executive Director stated it is ultimately the region’s responsibility to close out the
URI. The inspector determined the NRC had approved ABAQUS for the reactor coolant
system structural integrity analysis and without more information he could not provide
an answer regarding the adequacy of the region’s closure.
Findings
OIG found that although Region IV AIT, established to assess the circumstances
surrounding the tube leak and unexpected wear of tubes in the Unit 3 steam generators,
included a review of the SONGS 50.59 steam generator package to determine whether
SONGS needed a license amendment prior to installing the new steam generators, the
AIT did not document an answer to this question. In its initial July 18, 2012, inspection
report, the AIT communicated that the NRR Project Manager assigned to perform the
review identified one unresolved item (URI number 10, “change of methodologies
associated with 10 CFR 50.59 review”) for which additional information was needed to
determine if performance deficiencies exist or if the issues constituted violations of NRC
requirements. The URI described two instances that failed to adequately address
whether the change involved a departure of the method of evaluation described in the
UFSAR. Although NRC’s November 9, 2012, AIT followup report documented the
closure of this URI, and stated that neither change would have required a license
amendment, it did not answer the overall question of whether a license amendment was
required.
The AIT Team Leader and the current Region IV Deputy Regional Administrator told
OIG that based on what NRC reviewed during its inspections, the conclusion was that a
license amendment was not needed, although each allowed that the sampling approach
used to perform this assessment could have missed something. The Acting NRR
Director said he could not determine if an amendment was needed or not due to the
gaps that may exist regarding items that may require screening and/or evaluation. The
current Region IV Deputy Regional Administrator said additional inspection would be
required to answer whether a license amendment was required, and questioned
whether it would be a prudent use of resources to go back and accomplish that. The
former Region IV Deputy Regional Administrator said that in hindsight, he believes that
SONGS should have requested a license amendment from NRC prior to making the
change. He also believes the steam generator design was fundamentally flawed and
would not have been approved as designed. He said the AIT discussed a potential
50.59 criteria violation because of the design issues; however, the AIT ultimately
identified a design control violation.
OIG found that NRC’s justification for closing out URI number 10 does not align with
specific language in 10 CFR 50.59 concerning NRC approval for a change in
methodology, but was based instead on Region IV’s interpretation (in consultation with
46
Page 57-----------------------
NRR) of the rule. 10 CFR 50.59 (a)(2)(H) reflects that changes from a method described
in the UFSAR to another method are permissible without NRC preapproval if that
method has already been approved by the NRC for the “intended” application. In
closing out the URI, however, the AIT followup report determined the change of
methods would not have required a license amendment based on NRC’s approval for
the use of the method at other nuclear power plants in “similar” applications. OIG notes
that while the AIT characterized the issue as a change in methodology, it justified
closing the matter based on approval for a “similar” application rather than the
“intended” application as stated by the rule.
OIG also notes that while the AIT inspection report identified a URI pertaining to the
SONGS 10 CFR 50.59 screen and evaluation package, the NRR technical specialist
who reviewed this used a sampling approach and did not identify many of the
shortcomings described under issue 1 of this report.
ISSUE 3: NRC Oversight of SONGS UFSAR
Background
10 CFR Part 50, “Domestic Licensing of Production and Utilization Facilities,” Section
50.34, “Contents of Applications; Technical Information,” contains requirements for the
content of applications for construction permits and operating licenses for nuclear power
reactors. Per the regulation, an application for a construction permit must include a
preliminary safety analysis report (SAR), and an application for an operating license
must include a final SAR (FSAR). Section 50.34 states that the FSAR is to include
information that describes the facility, presents the design bases and the limits on its
operations, and presents a safety analysis of the structures, systems, and components,
and of the facility as a whole. The FSAR and the plant’s license and associated
technical specifications are the principal regulatory documents describing how the plant
is designed, constructed, and operated. The FSAR is a key reference document used
by NRC inspectors during plant construction and operation.
In 1980, NRC issued the FSAR update rule, 10 CFR 50.71(e), to ensure that licensees
maintain the information in the UFSAR to reflect the current status of the facility and
address new issues as they arise, so that the UFSAR can be used as a reference
document in safety analyses. Information to be maintained includes all changes made
in the facility or procedures as described in the FSAR; all safety analyses and
evaluations performed by the licensee either in support of approved license
amendments or in support of conclusions that changes did not require a license
amendment. 10 CFR 50.71(e) specifies the type of information that must be submitted
and states that licensees must submit a UFSAR update annually or within 6 months
after each refueling outage provided that the interval between successive updates does
not exceed 24 months.36
36AS noted in the background and chronology section of this report, licensees have similar reporting requirement
(every 24 months) for activities implemented under 10 CFR 50.59.
47
Page 58-----------------------
Inspections in the 1996-1997 timeframe by NRC and licensees identified numerous
discrepancies between updated FSAR information and the actual plant design and
operation, and these findings raised questions about possible noncompliance with
10 CFR 50.71 (e). In June 1999, NEI issued NEI 98-03, “Guidelines for Updating Final
Safety Analysis Reports,” to provide licensees guidance for updating FSARs consistent
with the requirements of 10 CFR 50.71 (e). In September 1999, NRC published
Regulatory Guide 1.18137 “Content of the Updated Final Safety Analysis Report in
Accordance with 10 CFR 50.71(e),” which endorsed NEI 98-03.
During this same period of time, the NRC promulgated the revised 10 CFR 50.59 rule.
Consistent with the requirements of 10 CFR 50.59(d)(2), the licensee is required to
submit a summary description of any changes to the facility as described in the FSAR
that require an evaluation using the criteria defined in 10 CFR 50.59(c)(2). This report
is required to be submitted to NRC on a not-to-exceed 24 month frequency, and is a
standalone report and different from the UFSAR reporting provided to the NRC under
10 CFR 50.71(e).
NRC Oversight Responsibility for UFSAR and 59.59 Report Reviews
NRC does not have a specific ROP inspection for assessing whether licensees are
appropriately maintaining UFSARs; however, the agency has two primary means for
providing oversight of UFSARs. First, embedded in certain ROP inspections are
requirements to verify that licensing basis documents are updated appropriately. For
example, 1 7, Evaluations of Changes, Tests, and Experiments and
IP 711 1.1
Permanent Plant Modifications, directs inspectors to verify that license basis
documentation have been updated accordingly and are still consistent with the new
design. An example identified was the UFSAR. Another example includes IP71111.18,
Plant Modifications, which directs inspectors to verify that design and licensing
documents have either been updated or are in the process of being updated to reflect
the modifications. Examples of design documents which could be affected by
modifications include the UFSAR.
The second primary means for providing oversight of UFSARs is provided by NRR
project managers,33 who are required per the NRR Division of Operating Reactor
Licensing (DORL) Handbook39 to perform a review of UFSARs within 90 days of receipt
37 NRC Regulatory Guides are issued to describe and make publicly available such information as methods
acceptable to the NRC staff for implementing specific parts of NRC’s regulations, techniques used by the staff in
evaluating specific problems or postulated accidents, and data needed by the NRC staff in its review of applications
for permits and licenses.
38 NRR Office Instruction LIC-100, Revision 1, states that project managers generally coordinate NRR staff efforts for
an assigned facility, a generic issue, or a policy issue to ensure that the outputs are complete, accurate, and timely.
Project managers also serve as the point of contact with licenses for assigned facilities and they are generally
responsible for managing the licensing agenda for assigned facilities and resolving issues about licensing bases for
assigned facilities. The NRR Division of Operating Reactor Licensing (DORL) Handbook, states that the PM serves
as the licensing authority regarding the maintenance and amendment of a site's licensing and design basis.
39 The DORL Handbook is an online NRC resource intended to provide easy access to NRC regulations, procedures
and documents in order to help NRR Project Managers, Licensing Assistants, and others complete various licensing
actions and tasks.
48
Page 59-----------------------
to, among other tasks, check that the UFSAR revisions includes the following, as
applicable:
• New regulatory requirements.
• Changes to the facility or procedures. (Per NEI 98-03, 40this includes changes
implemented under 10 CFR 50.59.)
• Analysis of new safety issues.
• Long-term temporary modifications.
• Discrepancies between the facility and the UFSAR.
These expectations are listed in guidance titled “UFSAR Changes - 10 CFR 50.71 (e),”
located within the “Design Bases” section of the DORL Handbook. The guidance
(under project manager “actions”) also states that the results of the project manager’s
review are to be documented by a memorandum to file and that if there are any
significant findings, the project manager should consult with regional representatives for
possibly including findings in an inspection report. The guidance also includes an
“additional guidance” section, which states that any exceptions to the 90-day review
requirement should be negotiated with the respective project director and estimates that
the project manager’s review of UFSAR changes should take approximately 8 hours.
The “additional guidance” section also states that the review should “determine that, for
those FSAR update changes that the project manager is familiar with, these changes
are appropriately addressed by licensing actions (changes to the facility or procedures
previously described in the FSAR), 10 CFR 50.59 submittals, or regional inspection
activities.” It also states, “a representative sample should be chosen to perform this
review.”
OIG noted that a different DORL Handbook section, titled “50.59 Evaluations,”
specifically describes the project manager’s role in reviewing updated FSARs for
changes subsequent to 50.59 evaluations during the project manager’s review of
updated FSARs. The guidance states:
Under the provisions of 10 CFR 50.59, licensees are permitted to
make changes to the facility and procedures, as described in the
safety analysis report (SAR), and to conduct tests or experiments
not described in the SAR, without prior NRC approval, provided a
change to the technical specifications is not involved or the
proposed change, test or experiment does not meet the criteria of
10 CFR 50.59(c)(2). Licensees must maintain records of such
changes, supported by a safety evaluation which provides the basis
for determining that the change, test, or experiment does not meet
the criteria of 10 CFR 50.59(c)(2) and report such changes to the
NRC in accordance with 10 CFR 50.59(d)(2). A license amendment
must be prepared for changes that meet the requirements of 10
CFR 50.59(c)(2).
40 Revision 1 of NEI 98-03, “Guidelines for Updating Final Safety Analysis Reports”, dated June 1999, provides
methods that are acceptable to the NRC staff for complying with the provisions of 10 CFR 50.71(e).
49
Page 60-----------------------
Under the current Reactor Oversight Process (ROP), Region-based
inspectors have been given the responsibility to perform an
inspection module/procedure on facility changes made by licensees
without prior NRC approval in accordance with 10 CFR 50.59.
Consequently, DORL Project Managers do not directly review the
50.59 screens or evaluations. However, DORL Project Managers
are responsible for the review of those changes to the UFSAR not
requiring prior NRC approval as part of the project manager
review done of the summary reports provided under 10 CFR
50.71(e).
OIG Review of NRR Oversight of SONGS UFSAR and 50.59 Submittals
OIG compared the number of UFSARs submitted by SONGS to NRC between 2001
and 2011 with the number of documented NRR reviews of those submittals and learned
that while SONGS submitted six updates during this timeframe, NRC could provide
documentation to support two NRR reviews of the six submittals. One review,
documented in a memorandum dated December 31, 2012, reflected that an NRC
Project Manager reviewed a June 10, 2011,41 updated FSAR submittal from SONGS
and found the submittal met 10 CFR 50.71(e) requirements. The memorandum stated
that the review encompassed a sample of potential changes to the UFSAR as a result
of license amendments to ensure the revised UFSAR reflected those changes. The
memorandum had no reference to the other tasking areas (for example, the 50.59
changes or new regulatory requirements) as being reviewed by the Project Manager.
The other memorandum, dated November 15, 2001, documented a limited review of
five licensing actions completed during the period of time covered by the licensee’s
UFSAR revision. The Project Manager who conducted the review verified that two of
the five licensing actions affected the information in the UFSAR and the changes were
reflected in the UFSAR. Similarly, this memorandum had no reference to the other
tasking areas (for example, the 50.59 changes or new regulatory requirements) as
being reviewed by the project manager.
OIG Review of NRR Oversight of Other Updated UFSAR Submittals
OIG determined between April 2010 and March 2014, two NRR branches received
updates for 21 nuclear power plants. Of the 21 submittals, 5 reports were reviewed by
NRR within the 90-day timeframe specified in the DORL Handbook, 7 reports were
reviewed between 90 days and a year after receipt, and 9 reports were reviewed more
than a year after receipt.
41 OIG learned that although SONGS turned in its April 2009 update on June 10, 2009, the NRC Document Control
Desk rejected the submittal and SONGS allegedly did not resubmit an acceptable format for ADAMS. A note in
NRR’s files, dated February 15, 2011, stated the Project Manager would wait until the April 2011 update to perform a
UFSAR review.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 61-----------------------
Additionally, OIG noted that 2 of the 21 nuclear power plants that were reviewed by
NRR had a reference to 50.59 evaluations, whereas 3 of the 21 specifically stated that
“no review” of 50.59 was done and 16 of the 21 had no discussion of 50.59.
OIG Interviews of NRR Staff
The NRR Project Manager who wrote the December 2012 memorandum documenting
review of the SONGS June 10, 2011, UFSAR, told OIG that he was tasked in the
October/November 2012 timeframe, by an NRR Branch Chief to conduct the review
because it had not been reviewed for a while. The Project Manager had recently been
assigned to SONGS and had no previous experience with the station. To conduct the
review, the Project Manager pulled all license amendments from 2002 forward, took a
random sampling, and compared them to the FSAR. He recalled only one discrepancy
between the facility and the FSAR, which pertained to power level. The project
manager did not compare the 50.59 change report to the FSAR and said that as a
project manager he would not typically do a 50.59 review. He thought that another NRR
group, the Generic Communications Branch, reviewed changes under 50.59. The
Project Manager said he did not question the multiple methodology changes that OIG
identified in Section 3.9 of the 2011 updated FSAR as discussed under Issue I of this
report. He said, “I didn't specifically go page by page looking for change bars and then
work backwards. I worked from the license amendments and moved into the FSAR for
review. That was really where my focus was.”
The Project Manager said that the UFSAR reviews by project managers are a low
priority and he was not sure if they could be given a higher priority because project
managers have a lot of work already. With regard to the 10-year period in which no
review was documented by NRR of SONGS’ UFSARs, the Project Manager said that a
lot of changes occur over such a timespan and a reviewer will likely miss some of them
because “the change bars roll off from one year or one change to the next."
A different NRR Project Manager who was responsible for SONGS in 2009 told OIG
there is a general expectation that the project managers review updated FSARs upon
receipt. However, he believed there were inconsistent practices within NRR as to the
level of review. He believed project managers who completed reviews documented
them in a memorandum to file. Although he remembered receiving the 2009 and 2011
FSARs from SONGS, he did not remember doing a review for either one. He was not
aware if the SONGS Project Manager before him reviewed or documented a review.
He said that even if there is some guidance to project managers on these reviews, it
was his impression from talking to peers, that the UFSAR reviews were not consistently
performed. He was not sure if this was a conscious decision by project managers or if
the requirement was not reinforced by management. He thought the review was
considered a low-value activity.
As far as the 50.59 summary document submitted by the licensee, this Project Manager
described the reports as being brief and containing a summary of the licensee’s basis
for determining that the change can be made under 50.59. He was not aware nor had
51
Page 62-----------------------
he done a formal review of those 50.59 documents. He believed they were used as a
vehicle to communicate with regional inspectors to identify any areas of concern. He
said project managers used to have just one site to monitor, but since the 1990s, project
managers have multiple sites. With one site, he said, the project managers had more
opportunity to better understand the licensing basis for a plant and to assist with onsite
inspections of the licensee’s 50.59 activities. Today, regional inspectors implement
50.59 inspections in the course of their “core" inspections and NRR project managers
are not involved in the process.
An NRR Branch Chief involved with SONGS learned from a Project Manager that the
SONGS UFSAR review had not been performed for more than a year. The previous
NRR Branch Chief tasked the Project Manager to conduct the review of the UFSAR
submitted by SONGS in 2011. He did not think the 10 CFR 50.59(d)(2) submittal had
been included in the project manager review. He stated the Project Manager had done
a “comprehensive” review of the document sometime in the June through November
2012 timeframe. He stated the Project Manager identified all the changes, including
going back to previous changes not reviewed, and documented his review. It was his
understanding that the only new information in the steam generator section of the 2011
UFSAR was related to the technical specifications. Additionally, he discussed the
process that is generally followed. He described that headquarters program managers
review the UFSAR submittals by the licensees to understand the changes made. He
said they review the 10 CFR 50.59(d)(2) submittals to make sure it makes sense and a
license amendment was not required. If necessary, the headquarters project managers
contact the region and pursue a more detailed evaluation.
OIG interviews of NRC Senior Managers
Former Region IV Deputy Regional Administrator
The former Deputy Regional Administrator told OIG that during the AIT, Region IV staff
reviewed the original SAR and noted that the licensee had made many changes to the
steam generators over a 25-year period, which were no longer reflected in the UFSAR
or consistent with the original SAR. When a licensee then goes to replace the steam
generators, they are then comparing to whatever existed just before the replacement.
All the changes have already occurred and were never updated [in UFSAR].
Current Region IV Deputy Regional Administrator
The current Region IV Deputy Regional Administrator, stated that his general
expectation is that all material submitted to the NRC from a licensee would be reviewed
by headquarters staff. He said region staff does not review the UFSAR except during
periodic 50.59 inspections. He said the UFSAR is important to regional inspectors
during inspections because it describes the licensing and design basis of nuclear power
plants.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 63-----------------------
Former NRR Director
The former NRR Director advised OIG that he expected project managers to review
revisions the UFSAR submitted by licensees under the 10 CFR 50.71(e) process and
verify that changes made by licensees through various processes such as the 50.59
evaluations, license amendments, and licensee commitments, have generally been
updated into the UFSAR. The project manager should document the review, which
usually should take anywhere from 60 to 90 days.
Regarding OIG’s observation that the SONGS FSAR was reviewed twice during an 11-
year period and that review of the 2011 submittal was completed in December 2012, the
former NRR Director acknowledged that this was unacceptable to go for such a long
period of time without FSARs being reviewed.
Regarding the NRR staffs review of FSARs submitted by other nuclear power plants,
the former NRR Director said NRC should either change the requirement to complete
such reviews within 90 days or do a better job to accomplish the requirement. The
FSAR review is a self-imposed requirement and if the agency was not meeting its own
internal guidance, then the agency should change the guidance and consider what
really makes sense based on safety significance. He noted that there was nothing
significant about completing the review in 90 days. He would need to review the safety
benefit of doing the FSAR reviews and whether there was a better periodicity for doing
these reviews. The former NRR Director said that the staff should review the issue and
decide whether it makes sense to conduct the reviews on a yearly basis or every 2
years and develop a mechanism for tracking such reviews.
The former NRR Director said that the project manager’s review of the UFSAR is an
administrative task. Project managers should know all the license amendments that
have been submitted over the past 18 months and ask themselves if the amendments
affected the FSAR and if so, confirm that these changes are reflected in the FSAR. He
noted that this review is a bookkeeping exercise and it is not a technical exercise. The
former NRR Director said the technical review was completed when the staff reviewed
the license amendment request. Regarding project manager review of the 50.59 (d)(2)
licensee submittals, the former NRR Director told OIG that the review is also an
administrative review and more or less based on the experience of the project manager.
The project manager should question whether the 50.59 changes made sense, whether
there was any question as to why the change was made under the 50.59 process, and
why the change did not trigger a license amendment request. He noted the project
manager may question whether there was a need to follow up on an issue with a
resident inspector or from an inspection standpoint. Such questions are kind of a
judgment call made by project managers when they review their list of 50.59 changes.
Examining whether a change should have screened in for evaluation would require a
more in-depth review to look at what was done behind the 50.59 (criteria i-viii).
53
Page 64-----------------------
Deputy Executive Director for Reactor and Preparedness Programs
The Deputy Executive Director told OIG that NRC oversight of 10 CFR 50.71(e) is
critical since it enables the NRC to know whether or not a plant is in compliance with its
licensing basis. He added the purpose of 10 CFR 50.71(e) is to ensure that changes
either approved by NRC or made by licensees made under 10 CFR 50.59 process are
captured in the UFSAR.
He considers it a priority to review the UFSAR. Because it is not a baseline inspection
in ROP, headquarters project managers are tasked with reviewing the 50.71(e) reports
submitted by licensees. If a project manager finds an issue, the process is for the
project manager to contact the branch chief and get inspection follow-up on this issue.
The Deputy Executive Director thought that project managers were checking the
10 CFR 50.59(d)(2) reports to assess whether 50.59 changes are in the UFSAR.
Findings
OIG found that NRC does not consistently use one of its primary oversight methods to
assess whether licensees are keeping their power plant licensing basis documentation
up to date as required by 10 CFR 50.71(e). Although licensees are required, per 10
CFR 50.71(e), to biannually submit UFSAR updates reflecting the current status of the
facility so that the document can be used as a reference document in safety analysis,
the NRR project managers tasked to review these submittals do not always conduct the
reviews within the required 90-day timeframe. Moreover, although licensees also must
biannually submit, per 10 CFR 50.59(d)(2), information concerning changes made
under 10 CFR 50.59 without NRC prior approval, NRR project managers -who are
instructed to consider this information during their review of 10 CFR 50.71(e) submittals
-do not always take the 10 CFR 50.59(d)(2) information into consideration during their
reviews. OIG found that while NRC expects a plant’s UFSAR to accurately reflect a
plant’s licensing basis, the former Region IV Deputy Regional Administrator said that
during the SONGS AIT, Region IV staff noted the licensee had made many changes to
the steam generators over a 25-year period that were not reflected in the UFSAR or
consistent with the original SAR.
OIG reviewed documentation of project manager reviews in two NRR branches and
found project managers reviewed only 5 of the 21 most recently received licensee
UFSAR submittals within the 90-day timeframe, while 7 were reviewed between 90 days
and a year after receipt, and 9 reports more than a year after receipt. Moreover, only
two of the project manager reviews contained a reference to review of 10 CFR 50.59
documentation submitted by licensees even though project manager guidance directs
that this occurs. OIG also found that over a 10-year period, NRC staff documented two
reviews of changes to SONGS’ UFSAR, although the licensee submitted six UFSAR
updates during this period as required, and neither NRC review mentioned
consideration of 10 CFR 50.59 changes.
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
Page 65-----------------------
Although senior NRC managers expect the project managers to conduct the reviews
within the required timeframe, and to consider changes made under 10 CFR 50.9 as
part of that review, two NRR project managers interviewed said the reviews are
considered a low priority. Neither of the project managers included the 10 CFR 50.59
information in their reviews of 50.71(e) submittals; one thought this review was
conducted by a different NRR group and the other thought the 10 CFR 50.59
information was used by regional inspectors for a different purpose.
In contrast, the Deputy Executive Director for Reactor Preparedness Programs
considers NRC’s oversight of 10 CFR 50.71(e) to be critical for enabling NRC to know
whether a plant is in compliance with its licensing basis, and considers the project
manager review of 50.71(e) submittals to be a priority. While the former NRR Director
also expected project managers to conduct the required reviews to assess whether
changes made by the licensees have generally been updated into the FSAR, he viewed
the project manager’s review as a bookkeeping exercise that is based on the
experience of the project manager. He noted that the FSAR review is a self-imposed
requirement and if NRC is not meeting its own internal guidance, then it should either
meet the requirement or change the guidance based on safety significance.
55
Page 66-----------------------
Page 67-----------------------