ML20116M889
| ML20116M889 | |
| Person / Time | |
|---|---|
| Issue date: | 09/30/1990 |
| From: | Office of Nuclear Reactor Regulation |
| To: | |
| References | |
| NUREG-BR-0125, NUREG-BR-0125-V02-N2, NUREG-BR-125, NUREG-BR-125-V2-N2, NUDOCS 9608210146 | |
| Download: ML20116M889 (6) | |
Text
. _ _ _. _ _
1
)
i l/
DFFICE hp 1 CLE AR RE ACTOR REGULATION l
U.S. NUCLE AR REGULATORY COMMISSION 4
E M
E I
i E
M E
E i
D k
M M
N B
N m
w i
TECHNICAL
!=!
nuase/sa.0,2s NEWSLETTER 5i seJAMM Using Lessons from Opera $ng; Experience l
For Future LWRs i
j by Thomas E. Murley l
The technical regulations governing safety design features of Severe accident policy statement currently operating reactors are requirements that have grown l
over the past three decades. Many of the procedural aspects
% addition, the designers are reviewing all generic letters, bulle-1 of reactor licensing were set forth in 10 CFR Part 50 in 1956, t,ns, information notices, and circulars to show that the new designs i
but the first real technical rule was Part 100," Reactor Site deal with the issues in those NRC issuances. Earlier this year, the l
Criteria," issued in 1%2. In 1971, the AEC published the staff received policy guidance from the Commission on 15 issues,
}
General Design Criteria. As safety reviews, research results, mostly severe accident topics, where the staff was proposing to l
and operating experience accumulated over the years, the depart from current regulations. Other areas where requirements NRC dealt with new safety issues by promulgating new regu-remain to be specified are the technical specifications and the j
lations. In recent years,the more significant regulations were reliability assurance plan that is yet to be submitted. With these the ATWS Rule in 1984, the Pressurized Thermal Shock Rule requirements implemented, we will have applied the experience of i
in 1985, and the Station Blackout Rule in 1988.
1450 reactor-years' operation toward making the next generation l
of reactors safer than current plants.
4 At the time of the TMI-2 accident in March 1979, the nuclear I
industry in this country had about 450 reactor-years of operat-We are confident that the designs for the large evolutionaryreactor ing experience. Today, we have about 1450 reactor-years designs will be safer than the current designs for operating plants.
experience. Manylessons were lcarned as a result of the TMI-This confidence is based on the fact that all of the operating j
2 accident and from safety reviews, research, and operating ex-experience is directly applicable to the evolutionary plants.
j periencein the ensuingyears. Now,we must consider how to
'j apply all of our experience in setting safety requirements for However, the situation with regard to passive plant designs is not the next generation of reactors.
as clear, partly because of design philosophy and partly because of economic considerations. The design philosophy appears to be The staff has been following Commission guidance in using one ofimproved safety through simpler designs, larger operating the following bases for future plant requirements:
margins, and passive safety features (e.g., gravity-fed emergency NRC regulations Continwd on Page 2 Post-TMI requirements, as embodied in Section 50.34(()
What's in this ISSUE?
Technical resolution of all unresolved safety See Page 2 issues and the medium-and high priority generic safety issues p ';,
9608210146 900930 PDR NUREG BR-0125 R PDR ai, dl
~
After review and inspection by the Office of Nuclear Reactor IN THIS ISSUE Regul 6 n (NRR), Region 1, and NRR amJ' acta Battde Memorial,
Institute - Pacific Northwest Laboratory (PNL), the facility was Using Lessons from Operating Experience allowed to transition from mode 1 (defueling) to mode 2 (defueling for Future LWRs complete) and to mode 3 (defueling and fuel shipments complete) by Thomas E. Murley.........
.I n April 27,1990. Mode 3 has fewer restrictive technical specifi-cations than mode 1 and does not require the licensee to staff the Three Mile Island Unit 2 Completes Defueling control room with licensed operators.
bylee Thonus and Michael Masnik 2
The completion of defueling ends an 11-year effort of cleanup and Cracking in Combustion Engineering
$2 defueling following the March 28,1979, accident. The total cost of Steam Generator Tubes the cleanup and defueling has been approximately $1 billion. After by R. W. Winters.
.i
.3 the accident, GPUN initially performed activities to stabilize and cool the reactor. Once the reactor was stable and cool, GPUN j
Region V Engineering Managers Forum 7 proceeded to decontaminate approximately 1 million gallons of by F. Randall Huey..
.4 highly contaminated water. This activity began in November 1979 and was completed in May 1982. Then, after a period of data NEWSLE' ITER CONTACT:
Sathering, analysis, preparation, and construction, GPUN began i
defueling the reactor in October 1985.
Valeria Wilson, NRR 4921208 v
Requiring 4-1/2 years, defueling included the following principal activities in sequence: rubble digging, drilling, and plasma arc cutting. In February 1990, the licensee submitted the Defueling Using Lessons Compietion Report and then in Aprii 1990 submitted suppiemen-1 tal information, including the results of the final vacuuming and contin =d from Pase 1 video examination. Region 1 and PNL assisted the NRR staffin performing a detailed evaluation of the completion of defueling coolm.g systems and natural circulat. ion shutdown cooh.ng systems). These des,gn choices require smaller plants (600 and the justification for the change from mode 1 to modes 2 and 3.
i j
MWe) that, to be competitive, must have a capital cost of about one half that of the larger evolutionary plants. This,in The TMI 2 technical specifications required that the following l
turn, has led designers ta corsider ngorous cost-cuttmg measures.
three conditions be met for the licensee to transition from mode 1 to mode 2'-
The combination of this design philosophy and these cost-cutting measures has caused designers to eliminate such o
The reactor vessel and reactor coolant system are systems as safety-grade emergency electrical power and con-tainment spray systems. Such fundamental departures from defueled to the extent reasonably achievable.
I j
past designs raise questions about the extent to which current The possibility of criticalityin the reactor buildingis o
regulations and past operating experience can be applied t passive designs. These questions will be a big challenge as we precluded-initiate the detailed safety review of passive plants.
o No canisters containing core material are in the reactor building.
THREE MILE ISLAND The additional requirement for transiti n t m de 3 required that no canisters containing core material be stored on the TMI-2 site.
UNIT 2 At the end of defueling, the licensee estimated that approximately 608 Kg of fuel (UO,) remained, out of an original core load of COMPLETES DEFUELING 94,200 Kg.
by Lee Thonus and Michael Masnik, PD14 The remaining fuel was located in inaccessible areas that would be difficult to defuel or was in the form of thin surface I'dms. The licensee's estimates were based op gamma spectrometry of fuel In March 1990, defueling crews of the General Public Utilities analogues (principally Ce.144), active neutron interrogation, and Nuclear Coorporation (GPUN the licensee) performed a final visual estimates of fuel volume. A combined staff from NRR, flush and vacuuming to complete the defueling of the reactor Region 1, and PNL reviewed and analyzed a selected sample of vessel at Three Mile Island Unit 2 (TMI-2). Following this GPUN's video tapes, measurements, and calculations. The staff activity, GPUN performed a video examination to verify the concluded that the calculationswere accurate and the residualfuel efficacyof the effort and to quantify the remaining residual fuel. estimates were conservative. An independent verification of the 2
J
- licensee's criticality'models and calculations conntmed that they outside diameter of the tubes at the top of the tubesheet. Three were conservative and that the cdculated value for K,was less tubes were removed for andysis to determine the cause of this thin 0.95. The reactor building, fuel handling building, and rail cracking. As a result of this analysis, the licensee determined the facilities were inspected by onsite Region I personnel to verify cause of this cracking was caustic-induced secondary side stress that no fuel-bearing canisters or casks remained in storage or corrosion. In March 1988, during the fuel cycle following this de-transit. Upon completion of these verincations, the staff con-termination, the licensee began treating the secondary water with cluded that the facility had been defueled to the extent reasona-boric acid to minimize the caustic-induced cracking.
bly achievable, that criticality was precluded, and that no fuel-bearing canisters remained on site. The staff thus concurred In the 1%ruary 1989 refueling outage, the licensee found 309 with the licensec's action of passing from mode 1 to modes 2 and additional tubes with circumferential cracks in this same area. The 3.
Licensee determined that the identification of the large number of additional cracks resulted from improved eddy current analytical The completion of defueling at TMI 2 marked the end of a long, techniques (i.e. frequency mixing), and that most of the cracks expensive,and arduous effort by GPUN. Future efforts at TMI-found were present during the previous inspection but not identi-2 include the disposal of the 23 million gallons of stored fled with the techniques available during that inspection. These tecident-generated water and the long-term storage of the con-cracks that occur in the transition zone where the tube diameter is taminated reactor facility. A plan to evaporate the accident-changing are virtually transparent to eddy current bobbin coil gercrated water was approved after a public hearing, and the techniques. Analysis is further complicated by interference of the ev*poration will begin this fall. GPUN has proposed to place sludge pile with the eddy current signal. At the request of the TMI-2inmonitoredstorage,termedbythelicensee Post-Defu-NRC, the licensee agreed to perform an in-cycle inspection during cling Monitored Storage
- or PDMS, until TMI 1 is to be decom-October 1989. During this inspection, the licensee found an missioned, at which time GPUN will decommission both units additional 104 tubes with circumferential cracks on the tube simultaneously.
outside diameter at the top of the tubesheet. The number of cracks correlated well with the number predicted by the licensee before the inspection. Tbc reduced number of cracks in the tubes CRACKING IN indicated that the borte acid treatment was successfut in mitigating l
the cause of the failures. In addition to the eddy current examina-COMBUSTION ENGINEERING tion, the licensee aiso used uitrasonic inspection techniques to
- **" i* * ' *' i d * """* d b
- d d' *"" * "' * *'"'""""
Y STEAM GENERATOR TUBES In June 1990, while Millstone Unit 2 was shut down, the licensee by R. W. Winter, Region i elected to perform another mid-cycle inspection of the steam generators. During this inspection, the licensee found 23 tubes A new type of cracking has developed in the steam generator with circumferential cracks on the tube outside diameter at the top tubes in nuclear power plants designed by Combustion Engi-of the tubesheet. Using further improved eddy current analytical neering Incorporated (CE). In the roll transition area located techniques, the licensee determined that 22 of these cracks had immediately above the tube sheet, circumferentially oriented exist ed during the October 1989 inspection, and thereforc only one crteks have developed that are not detectable using the eddy additionalcrack had developed in the period between thesc inspec-current bobbin coil method of examination. In the past, exami-tions. Thus, the growth of these cracks a ppears to slow significantly nations of the steam generator tubes have not included the after initiation as the stresses are relieved by the crack initiation.
section of the tube immediately adjacent to the tube sheet because the test methods cannot detect flaws in this area. All of the tube cracks foundin the Millstone steam generatorswere Additionally, the failure mechanisms are not the same in all located in the sludge pile and initiated on the outer tube surface.
cases and can result in cracks that propagate from either the inside or the outside diameter of the tube. In January 1987, At the Maine Yankee Atomic Power Plant,another CE plant,the I
approximately one month after startup from a refueling outage, Maine Yankee Atomic Power Ccmpany (the licensee) also found a smallleak (0.1 gallons per minute) developed in one of the two circumferential cracks at the top of the tubesheet in the transition steam generators at Millstone Nuclear Power Station Unit 2.
area. Initially,it appeared that the same mechanism was at work The unit was subsequently shut down, and the leaking tube was in the Maine Yankee steam generators as at Millstone. However, identified. Eddy current examination using a rotating pancake upon removal and examination of a tube, the licensee established coilidentified a circumferential crack in the roll transition at the that the cracks at Maine Yankee initiated from the inside diameter top of the tubesheet. The tube was plugged, and the unit of the tube (the primarywater side). Allbut one of the cracks were continued operation until the next refueling outage.
found in tubes within 10 rows of the outer diameter of the steam generators. This area was substantially outside of the sludge pile.
During the next refueling outage in January 1988, the Northeast Reanalysis of the eddy current data confirmed this for the remain-Nuclear Energy Company (the licensee) found 26 additional ing tube s. M aine Yankee also used ultrasonic techniques to estab-I tubes that had circumfe rential cracks in the roll transition on the lish the size of the cracks found by eddy current examination. Un-l
O fortunately, during the ultrasonic examination the licensee as-original design bases for their plants, and when their engmeering sumed that the cracks initiated on the outer tube surface, and organnations inadequately participated in the plant maintenance the relevant information to allow analysis for inside cracking and modification processes after the initial plant construction.
was not preserved during the data collection, so no reanalysis of the ultrasonic data is possible.
In June 1988, the Region V licensees established the Region V Engineering Managers Fourm. This forum was antaWidal to As of September 1990, Maine Yankee is studying the conditions address the findags from several SSFI inspections conducted by that could have caused the cracking of these steam generator licensees and to address the need for their engineering organiza-tubes. Some of the conditions under consideration include tions to actively intrude into and participate in the support of primary and secondarywater chemistry, operating temperature, nudear plant operations. Since estabEshag the forum, the members and other operating parameters. In addition, the licensee is con-have held six meetings of the main group and numerous subcom-sidering whether the same mechanism caused the failures expe-mittee meetings. These meetings have resuhed in three formal rienced in the control element assemblies.
documents, as A-d below.
At the Calvert Cliffs Nuclear Power Plant, the Baltimore Gas CHARTER (issued December 1989) and Electric Company (the licensee) recentlyinspected its CE steam generators in Unit 1. During this inspection, the licensee The charter defines the purpose, objectives, and organivation of examined approximately 50 percent of the tubes in the steam the forum.
generators at the top of the tubesheet in the sludge pile and a 10 percent random sample of other tubes using rotating pancake The stated purpose of the forum is to address generic issues coil techniques. This inspection was made to determine whether related to engineering activities at nuclear power plants. The the Calvert Cliffs steam generators were experiencing the same forum provides an environment for efficient evaluation of joint type of tube degradation found at Millstone and Maine Yankee. engineering related issues and enables licensees to share knowl-During this inspection, the licensee found no circumferential edge, thus enhancing the quality of the final products of the forum.
cracks, indicating that the steam generators at Calvert Cliffs were not susceptible to the same phenomenon or that the The primaryobjectives of the forum are to enable the members to I
degradation had not yet started.
evaluate and discussjoint engineering-related issues; share knowl-edge, experience, and lessons learned; develop unified position In summary, from the reduction of the number of cracked tubes statements and guidance on related engineering topics; and im-in the Millstone steam generators, the failure mechanism ap-prove communication among member utilities.
pears tobe caustic.ind uced intergranular stress corrosion erack-ing and has been controlled by the addition of boric acid. The The organization of the forum includes the corporate engineering failure mechanism at Maine Yankee has not yet been defined. manager and designated alternates from each member utility.
The steam generator tubes at Calvert Cliffs hau not shown The forum selects a chairman, vice chairman, and secretary at signs of either failure mechanism and may not have the condi-periodic (usually annual) elections. The forum may appoint tions that caused the failures at Millstone and Maine Yankee.
subcommittees to analyze specific engineering issues and report findings and recommendations back to the forum.
DESIGN B ASIS DOCUMENT PRO-GRAM GUIDELINES (issued May1990)
This guideline document was developed by members of a forum REGION V ENGINEERING subcommittee who were members of the Nuciear Management
-MANAGERS FORUM
- '"'""'"("""^*)
- 'i'""'"*'""**"")
working group. The Region V document was Inc foundation for by F. Randall Huey, Region V The guideline document defines the scope of DED progruns and During the last few years, the NRC has focused increasing the evaluations needed to develop design-basis information. In attention on the assessment of the quality of engineering and particular, the document establishes criteria for DBD scheduling, technical work being performed by licensees. Primarily, this, qualityassurance, verification,and use. It also addressesthe need increased attention resulted from the findings of safety system for open items tracking, relationships between licensing bases and functionalinspections (SSFis). The SSFl is particularly effective design bases, use of pre-Appendix B design information, and at highlighting implementation weaknesses in licensee engi. training of plant personnel on DBD use. The document provides neering programs. Among most noteworthy finding of past examples of typical design-basis and design-output documents l
SSFIs, the teams performing thesc inspections consist entlyiden. and describes mechanisms for configuration management, to tified problems that resu'ted when licensees lost control of the ensure continued updating of design-basis information.
s s
1 PROACilVE ENGINEERING GUIDELINES Ensure an appropriate design safety margin in all design o
(issued Mkrch 1990) activities.
This document provides a format for promoting active, systematic, o Ensure that the front-line engineering and technical sup-and timely involvement of the engineering organization it. overall port personnel have quick access to design-basis
- forma-m plant operations. The document has been prodded to and favora-tion. Ensure that the design-basis document is continu-bly commented on by the Institute for Nuclear Power Operations ously used to support engineering activities.
(INPO), as well as engineering management counterparts in other Regions.
Ensure effective and timely involvement of qualified per-o i
sonnel from both the engineering and operating staffs in i
The primary strength of these guidelines is the recognition of the the budgeting process.
following:
Develop an effective process for periodic design self-as-o Fundamentally, nearly every decision that affects plant sessment. Engineering management must receive fre-operation has a design or engineering element. Therefore, quent feedback on the effectiveness of initiatives to im-
)
it is beneficial for the engineering organizations to be prove engineering.
I actively involved in plant operation. Plant operation within the design bases is a shared responsibility between the F"$g TasmwortandImohensait engineering and operation organizations. Three factors are availabic to promote active and systematic involvem ent:
Structure the engineering organization to promote the o
team approach. Ensure that personnel on the site and at (1) Processes and Procedure: Selection of the processes the corporate offices are working as a team.
at the plant where engineering can be effectively and effi-ciently utilized is the first step, and implementation in ac-Provide job description for engineers that clearly reflect o
cordance with the licensee's procedures assures consistent the proactive engineering approach. Ensure that the application.
first-level supervisor has an appropriate role in promot-ing teamwork.
(2) Availability: Availability of the engineering resources facilitates prompt participation.
Establish rotational and cross-training assignments.
l I
(3) Culture: Teamwork is essential and is perpetuated by Provide for periodic joint walkdown inspections by the i
the culture endorsed by management. Corporate policies system engineer and design engineer of plant systems.
n tin te.
Plant Needs 9'"#'"T Some of the highlight recommendations of the document include the following:
o Ensure that a credible working relationslu.p exists be-tween the engineering organization and its operating F**My Gear Responsibilitier and Accommtability plant counterpart.
o Provide cognizant engineers with on-calllists.
Provide a formal charter to clearly establish responsibility o
and accountability and define relationships and interfaces o
Ensure that appropriate engineering personnel are lo-between engineering and technical support organizations and their plant operating and maintenance counterparts.
cated at the plant.
Provide for periodic, joint review of engineering backlogs Provide focused engineering resources. Establish effective o o
long-range planning and ensure that engineering resources by the engineering and operating organizations.
are efficiently applied to the right efforts at the right time.
Estabissh Efectiw Canmsukann Estabinshim Efectiw Plant Support Prognmss and Processes Prov.de for appropriate engineering partmpation on o
i P ant review committees, as well as the following daily l
Provide a root cause analysis program that is recognized as o
3 P ant status meetings, restart readiness reviews, signifi-l i
being effective in precluding recurrence of plant prot.lems.
cant event reviews, and plant management reviews.
Develop an integrated design control process that can o
j accommodate both the simplest and the most complex Provide an NRC non-conformance evaluation program o
i changes to the plant in a disciplined and controlled manner.
that is clearly defined, easy to use, and rigorously tracked.
4 w
+-
s
~
o Est:blish a process for first-level supervisory interde-partmental meetings.
o Promote the use of electronic communication systems.
The forum is currently working on other initiatives such as Spare Parts, Backlog of Engineering Work, and Qualification Standard for Engineers. The members of the forum will work closely with and involve INPO in all of their initiatives. To the degree that initiatives involve generic issues, requiring close interface with the NRC, the forum members will involve NUMARC (as was the case with the guideline for the design basis document).
Tb?. members of the Regko V Forum are enthusiastic and con-fident that their efforts will go far toward improving overall ruclear plant operations. They are actively interfacingwith their counterparts in other Regions and have indicated that utilities in both Region 11 and Region III are establishing similar forums.
The ultimate measure of the success of these initiatives will be the success with which the individual licensees translate the guidelines into clearly defined responsibilities and management expectations. The efforts by the members of the Region V Forum appear to be a good first s;ep.
l 6
l l
sl s
I 1
i l
l 1
4 1
//
/
l I
i I
THIS DOCUMENT WAS PRINTED USING RECYCLEO IAPER.
i l
l l
l l
l