ML19332F480
| ML19332F480 | |
| Person / Time | |
|---|---|
| Site: | Oyster Creek |
| Issue date: | 12/11/1989 |
| From: | Tosch K NEW JERSEY, STATE OF |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| Shared Package | |
| ML19332F481 | List: |
| References | |
| NUDOCS 8912150064 | |
| Download: ML19332F480 (22) | |
Text
y 1 r,w i
- la ==* ' State of New Jersey )
DEPARTMENT OF ENVIRONMENTAL PROTECTION i n DMSION OF ENVIRONMENTAL QUAUTY j cN 415 a Trernon, N.J. 08625 0415 (609)967 6402 Fax (609)987-6390
.lill Lipotl, Ph.D., Assistant Director l Radiation Protection Programs December 11, 1989 i U.S. Nuclear Regulatory Commission '
Attention: Document Control Desk Washington, D.C. 20555 Gentlemen:
Subject:
Oyster Creek Nuclear Generating Station i Docket No. 50-219 Draft Safety Evaluation Report (SER)
The . Department of Environmental Protection, Radiation 1 Protection. Program's (RPP), Bureau of Nuclear Engineering's - (BNE)-
f staff ' has reviewed the Draft Safety Evaluation Report (SER),
prepared by the NRC, for the Full Term Operating License (FTOL) application filed by GPUNuclear Corporation (GPUN) for the Oyster Creek Nuclear Generating Station (OCNGS). On the basis of the Draft SER - (NUREG-1382) prepared in September, 1989 and dated October, 1989, the U.S.' Nuclear Regulatory Commission (NRC) staff !
has concluded that the'OCNGS can continue to be operated without endangering'.the health and safety of the public. The BNE recommends that the SER, which is largely dependent upon .the Systematic Evaluation Program.(SEP), has several areas where more study of the technical basis is needed.
A review of the Draft SER provided little in the way of coherency to the-piecemeal process of evaluation reflected by the SEP and Integrated Plant Safety Assessment RE .rt (IPSAR) for Oyster Creek. The Draft SER contains many discreet evaluations p that have-been performed by the ever changing NRC licensing staff i over a ten year period. Many SEP topic reviews are vague in documenting how safety evaluations were performed, who performed them, and the degree to which they reflect a compromise of the specific intent of each SEP topical review. In some cases there js a question about the accuracy and/or correctness of the information used to reach a safety related conclusion. Currently, L there are eight open SEP topics and six Unresolved Safety Issues associated with the draft SER.
Several steps were taken to review the Draf t SER. These steps were an overview of the Provisional Operating Licence /FTOL process, SEP, an overall review of SEP by an independent contractor, detailed BNE technical review of SEP topics and BNE comments.
Coo l New Jersey is an Equalopportunity Employer g g
l SA 891211 '
P CK 05000219
- _PD C_ _ ,, , _ _
l l
l USNRC December 11, 1989 1
History Jersey Central Power & Light (JCP&L) received its Provisional '
Operating License (POL) for the OCNGS on April 9, 1969 and filed an application to convert POL DPR-16 for Oyster Creek to a Full Term OpertSting License (FTOL) in a letter dated March 6, 1972.
From 1959 to'1971, the U.S. Atomic Energy Commission issued POLS to 15 power reactors for periods of up to 18 months as an l intermediate stage before issuing an FTOL. The purpose of the POL ,
was to provide an interim period of routine operation during which 'l the licensee and NRC staff could assess plant operating parameters and performance' against predicted values and resolve generic concerns identified during the licensing process. 1 In 1977, the NRC staff recommended to the Commission that nuclear power plants with Provisional Operating Licenses, be I' included in phase II cf the Systematic Evaluation Program (SEP),
because much of the review necessary for conversion of the POLS was !
similar to the scope of the review of the designs of the 11 older l operating nuclear power plants, including the OCNGS. This l racommendation was performed in order to reconfirm and document their riafety. That recommendation was adopted, and the major portion of the technical input supporting the Draft SER comes from I the SEP topic evaluations which were incorporated into the IPSAR for Oyster Creek (NUREG-0822).
Overview of the SEP for OCNGS 1
-l The SEP review provides (1) an assessment of the significance !
of dif ferences between current technical positions on safety issues l
- and those that existed when a particular plant was licensed,.(2) l a basis for deciding how these differences should be resolved in I an integrated plant review, and (3) a documented evaluation of ,
plant safety, i The SEP review covers 137 topics but for Oyster Creek's review i 54 of the topics were deleted from consideration because:
- 1) A review was being conducted under other programs i.e.,
Unresolved Safety Issues (USI) or Three Mile Island Action Plan Tasks (NUREG-0737);
- 2) The topic was not applicable to Oyster Creek;
- 3) The items to be reviewed under the topic did not exist at the plant.
J
s=
USNRC December 11, 1989 Therefore of the original 137 topics, 83 were reviewed for the OCNGS. Of these topics:
- 1) 43 met current criteria or were accepted on another defined basis.
- 2) From a review of the 40 remaining topics, certain aspects of plant design were found to differ from current criteria. These 40 topics were considered in the IPSAR.
The IPSAR covered these 40 topics by evaluating the safety -
significance and other factors of the identified differences from current design and arrived at decisions on whether modification was necessary on an overall plant safety viewpoint. To arrive at these '
decisions, engineering judgement was used as well as the result of a limited probabilistic risk assessment.
l In general, the staff's position on the 40 topics fell into one or more of the following needs: (1) equipment modification or addition, (2) procedure development or Technical Specification changes, (3) refined engineering analysis or continuation of ongoing evaluation, and (4) no modification necessary.
For those positions classified as either Category (1) or (2),
the IPSAR lists the scheduled completion dates agreed upon by the staff and the licensee. NRC Region I staff has verified or is
-verifying the implementation of these positions. For those positions classified as category (3), the licensee has provided the results of the ongoing evaluation to the NRC staff for review.
The IPSAR for the OCNGS was issued in January 1983 and identified those issues and SEP topics that were " resolved." A supplement to the IPSAR (NUREG-0822) published in July 1988 to provide the staff's evaluation of the Category (3) issues and to summarize the statues of all actions to be implemented as a result of the SEP review.
I. Overall review (Independent Contractor)
In 1985 the Nuclear Engineering Section (NES) of the Bureau of Radiation Protection hired an independent contractor Mr. Peter Davis to review the IPSAR. The NES hired Mr. Davis because of his experience in the application of Probabilistic Risk Assessment techniques. There were three tasks assigned for analysis specific to the IPSAR and a two phase report was generated.
The major conclusion from Mr. Davis's evaluations relevant to this discussion are summarized below.
From Final Report Phase I, "A Review of the Use of PRA in the Integrated Plant Safety Assessment Study for the Oyster Creek Reactor":
USNRC December 11, 1989
- 1. For a large number of deleted SEP topics, the NRC evaluation does-not provide the basis on which they were deleted for the OCNGS. ,
- 2. The quantitative probabilistic risk evaluation of SEP topics was a very minor consideration in deriving backfitting Instead, it appears that the primary basis for requirements.
decisions on backfit requirements was engineering judgement coupled with the extensive experience base developed by the NRC in using design basis accident concept in conjunction with deterministic analysis.
- 3. The risk assessment evaluation of 20 SEP topics as provided in the Attachment Section, contains several apparent discrepancies. In no case, however, were such discrepancies found to have an adverse influence on final decisions regarding backfit requirements. The qualitative results used for backfit decisions are, therefore, considered valid.
- 4. Several topics were identified which appear to have an influence on the probability of causing releases below levels considered in PRAs, but in excess of Protective Action Guides.
From Final Report Phase II, "An Assessment of Risk Significance of SEP Issues for Oyster Creek,":
- 1. The Oyster Creek core melt probability (4.4E-4 per l Sec'c ion IIA) as estimated from revising the results of the l Millstone Point Unit PRA is relatively high, as is the frequency of major release probability when compared to other PRA results.
l These results should be viewed with caution, however, due to the l inherent uncertainties in PRA results as well as variations in rigor and methodology employed by the PRA studies and uncertainties in extrapolating the Millstone results to Oyster Creek. An estimate of risk from Oyster Creek due to core melt accidents is provided in the Attachment Section.
1
- 2. Of the 87 issues, some judgement of potential risk significance could be made on only 48. The remainder were related to external events not considered in PRAs. Only 10 of the 48 were found to have potential risk significance. Of the 10, backfit requirements are being imposed on 9. A summary of these results as tabulated in the report are shown in Table 1.
Note: The phase II report also provided a comparison of the risk assessments of nuclear plants similar in design to Oyster Creek and provides some insight into the importance of plant-specific analysis, see Attachment Section.
y 9
USNRC Decem%r 11, 1989 e-L Table 1 i
SUMMARY
OF ISSUE EVALUATIONS No. of No. of issues-Risk Significant Issues Requiring Backfit Comment i
i Undetermined 35 22 3 others partly l-covered under '
i related backfit None- 35 14 3 others partly covered under ;
related backfit LLow- 7 3 1 other partly covered under related backfit i
Possible 9 8 No backfit !
L required for reactivity.
control issue
( (#48 from Table IV-1)
High 1 1
- Reliability of
'RHR (#54 from i
E Table IV-1) j Total 87 48 ;
f
- See Attachment Section
USNRC December 11, 1989 II. BNE In-house Review 1
- 1) SEP Topic II-1.C: " Potential Hazards or Changes in Potential Hazards Due to Transportation, Institutional, and Military Facilities."
NRC SER' Analysis: "The staff reviewed the potential hazards to safety-related structures, systems and components resulting from nearby transportation, institutional, industrial, and military facilities." 1 "The nearest transportation route to the station is U.S. Route 9, which is located approximately 0.25 east of the reactor building. However, Route 9 is not heavily used for shipping in the locality. There are no industries in close proximity to the plant site that are expected to use or store large amounts of explosive or hazardous material...Through traffic generally uses the Garden State Parkway, a limited access toll road that runs parallel to Route 9. The Parkway is about 1.25 miles west of the plant and is open to truck traffic. The. separation distance between the highway and the plant exceeds the minimum distance criteria given in Regulatory Guide 1,91 for truck-size shipments of explosive material. Therefore, in a letter dated February 4,1982, the staff concluded that the transportation of hazardous materials on U.S.
9 poses no significant-hazard to the plant."
"In a letter dated February 4, 1982 the staff concluded that the licensee's exclusion area authority has the proper authority to determine all activities within the exclusion area, as required by 10 CFR Part 100."
Discussion: Commercial vehicles are restricted from using the Garden State Parkway from its northern origin in Montvale, N.J.
south to exit 105 near Eatontown, N.J. In practice, most long-haul commercial traffic utilizes the N.J. Turnpike and the majority of local commercial traffic along the eastern shore utilizes Route 9 south from Perth Amboy. In addition, since 1982 the areas immediately north and south of Oyster Creek have been significantly developed and Route 9 has become a thoroughfare i.e.,
transportation route for commerce as well as for public use.
Route 9 is on the border of the Exclusion and low Population Zones around Oyster Creek and is subject to their definitions and criteria as defined in 10 CFR 100. The question is whether the OCNGS site meets those criteria in 1990 rather than 1982 since the '
FTOL evaluation should reflect the present condition of the plant and its surroundings.
j i
USNRC December 11, 1989
- 2) SEP Topic II-1.B: Population Distribution NRC SER Analysis: "The peak seasonal population (defined as the sum total.of permanent and transient population groups on an average day during peak summer season) within a 10-mile radius of the plant is expected to be 179,840. The permanent resident population within the 10 miles is 66,815 (1980 census) . "
"The nearest population centers with more than 25 residents are Dover Township, Gilford Park, and several smaller cities. The '
population in 1980 was approximately 64,445."
In a letter dated February 4, 1982, the staff concluded that the low population zone and population center distances specified for the Oyster creek site are in conformance with 10 CFR Part 100."
Discussion: A population distribution projection for the 50 miles circumferential to the OCNGS, was developed as an integral part of the " Base Document for Considering the Full Term Operating License for the Oyster Creek Nuclear Generating Station." It utilizes the OCNGS Updated Final Safety Analysis Report (FSAR-1984) and " Technical Guidance for Siting Criteria Development" NUREG-2239 which was prepared for the NRC by Sandia National Laboratories. Data from pages 59-64 from the Base Document in the Attachment Section follows this discussion and suggests that the #
data base from which they draw their SER conclusion. This would seem to be intolerable both as it may mislead those reviewing the SER i.e., Advisory Committee on Reactor Safeguards-(ACRS) and the State of New Jersey who is responsible for emergency preparedness around Oyster Creek. A comparison of the SER population
-projections and those in the Base Document are shown below:
Permanent and Transient Permanent 10mi radius NRC SER 179,840 66,815 RPP Base Document 223,330 132,755 Some other exacerbating considerations are defined in the attachment and of particular importance is that over 70% of the total population resides in 4 out of the 16 sectors within 10 miles around the OCNGS and population densities exceed most, if not all, siting criteria guidance published by the NRC.
To demonstrate the vagueness of the NRC staff's evaluation, it is difficult to understand the significance of a population in 1980 of approximately 64.445 for Dover Township, Gilford Park, and conclusion that population center distances specified for the Oyster Creek site are in conformance with 10 CFR Part 100.
1 i
USNRC December 11, 1989 It seems reasonable to ask for a more accurate population I center analysis for 1990. An analysis needs to be shown in the SER that defines the populations in incremental distances between 10 and 50 miles of the site considering that a population over 500,000 presently live within 30 miles of the site and over- 3,000,000 live within 50 miles.
- 3) SEP Topic III-5.B: " Pipe Break Outside Containment" NRC SER Analysis: "10 CFR Part 50 (GDC4), as implemented by SRP Sections 3.6.1 and 3.6.2 and Branch Technical Position MEB 3-1~'and ASB 3-1 (NUREG-0800), requires, in part, that structures, systems and components important .to safety be designed to accommodate the dynamic effects of postulated pipe ruptures. The safety objective of the review under this SEP topic is to ensure that if a pipe should break outside the containment, the plant could be safely shutdown without a loss of containment integrity."
"In IPSAR Section 4.10(2), the staff identified concerns associated with emergency condenser isolation. In Supplement 1, the staff indicated that the licensee would submit information on ,
this matter for review. In a letter dated July 27, 1988, the licensee described plans to replace all four isolation condenser penetrations. Additionally, all isolation condenser pip.ing on the 75-foot elevation will be replaced with Nuclear Grade 316 stainless steel piping and penetration material. To provide time for design review, equipment procurement and -logistical optimization of implementation, the licensee has proposed a deferment in the schedule (from the Cycle 12 refueling outage to the Cycle 13 refueling outage) for the resolution of this issue. The staff finds this change in schedule acceptable. It will review the licensee's final design when it is submitted."
Discussion: Two inhouse reports, both engineering and radiological consequence, are located in the Attachment Section.
The conclusions were as follows:
In the present configuration of the isolation valves of the isolation condenser line, does not meet the NRC General Design Criteria on " Environmental and missile design basis" and on
" Reactor coolant pressure boundary penetrating containment".
Neither a leak nor a break could be tolerated at certain critical locations without significant radiation release to the environment.
The austenitic stainless steel exhibits leak-before-break under normal loading. However, under adverse loading conditions, comprising of seismic and/or water hammer loads, the isolation condenser piping may have a major rupture.
USNRC ' December 11, 1989 Jet impingement of steam at operating temperatures onto the valve -operator and its control circuitry, raises environmental .
qualification concerns for the isolation valves. l A loss of coolant accident (LOCA) outside the drywell has l severe radiological consequences to the public, see report in the l Attachment Section. An automatic isolation valve inboard to the ;
containment would preclude a LOCA outside the primary containment. l The necessary restraints to prevent pipe whip, which is i required per NRC design criteria are not provided.
l
- 4) SEP Topic III-6: "New Seismic Floor Response Spectra" NRC SER Analysis: "Several different floor response criteria l were used in the design of the Oyster Creek station. This contributed to the difficulties associated with the Oyster Creek seismic design basis in subsequent applications. In July 1987, the licensee proposed to develop new standardized seismic floor
response spectra for futurc applications at Oyster Creek, including L the resolution of some of the issues associated with NRC IE Bulletin 79-02 and 79-14."
l "By letter dated September 19, 1988, the licensee summarized
- the history and status of development of these new seismic spectra.
l By letter dated October 17, 1988, the staff indicated that a number of issues still had to be resolved before the new floor response i
spectra, including the use of site-specific data, were approved.
Development of the new seismic floor response spectra continues, l and they will be reviewed by the staff when they are submitted."
Discussion: The NRC letter dated September 15, 1989 required GPUN'to implement the modificaticas of all supports, meeting the requirements of IE Bulletin 79-02 and 79-14. However, GPUN l indicated in their November 1, 1989 letter to the NRC that the required modifications will be difficult to implement by 13R because of schedule problems. The GPUN believes that because the New Seismic Response Spectra presently being developed the modification to the remaining 23 supports in question may not be required.
The BNE has evaluated NUREG-1150, " Severe Accident Risks: An Assessment for Five U.S. Nuclear Power Plants," (Second Draft for Peer Review, NRC, June 1989) as a resource document on this topic.
As a result of that evaluation NUREG-1150 states that seismic events can contribute significantly to the overall risk of initiating nuclear power plant core-melt &ccidents.
In NUREG-1150 Volume 2 Appendix C, " Analysis of Seismic Issues," has several points which are relevant to this issue.
USNRC December 11, 1989
- 1. Since the mid-1970's some 25 plant-specific seismic PRAs have !
been completed and have shown that seismic events can be significant contributors both in terms of core damage frequency and the potential for releases of radioactive material.
- 2. Two major programs have been undertaken in the past few years !
to develop methods and data banks to estimate the seismic hazard !
at all locations of the U.S. east of the Rocky Mountains. Lawrence i Livermore National Laboratory (LLNL) conducted one of those programs entitle, " Seismic Hazard Characterization Project" and j referenced specifically as " Seismic hazard Characterization of 69 Nuclear Power Plant Sites East of the Rocky Mountains," NUREG/CR- l 5250, Vols.1-8, UCID-21517, January 1989. Electric Power Research Institute (EPRI) conducted the other program.
- 3. The hazard was calculated for each of the seismic sources and combined for all sources. For each site, typically, 2,750 curves were developed.
- 4. In addition to the hazard curves for the 69 nuclear plant sites east of the Rockies, LLNL project also generated uniform hazard spectra for various return periods for each site,
- 5. The seismic hazard spectra based on eastern earthquakes are higher at high frequencies and lower at low frequencies.
l
- 6. Peak Ground Acceleration (PGA) have not been shown to be good 3 l indicators of the damage potential of an earthquake for ductile l l structures / components, since low magnitude events can produce a l large PGA but little damage. This concern has been alleviated to some extent in the LLNL and EPRI studies by the use of a minimum <
magnitude of 5.0, the magnitude below which damage to the l engineered structures is considered unlikely.
The BNE staff wants to know what seismic criteria were used when OCNGS was designed and constructed, compared to the seismic l criteria used in the design of modern nuclear plants in the eastern l U.S. Additionally, since GPUN feels that this issue does not need prompt attention, the BNE will request that an independent seismic evaluation of the OCNGS so that the DEP can review the methodology used by the licensee.
The BNE staff recommends that the repair of the remaining 23 supports in question should not be postponed.
l III. BNE SEP Topic Comments The Draft SER did not contain enough detail for BNE to thoroughly analyze the following 3 topics. These topics are of concern to the BNE. The following three SEP topics need additional clarification for the staff to finalize its analysis.
USNRC December 11, 1989 ,
A. Item V-6, Systematic Evaluation of Plant (SEP-NUREG 0822)
" Reactor Vessel Integrity" states that the base Reactor Vessel material properties are not available.
Question: How does one know the most limiting material in the -
belt line region-of the vessel and determine its reference Nul Ductility transition temperature to establish the Heat-up and Cool-down curves?
The SER paragraph 5.3 (NUREG-1382) states that the Heat-up and Cool-down limit curves are now valid through 15 Effective Full Power Years. Please provide the BNE with the necessary information to perform an independent analysis.
B. The' welds in the Reactor Vessel belt line region and other critical-locations such as nozzles have not been inspected and can not be inspected per A.S.M.E.Section XI code during the life of the plant due to lack of accessibility. We understand that there was a waiver granted to the licensee on this inspection.
Question: How does the vessel integrity get verified without the inspection of these critical areas?
Please provide the BNE with additional information which led to the waiver being granted to the licensee.
C. The SER does not address the requirement the of 10 CFR 50.44 amendment, requiring hydrogen recombiners. GPUN believes, based on a study performed by the BWROG and documented in NEDO-22155 entitled, " Generation and Mitigation of Hydrogen in Inerted Containment," hydrogen recombiners for the containment are not needed. The BNE staff believes that GPUN may be misinterpreting the amendment requirement. Please provide clarification to the BNE on the NRC's interpretation of the intent of the requirement.
==
Conclusion:==
The BNE is concerned that the adoption of this draft SER for OCNGS is not representative of the plants actual operational safety. Given OCNGS present age, potential degraded condition, its unique design (lack of certain standard plant safety systems),
and the increasing population of the area, the BNE feels that a more rigorous review of the SER by the NRC staff is warranted.
We emphasize that since the SER is the definitive safety evaluation for considering OCNGS to pursue a FTOL and plant life extension, it needs a more thorough up-to-date technical review.
'l USNRC December 11,1o89 Three of the inhouse SEP Topic reviews demonstrate that a potential accident could arise and have a greater impact than expected. The plants ability to withstand and mitigate certain accidents is in question along with the calculated off-site dose consequence (population exposure). Using three of the SEP topics
" Seismic Floor Response Spectra", " Pipe Break Outside Containment",
and " Population Distribution" shows the relationship and relevancy of SEP Topics. The BNE could not reach the same conclusion as the' NRC - that OCNGS can continue to operate without endangering the health and safety of the people - without further information and clarification.
Sincerely, r v& W Kent W. Tosch, Chief Bureau of Nuclear Engineering
- c. Gerald P. Nicholls, Deputy Director Jill Lipoti, Assistant Director l
I l
l 1
l l
l l
l
VI ATTACHMENTS i
'l
-l',
i 3.
e 4
I.
APPLICATION OF LEAK-BEFORE-BREAK ANALYSIS FOR ISOLATION CONDENSER PIPING AT OYSTER CREEK NUCLEAR GENERATING STATION Oyster Creek has two Isolation Condenser steam lines that carry steam from the reactor vessel to the Isolation Condenser to remove decay heat and depressurize the reactor in the event that the turbine generator and the main condenser are unavailable as a heat sink. The attached figure shows the steam flow diagram for the Isolation Condenser and the reactor. There are two isolation valves on each of the steam lines before the Isolation Condenser. These valves are located-outside the drywell.
Considering the following scenarios,
- 1. A break between the two isolation valves with a failure of the first valve to close.
- 2. A pipe break between the second valve and the condenser resulting in pipe whip so that the isolation valves would not close.
- 3. A break between the first isolation valve and the drywell (not l considered as a design basis).
I Each of the above would result in a loss of coolant accident outside the drywell with significant radiological consequences. The I
existing arrangement of isolation valves violates the NRC design criterion #55 on " Reactor coolant pressure boundary penetrating containment", outlined in 10 CFR 50 Appendix A. Further, the design 1
k i
l O. -
i
/
.., I 9 9 (d, W,ll L
!,,s
! as ,
.., A. _ l 6
, . _~ ;
i U V s
L r'
g,,,,,,,,,,,,,~,,,,,,,,,,,,,~,,,,,,-,,,,,,~........,,,,,,,,w
-- - =
- ,,,,
r ,,, S l r r e, .,__ __________.
. @.i, ( (Ip._s l $ l HG H@ v u a 01 l -D <H> <3
- n s
= '
\ Q l , _ _ _ _ _ _ _ _ _ _ .. _.
l
\ q ll l I
%- ) .
- a g I l =>= 7 mo 1 ll + =
(y O l l
[ ) *<
, , y_________ __
y l llliG HG v 8
= ll l I l
I
><H>& n !A =
, , g_________ __
i k _s 8 i
444=i gwE g
- = '
3iIi2< .
O l
1
of the piping outside the drywell does not provide adequate restraints to accommodate the dynamic effects of postulated pipe ruptures. This is also in violation of NRC General Design Criterion
- 4 on " Environmental and missile design basis" outlined in 10 CFR 50 Appendix A. To comply with the NRC design criteria as stated above, Oyster Creek must instal an automatic isolation valve inside drywell and provide necessary restraints on the piping outside the drywell. However, the physical arrangement and space availability ,
preclude installation of restraints and make it difficult to instal an isolation valve inside'the drywell.
In view of the fact that a pipe break outside the drywell at the locations stated above, will have significant radiological consequences.and prevention of this scenerio would be extremely difficult, the licensee wants to justify that the Isolation Condenser line would leak a detectable amount well in advance of any crack growth that could result in a sudden catastrophic break.
This - is the leak- before-break (LBB) concept. The licensee has stated that the leak rates from postulated cracks are sufficiently high so that visual monitoring is an acceptable method of leak detection and there would be sufficient time to take appropriate actions (i.e. shutdown or isolate the affected condenser) between the time of leak detection and the time that a crack would grow to an unstable size. Although the leak-before-break (LBB) concept is well established and proven, the methodology is used with caution. Firstly, in the design of emergency core cooling system, 2
containment or other engineered safety features LBB is not acceptable as a substitute for the analysis of the double-ended guillotine break. Secondly, the LBB methodology is not applied to systems in which excessive or unusual loads or cracking mechanisms can be present because the phenomena adversely affect the piping behavior. The excessive / unusual loads or cracking mechanisms of concern include waterhammer and corrosion (particularly, intergranular stress corrosion cracking - IGSCC) erosion, creep, fatigue and brittle fracture. In any event, the stainless material of the isolation condenser-piping, due to its high ductility, usually will fail in ductile mode in the absence of any unusual loading as stated above. The pipe will tolerate large circumferential cracks without failure. There will be a tendency for the pipe to leak prior to crack extension in the
_ circumferential direction, thereby indicating a strong trend towards a leak-before-break. Under normal operating condition a 304 stainless steel pipe designed to ASME codes, will tolerate the following cracks before any catastrophic failure, assuming no seismic or waterhammer loads are imposed.
e A through wall crack extending 120 degrees over the circumference in the presence of pressure and bending stresses only.
l e A crack extending over the entire inside periphery l
l upto 65% of wall thickness, in the presence of pressure and bending stresses only.
1 1
3 v.
__ m _ _. ,_ _ __. _
O 0 l
However, considering Oyster Creek situation, the following constraints may have-a direct impact on the use of LBB methodology to the isolation condenser piping.
e The BWR coolant with high oxygen concentration is conducive to I IGSCC in the presence of a tensile stress field. In case of Oyster Creek the material being one of the common type stainless steels, is.very susceptible to IGSCC.
1 e The piping has been in service since the initial operation of the plant. The number of stress cycles perhaps have been high causing the IGSCC flaws to have grown in size. [
- The cracks due to IGSCC are difficult to characterize during l
ultrasonic examination and some times the test results may vary '
for a particular weld with different examiners, j l
e The piping was originally designed to ANSI B31.1 which did not require a dynamic fatigue analysis. A modern plant built to present day standard will meet the rules of ASME section III. With the conservatism of ASME section III design along with the criteria of recent 10 CFR 50, one could take credit for LBB.
l e There is a possibility of waterhammer in the isolation condenser piping if the water level in the reactor vessel is high and the normally closed isolation valve on the return line is suddenly 4
_ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ - . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - - - _ _ __ J
I
, .;
opened. The loads imposed during waterhammer may cause a rupture of the line in the presence of existing IGSCC crack.
e If there is a major leak resulting in a jet impingement of i I
saturated steam on the valve operator of the isolation valve I upstream of the discharge, there is a possibility that the isolation valve will fail to close. Is there an environmental !
qualification performed on the operator to address this question?
How does the licensee plan to mitigate the loss of coolant accident outside the containment should the isolation valve fail i to close per above scenario? The basic issue here is not the )
application of LBB, it is the severity of an uncontrollable j' discharge due to a major leak outside the containment, i
I LBB Analysis l
For LBB analysis, the flow rate through the leak determines the crack size which is then evaluated against,the critical crack i
size for the piping. In case of an uncollectible leakage such as u would occur in case of an isolation condenser line leaking, the crack size can not be estimated. Hence the stability of the crack can not be determined. If the crack size is estimated by the amount of make-up flow into the core, its accuracy is questionable. In order to prevent a crack on the isolation condenser line to grow to an unstable size, it becomes necessary to isolate a line if L there is an evidence of leak irrespective of the amount of leakage.
5
+
Hence, leak detection of the line during normal operation is of paramount importance. Either a leak detection-device for on-line monitoring of the line or continuous operator surveillance is necessary for safety of operation of the line.
The analysis requires an input of existing material properties including its ~ fracture toughness to establish the stability of crack. It is unknown if data pertaining to fracture toughness is available for aged and sensitized' stainless steel in its present condition.
Conclusion:
}
e In the present configuration of the isolation valves of the '
the isolation condenser line which does not meet the.NRC General Design Criteria on " Environmental and missile design basis" and on " Reactor coolant pressure boundary penetrating containment" neither a leak nor a break can be tolerated at certain critical l
locations without significant radiation release to the environment.
e The austenitic stainless steel exhibits leak-before-break l under normal loading. However, under adverse loading conditions on 1
the isolation condenser piping, comprising of seismic and/or waterhammer loads a major rupture may occur.
l e Jet impingement of steam at operating temperatures on to the 1
6
valve operator and itr control circuitry, raises environmental qualification concerns for the isolation valves.
e A loss of coolant accident (LOCA) outside the drywell has severe radiological consequences to station personnel. An. automatic isolation valve inboard to the containment would preclude a LOCA outside the containment.
e Necessary restraints to prevent pipe whip is required per NRC
, design criterion and should be provided, i l
I BEZERENCES l 1 B.F.Beaudoin, T.Hardin, and D. Quinones, " Leak-Before-Break Application in' Light-Water-Reactor Plant Piping". Nuclear Safety,Vol.30, No.2, April-June 1989.
- 2. Code of. Federal Regulations, 10 CFR 50 Appendix A.
l
- 3. American Society of Mechanical Engineers Boiler and Pressure Vessel Code,Section III, 1983 edition, Nuclear Power Plant l Components.
l l
l I
7
y Off-Site Dose Consequences Resulting from a Pipe Break Between the Drywell and' Isolation Condensers
.at the Oyster Creek Nuclear Generating Station
?ummary The purpose of this study was to determine first order estimates for off-site radiological doses at the Oyster Creek Nuclear Generating Station (OCNGS) from a pipe line break between the outside of primary containment (drywell) and the isolation valves for the isolation condensers. Such a pipe line break at OCNGS could result in the release of primary coolant from the reactor into secondary containment due to the lack of isolation ;
valves inside the drywell. The potential off-site radiological doses from a low probability accident from a single seismic event i would exceed the Nuclear Regulatory Commission's (NRC) reactor siting criteria (25 rem to the whole body, 300 rem to the <
thyroid) found in 10 CFR 100. A single failure break where the plant's safety systems are operational, the potential off-site doses under certain atmospheric conditions could result in thyroid doses in excess of the State's Radiological Emergency Response Plan (RERP) off-site protective action recommendation guidelines for evacuating the public (1 rem for whole body, 5 rem '
to the thyroid) which are a fraction of the 10 CFR 100 values.
These off-site doses could be substantially reduced or avoided by ths installation of isolation valves inside the drywell (primary containment).
The niajor findings from this study are summarized below.
- 1. Off-site doses from a low probability accident due to a single seismic event could possibly result in off-site doses with severe radiological consequences (i.e. stochastic and non-stochastic health effects from exposure to radioactive material) to the public in the vicinity of OCNGS.
- 2. A more probable event would be a single pipe line break with all of the plant's engineering safety systems operational.
Even for this type of accident, the resulting off-site doses could require the Department of Environmental Protection to recommend protective actions for the public.
[
- 3. The Licensee's most conservative meteorological conditions for accident assessment are not the most frequently occurring condition in the vicinity of OCNGS. Meteorological conditions used by the licensee include direction dependant I
atmospheric dispersion factors, higher than the average wind speeds and less stable atmospheric stability. These conditions result in off-site radiological doses that are approximately an order of magnitude lower than those doses calculated from the most frequently occurring meteorological conditions around OCNGS.
1 l
l l