ML20205K846
| ML20205K846 | |
| Person / Time | |
|---|---|
| Issue date: | 03/14/1986 |
| From: | Minogue R NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Sniezek J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| Shared Package | |
| ML20204J261 | List: |
| References | |
| FOIA-87-714, RTR-REGGD-8.032, TASK-OP-713-4, TASK-RE NUDOCS 8603250396 | |
| Download: ML20205K846 (2) | |
Text
- _ _
i ' g 6 - N g u.s m otstribution
/
- ogeg5 WKT ,
^ rD m ,+
g
..- .. w 'nRANs R0/suBJ -
, ' RMinogue
~O ORoss s .
KGo11er, D e
~y e Econti. , J. As :f. ,
Y# 14 D36 ,
der v erodsky ;:
4 . v.i .
MEMORANDUM FOR: James H. Snierek Acting Deput.y Executive Ofrector for
- Rcgional Operations & Generic Requirements .
FROM: Robert B. Minogue Director ~ . . , , , _
Office of Nuclear Regulatory Research
SUBJECT:
CRGR REVIEW OF REGULATORY GUIDE. OP /13-4, "CRITERIA FOR ESTABLISHING BI0 ASSAY PROGRAMS FOR TRITIUM" We would like to publish the subject regulatory guide in final form; a copy of che guide is enclosed as Enclosure A. This guide was not submitted to the CRGR for review befo.e publication for coment because it does not contain new recomendations that would place an additional burden on agency or licensee resources. The guide in fact contains recomendations that would alleviate the netd for most of the tritium bioassays now being perforced at U.S. nuclear power plants. However, I am now submitting this guide to you, as advised by Walter Schwink, to provide you the opportunity to decide shether a CRGR review is required before final publication.
There is now no official generic licensing guidence for power plants specifying coolant f ritfuni concentratfon values balow which bicassay for tritium is unner.essary, ih? proposed guide would sonify such ccneentrations (shown in the right hand colunms cf Tavle 1 (Enclosure 8) from the guide) above which bicassay programs should be provided. Thes, logically the same numbers provide concentrations below which tritium bicausy is corsidered unnecessary. These concentrations were derived from nbservations of the amounts of tritium in the urine samples of workers compared to the concentrations in reactor cooltnt, as described in the at.tached reference fmm Health Pnysics 33, pp. 9198,197/
/ Enclosure C). As indicated by the circled informtion in Table 1 of Enclosure C, U.S. nuclear power p hnts would not be required to coesider tritium bicassay ceasurements for workers if this guidance were fort.' ally
- romulgated by the NRC, since concentrations of tritiun in LWR coolat.t do not generally reach the concentration levels of Table 1 of the guide.
These concentration levels were originally derived while the PES staff was preparing a staff licensing position for NMSS (Enclosurc 0). RES staff at that time realized that NRR was requiring tritium bicassay in some technical specifications for nuclear power plants, and that many power plants were carrying out tritium biotssay of workers but finding very small or zero concentrations in urine. Therefore, RES staff carried out the evaluation of tritium concentration levels, as well as total activity amounts being handled when dilutter. is not sufficient to reduce probabilities of intake, for both NRR and NHSS licensing guidance. The NRR staff was included in the coment and concurrence stages _of_developrent for both the 1977 NMSS staff position and the OFC: ( [ 3. 86 fMh XA
.............). h.............:............:............:..........
NAME: \ : .
) : ) : :
DATE:
- '~m " : : : :
??..
MAR 14 C46 2-proposed guide. Coments of hMSS, NRR and IE have been satisfactorily incorporated into the proposed guide.
There is an aoJitional alleviatory aspect of the proposed guide for NMSS licensees. Since NMSS requested in 1981 that the 1977 staff position be reevaluated and developed as a formal regulatory guide, RES staff had to carry out studies of the radiotoxicity and hazard of working with many other types of tritiated compounds than HTO. The literature review, rationale for the approach in the currently proposed regulatory guide, and methods for converting urine concentrations and air concentrations to estimates of internal radiation dose, are presented in NUREG 0938 (1983) (Enclosure E). The literature review and analyses presented in NUREG-0938, which were discussed by RES staff with some of the nation's authorities in tritium radiobiology, provided justification for raising the levels of activity of tritiated nucleotide precursors for which bicassay should be conducted by a factor of ten, so the) could simply be combined with HTO into one category. (Compare Table 1 of the regulator guide (Enclosure B) with the table taken from the 1977 AMSS staf f position Enclosure F).)
Therefore, we are submitting Regulatory Guide OP 713 4 to the CRGR for infomation only. As advised by Walter Schwink, if we are not contacted by the CRGR within 15 days, we will proceed with publication.
Orisimal s! csd byi RC32"4T 3. V'.W93 Robert B. Minogue, Director Office of Nuclear Regulate *y Research
Enclosures:
A. Oraft Regulatory Guide OP 713 4 B. Table 1 C. Reference from Fealth Pnysics 33 ,
D. Staff Licensing Position for NHSS E. NUREG.0938 F. Table from NMSS Staff Position
- 9 cc: E. Hill i
n anna ...
(
......g OFC: RES:RRANB : RESi; S: RES S:R : :
... . . . . . . . . . . . . . . . . n.........
., 8 : R E S :....
.... 0 ......... ,
NAME: ABro k : RMA . . . . . . . .ti. . . . . . . . . . . . . :. DR
- KGoller . . . .s . . . . .: . . . .....
. . .. . . .k. . . oped : p.
. . . . . . . . . .tp. . .y........\%a nder : EC f DATE: 3/4/86 : 3/g/86 : 3/ /36 : 3//J/86 : 3@86 : 3/g/86 :
b
'L
. -. a. . . . _.
f-
.~
l ,
/ t '>
Ui 6b r i L
I
' ASSESSMENT OF YALUE-IMPACT ASSOCIATED WITH .".6IMIN ATION J
OF AR3TTRARY INTERMEDIATE BREAKS IN TIIE DESIGN OF NUCLEAR POWER PLANTS f}
I i q
i L Garry S. Holman i
I
, t l
l l
Nuclear Systems Safety Program 6
Lawrence Livermore N ational Laboratory I
f h
April 23,1986 f
- G%0 Y 'I9
.i _ a__ _ - - - - - - - - - -
_ _1_ -
. . . - I j * ~
g DRAFT Rev. O i ,
i r
t L'
i ABSTRACT i i Current regulations governing the design of nuclear power plants require that i j safety-related structures, systems, and components be designed to withstand the effects l l
i of loss-of-coolant accidents. To account for such effects as pipe whip, jet impingement, ;
and changes in local environment, nuclear power plant designers have been required to !
postulate ruptures in certain high-energy systems such as reactor coolant piping. The !
U.S. Nuclear Regulatory Commission, through its Standard Review Plan, rcquires that i j
pipe ruptures be ptulated in these systems at terminal ends and at all intermediate i locations where specific stress and fatigue limits are exceeded. If these limits are not l exceeded anywhere along a piping run, two "arbitrary intermediate breaks" must be j;
- i postulated at the points of highest stress even though these stresses may be much lower than allowable design values. Arbitrary in:ermediate breaks may represent over one hsif l
lf i4 of the total postulated breaks in a typical nuclear power plant, and have led to the installation of numerous "protective devices" such as jet shields and pipe whip restraints. An NRC task group evalueting the potential for pipe break recently j concluded that protective devices associated with arbitrary intermsdiate breaks can i
l
' , adversely affect plant operation without contributing to plant safety, and recommended
. that the arbitrary intermediate break requirement be removed. Tha NRC is presently
{ , proposing to revise the Standard Review Plan accordingly. Under the proposed setion,if ;
4 intermediato breaks could not be located according to present stress or fatigue limits, '
! arbitrary intermediate breaks would no longer be required. This would allow the dynamic l , effects of these breaks -pipe whip and jet impingement - to be disregarded as a basis for plant oesign. The proposed action would, however, still require the cafety-related equipment located near the affected piping be qualified for the "non-mechanistic" (
environmental effects of a postulated becak. This report presents a detailed analysis of
, g costs and benefits associated with the proposed action. The results of this analysis l
Indicate that substantial cost savings and reductioas in occupational radiation exposure ;
j would result from the proponed action without detrimentally affecting public health and ,
! , safety. !
- ' t
)
i
. i i
l !
i e l
I I'
i i
i I
1l-1 l 04/01/86
!! 586-044 i!
l' i
2 JL
% _._.. 1.;.1.i _ _ _
u __ - -
-o . _ _.
- f. . .
- - R ;v. O t
g DRAFT s
t CONTENTS l
A B ST R A CT. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . , . . . . . . . . . . . . . . . . . . . . . 11 r
LIST O F T A B L ES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv A C K N O W L E D G E M E N TS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
v
- 1. IN T R O D U CT I O N . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . '.
5
- 2. PROPOSED ACTION AND POTENTIAL ALTERN ATIVES . . . . . . . . . . . . . . .
7 i
- 3. AF F E CT E D D E CISION F A CTO RS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
8
9
- 5. UN QU ANTIFIED RESIDU AL ASSESSMENT . . . . . . . . . . . . . . . . . . . . . . . . . . . .
10
' 6. DEVELOP MENT OF QU ALIFICATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
11 l 6.1 P ubl i c H eal th . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Occupational Exposure - A ccidental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 6.2 O ccu pational E xposure - R outine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
) 6.3 O f f si te P roper ty D am age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 6.4 O nsi t e P roper t y D am age . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 6.5 Industry Im plem ent ation Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46 6.6 Industry O per ating C os ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 6.7 NRC Developm ent and Implementation Coets . . . . . . . . . . . . . . . . . . . . . 57 6.8 N R C O per ati on C os t . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57 6.9 54
, 7.
SUMMARY
54 f 7.1 D iscussi on o f R es ul ts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
E f f ect of M odif ying U sage F actor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
, 7.2
............. 60 R E F E R E N C ES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
APPENDIX At Proposed Revision to SRP 3.6.2 l
APPENDIX B: PlantsIncluded in Assessment i
k
\
f S86-044 - til - 04/01/f6 I
l l
, a
cn - -_;;;._ ; x.- _ =_- ~ - - - -
'o . e
' - Rcy.0 3 DEAFT LIST OF TABLES
- 1. Breakdown of excluded AIB locations for a sample PWR plant, including pipe whip
'I restraints and jet shields.
- 2. AIB locations and related protective devices excluded from plants in PWR data base.
- 3. AIB locations and related protective devices excluded from plants in BWR data base.
- 4. Arbitrary intermediate break initiating frequencies.
- 5. Accident sequences and release entegories used in AIB risk analysis (PWR plants).
[ 6. Accident sequences and release categories used in AIB risk analysis (BWR plants).
- 7. Occupational radiation exposure due to accidents.
- 8. Summary of plant maintenance activities affected by removal of pipe whip restraints, with related radiation dose estimates.
- 9. Overall avoided occupational radiation exposure.
- 10. Offsite (public) property damage,
- 11. Onsite property damage.
- 12. Summary of implementation costs for pipe whip restraints and jet shields.
i
! 13. Total avoided implementation costs.
F i 14. Summary of actual pipe whip restraint costs for a sample PWR plant.
- 15. Summa y of plant maintenance activities with related cost estimates.
- 16. Total avoided operating costs.
- 17. Summary of value-impact.
h f
S86-044 - iv - 04/01/86 l
..- ~ .,~e ~ ;;._ 2____2 __.; -
4
' - Rsv. 0 r(-DRAFT t
ACKNOWLEDGEMENTS i
This work was supported by the U.S. Nuclear Regulatory Commission Office of Nuclear Regulatory Research, Mcchanical/ Structural Engineering Branch. Dr. John A.
O'Brien was the NRC monitor for this project.
The value-impact assessment described in this report was based largely on information supplied to us by various scurces within the nuclear industry. In some cases,
{
considerable effort was spent on our behalf to develop information specifically for this study. The author would particularly like to thank the following organizations for their contributions:
i The Atomic Industrial Forum Public Service Electric Carolina Power & Light Sargent & Lundy Combustion Engineering Commonwealth Edison Southern California Edison i < Consumers Power Stone & Webster Duke Power Tennessee Valley Authority t
General Electric Texas Utilities Northeast Utilitics Virginia Power Pacific Gas & Electric Westinghouse Public Service of New Hampshire Washington Public Power i Supply System I
Their participation and continued interest in our tsuch programs is gratefully acknowledged.
The author would also like to thank Raman Pichumani of the NRC Office of Nuclear Reactor Regulation for his assistance in compiling past Industry requests for exemption from the arbitrary intermediate break design requirement.
i 5 I
i I
S86-044 -v- 04/01/86 i
i I
_d
RGv. 0
- - g DRAFT 4
- 1. INTRODUCTION The Code of Federal Regulations, Title 10, Part 50 (10CFR50), requires that r structures, systems, and components important to the safety of reaclear power plants in the United States be designed to withstand the effects of natural phenomena and the effects of normal and accident conditions [1]. The U.S. Nuclear Regulatory Commission (NRC), through its regulations, Regulatory Guides, Branch Technical Positions and the Standard Review Plan, has required that various accident loads and loads caused by natural phenomena be considered, both individually and in appropriate combinations, in the analysis of safety-related structures, systems, and components.
Designing safety-related structures, systems, and components to withstand the effects of loss-of-coolant accidents (LOCAs) is one important load requirement. To account for these effects, the design basis of nuclear power plants has historically included postulation of double-ended guillotine breaks (DEGBs) in certain high-energy systems such as reactor coolant piping. Responding to increasingly detailed regulatory criteria has required significant amounts of analysis to evaluate hydrodynamic loads and the resultant response of structures and mechanical components, and has led to the placement of massive pipe whip restraints and jet impingement barriers (or "jet shields")
near piping as protection against the dynamic effects of postulated pipe breaks. Nuclear power plant designers have generally contended that under most circumstances the likelihood of such pipe t>reaks is so low as to be considered incredible, and that their i postulation places unrealistically severe constraints on plant design. Many experts also believe that certain "protective devices" (i.e., pipe whip restraints and jet impingement barriers) may actually decrease the reliability of piping systems by limiting access to pipe welds, therefore reducing the effectiveness of in-service inspection, and by potentially restricting the movement of piping during routine operation and thus increasing pipe stresses caused by restrdnt of thermal expansion.
The NRC position on postulation of pipe ruptures is presented in the Standard Review Plan, Section 3.6.2, which includes Mechanical Engineering Branch Technical Position MED 3-1, "Postulated Rupture Locations in Fluid Sptem Piping Inside and Outside of Containment" [2,3]. The rules stated in this position are intended to utilize available piping design information for postulating pipe rupture at locations having relatively Sigher potential for failure, such that an sdequate and practical level of paotection may be achieved.
As discussed in the report of the NRC Piping Review Committee, Task Group on Pipe Break, years of operating experience indicate that piping failures generally occur at high stress and fatigue locations - terminal ends, weld joints, fittings, or component connections - where high stress concentrations and the most severe cyclic fatigue effects are expected (4). Many piping failures result from design, construction, or operational errors, or from unanticipated situations not originally designed for such as hydrodynamic loads (e.g., water or steam hammer) and corrosive environments. When originally developed, the position combined operating experience data with the then state-of-the-art understanding of pipe failure mechanisms.
The present NRC position requires that pipe breaks be postulated 0 in all "high-
- energy" piping systems (i.e., where design temperature exceeds 200 F or design pressure j exceeds 275 psig) at certain specific locations. For ASME Class 1 piping, these include the following:
S86-044 04/01/86
a .
yg a h*~\ DRAFT Rev.O t
e at all terminal ends, such as component connections, e at any location where the calculated primarpplus-secondary stress exceeds a prescribed limit, namely 2.4 times the allowabl,, stress intensity (Sm) as defined by Section111 of the ASME Code (5),
2L e at any location where the cumulative usage factor (which provides a measure of how much fatigue loading the pipe is subjected to) exceeds the prescribed limit of 0.1.
t f These criteria were devised to take into account the combined effects of stress and fatigue. Similar criteria are specified for ASME Class 2 and 3 piping (and to all seismicallrdesigned non-nuclear piping as well) except that fatigue is not a design consideration.
In the event that two intermediate break locations cannot be identified according to these criteria, SRP 3.6.2 requires that no fewer than two separated break locations (or t only one if the pipe run has no more than one change in direction) be chosen on the basis of highest stress. This "arbitrary intermediate break" (AIB) requirement was originally intended to compensate for any lack of knowledge about the precise location of breaks and about factors such as unanticipated loads or design and construction errors that muld affect the potential for pipe breaks. As a partial "fall anywhere" concept, the
' requirement for protection against at least two intermediate breaks was believed to enhance plant safety even though stresses at postulated AIB locations are of ten well below ASME Code allowable stresses.
The impact of this requirement on the design, construction, and operation of nuclear power plants has been substantial, and has brought into question whether the requirement actually contributes to overall plant safety. The major issue centers on "protective devices" (pipe whip restraints and jet shields) required near cach break location to protect structures and equipment from the highly dynamic phenomena that r would result from a postulated pipe bresk. Direct impacts of the arbitrary intermediate break requirement include:
o Complications in piping system design. Of the approximately 300 postulated break r
points typic'd of a modern PWR plant, over half result from the arbitrary break requirement (6). Although the associaled pipe whip restraints also represent over i half cf the total in the plant, the associated effort - and therefore the cost - of their design and installation represents a much higher percentage of the total cost for protection against pipe. This is because locating the highest stress points, and therefore the postulated breaks, is a highly iterative process. Unlike terminal end breaks, whose location is fixed as soon as plant layout and preliminary stress f analyses are completed, the intermediate break locations are first known only af ter i detailed piping and hanger design are finished. Even then, the high-stress locations I will of ten shif t during construction as piping layout or hanger details are changed
! to accommodate field interferences. This compileates the locating of structural embedments for pipe whip restraints, the allocation of space for supporting steel, and the location of safety-related equipment away from break locations.
S86-044 04/01/86 I
i
E._ - _ . . . - . . _ . __ - - - - - - - - -
3o o DRAFT Rev. 0 0
e Furthermore, the postulated break locations will depend to some degree on the mathematical model used to predict stresses, particularly its ability to treat branch piping, and the extent to which branch lines are included in the analysis of the piping system in question.
o Although SRP 3.6.2 provides criteria intended to minimize the need to relocate postulated break locations when the high-stress points shift, these criteria provide little relief in practice. The designer is still responsible for justifying that the shif t will not result in reduced plant safety, which leads to additional analytic effort as well as to possible redesign, f abrication, and installation of pipe whip restraints.
l e Increased plant contestion. Pipe whip restraints and jet shields are often very J complex, very mass ve steel structures that further congest the already cramped I confines of a typical nuclear resotor containment. As a result, plant maintenance activities, particularly routine imservlee inspection (ISI) of pipe welds, become less
, efficient and related costs and occupational radiation exposure (ORE) increase accordingly. In some cases. restraints must be removed to gain access to pipe welds or plant equipment and thee: --Installed, of ten to close tolerances.
3 Deoending on restraint size, type of mounting, rigging and scaffolding required and general accessibility, removal and reinstallation alone can seriously increase both maintenance costs and radiation doses to plant personnel; in some cases, they may f ar exceed those for the maintenance activity itself.
Increased plant congestion would also hamper recovery from unusual plant conditions. In the event of a radioactive release or spill, for example, decontamination operations would be more effective if protective device support structures, with their complex shapes, could be eliminated. Access for fire control '
in certain plant areas would also be enhanced by reducing the number of protective
! devices.
e Reduced pipin.t rel ability. Many experts argue that pipe whip restraints and jet ;
f impingement barriers may actually decrease the reliability of piping systems by limiting access to pipe welds, therefore reducing the effectiveness of imservice inspection, and by potentially restricting the movement of piping during routine
, , operation and thus increasing pipe stresses caused by restraint of thermal
! expansion. This latter situation could occur, for example, if pipe whip restraints .
l (which are designed to accommodate thermal expansion displacements without I
binding, and are later shimmed according to actual displacements measured during plant startup) were to be installed liieorrectly. The results of a separate NRC l ,
research program at LLNL indicated that piping system reliability does indeed increase with flexibility (7).
l '
- i; Increased heat loss. Thermal movement of piping during heatup and cooldown could
- o deform piping insulation through contact with a pipe whip restraint. To prevent this from occurring, insulation must be cut back in these areas, which creates
' convection gaps through which heat escapes to the containment interior. The heat loss from 6 inches of uninsulated pipe is equivalent to that from about 100 feet of insulated pipe. Consequently, the heat Icss associated with pipe whip restraints may as much as double the total heat loss inside of containment, not only reducing
- ' plant efficiency, but elevating the temperatures inside many containments to near lj technical specification limits.
S86-044 04/01/86 I
1
L _..... :_.. -- - - - - - - - - -- -
t, .
$ DRAFT Rev.0 These and other considerations have already led several nuclear [.ower plant licensees to request exemptions from the arbitrary break requirement. In the industry view, present knowledge and operating experience both indicate that designing for arbitrary intermediate breaks does not enhance plant safety and that therefore the associated costs and added personnel exposure are not warranted.
The Pipe Break Task Group concurred in evaluating the potential for pipe break that certain protective devices, particularly those associated with arbitrary intermediate
! breaks, can adversely affect plant operation without contributing to plant rafety.
' Furthermore, the Task Group specifically recommended that the arbitrary intermediate
[ break requirement be removed. The NRC is presently implementing this
}
recommendation through proposed modifications in the Standard Review Plan.
i This report presents a regulatory analysis of the proposed SRP modification, I following accepted NRC guidelines (8). The purpose of the regulatory analysis is to
,f evaluate costs and benfits associated with a specific regulatory action under the premise that a suitable technical basis exists for implementing the action. Although portions of the analysis (particularly estimated rLik associated with the propmed action) may also draw on this technical basis, the regulatory analysis itself is not intended as technical jmtification for the proposed regulatory action.
I t
i L
4 S86-044 04/01/86 P _ _
T_ . -. . - . . . . - _ . . . . - -. - -- -- -
?. -
\ DRAFT Rev.O r
- 2. PROPOSED ACTION AND POTENTIAL .6 LTERN ATIVES The NRC has proposed a revision to the Standard deview Plan, Section 3.6.2, "Determination of Rupture Locations and Dynamic EffIcts Associated with the l
Postulated Rupture of Piping," that would delete the present arbitrary intermediate
[ break requirement for all ASME Class 1, 2, and 3 piping ',ptems, as well as for other i seismically designed "high energy" piping. Under the preposed action, if intermediate
' break locations could not be identified according to preseat stress or (for Class I piping) usage factor limits, Branch Technical Position MEB 3-1 would no longer require postulated breaks at the two locations of highest stress. Appendix A presents the relevant sections of MEB 3-1 that would be modified; all other provisions of SRP 3.6.2 -
such as the requirement for postulated breaks at terminal ends- would remain in effect.
l One alternative to the proposed action would be to grant partial exemptions to the
! current Standard Review Plan; as mentioned earlier, several licensees have already applied for - and in some cases been granted - such exemptions. It can be argued, i j however, that extensive elimination of arbitrary intermediate breaks in this manner
- would not be an appropriate use of the exemption process. Pact operating experience
' combined with advances in analytical techniques suggests that it would be more
' appropriate to make general revisions in the Standard Review Plan at this time.
Another alternative would be to take no action pending implementation of recent ulemaking actions initiated by the NRC. The NRC has proposed modifying the requkements of 10 CFR 50, General Design Criterion 4 (UDC-4), "Environmental and Missile Design Bases," to exclude dynamic effects associated with postulated pipe ruptures from the plant design basis. This rulemaking is proceeding in two phaseJ beginning with an "interim" rule change applicable only to PWR reactor coolant loop t'
piping [9). Current NRC planning calls for eventually extending the interim rule into a "broad-scope" rule which would si nilarly apply to any piping system for which the probability of rupture could be shown to be negligibly low. The effects of the proposed y
rule change, including associated costs and benefits, were evaluated in a regulatory analysis performed by LLNL [10).
The proposed broad-seope rule, however, would exclude all breaks in a piping system, including those "high stress" locations unaffected by the proposed revisions in SRP 3.6.2, and would therefore require extensive analyses to demonstrate compliance l
with the criteria for break exclusion. The proposed Standard Review Plan revision, on the other hand, could provide substantial- and perhaps most importantly, immediate -
relief to the industry, technically justified by information now available.
Equipment Qualification One issue raised in connection with the proposed SRP revision is the cf tect that
. elimination of arbitrary intermediate breaks wBl have on the qualification of safety-related mechanical and electrical equipment. Under the proposed action, dynamic effects associated with arbitrary intermediate breaks, such as pipe whip and jet impingement, would no longer have to be considered. The proposed action would, however, stDI require that safety-related equipment located near the affected piping system be qualified for the "non-mechanistic" effects of a postulated rupture.
S86-044 04/01/86 4
k
g _ _ _ _ _ . . _ - -
.e Revo 0
)* ; -DRAFT l .
t Non-mechanistic effects include the changes in sub-e mpartment environment -
bulk pressure, temperature, and humidity - that would occur if a pipe rupture occurred at some unspecified location along the piping run within the compartment. Current industry practice is to que.11ty equipment on the basis of that break location which causes i the worst environmental effect on safety-related equipment. The time-dependent t changes in local environment are calculated by performing a mass and energy balance on the affected region, based on "characteristic" mass and energy release rates such as
- those presented in the American Nuclear Society Standard ANS-58.2 [11]. The release rates in ANS-53.2, for example, are transient profiles based on the layout and geometry i of the affecting piping run as well as on thermal-hydraulle conditions. Such profiles are t not intended, however, for long-term mass aM energy releases, or for thrust loads and jet impingement forces, neither of which would be considered under the propmed action.
Because the proposed qualification requirement already reflects current industry l
practice, the proposed SRP revision would have no real effect on equipment qualification. Consequently, no values or impacts associated with equipment qualification are considered in this assessment.
I k
l h
l I
i f
i S86-044 04/01/86 B
A
w .... - - .. -- .: -
s *
'. DRAFT Rev.0
, 3. AFFECTED LECISION FACTORS l Major deelslon factors affected by the propmed regulatory action and included in
- this value-impact assessment are summarized as follows
Causes Causes Causes I Decision Factors Quantified Unquantified No Change Change Change I Public Health X l Occupational Exposure (Accidental) X Occupational Exposure (Routine) X
, Public Property X X l Onsite Property 1 Regulatory Ef ficiency X Improvements in Knowledge X l
Industry Implementation Cost X Industry Operating Cost X l X NRC Development Cost NRC Implementation Cost X NRC Operating Cost X i
4 i
\
)
S86-044 04/01/86 t
, L,___ _ - _ _ 2, __ - --
= __ __ . .
n .
i 1 DRAFT R0v. 0 e
i
- 4. VALUE-lMPACT ASSESSMENT
SUMMARY
Best High Low Estimate Estimate Estimate Values (marr rem)
Public Health -1.6 -7.Re+ 2 -3.9 e-2 Occupational Exposure (Accidental) -7.3e-2 -1.le+2 -8.8e-4 Occupational Exposure (Routine) 1.7 e+ 4 4.2e+ 4 2.2e+ 3 Values Sub-Total 1.7 e+ 4 4.2e+ 4 2.2e+ 3
_- t i l i
Impacts ($) j Industry Implementation Cost -15e+6 -35e+ 6 -3.1 e+ 6 !
, Industry Operating Cost -2 6e+ 6 -88 e+ 6 -2.5e+6 .
NRC Development,Implemt.ntation 0 0 0 NRC Operating Cost 0 0 0 Power Replacement Cost 0 0 0 t Off aite Property Damage 1.7 e+ 3 8.7e+ 6 2.le+1 [
t Onsite Property Damage 1.0e+3 1.5e+6 1.3e+ 1 i impacts Sub-Total -41 e+ 6 -105e+ 6 -5.6e+3 1 !
l ;
Notes:
I
- 1. Total for 115 plants. See text for discussion of relevant assumptions.
I 2. Property damage costs reflect a 10% discount rate. !
f !
i.
t I
i t ,
S86-044 04/01/86 !
I 4
l h
- DRAFT Rcy.0
- 5. UNQUANTIFIED RESIDUAL ASSESSMENT The following decision factors were not quantified in the assessment of the proposed action:
o Regulatory effielency. The current use of plant-specific a .emption* to the regulations entails a significant allocation of NRC resources for bot: te& deal ar i j legal effort. It is anticipated that substantially less effort would be requi .c.Iader
, the proposed action, resulting in improved regulatory of ficiency.
o Improvements in knowledge. The proposed SRP revision would offer substantial motivation to the industry to eliminate postulated pipe breaks now reqt. ired under current regulations. Implementation of the proposed action would promote further investigations to determine under what conditions and for which # ping systems artatrary intermediate breaks could be eliminated. Such investiget..ons could also provide important insight into the extent to which more general exclusion of pipe breaks - such as would be permissible under the broad-scope GDC-4 rule change -
could potentially be applied.
r r
P i
t r
u i
! S86-044 04/01/86 i
f 5
Ji
i .. . _ i . _ ._ . -_ ~ - _
r i DRAFT Rov. 0
- 6. DEVELOPMENT OPf0 ALIFICATION, j This section presents the detailed development of value-impact for the proposed
" action. The assessment is based on the following general assumptions:
I (1) A total of 80 PWR plants are considered, of which 62 are operatliig and 18 are under construction effective December 31, 1985. Appendix B gives a vendor-by-vendor breakdown of operating plants and plants under construction. The 80 plaris have a total remaining lifetime of 2689 plant-years (py), assuming a design lifetime of 40 years, and an overall average remaining lifetime of 33.6 years (31.8 years for j operating plants only). Plants under construction have an average forward-fit (time to operation) of 1.7 years.
(2) A total of 35 BWR plants are considered, of which 31 are operating and 4 are under construction effective December 31, 1985. Appendix B gives a vendar-by-vendor breakdown of operating plants and plants under construction. The 35 plants have a total remaining lifetime of 1157 py, and an overall average remaining lifetime of 33.1 years (32.2 years for operating plants only). Plants under construction have an average forward-fit of 1.3 years.
(3) The combined population of 115 plants has a total remaining lifetime of 3846 plant-years, and an overall average remaining lifetime of 33.4 years (31.9 years for operating plante only). The low forward fits for both PWR and BWR plants imply that all nort-operational units are in advanced stages of construction; therefore, for purposes of this regulatory analysis, the average forward-fit for all plants has been assumed to be zero.
i (4) The only values and impacts considered in this analysis are those uniquely associated with elimination of the present AIB regulrement. Values and impacts f resulting from the general elimination of pipe breaks as a plant design basis (such as avoided costs r.nd ORE associated with pipe whip restraints at nort-AIB locations) are addressed in the GDC-4 regulatory analysis and are not included here.
It is also important to note that, as mentioned earlier, the utility owilers of several
- plants have already applied for, ard in some cases been granted, exemptions from the arbitrary intermediate break requirement (12-23]. Strictly speaking, these plants should not be included in the regulatory analysis because they would not be affected by the proposed action unless additional AIB locations (i.e., not included in the exemption request) were later eliminated. Several utility owners responding to our request for supporting data also indicated that they did not foresee taking any action at this time even if the SRP revision were implemented. Strictly speaking, these plants sMuld also be excluded from the sample considered. Ilowever, because we do not have detalled data l, for all plants concerning which would and which would not take advantage of the SRP modification, and because AIB identification is an ongoing process subject to future change, we have elected to retain both groups of plants in the sample to obtain an indication of the potential implications of the proposed action.
S86-044 04/01/86 t
U
~ --
~C._._ _a-=
L. .
'. D2 AFT Rev.0 6.1 Public Health The changes in risk and core melt frequency associated with the proposed action were developed by first estimating the probability that a pipe would rupture at a postulated arbitrary intermediate break location. This result was then applied to appropriate accident sequences to estimate the probability of core melt and the resultant pubile risk from radioactive releases.
We based our assessment of risk r",ociated with elimination of arbitrary intermediate breaks on the following general assumptions e the risk associated with any given AIB depends solely on the diameter of the affected piping. This approach is analogous to that taken towards lose-of-coolant accidents in WASH-1400 [241, which divided pipe breaks into three size ranges
~
(1/2" 1 D 1 2"), These size ranges were used ln the present analysis, which allowed the generic accident sequences and release categories from WASH-1400 to be applied in the present analysis, e for any given plant, the probability of having at least one AIB in a given size range is represented by the number of postulated AIBs in that size range multiplied by a generic break probability dependent only on p(pe diameter.
e because plant safety systems are presumably designed against pipe break effects (such as changes in containment environment), an A!B itself contributes no
> additional risk. In this regard, it is important to note that the proposed action st01
> requires that safety related equipment near the break location be qualified for non-meet anistic break effects (i.e., bulk pressure, temperature, humidity). Dynamic i effects associated with the AIB, such as pipe whip and jet impingement, would no longer need to be considered.
only breaks inside of containment contribute to plant risk. l e
The number of A!B locations generleelly applicable to all plants was developed separately for PWR and BWR plants (including uncertainty bounds) based on plant-specific estimates supplied to us by several plant owners. Note that the approach in effect adds a set of
new" break locations to the total population of potential breaks inside of containment.
Thus, the resultant risk represents a true "delta" risk associated with the proposed action.
I
- It would, in principle, be "more accurate" to evaluate risk on a system-brsystem basis. This, however, would ideally require for each system that (1) a detailed fracture mechanies evaluation be performed to estimate rupture probabilities based on normal operating and postulated accident conditions, and (2) the effect of postulated breaks on
, overall plant safety be assessed. Table I shows a sample - but not necessarily I representative ~ system-by-system breakdown of AIB locations and related protective t
devices excluded from one of the PWR plants used as a basis for our assessment. Tables 2 and 3, whleh we developed from industry exemption requests and from detailed data j supplied to us show for PWR and BWR plants, respectively, how the number and sizes of excluded breaks (and the number of excluded protective devices as well)can vary widely 9
S86-044 04/01/86 f
s I
- Rcy. 0
, g -DR AFT from plant-to-plant. The number of systems potentially involved, as well as plant-
! specific variations in both the number and characteristics of the affected systems (which, in turn, affects the number of pmtulated AIBs) clearly make such an approach impractical.
1 The break size distributions in Tables 2 and 3 form the basis on which we estimated the number of excluded AIDS in our subsequent value-impact evaluations. The plants listed in Tables 2 and 3 are those for which data was provided in sufficient detail to develop the AfD size distributions; they are not, however, the only plants that we f received AIB data for. A comparison of the detailed data in Tables 2 and 3 against e equivslent but less detailed information for other plants (e.g., total excluded AIBs, total I protective devices eliminated) indicated that these distributions should be reasonably j representative of all PWR end BWR plants.
Probability of Pipe Break Wo based "best estimate" probabilities of rupture at postulated AIB locations on the results of prior LLNL probabilistic studies of reactor coolant piping reliability in PWR and BWR plants (25-28). These evaluations focussed on estimating the probability of a
, double-ended guDlotine break (DEGB) in PWR reactor coolant loops and in BWR main steam, feedwater, and recirculation loop piping. To estimate the probability of DEGB, LLNL considered two causes of pipe breaki pipe fracture due to the growth of cracks at welded joints ("direct" DEGB), and pipe rupture indirectly caused by the seismically-induced failure of critical supports or equipment ("indirect" DEGB).
The probability of direct DEGB was estimated using a probabilistic fracture mechanics model that calculated the growth of as-fabricated surface flaws at welded
- joints, taking into account loads on the piping due te normal operating conditions as well
! as seismic events. Other factors, such as the capability to detect cracks by non-j destructive examination and the capability to detect pipe leaks, were also considered. A detailed evaluation of Westinghouse plants yielded a best-estimate system probability of direct DEGB of 1.0e-12/py for plants east of the Rocky MDuntains, with a 90th-percentile value (i.e.,90% confidence limit) of 1.0e-10/pyl this latter value also bounded the direct DEGB probabilities for west coast plants and for Combustion Engineering plants. Although the probability of direct DEGB was not explicitly estimated for
( Babcock & Wilcox plants, a review of reactor coolant loop stress information inferred that the probability of crack-induced pipe break should be similarly low. Therefore, this estimate was assumed applicable to all PWR plants.
The results of the LLNL fracture mechanles evaluation of reactor coolant piping in the Brunswick Mark I plant indicated that for main steam and feedwater piping, as well as for recirculation loop piping without intergranular stress corrosion cracking (IGSCC),
the system probabilities of direct DEGB are similar. The results of the Brunswick evaluation are of interest in this regulatory analysis because several sizes of large-bore piping, ranging from 12 to 28 inches nominal diameter, were considered as well as the smaller-diameter (4 inch) recirculation bypass piping. Based on these results and on our past experience with related probabilistic evaluations, the following appear to be reasonable estimates of the annual probability of rupture per excluded break location:
S86-044 04/01/8G e
u
- - - - - - ~ - - - - - ~ - - - ~
?, DRAFT Rev.0 t
e D > 6" 1.2e-12/AIB yr o 2" 1 D 1 6" 1.4e-10/AIB-yr e 1.4e-9/AIB-yr f" 1 D 1 2" The large- and small-bore break probabilities are taken directly from our PRAISE evaluations of BWR recirculation piping (no IGSCC) and represent the highest estimated break probability in each size range. Because we presently have no PRAISE results for piping less than four inches in diameter, we assumed - based on our judgement - that
, the probability of a small-small AIB is one order of magnitude higher than that for the I
smallest pipe diameter for which PRAISE results are available.
Our PRAISE results showed that DEGB probabilities at certain weld joints are of ten two or more orders of magnitude less than the system DEGB probability. We therefore i
regard the above break probabilities as conservative because AIBs are postulated only at those locations in a piping system , defined (by not exceeding specified stress or usage factor limits) as being least susecptible to the high thermal fatigue stresses that i contribute meet to the likelihood of pipe rupture. In other words, a piping system would
[
be more likely to fall at some postulated location other than an AID, a location (such as a terminal end) that would remain unaffected by the proposed SRP revision. To account i for this effect, we assumed that the low estimate of break probability for each size range was simply one order-of-magnitude less than the corresponding best-estimate probability.
The high estimate of break probability is based on t!.e NRC position stated in the proposed GDC-4 rule change. The GDC-4 would allow exclusion of dynamic effects associated with double-ended pipe breaks when the their probability of break can be demonstrated to be negligibly low. In this case, "negligibly low" is defined as a s3 stem i rupture probability on the order of le-6 per year. We divided this probability by four break locations (two AIDS plus two terminal ends) to establish the higtrestimate of per-AIB break frequency, regardless of pipe diameter. Note that this contervatively applies equal weight to the AIB and terminal end break locations.
We then developed overall probabilities of AIB occurrence (with upper and lower bounds) by multiplying the s ngle-break probability for each :ize range by the corresponding number of excluded breaks gi'en in Tables 2 and 3 for PWR ard BWR I plants, respectively. The resultant "initiating event" frequencies, summarized in Table 4, therefore take into account both the number of breaks ard the single-break probability
- for each size range. Note that the best-estimate initiating event frequencies in Table 4
- reflect the "best-estimate" per-AIB probabilities combined with the mean number AIDS excluded from the plants in our sample. The higtr and low-estimate initiating event frequencies in Table J combine, respectively, the high- and low-estimates of single-break
+ probabuity disetssed earlier with the higtr and low-estimates (mean value f, one standard
+
deviation) of the number of excluded breaks.
l It is important to note that although we regard these initiating event frequencies as "generic", the frequencies for different size ranges cannot be combined into a meaningful single frequency for an individual PWR or BWR plant, nor can PWR and BWR results be combined into a meaningful single frequency for all plants. The frequeneles must be S86-044 04/01/86 l
Ji
a___ _ _ _ . _- _
t, ,
', DRAFT R ;v. 0 I
considered separately because the accident sequences and release categories used to estimate core-melt freg'iency and public risk are different for each plant type and each size range.
Core Melt Frequency These results were combined with WAsil-1400 accident t,equences for large, small,
- and small-small LOCAs to estimate the change in core melt frequency, 6 F, resulting I from implementation of the proposed action.
For purposes of illustratiori, given that a large LOCA occurs inside of a PWR
' containment, the relevant accident sequences leading core melt, with their per reactor-year frequencies of occurrence, a:e as follows (see Table Sh AB with frequency = 1.2e-9/ry ACD with frequency = 6.0e-ll/ry AO with frequency = 9.le-9/ry AliF with frequency = 1.2e-10/ry AD with frequency = 2.0e-6/ry All with frequency = 1.0e-6/ry AF with frequency = 1.0e-8/ry ADF with frequency = 2.0e-10/ry The overall probability of core melt due to a large LOCA is therefore 3.le-6/ry. WASil-l 1400 assumes a large LOCA frequency of 1.0e-4/ry; the conditional probability of core melt is therefore 0.031 per large LOCA event.
I Similarly, the overall WASil-1400 probabilities of core melt due to a small LOCA
! and a small-small LOCA are, respectively, 6.le-6/ry and 1.7e-5/ry (see Table 5). Ilere WASil-1400 assumes a small LOCA frequency of 3.4e-4/ry and a small-small LOCA frequency of 1.0e-3/ry (see WASil-1400, Appendix 111). Thus, the conditional probabilities of core melt due to a small or a small-small LOCA are about 0.0T8 and 0.017, respectively.
Multiplying these conditional probabilities of core melt by the large , small- and i small-small AIB frequencies from Table 4 therefore yields best estimate core-melt frequencies of i
i e D > 6" 1.le-12/ry e 2" 1 D 1 6" 8.0e-ll/ry 8.3e-10/ry 31" 1 D 1 2" e
for a total best-estimate core-melt frequency of 9.lt-10/ry for PWR plants.
The estimated frequencies of core melt resulting from large , small- and small-small AIBs for BWR plants are similarly derived from the WASil-1400 accident sequences in Table 6. Combining these with the appropriate estimated break probabilities yields best-estimate values of:
S86-044 04/01/86 1
y p.__ _ _ . . _ _ _ _ -u.-___. --
-- _ -m. - _ . . _ _ _ . _ _ ___
Rey, 0 g DRAFT e D > 6" 4.0e-14/ry e 2" 1 D 1 6" 8.le-13/ry e O f"1 D S 2" for a total best-estimate core-melt frequency of 8.5e-13/ry. Note that this result is .
significantly lower than that for PWR plants, reflecting not only the greater number of l AIB locations excluded from PWR plants in general, but also the fact that no small-small l AIBs (which dominate the overall probkbility of core melt) were identified for BWR plants. [
[
Summing the results for PWR and BWR plants yields the following core--melt frequencies associated with the proposed action l Best estimate = 9.l e-10/ry l i
High estimate = 8.5e-7/ry Low estimate = 2.3e-ll/ry i I where the high and low estimates reflect the high and low estimates of large , small ,
and small-small AIB frequency previously derived. i i
f Public Risk We estimated public risk associated with the p oposed action by assuming that the ;
WASH-1400 release sequences for large , small , and small-small pipe breaks inside of [
PWR and BWR containments apply. As an illustrative example, the dominant large i i LOCA release sequences for PWR plants (from WASH-1400, Table V.3-14, Appendix V) are:
ABS (PWR-1) with frequency = le-ll/py , t ACDs (PWR-1) with frequency = 5e-ll/py AGs (PWR-1) with frequency e 9e-ll/py AFs (PWR' 1) with frequency = le-10/py r
AB7 (PWR-2,' with frequency = le-10/py l AB-6 (PWR-2) with frequency = 4e-ll/py !
AHF7 (PWR-2) with frequency = 2e-ll/py ads (PWR-3) with frequency = 2e-8/py l AHe (PWR-3) with frequency = 1e-8/py l' AF-6 (PWR-1) with frequency = le-8/py AG-6 (PWR-3) with frequency = 9e-9/py ACD4 (PWR-4) with frequency = le 11/py I 1 I g ADS (PWR-5) with frequency = 4e-9/py t AH4 (PWR-5) with frequency = 3e-9/py [
l 886-044 04/01/86 l
i I
I
L._ - - - . . . _ _- -
!' . DRAFT R;v.0 I
AB-c (PWR-6) with frequency = le-9/py AliF- c (PWR-6) with frequency = 2e-10/py i A DF- c (PWR-6) with frequency = 2e-10/py L
A D- c (PWR-7) with frequency = 2e-6/py AE- c (PWR-7) with frequency a le-6/py l WASil-1400 assumes a median large LOCA frequency of 1.0e-4/py. If we replace this value with the previously estimated probability of a large-bore AIB (3.Ce-ll/ry), the best estimate risk from the occurrence of at least one large-bore AIB becomes:
Risk = (9.5e-17/ryX5.4e+6 man rem) + (6.le-17/ryX4.8e+6 man-rend +
' (1.9e-14/ryX5.4e+6 man-rem) + (3.8e-18/ryX2.7e+6 man rem) +
' (2.6e-15/ryXI.0e+6 man rem) + (4.9e-16/ryX1.5e+5 man-rem) +
! (1.le 12/ryX2.3e+3 man-rem)
= 1.le-7 man-rem /ry The best estimates of risk associated with small- ard small-small AIDS in PWR piping are similarly developed using the small- and small-small LOCA release frequencies, respectively, from WASil-1400 (see Table 5). As for estimating coce-melt frequency, the W ASil-1400 small- and small-small LOCA frequencies (3.4e-4/ry and 1.0e-3/ry, respectively) were replaced in the present analysis by the appropriate AIB frequencies derived from the PRAISE results, 6
The best-estimate of per-plant risk due to any arbitrary break inside of a PWR l containment, regardless of size, becomes:
j I Risk = 1.le-7 man-rem /ry + 9.le-6 man-rem /ry +
6.0e-4 man-rem /ry
= 6.le-4 man-rem /ry Multiplying this result by the total remaining lifetime of PWR plants (= 2689 py) results in a best-estimate total public risk from the propmed action of 1.6 man-rem.
Equivalent results for BWR plants were developed by combining the estimated BWR r break frequencies with the appropriate BWR release sequences from WASil-1400 (see Table 6). The resultant best-estimate of per plant risk is 4.9e-6 man-rem /py, which when multiplied by the total remaining lifetime of BWR plants (= 1157 py) yields a total public risk of 5.7e-3 man-rem from AIDS in BWR piping.
Summing the results for PWR and BWR plants therefore yields the following incremental increases in total public risk resulting from the proposed actioru Best estimate a 1.6 man-rem s
liigh estimate = 7.8e+ 2 man-rem Low estimate = 3.9e-2 man-rem S86-044 04/01/86 6
A _ . .. .
- . '. DRAFT R;v. 0
'i .
Note that this analysis results in an increase, albeit small, in the estimate of public risk. Note also, however, that the high estimate of risk reflects both NRC minimum reliability guidelines (i.e., system failure probability on the order of le6/yr) and the climination of more than 13,000 postulated breaks over the entire plant population considered.
I i
l k
i 9
i r
S86-044 04/01/86 4
i -
.- - ~ ~ , _ - _ = .
.' l
=: ? j
"~~
Table 1. Breakdown of escluded AIB locations for a sample PWR plant, l1 including pipe whip restraints and jet shields. f{
i %
Piping System Location Pipe Safety Material design Design Breeks Restraints Jet Shields Size Class Presrire Temperature Eliminated Eliminated Eliminated
- 1. REACTOR COOLANT 1.1 Surge line IC 14 1 SS 2485 600 1 0 0 1.2 Pressurizer sprey IC 4 1 SS 2485 650 3 0 0 1.3 Bypass piping N/A 1.4 Branch lines N/A 1.5 Safety /PORY inlets N/A 1.6 Other 1C 2 1 SS 2485 650 4 0 0
- 2. MAIN STEAM 2.1 Primary lines IC 30,31 2 CS 1185 600 8 14 4 2.2 Branch lines N/A 2.3 MSIV drains N/A 2.4 Auxiliary steem N/A 2.5 Atmospherie dump OC 6 2 CS 1185 600 8 1 0 2.6 Other N/A
- 3. FEEDWATER IC Set 8 12 4 3.1 Main feedwater 16 2 CS 1185 f 3.2 Aux feedwater OC 4 2 CS 1185 600 8 0 0 l 1
3.3 Other N/A
~
+- l Tatde 1 (cont.). Breakdown of excluded AIB locatione for a sample PWR plant, including pipe whip restralrts and jet shields. l b
Loestion Pipe Safety Material Design Design Breeks Restraises Jet Shields Piping System Eliminated Size Class Pressure Temperature Eliminated Eliminated i!
- 4. SAFETY INJECTION 4.1 Core spray N/A IC 6 SS 2405 A50 2 0 0 4.2 LPCI 1 IC 1-1/2 1 SS 2405 650 7 0 0 4.3 HPCI 0 OC 4 2 SS 2485 650 2 0 4.4 RCIC N/A IC SS 2405 650 5 0 5 4.5 St Accumulators le 1 4.6 Other N/A
- 5. RESIDUAL HEAT N/A REMOVAL
- 6. CHEMICAL AND VOLUME CONTROL IC 3 2 SS 2405 650 6 0 0 6.1 Charging 2 SS 2405 650 5 0 0 OC 4 SS 2405 650 1 0 0 .
OC 3 2 2 SS 2405 650 6 0 0 OC 2 2445 650 2 0 0 6.2 Letdown IC 3 1 SS h
IC 4 2 O SOS 400 2 0 0 600 400 0 0 IC 3 2 SS 1 IC 2 2 SS 600 400 1 0 0
y =-n. .
. = - - -- - - , - . - . . _ _
~.+ 'r 1
3' *:
Tetne 1 (cont.). Breakdown of excluded AIB locations for a sample PWR plant, including pipe whip restraints and jet shields. Il Piping System Location Pipe Safety M aterial Design Design Breaks Rest v.ts Jet Shields Size Class Pressure Temperature Eliminated Elimas.4ted Eliminated l.
OC 3 2 SS 600 400 6 0 0 Seal injection IC 2 SS 2485 650 4 0 0 6.3 1
!C 2 2 SS' 2485 650 12 0 0 4; OC 3 2 SS 2485 650 3 0 0 OC 2 2 SS 2485 650 10 0 0 6.4 LoopfE1 N/A 6.5 Other N/A
- 7. REACTOR WATER N/A CLEANUP STEAM GENERATOR IC 3 t CS 1185 600 8 0 1
- 8. ,
600 20 0 1185 1 l BLOWDOWN IC 2 2 CS OC 3 2 CS 1185 600 5 0 0 OC 2 2 CS 1185 600 2 0 0 OC 4 2 CS 1185 600 6 0 0
- 9. STEAM GENERATOR WET LAYUP l t
- 10. WASTE EVAPORATOR N/A STEAM SUPPLY h
- 11. OTHER SYSTEMS N/A
- i1
. - - _ - . , . . . - . . - - . . , - , , - - , . - , - - - _ , - . - , -------,----n.,~ ~- , -.- , -. , - -
._.----. -- n ,
q.
4 .f Tatde 1 (cont.). Breakdown of exclanded AIB locations for a semple PWR plant, fr.du-f g pipe wh;p restreints and jet shields. }
i a.
Breaks Restraints Jet Shields Eliminated Eliminated Eliminated f t
IC D>S 22 fg6 13 l, 2<D<6 23 0 2 i IC 48 0 0 IC 1/2 1 6 12 f CUB-TOTALS 0 0 0 OC D>5 OC 2<D<5 44 I O 18 0 0 OC 1/2 1 6<2 1
155 27 15 II TOTALS
Ii
!I i!
N(YTES:
Safety class: 1, 2, 3, or NNS Location: Inside ewtainment GC) or oestside containment (OC) ll Pipe size: Nominal diameter in w% '
Pipe material: Carbon steel (C3) or stanc*ess steel CBS)
Temperature: Design temperature in *F ,i Pre-sure: Design pressure in poig
(
l I
_~__ _ , _ .- ,_ _ _ __ -_ _ _
t Table 2. AIB locations and related protective devices excluded from plants in PWR data base.
AIB Locations GC)
Plant Name Units AIB Locations (OC) Total AID locations ;
LLOCA SILOCA S2LOCA Total 20 38 8 66 123 189 Beaver Valley 2 1 40 18 70 128 30 158 Eyron 1,2 2 40 18 70 128 30 158 ,
Draidwood 1,2 2 1
32 78 5 83 j Catawba 1,2 2 25 21 i
25 21 32 78 5 83 McGuire 1,2 2 22 23 48 93 62 155 Seabrook 1 I i
62 38 4 104 132 236 South Texas 1.2 2 17 68 7 92 90 182 Vogtle 1,2 2 26 51 21 98 5' 152 CNP1 1 i
32 34 98 55 153 Hest estimate (per unit) 32 ,
50 60 157 102 259 liigh estimate (per unit) 47 47 Low estimate (per unit) 17 14 8 39 0 Notes:
- 1. Location: Inside containment QC), outside containment (OC)
Size: LLOCA (D > 6"), S11LXX (2" < D < 6"), S211XA (1/2" < D < 2") ~- ~
- 2. liigh and low estimates refle< t+ lo averaged over all tiiilts
. ._ . l _. . __ L . _ _ . _ _ _ . . _ ,,_ _ .__
f 4,
Table 2. AIB locations and related protective devices excluded from .
i plants in PWR data base. !
l f
Pipe Whip Restraints GC)
Units Restraints (OC) Total Restraints Plant Name t
LLOCA SILOCA S2LOCA Total 10 9 47 52 99 ;
Beaver Valley 2 1 28 i
4 42 23 65 l Byron 1,2 2 37 1 l
4 42 23 65 Braidwood 1,2 2 37 1 34 5 40 18 58 Catawba 1,2 2 1 34 5 40 18 58 CcGuire 1,2 2 1 0 0 26 1 27 Seabrook 1 1 26 24 22 0 46 N/,. N/A South Texas 1,2 2 39 2 89 21 110 Vogt!e 1,2 2 48 N/A N/A N/A N/A N/A N/A c P1 1 t
19 4 48 16 64 !
Best estimate (per unit) 25 33 6 84 30 114 11Igh estimate (per unit) 42 11 2 13 Low estimate (per unit) 8 2 1 Notes:
- 1. Location: Inside containme-t GC), outside containment (OC)
Size: LLOCA (D > 6*), S111XA (2" < D 16"), S21 IIA (1/2" 1 D 12")
- 2. liigh and low estimates refleet +15 averaged over all units
- 3. N/A not determined at present
- n -- ...-.- -.-- _.- .- - _-
. l Tatde 2. A1B locations and related protective devices excluded from I
plants in PWR data base. [
i Jet Shields GC)
Plant Name Units Jet Shidds (OC) Total Jet Shields LLOCA SILOCA S2LOCA Total Beaver Valley 2 1 15 7 3 25 18 43 2' e 0 e e e o Oyron 1,2 Braidwood 1,2 2 0 0 0 0 0 0 Catawbe 1,2 '
, 5 4 1 18 5 15 McGuire 1,2 2 5 4 1 10 5 15 13 2 0 15 0 15 Seabrook 1 1 3 South 1 exas 1,2 2 8 9 L 0 N/A N/A Vogtle 1,2 2 N/A N/A N/A N/A N/A N/A WMP1 1 N/A N/A N/A N/A N/A N/A i Best estimate (per unit) 4 2 1 7 4 !! ,
d High estimate (per unit) 9 4 2 15 3 24 Low estimate (per unit) 0 0 0 0 0 0 i;
Notes:
10 Location: Inside containment OC), outside e.:etaineneet (OC) s*
Size: LLOCA (D > 5"), 3114XA (2* < D < 4"), 1210CA (1/2" < D < 2")
~ - -
- 2. High and low estimates reflect + 1 o averaged over all units
- 3. N/A not detertained at present
~ . , - . - . - . _ _ __
,o -1
_ _ __ __ ~ -
o .
P Table 3. AIB locations and related ges xtive devices emeluded from plants in BWR data tune.
AIB Locations GC)
Units AIB Locations (OC) Total AIB Locations Plant Name LLOCA SILOCA 52LOCA Total 33 12 0 45 0- 45 Clinton i 27 4 0 31 50 01 Hope Creelc 1
/
6 4 0 10 0 10 ;
La Salle 2 i 6 0 24 25 24 Beet estimate fper unit) 16 i
32 10 0 42 50 42 i Hl.h estimate (per ufA) 6 Low estimate (per unit) 4 2 0 6 0 Notes: ,
- 1. Location: Inside containment GC), ovalde contalmnent (cc)
Size LLOCA (D > s'), SitiXA (2" < D 1 s'), 7Mf1FA (1/2"1 D < 2")
i
- 2. High and low estimates refleet +b averaged over all units
,e-&+w m +---ee----ww,r. rr - . , .- - - - - - - --v--_ --,,-----,.e- --=r,r e -<-r- ,- - ,---wer_ ~v ---m-~ -w-. . - - - - - + . ---31.ym--- ,. -,--#- .. ,
.- , _ -~ - ,_ _ ..-.- . _ _ , . _
Table 3 (cont.). AIB locations and related protective delices excluded from plants in BWR data base. I
\ .
Pipe Whip Restraints GC)
Restraints (OC) Total Restraints Plant Name Units LLOCA SILOCA S2LOCA Total i
2 0 37 0 37 Clinton 1 35 0 0 12 12 Hope Creek 1 0 0 0 11 2 6 5 0 11 La Selle 15 3 18 ~
Best estimate (per unit) 12 3 0 0 13 9 22 High estimate (per unit) 8 5 -
0 0 1 0 1 Low estimate (per unit) 1 t
Notes: ,
- 1. Location: Insi& containment GC), outside containment (OC)
Size: LLOCA (D > 6"), SilJ00L (2" < D 16"), S2tiXA (1/2" 1 D 1 2")
- 2. High and low estimates reflect +10 averaged over all units
~
l t
i i
.. _. _ . , , - ~. . , - -_ -. _ . .
p Table 4. Arbit ary intermediate break initiating frequencies.
Size Range Total AIB j Locations S2LOCA 0 "IY}
LLOCA SILOCA AIB locations per PWR plant (by size range) 32 32 34 98 Best estimate 47 50 60 157 liigh estimate 8 39 17 14 Low estimate AIB locations per BWR plant (by size range) 18 6 0 24 Best estimate 32 10 0 42 liigh estimate 0 6 4 2 Low estimate Estimated break frequencies UAIB yr) 1.2e-12 1.4e-10 1.4e-9 Best estimate 2.5e-7 2.5e-7 2.5e-7 liigh estimate Low estimate 1.2e-13 1.4e-Il 1.4e-10 Initiating event frequencies Ury), PWR plants 2.8e-11 4.5e-9 4.8 e-8 :
Best estimate 1.2e-5 1.3 e-5 1.50-5 liigh estimate 1.le-9 Low estimate 2.0e-12 2.0e-10 Initiating event frequencies Ury), BWR plants 2.2e-l l 8.4e-10 0 Best estimate 8.0e-6 2.5e-6 0 liigh estimate Low estimate 4.8c-13 2.8e-Il 0
V r
Table 5. Accident sequences and release categories used in AIB risk -
analysis (PWR plants).
L i
I LARGE LOCA ACCIDENT SEQUENCES AND RELEASE CATEGORY FREQUENCIES (D > 6")
Release Category Accident WASH-1400 Sequence J Accident *W PWR-5 PWR-6 PWR7 I PWR-1 P WR-2 P WR-3 PWR-4 le-9 1.2e-9 AB le-Il 1.4e-10 le-11 Se-Il !
ACD Se-Il 9.le-9 AG 9e-Il 9e9 le-10 1.2e-10 AHF 2e-Il 2e-6 2e-8 4e-9 2e-6 AD le-6 le-6 le-8 3e-9 AH le-8 !
ADF
'4.9e-8 le-Il 7e-9 1.3e-9 3e-6 3.le4 Totals 2.5e-10 1.6e-10 WASH-1400 SILOCA fi%.xy Ury) 1.0e-4 LLOCA core melt frequency Ury)e 3.le-6 Conditional core reelt probability: .031 J
!, , , - - , ,wn,- -.
- - , --.r. :--- - , - , . - ,- , , , - . < , .-- r v- . . - - , - . , , -, . . . -. - - --
-_-=-;-
c .
,,,_y
/ 0 Table 5 (cont.). Accident sequences and release categories used in AIB l risic analysis (PWR plants).
4 SMALL LOCA (SI) ACCIDENT SEQUENCES AND RELEASE CATEGORY FREQUENCIES (2" < D16") s.
Release Category Accident l WASH-1400 Sequence Accident N "' N PWR-4 PWR-5 PWR-6 PWR-7 rEl PWR-2 PWR-3 2e-9 2.5e-9l SIB 3e-11 Se-10 8e-11 7e-11 1e-11 SICD 3e-8 SIF 3e-10 3e-8 3e-8' SIG 3e-10 3e-8 ' 4.6e-10 6e-11 4e-10 SlHF 6e-9 3e-6 3e-6 SID 3e-8 Se-9 3e-6 3e-6 SlH 3e-8 3e-10 3e-10 ,
SIPF 1.le-8 2.3e-9 6e-6 6.le-6 Totals 7e-10 5.6e-10 ' t.2e-7 le-Il t7 ASH-1400 SILOCA frequency Vry): 3.4H SILOCA core melt frequency Ury)c 6.lH Conditional core melt probability: .016 ,
4
. s.
~
Table 5 (cont.). Accident sequences and release categories used in AIB -
risk analysis (PWR plas.*1). .
l SMALL LOCA CSI) ACCIDENT SEQUENCES AND RELEASE CATEGORY PREQUENCIES (2" 1 D 16") !
l WASH-1400 Release Category Accident !
Accident Sequence l M1 PWR-2 PWR-3 PWR-4 PWR-5 PWR-6 PWR-7 NY S2B le-10 1.4e-9 8e-O 9.5e-9 S2F 1e-9 1e-7 i _y S2CD le-10 2e-8 2e-8 S2G SE-10 SE-8 9.lE-8 S2C 2E-9 2E-6 2E-6 S2HF 2e-10 le-9 1.2e-9 S2D 9e-8 2e-8 Se-6 9.le-6 S2H 6e-8 le4 6e-6 6.le-6 S2DG le-12 le-12 Totals 2.2e-8 1.6e-9 2.3 H le-12 3e-8 2.9e-8 1.5e-5 1.7e-5 UASH-1400 S2LOCA fr%acy Vry,\ le-3 S2LOCA core mielt frequency Ury): 1.7e-5
. Conditional core mielt probability * .017 l
l 1
Table 6. Accident sequences and release categories used in AIB risk -
analysis (BWR plants).
LARGE LOCA ACCIDENT SEQUENCES AND RELEASE CATEGORY FREQUENCIES (D ) 6")
WASH-1400 Release Category Accident Sequence Accident
"'"'I 8*9"*" BWR-1 BWR-2 BWR-3 BWR-4 2e-9 4e-8 le-7 1.4e-7 AE le-10 2e-9 1e-8 1.2e-8 AJ le-10 2e-9 le-6 1.2e-8 AH1 le-10 2e-9 le-8 1.2e-8 Al 6e-I1 AGJ 6e-11 7e-10 7e-10 AEG AG5fl Se-Il 6e-Il 2.3e-9 4.6e-8 1.3e-7 8.2e-10 1.8e-7 Totals WASH-1400 LLOCA frequency (/ry): 1.0e-4 LLOCA core melt frequency Ury): 1.8e-7 Conditional core melt probability .0016 I
^ ^ ' ^ ' ' - ^ -
--- --m_____ _ W
Table 6 (cont.). Accident sequences and release categories used in AIB risk analysis (BWR plants).
i SMALL LOCA (SI) ACCIDENT SEQUENCES AND RELEASE CATEGORY FREQUENCIES s2" < D (6")
WASH-1400 Release Category Accident Sequence Accident F m uewy J 8*9""** BWR-1 BWR-2 BWR-3 BWR-4 2e-9 8e-8 te-7 1.8e-7 SIE 3.7e-8 i
4e-10 7e-9 4e-8 4.7e-8 l Sll 4e-10 6e-9 2e-8 2.6e-8 SlHI 2e-10 SIC 2e-10 2e-10 2e-10 SIGJ le-10 SIG1 te-10 2e-10* 2e-10 SIEG 2e-10 SIGH 1 2e-10 3.le-9 le-7 1.9e-7 Se-10 2.9e-7 Totals WASH-1400 SILOCA frequency Ury): 3.4e-4 SILOCA core melt frequency Ury): 2.9e-7 Conditional core melt probability .00086 u
- - - - - - . -n..-. . . . . . . . -- -- - _
Table 6 (cont.). Accident sequences and release categories used in AIB l'
risk analysis (BWR plants).
l 1
In SMALL-SMALL LOCA (S2) ACCIDENT SEQUENCES AND RELEASE CATEGORY FREQUENCIES (3 1 D Release Category Accident l WASH-1400 Sequence Accident F m wney N""** BWR-1 BWR-2 BWR-3 BWR-4 i
i le-9 2e-8 8e-8 le-7 j S2J 1.l e-7 I
S 21 le-9 2e-8 9e-8 2e-8 9e-8 1.le-7
8e-9 8e-9 S2C l S2CG 6e-Il 6e-ll 6e-10 6e-10 l 32GHI 3e-10 3e-10 S2EG 6e-10 i
6e-10 i S2GJ 2e-10 2e-10 1 S2GI
)
3.5e-9 7.4e-8 3.le-7 1.8e-9 0.9e-7 Totals WASH-1400 S2LOCA frequency Ury): le-3 S2LOCA core melt frequency Ury): 3.9e-7
- Conditional core melt probability: .00039 1
l I
f I
l
,_ . . _ _ _ . . . . _ _ . . . _ _ - . . _ _ . _ _ . _ _ . _ _ .: ._ . _ ~ r- '
DRAFT Rev.0 6.2 Occupational Exposure - Accidental The increased occupationcl exposure from accidents can be estimated as the product of the change in total core-melt frequency and the occupational exposure likely to occur in the event of a major accident. The r.ominal reduction in core-melt frequency was estimated to be 9.le-10/ry. The occupational exposure in the event of a major
, socident has two components. The first is the immediate exposure to personnel on site
' , during the span of the event and its short-term control. The second is the long-term exposure associated with cleanup and recovery from the accident. The incremental occupational exposure due to an accident is calculated as follows:
DTOA = NTDOA; DOA = (A F)(DIO + DLTO) where:
DTOA = total accidental occupational dose N = number of af fected facilities .
T = average remaining lifetime DOA = accidental occupational dme per plant-year AF = change in core-melt frequency {
DIO = immediate occupational dose i DLTO = long-term occupational dose I.
Table 7 presents the resultant occupational exposure due to accidents, based on cleanup and decommissioning estimates given in NUREG/CR-2800 [29].
i i
I I;
l S86-044 04/01/86 l
p-. - - - - - . _ . -
r, .
DRAFT Rev. 0 Table 7. Occupational radiation exposure due to accidents.
i Best High Low Estimate Estimate Estimate Increase in core-melt frequency (/py) 9.le-10 8.5e-7 2.3e-11 Immediate dose (man-rem / event) le3 4e3 0 I
,! Long-term dose (man-rem /e fent) 2e4 3e4 le4 j - . .
I Total exposure (men . rem) 7.3e-2 1.1 e2 8.8e-4 e
h h
l i
? !
h ,
?
I.
I S86-044 04/01/86
{. .
- DRAFT Revo0 6.3 Occupational Radiation Exposure - Routine The proposed action would allow elimination of jet shields and pipe whip restraints now required as protection against pwtulated AIBs. This would result in potentially a significant reductions in occupational radiation exposure (ORE), in particular by (1) avoiding, where applicable, the need to remove and replace pipe whip restraints blocking access to welds during in-service inspection (ISI) and (2) by generally relieving plant congestion, making maintenance operations more ef ficient.
Quantitative estimales of routine avoided ORE are presented only for elimination
} of pipe whip restraints; in general, the industry sources providing input for our study did '
not quantify dose reductions specifically associated with elimination of jet shields. Note that only restraints excluded inside of containment would contribute to reductions in personnel radiation exposure.
Table 8 summarizes the sources of routine occupational expmure considered in this l evaluation; it is not, however, necessarily intended as an exhaustive list of all sources of avoided ORE resulting from the propmed action. The specific activities described below and their associated dme estimates represent a composite of information provided by ,
several industry sources - most of whom only quantified overall ORE reductions - and from past LLNL value-impact assessments. Note that the following dose values are best estimates; high and low estimates in Table 8 reflect a 50 percent uncertainty range, ;
except where noted otherwise. l i
In-service inspection of piping welds. Pipe whip restraints restrict access to piping welds for routine in-service inspection. As a result, personnel exposure is increased because the restraints must of ten be removed and then reinstalled to perform ISI. Even when restraints are specifically designed so that ISI can be performed without their removal (as is the case for severallater-generation plants), their presence still reduces efficiency and therefore causes workers to remain longer in high radiation areast Based on dose values used in the prior ODC-4 evaluation lid, we have assumed that each excluded restraint would reduce exposure (due to either cause) by 1.0 man-rem per ISI.
Assuming a 10-year inspection interval implies that about 4.0 man-rem would be avoided .
per excluded restraint over a 40 year plant lifetime, or about 200 man-rem total if 50 restraints were eliminated inside of containment.
Routine restraint maintenance. It is anticipated that restraints wD1 be visually inspected t
i once every five years, resulting in 0.5 man-rem per restraint total exposure over the 40-1 year plant lifetime based on the experience of one plant owner providing Input to our j study.
Restraint gap verification. It is anticipated that restraint gaps will be verified every ten l years. Assuming that exposure averages 0.1 man-rem per verification implies an dme of 0.4 man-rem per restraint over a 40-year plant lifetime [ld. )
On the whole, these results imply ORE reductions per excluded restraint of 2 to 6 man rem, which, when taken together with the projected numbers of excluded restraints, generally agrees with the overall dose estimates provided by the industry sources.
S86-044 04/01/86
, ', -_ DRAFT R 0v. 0 I
In eddition to these specific activities, elimination of pipe whip restraints and jet shields would relieve congestion inside of containment and generally improve the t efficiency of maintenance activities. The associated ORE reductions are difficult to quantify, being highly sensitive to plant-specific variations in the type and frequency of maintenance activities performed. Consequently, avoided ORE associated with reduced plant congestion was left unquantified in this study. It can be generally said, however, that the proposed action would reduce worker exposure from any maintenance activity by about ten percent (10).
Reduced plant congestion would also reduce personnel exposure in certain "non-routine" situations, such as recovery from unusual plant conditions. In the event of a radioactive release or spill, for example, decontamination oput.tions would be more port structures, with
- effective - and their complex personnel shapes, were exposure eliminated.les - if protective Access device w[n certain plant areas for fire control I would also be enhanced by reducing the number of protective devices.
l 1 The dose values in Table 8 apply both to operating plants and plants tmder construction, with the exception that the total per-restraint avoided dme would have to be reduced by a one-time dose incurred for restraint removal and disposal The best ,
estimate value in Table 8 (0.1 marr-rem) assumes that the restraint is abancbned in place. In other words, only the shims are removed - disabling its restraint function-along with limited structural steel if necessary to improve access for ISI. The high estimate in Table 8 is assumed to be twice this value. Note that if the restraint were to be removed along with its supporting steel the resultant ORE would be much higher (on the order of 1 man-rem or more). The low estimate assumes that the restraint is left alone until some other reason (such as scheduled maintenance) causes its removal, af ter which it would simply not be reinstalled.
l Development of Overall Avolded ORE
, We developed separate per-plant estimates of avoided ORE for PWR and BWR t plants by combining the per-restraint dose utimates described above with the respective
, numbers of pipe whip restraints excluded inside containment only (see Table 9). We then multiplied these results by the number of affecteo PWR and BWR plants to obtain the
- following overall estimates of avoided ORE resulting from the proposed action:
Best estimate = 1.7e+4 man-rem l High estimate = 4.2e+4 man-rem Low estimate = 2.2e+3 man-rem where the high and low estimates reflect unce:tainty in both the per-restraint dose values Q50 percent) and in the number of excluded restraints. Note also that in j developing the overall avoided ORE that the per-plant values Ior operating plants were prorated by the ratio of the number of remaining plant-years to total plant life (::33/40).
i S86-044 04/01/86
- ~
Table 8. Summary of plant maintenance activities affected by removal of pipe whip restraints, with related radiation dose estimates (man-rem / device).
Operating Plants Construction Plants Maintenance Activity Best High Low Best High Low Estimate Estimate Estimate Estimate Estimate Estimate
- 1. In-service weld inspection 4 6 2 ,
4 6 2
- 2. Routine restraint maintenance .2 .3 .1 .2 .3 .1
- 3. Gap verification .4 .6 .2 .4 .6 .2
- 4. Removal ar.S d.sposal .1 .2 0 0 0 0 (implemeatation dose)
Total avoidec ORE (man-rem / device) 4.5 6.7 2.3 4.6 6.9 2.3 Notes:
- 1. All values cumulative over a 40-year plant life.
- 2. Values given are for pipe whip restraints only.
Table 9. Overall avoided occupational radiation exposure.
PWR Plants BWR Plants Best High Low -
Best High Low Estimate Estimate Estimate Estimate Estimate Estimate Avoided ORE Excluded restraints 48 84 11 18 22 1 (in containment only)
Operating plants (man-rem / plant) 216 563 25 81 147 2.3 Construction plants (man-rem / plant) 212 580 25 83 152 2.3 Total Avoided ORE (man-rem) PWR Plants BWR Plants All Plants Operating plants 62 31 93 Construction plants 18 4 22 Average remaininglife 31.8 32.2 31.9 (operating plants only)
Best estimate (man-rem) 1.5e4 2.4e3 1.7e4 High estimate (man-rem) 3.8e4 4.3e3 4.2e4 Low estimate (man-rem) 1.7e3 5.4e2 2.2e3
~
_m _
- - . . 3m a 4 ea - --- '-m --- -- --
t
. , DRAFT Rev.0 I 6.4 Offsite (Public) Property Damage I
, The effect of the proposed action on the risk to offsite property is calculated by i multiplying the change in accident frequency by a generic offsite propety damage
. estimate. This estimate was derived from the mean value of CRAC2 calculations, i
! assuming an SSTI release (major accident), for 154 reactors [30]. CRAC2 includes costs
, for evacuation, relocation of displaced persons, property decontamination, loss of use of I contaminated property through interdiction, and crop and milk Icsses. Litigation costs, i
. Impacts to areas receiving evacuees and irstitutional costs are not included. The damage
. estimate is converted to present value by discounting at 10 percent. A five percent -
discount rate was also considered as a sensitivity case.
The following discounting formula is employed:
-It g -It f
D,e -e V 1
) . where:
[ D = discounted value
- i Y = damage estimate
, tg
= years before reactor begins operation; 0 for operating plants tg = years remaining until end of plant life I = discount rate For purposes of this assessment, no distinction is made between operating and planned reactors; the everage remaining life of the entire population of 115 reactors is 33.4 years. The 10% discount factor is therefore 9.6, the 5% discount factor 16.2. The risk to offsite property, VFP, is estimated as:
VFP = N(A F)D where N aM A F are, respectively, the number of plants and the change in core-melt frequency. The raults are summarized in Table 10; upper and lower bounds are values calculated in Ref. 30 for Indian Point 2 and Palo Verde 3 coupled with the bounds on the core-melt frequency, i
l l
S86-044 04/01/86 c
I e
i
- i. .
DRAFT Rey, 0
.j- .
Table 10. Offsite (public) property damage.
5% discount facton 16.2 10% discount facton 9.6 -
Best High Low Estimate Estimate Estimate i
i Change in core-melt frequency (/py) 9.l e-10 a.'.e-7 2.3e-11 Offsite property damage ($/ event) 1.7e9 9.2e9 8.3e8 Discounted property r}amage ($/ event) !
5% discount rate 2.8e10 1.5 ell 1.4e10 10% discount rate 1.6e10 8.9e10 8.0e9 Value of offsite propety damage ($)
5% discount rate 2.9e3 1.5e7 3.4el 10% discount rate 1.7e3 8.7e6 2.lel 886-044 04/01/86
DRAFT R ;v. O l
6.5 Onsite Property Damage The effect of the proposed action on the risk to onsite property is estimated by multiplying the change in accident frequency by a generic onsite peoperty cost. The /
generic onsite property cost in NUREG/CR-2800 was taken. Costs included are for t
interdicting or decontaminating onsite propety, for replacement power, and for the capital cost of damaged plant equipment. Onsite property damage oosts were discounted
, using the following formula:
f y=ml 2 I {,-It } {(1 - e-I*) [1 - e ]}
where:
D = discounted value V = damage estimate m = ' fears required for plant recovery (= 10 years) tg
= years before plant begins operation; O for operating plants tg = years remaining until end of plant life
! = discount rate
, For purposes of this assessment, no distinction is made between operating and planned reactors. The 10% discount factor is therefore 6.1, the 5% discount factor 12.8. The risk to onsite property was estimated as:
lyt YOP= N( AF)U t
l '
where:
VOP= value of avolded onsite property damage l N = number of affected facilities AP = change in accident frequency q
U = present value of property damage occurring with frequency AF i
The results are summarized in Table 11; the uncertainty bounds reflect af,50% spread in
- the generic propety cost coupled with the bounds on core-melt frequency.
t E
886-044 04/01/86 t
-3
- - -DRAFT - R ey, 0 4
Table 11. Onsite property damage.
Recovery time (yr): 10 5% discount facton 12.8 10% discount facton 6.1
. Best Higti Low Estimate Estimate Estimate l
l I
Change in core-melt frequency (/py) 9.le-10 8.5e-7 2.3e ll Onsite property damage ($/ event) 1.7e9 2.5e9 8.2e8 I
Discounted property damage ($/ event) ;
5% discount rate 2.2e10 3.2e10 1.le10 10% discount rate 1.0e10 1.5e10 5.0e9 Value of onsite property damage ($)
5% discount rate , 2.3e3 3.le6 2.6el 10% discount rate 1.0e3 1.564 1.3el i !
l I
f l
S86-044 04/01/86
\ '
i Y _ _ _ _ _ . . _ _ _ - -
i <
t.
DRAFT Rev.O s
6.6 Industry Implementation Costs i Significant cost benefits to industry are anticipated as a result of the proposed action, primarily for avoided costs associated with pipe rupture protection devices (pipe whip restraints, jet impingement shields). Table 12 presents a breakdown of estimated costs assumed appilcable to all protective devices, that is, jet impingement shields as
' well as pipe whip restraints. Cost items applicable to new plants or to plants under construction include the following:
1 e Design engineering. All engineering costs for device design, including civil engineering cmts, draf ting, and field follow during construction.
o Hazard engineering. Analysis costs for determining postulated AIB locations, evaluating pipe whip and jet impingement loads, and target response, as well as miscellaneous piping analysis. These costs include iterative analysis costs to redefine break points in the event, for example, that field interferences cause l piping to be rerouted. They do not include non-mechanistle analyses of i environmental ef fects (pressure, temperature, humidity) caused by pipe break.
h Quality assurance follow during construction, y e Other manpower costs. l miscellaneous manpower and paper emts. ,
e liardware ard fabrication. Device fabrication, including material and other g hardware costs.
i t
Installation. Device installation during construction. These costs rnay of ten not be e
incurred until very late stages of plant construction.
The per-device implementation costs presented in Table 12, including the high and low estimates, are generic estimates provided to us by indstry sources.
I l For operating plants, no implementation costs other than those for device removal j
{ and disposal would apply. Note in this case that this would be an actual one-time cost borne by the plant owner (i.e., a negative avoided cost). The best estimate in Table 12 i' assumes that pipe whip restraints would be abandoned in place, with only the shim packs removed to prevent any possible binding with the pipe during thermal espansion. The high estimate is assumed twice this value; note however that if the device were completely removed together with its supporting steel the cost would be much higher.
The low estimate of zero assumes that the device would be lef t in place until some other ,
reason (such as routine maintenance) required its removal, af ter which it would not be reimtalled. In this case, the removal cost would t.e regarded as an anticipated part of ,
the maintenance cost and therefore not associated with the proposed regulatory action. ,
For construction plants, we have assumed that break location is complete, and that protoative devices have been designed and fabricated but not yet installed. In other j words, all plants are assumed to be in late stages of construction, as implied by the low average forward fits for the plant population considered.
I S86-044 04/01/86 l
1
m
, DRAFT Rev. 0 Development of Overall Avolded Implementation Costs t
- We developed per plant avoided implementation costs by multiplying these estimates by the number of pipe whip restraints and jet shields presented in Tables 2 and 3 (see Table 13). These results were then multiplied by the number of affected PWR ar.d BWR plants to obtain the following estimates of overall avoided implementation ecsts:
Best estimate = $15 million i
. High estimate = $35 mDllon Low estimate = $3.1 million where the high and low estimates reflect not only the high and low cost estimates from Table 12, but the hiah and low estimates in number of excluded protective devices as '
well. ;
I As noted earlier, we estimated industry implementation costs on the basis of l generic per-device costs assumed applicable to all pipe whip restraints and jet i i impingement shields. In reality, the cost of pipe whip restraints can vary widely 4
- depending on size, complexity, and the operating characteristics of the piping system I
{ with which it is associated. Table 14 presents a summary of actual pipe whlp restraint costs for a sample PWR plant, all for high-energy piping systems. Note that even g restraints for small-diameter piping can cost much more than the approximately $50K per device assumed in developing the implementation costs in Table 13. Consequently, the values developed in this analysis offer a conservative estimate of the actual avoided l costs associated with the propcsed act:on.
Recall also that the avoided implementation costs were estimated assuming that all analysis as well as all design and fabrication of pipe whip restraints was complete. For new plaats, of course, this would not be the case. Using the per-restraint cost estimates from Table 12 and the excluded restraint distributions from Tables 2 and 3, we estimated the following per-plant cost savings for future plants Best estimate = $3.6 mDllon fligh estimate = $9.2 mulion Low estimate = $0.5 mD11on As before, the actual cost would depend heavily on the number of excluded break ;
locatious.
i l
l 886-044 04/01/86
.- . . _ . ... = - ._ - - -. .
~- - -
Table 12. Summary of implementation costs for pipe whip restraints and jet shields ($K/ device).
I Operating Plants Construction Plants New Plants Costitem Best High Low Best High Low Best High Low j stimate Estimate Estimate Estimate Estimate Estimate Estimate Estimate Estimate
- l. Design engineering Wa da da 1 2 0 7 10 5 j 2. Hazard engineering da da da 5 7 4 20 26 17 i
- 3. QA, design follow, da da da 2 4 1 2 4 !
l miscellaneous
- 4. Materials and da da da 0 0 0 3 12 8 fabrication Installation da da da 10 15 8 10 15 8 5.
- 8. Removal and -2 -4 0 Wa da da da da da dsposal ,
Total costs ($K/ device) -2 -4 0 18 28 13 48 67 39
}
1 Notes i 1. Costs apply to all protective devices (Le., pipe whip restraints, jet impingement shields,.
l 2. For construction plants, it is assumed that design, procurement, and fabrication are complete.
I I
4 l
I
, , . - - , , ,--,r ,r - - - . , -
, ,, e ., , - - - m. p -w w ,
m
- m. . -....,.r ...
~
Tatdo 13. Total avoided implementation costa. **
PWR Plants BWR Plants Excluded Devices Best High Low Best High -Low Estimate Estimate Estimate Estimate Estimate Estimate Restraints eliminated 64 114 13 18 22 1 Jet shields elimiested 11 24 0 0 0 0 Total protective devices 75 138 13 16 22 1 -
Impleasentation Costs ($K/ plant)
Operatlag plants -150 -552 0 -38 -88 0 Construction planta 1350 3864 169 324 616 13 New (future) plaats 3600 9244 507 884 1474 39 TotalImplementation Costs PWR Plants BWR Plants All Plants Operating planta 82 31 93 Constreetion ;plaats IS 4 M Best estimate ($) 1.5e7 1.Se5 1.5e7 High estimate ($) 3.5e7 -2.Se5 3.5e7 Low estimate ($) 3.0e6 5.3e4 3.1es e r--ww.---w- - - , - , , , , ,- ,, . - , - . - . y. _y _ _ , , . . , , , , , , , , _ _ _ _ ,,,7 , , . , , _ . , _ , _ _ . , , , _ , , , , _ _ _ , , ,_ _
- . DRAFT Revo0 Table 14. Summary of actual pipe whip restraint costs for a sample PWR plant.
Average High Low Pipe Diameter Restraints Ccet Cost C ost
($K) ($K) ($K) 2" 3 57.6 79.3 39.3 !
l 3" 4 80.3 126 31.4 4" 25 77.7 102 36.1 6" 17 44.8 80.5 17.2 12" -
2 96.2 100 92.0 16" 5 27.1 37.6 18.1 32" 3 124 168 95.8 l
l l
l .
l l
l l
886-044 04/01/86 :
t
F
- f. ,. Rev. 0
', -DRAFT
! l' G.7 Industry Operating Costs The proposed action would allow elimination of jet shields and pipe whip restraints now required as protection against postulated AIBs. This would result in potentially significant reductions in plant operating costs, in partleular by (1) avoiding, where j applicable, the need to remove and replace restraints blocking access to welds during in-i service inspection (ISI) and (2) by generally relieving plant congestion, making maintenance operations more efficient. Restraint exclusion would also re&ce accordingly the number of convection gaps in pipe insulation, minimizing heat loss and improving operating efficiency.
, Quantitative estimates of avoided costs are presentml only for elimination of pipe whip restraints; in general, the industry sources providin nput for our study did not quantify cost reductions specifically associated with elimi on of jet shields. Note that I restraints excluded both inside and outside of containment would contribute to cost reductions.
, Table 15 itemizes specific maintenance activities considered in this evaluation that j would be affected by the removal of pipe whip restraints; it is not, however, necessarDy
, intended as an exhaustive list of g avoided costs associated with the proposed action.
The specific activities described below and their associated cost estimates represent a
, composite of information provided by several industry sources - most of whom caly quantified overall operating cost reductions - and from past LLNL value-impact
- assessments. Note that the following cost values are best estimates; high and low estimates in Table 15 reflect a +50 percent uncertainty range, except where noted
- otherwise.
f In-service irspection of piping welds. Based on industry estimates, we have assumed that j each excluded restraint would reduce ecsts (either for restraint removal and reinstallation or by improving access for ISI) by $1K per ISI. Assuming a 10 year inspection interval implies that about $4K would be saved per excluded restraint over a 40-year plant lifetime, or about $600K totalif 150 restraints were eliminated.
J Routine restraint maintenance. It is anticipated that restraints will be visually inspected 0 once every five years, resulting in a total cost of 0.5K per restraint over the 40-year
[ plant lifetime based on the experience of one plant owner providing input to our study.
L Restraint rap verification. It is anticipated that restraint gaps will be verified every ten years. Assuming an average cmt of 0.2K per verification implies a total avoided cost of 0.8K per restraint over a 40-year plant lifetime [lG.
These results imply per-rmtraint cost reductions of $3K to $11K which, when taken i together with the projected numbers of excluded restraints, yield avoided costs of about
$450K to $1.6 mulion if 150 restraints are excluded. These values generally bracket the overall cost estimates provided by the industry sources. Improved insulation of piping ;
could save an additional $600K over the 40-ycr.r lifetime of each plant by reducing heat ,
loss at the excluded restraint locations (id, in addition to these specific activities, elimination of pipe whip restraints armi jet '
shields would relieve congestion inside of containment and generally improve the S36-044 04/01/86
r __ _
- f. .
R%0 DRAFT efficiency of maintenance activities. The associated cost reductions are difficult to quantify, being highly sensitive to plant-specific variations in the type and frequency of maintenance activities performed. Consequently, avoided costs due to reduced plant congestion were lef t unquantified in this study. It can be generally said, however, that the proposed action wWd reduce the cost of any maintenance activity by about ten percent [10].
~
The cost figures in Table 15 apply both to operating plants arxl plants under construction when figured over a 40-year plant lifetime. Note that costs of restraint removal and disposal are implementation costs included in Tables 12 and 13.
Development of Overall Avoided Operating Cuts We developed separate per-plant estimates of avoided operating costs for PWR and
- BWR plants by combining the per-restraint cost estimates dtscribed above with the respective numbers of pipe whip restraints excluded inside com tinment only (see Table 16). We then multiplied these results by the number of affected PWR and BWR plants to L obtain the following overall estimates of avoided ORE resulting from the propmed action
I Dest estimate = $26 million fligh estimate = $88 mallon '
i Low estimate = $2.5 million 2
! where the high and low estimates reflect the uncertainty in both the per-restraint costs ,
Q50 percent) and in the number of excluded restraints. Note also that in developing the j overall avoided ORE that the per-plant costs for operating plants were prorated by the 4 ratio of the number of remaining plant-years to total plant life (=33/40). -
Note that the overall avoided operating costs actually exceed the overall avoided l t !
l implementation costs (except for low estimate). This is because where only construction '
+ plants would benefit from avoided implementation costs, operating plants as well as construction plants would benefit from reduced operating costs (assuming that pipe whip
- restraints were removed from operating plants). The relatively few construction plants (as a percentage of the total plant population) keep the overall implementation costs low I even though the per-plant implementation costs are significantly greater than operating i costs.
i I
04/01/86 S86-044 6
- 4,7 1.,. .- . . . . . _ _ , . _ _
Table 15. Summary of plant maintenance activities affected by removal of pipe whip restraints, with associated ecsts ($K/ device).
Operating Plants Construction Plants .
Maintenance Activity __ _
Best High Low But High Low Es'imate Estimate Estimate Estimate Estimate Estimate In-service weld inspection 4 8 2 4 8 2 1.
.5 1 .3 .5 1 .3
- 2. Routine restraint maintenance
.8 1.6 .4 .8 1.6 .4
- 3. Gap verification 5.3 10.6 2.7 5.3 10.6 2.7 l Total avoided costs ($K/ device) l l
l Notes:
- 1. All values cumulative over a 40-year plant life.
- 2. Values given are for pipe whip restraints only.
i
. l'
-. ;- _ _ _ _7_ _ _
- - - - - . ~ .- - .,, . ... .. . _ , , , , _ _ _ _
Table 16. Total avoided operating costs. i' PWR Plants BWR Plants
, s-i Ber t High Low Best High Low i:
Estin ate Estimate Estimate Estimate Estimate Estimate l
S Restraint Costs ($K/ plant) 64 114 13 18 22 1 Restraints ucluded 339 1206 35.1 95.4 233 2.7 i Operating plants 339 1208 35.1 95.4 233 2.7 Construction plants Total Avoided Costs ($) PWR Plants BWR Plants All Plants 62 31 93 Operating plants Construction plants 18 4 22 l
i 31.8 32.2 31.9 l Average remaininglife 3
(operating plants only) i 2.6e7 1 Best estimate 2.3e7 2.Se6 8.le7 6.Se6 8.8e7 l l'$ ==timata 2.4e4
~
Low estimate 2.4e4 6.7e4 4
1 I.
1
. _ _ ,TT : r :: r. . :-. - - - . .
.=. -. _ - - - - . -. , . _ ., . , _. -.
I
[.
', DRAFT RCvo 0 6.8 NRC Development and Implementation Cost Development of the proposed SRP modification, including NRC staff review, is complete; no additional development costs and only mini:nal implementation costs are anticipated for final modification of the Standard Review Pla!L t
6.9 NRC Operation Ccet, ,
No additional NRC operating costs related to the proposed action are anticipated.
l It is anticipated that less effort would actually be required, and NRC operating costs would be reduced.
'0 t 4
i I
?
! i i
, i t
O I
4 4
- S86-044 04/01/86 i
l l
1
- \ DRAFT R;v.0
- 7.
SUMMARY
7.1 Discussion of Results The results of the value impact assessment are summarized in Table 17. In this L table, "value" represents reductions in occupational radiation exposure, "impact" the costs associated with the proposed action. The nominal estimates of oost and radiation do6e indicate that substantial overall reductions in each could result from implementation of the proposed action. In addition, the proposed action would increase i
the best-estimate of public risk only negligibly, by less than 2 man-rem over the entirc,
- population of plants considered.
! Together with the specific results in Table 17, the following general observations are noteworthy:
e The proposcJ action would eliminate dynamic effects associated with postulated AIBs from the plant design basis. As a result, ploe uhlp and jet impingement would not have to be considered in the qualification of safety-related mechanleal equipment. The proposed action would, however, still require that safety-related i equipment located near the affected piping system be quallfled for the "non-mechanistic" environmental effects (such as changes in local bulk pressure, temperature, and humidity) of a postulated rupture. This requirement already refleets current industry practice; consequently, the poposed SRP revision would have no effcet on equipment qualification, e Of the plants currently in operation or under construction, only construction plants would potentially see any immediate benefit following implementation of the proposed action. Pipe whip restraints and jet shields fabricated but not yet installed would no longer be required at AIB locations, resulting in substential savings in avoided installation costs. The actual avoided costs would depend on plant construction status, particularly the extent to which protective devices had ti l already been installed at the time the proposed action took offeet.
n For operating plants, immediate removal of protective devices in response to the proposed action would actually result in implementation costs being incurred -
about $150K per plant - as well as in additional worker radiation exposure. Most owners of operating plants responding to our information request indicated that j they would take no immediate action if the proposed action were implemented.
'i e With regard to avolded operating costs, operating plants and plants under construction would benefit about equally if the AIB requirement were eliminated.
- Protective devices removed during the course of routine in-service weld inspection would no longer have to be reinstalled, resulting in lower operating costs and ,'
reduced personnel radiation exposure. This would also hold true even when ,
restraints have been specially designed so that their removal is not required, as ,
their presence can stD1 restrict weld access and therefore hinder ISI. Plant congestion would also be relicved in general, improving the efficiency of routine plant maintenance as well as recovery from potential unusual situations such as fires or radioactive spills.
i S86-044 04/01/86 t
i h
i
bo
'.._ DRAFT. Rev. 0 l l[ .
i
- ' For plants considered in this assessment, it is interesting to note that the relatively few
+ construction plants (as a percentage of the total plant population) keep the overall
+ implementation costs low even though the per-olant implementation oosts sre
- significantly greater than operating costs.
) The actual cost and ORE reductions experienced by ary one plant will, of course,
- depend heavDy on the number postulated breaks eliminated and the asacolated number of protective devices affected. It can, however, be generally said that the proposed action will save on the order of $2 mBilon per PWR plant in implementation easts (construction
, plants only) and between $300K and $400K over plant life in maintenance oosts assuming
! 150 arbitrary intermediate breaks are eliminated throustrout the plants if 100 of these i breaks were inside of containment, ORE would be reduced by about 200 marrrem over ,
plant life. Improved plant officioney due to reduced heat loss at excluded restraint
- locations could save an additional $600K over 40 years. Equivalent oost and ORE values
! for BWR plants would be lower, due to the fewer number of excluded breaks indleated by t our assessment.
'6 Recall though that the assessment for existing plants assumed that AIBs had 1 already been located and protective devicer already designed and fabricated. If A!Bs are C
, j eliminated, these costs would no longer be incurred as a part of new plant design and
!q construction. Therefore, the proposed action would have an even more significant effset j1 in the future, resulting in east savings on the order of $4 mBilon per plant assuming ia elimination of 150 postulated break locations. Avoided ORE and operating easts for 1/ future plants would be similar to those for existing construction plants.
1 7.2 Effeet of Modifying Usage Factor j As discussed in Section 2, the foregoing regulatory analysis was based on
- eliminating from SRP 3.6.2 (MEB 3-1) the ewrent "arbittery intermediate break" i requirement while leaving unehanged the present stress and usage factor eriteria that
- j. define where breaks must be postulated. Since commisaloning the regulatory analysis, the NRC has modified its proposed regulatory action to also include an increase in the
,s
[ usage factor limit. The proposed increase - from 0.1 to 0.3 - is based on an analysis by j; Rodabaugh showing that a usage factor threshold of 0.328 would be more consistent with ie the 2.45 m stress limit (31]. At NRC request, we have made a llmited assessment of how
,( this change might affect value-impact associated with elimination of the arbittsry inter-mediate break requirement.
l
! Under the preposed change in usage factor, many intermodlate break locations defined as mandatory under the present usage factor threshold would revert to breaks. This, in turn, would increase the total population of potential break "arbitrary"inside locations of containment above that considered in the original AIB regulatory analysis.
l l Given an increase in usage factor threshold, the speelfie systems that would be
! affected and the number of "new" arbitrary breaks that would result is dif fleult to assess l without a detailed surveys such a survey was outside the scope of the original regulatory analysis. We did, however, learn the following based on a brief review of information j l
from LLNL files and from a limited number of discussions with industry representatives: }
l I 366-044 04/01/66
- i I
1 I
I
'. '. DRAFT Rcy, 0 i
e according to one PWR vendor participating in our evaluation, increasing the usage f actor threshold from 0.1 to 0.4 (the value recommended by Rodabaugh) would most
' likely cause several intermediate break locations in pressurizer surge and spray lines to revert from mandatory to arbitrary. The vendor had not, however, estimated how many breaks would be affected, e for four of the PWR units in our AIB data base, according to another industry source, increasing the usage factor limit from 0.1 to 0.4 would eliminato at least 75 percent of the remaining mandatory intermediate breaks in Class 1 piping. For these units, usage factor (and not stress) now prevents these break locations from being categorized as arbitrary. The source providing this information did not, however, specify the preciso number of breaks involved, nor did a brief review of Final Safety Analysis Report (FSAR) Information fer these four units provide further insight.
e a review of PSAR information for a plant similar to those above but not included in the AIB data base indicated for the surge line that six intermediate break locations had usage factors greater than 0.1, of which four had a usage factor less than 0.3 (the proposed A1B threshold). Insufficient . stress information wrs avaDable to
- determine if these four locations would actually revert to the "arbitrary" category; the rough agreement with the figures for the four units above is nevertheless interesting. In all cases the breaks were in piping three inches in diameter and above.
s A review of simDar information for pressurizer spray line, pressurizer sarety and relief valve piping, and CYCS (Chemical and Volume Control System) piping did not produce any candidate locations that might revert to AIBs under the proposed change in usage factor, e a review of design information for two BWR units identified a total of 289 Class I breaks, about half of which were intermediate breaks. Of the intermediate breaks, 11 were mandatory based on stress limit and 49 were mandatory based on usage factor. Under the proposed change in usage factor, 21 of these 49 breaks would revert to the "arbitrary" category.
At first glance, these results might imply that relatively few mandatory breaks would revert to arbitrary breaks under the propcsed change in usage factor, flowever, it is important to note that the categorization of a postulated break as "mandatory" or "arbitrary" is usually based on design stresses and usage factors that often embody a algnifcant degree of conservatism. Changing the criteria for postulating break locations could offer incentive to perform less conservative analyses and thereby increase the number of breaks that could be excluded under the revised criteria.
Consider for a moment how usage factors are calculated as part of piping system design. The number of allowable load cycles at a given location (which defines the ,
denominator of the usage factor)is determined from the fatigue ("S-N") curves presented in Pigs.1-9.0 of the ASME Code, Section 111, as a function of peak stress intensity at that location. Design stress intensities are typically calculated from Code Eq. (11), whleh generally yields conservative results, if, however, a lower stress intensity is estimated at the break location by using more exact (i.e., less conservative) analytic methods, the l
S86-044 04/01/86
f '
DRAFT Rt.v.O I '. .
s
- 4 corresponding number of allowable load cycies given by Figs.1-9.0 will be accordingly higher and the usage factor (for a given number of duty cycles) lower. Depending on the range of allowable load cycles consited, relatively small reductions in omiculated stress can damatleelly re&oe the associated usage factor.
According to one architect-engineer source, the amount of effort required to reduce a high usage factor (e.g., in the 0.7 to 1.0 range) to the present SRP 3.4.2 ,
threshold is not warranted by the benefit to be derived (i.e., elimination of the postulated break), assuming such a re&ction is achievable in the first place. Accordng to this name source, the "break even" usage factor threshold - that whieh wodd justify additional analytle offart -is in the 0.3 to 0.4 range. Consequently, the proposed increase in usage factor would provide substantial incentive for plant designers to identify, through ,
detailed analysis, additional postulated break locations which could revert to the "arbitrary" estegory. These could then be excluded under the proposed regdatory action ,
described in Beetion 2 of this regulatory analysis. ,
The fdlowing discussion addresses how these additional arbitrary break locations would affeet the results of the foregoing regulatory analysis. ;
Public Risk f As Section 6 discusses in detall, two factors affect the estimated risk associated with the elimination of arbitrary intermediate pipe breaks the number of breaklocations eliminated ami the probability of break at a given location. Although relaxation of the current usage factor threshold would almost certainly increase the total number of ;
emeluded breaks, in our opinion this change would only negligibly affect the results of the foregoing regulatory analysis. This judgement reflects the fdlowing considerations
. (1) any change in usage factor threshold would, by definition, affect only break .
locations postulated in ASME Class 1 piping systems. The number of AIB locations postdated under current criteria for Class 2,3, and non-nuclear elass piping, which constitute the majority of excluded breaks.in the foregoing regulatory analysis, would remain unaffected. Therefore, even though a substantial percentage of Class 1 intermediate breaks mandatory under owrent requirements might revert to -
the "arbitrary" eategory, thu number of additional breaks as a percentage of the total population of excluded breaks inside of containment (i.e., those assumed to contribute to risk) would remain small. ,
Recall also that of the approximately 300 postdated break locations in a typical PWR plant, about fifty percent are A1Bs that could be eliminated without any change in the present stress and usage factor limita. Therefore, the proposed change in usage factor could theoretleally increase the total number of excluded breaks by at most a factor of two. The actual increase would be much less af ter breaks in non-Class I systems and Class 1 breaks that would remain mandatory, such as those at terminal ends and at intermediate locations exceeding the stress or (revised) usage factor limits, were accounted for.
i As was the case for general exclusion of AIBs, the actual number of "new" A1B locations resulting from the proposed change in usage factor would most likely very i widely from plant to plant. On the average, however, we would expect the 586-044 04/01/86 1
. . _ _ _ _ _ _ _ _ _ _ . . _ _ _ _ _ _ _ _ _ _.___.___________________J
- - DRAFT Revo0 s
resultar.t total per-plant break population to remain within the high and low estimates for the plants in our AIB data base (see Table 2).
(2) most of the currently mandatory intermediate breaks reverting to arbitrary breaks would be in intermediate- and large-diameter piping (i.e., nominal diameter greater than two inches). The corresponding "best-estimate" break probabilities assumed in the foregoing analysis are, respectively, on the order of one and these orders of magnitude less than that for small-diameter piping. Consequently, the percentage increase in total risk (i.e., over that estimated for general elimination of AIBs under the present criteria) would be much less than the peecentage increase in total break population.
These considerations imply that the propmed change in usage factor would cause no significant (i.e., "order of magnitude")increues in our estimates of public risk and core l l melt frequency. ;
1 Assessment of Value-impact The proposed change in usagt factor, by causing certain mandatory Class 1 breaks l
to revert to AIBs, would allow elimination of pipe whip restraints and jet impingement barriers associated with those breaks. As discussed in the detailed regulatory analysis, removal of such "protective devlees" would reduce related implementation and operating costs as well as occupational radiation expmure. Quantifying cost and ORE reductions specifically assoelated with the propmed change in usage f actor was outside the scope of this analysis; however, arg reduction in the number of required protectivo devices clearly benefits the industry from a cost and ORE standpoint.
As noted earlier, we expect that the proposed change in usage factor would only negligibly increase core melt frequency over that estimated in the detailed regulatory analysis. Consequently, those parts of the value-impact assessment dependent on core melt frequency (i.e., occupational exposure due to accidents, on- and off-site property damage) would similarly be only negligibly affected.
Table 17. Summary of value impact (total for 115 plants). j I
Dest liigh Low Estimate Estimate Estimate Public health (man-rem) - 1. 6 -7.8 e+ 2 -3.9 e-2 Value (man-rem) 1.7 e+ 4 4.2e+ 4 2.2e+ 3 Impact ($) -41 e+ 6 -105e+6 -5.6 e+ 6 ,
l h
S86-044 04/01/86
. . DRAFT ROvo 0 0
REFERENCES
- 1. "Design Baser, for Protection Against Natural Phenomena," Code of Federal Regulations, Title 10, Part 50, Appendix A.
- 2. U.S. Nuclear Regulatory Commission, "Determination of Rupture Locations and Dynamic Effects Associated with the Postulated Rupture of Piping" 8tandard Review Plan, Report NUREG-0800, Section 3.6.2 (July 1981).
- 3. U.S. Nuclear Regulatory Commission, "Postulated Rupture Locations in Pluid System Piping Inside and Outside of Containment," Branch Technical Position MEB 3-1 (July 1981).
- 4. U.S. Nuclear Regulatory Commission, "Evaluation of Potential for Pipe Breaks",
Report of the USNRC Piping Review Committee, Report NUREG-1061, Vol 3 ,
(1985).
- 5. Amerlean Society of Mechanical Engineers, BoDer and Pressure Vessel Code.
- 6. R. Schmitz, "Proposed Changes in Intermediate Pipe Break Criteria," Bechtel Power Corporation (September 1983). Presented at the CSNI Meeting on Leak- '
Before-Break in Nuclear Reactor Piping Systema, Monterey, Californla, '
September 1-2, 1983.
- 7. S. Lu and C.K. Chou, Reliability Analysis of Stiff Versus Plexible Piping - Final ,
Project Reprt, Lawrence Livermore National Laboratory, Report UCR s-20140, j N UREO/CL-4263 (April 1985). ;
- 8. S. Heaberlin, et at, A Handbook for Value-Impact Assessment, Pacific Northwest Laboratory, Report PNL-4646, N UREO/CR-3568 (December 1983).
- 9. "Modification of General Design Criterion 4 Requ!renents for Protection Against Dynamic Effects of Postulated Pipe Rupture," Federal Register, Vol 50, No.126 (July 1,1985). ;
- 10. O. Holman and C.K. Chou, Assessment of Value-impact Assoelated with the Elimination of Postlulated Pipe Ruptures from the Design Basis for Nuclear Power li Plants, Lawrence Livermore N ational Laboratory, Report UCID-20397
'(M arch 1955).
- 11. American Nuclear Society, Design Basis for Protection of Light Water Nuclear Power Plants Against Effects of Postulated Pipe Rupture, Alt S Standard N58.2-1980 (1980). f
- 12. Duquesne Light Company, AIB exemption request for Beaver Valley Unit 2 (M arch 1985). l
- 13. Commonwealth Edison Company, AIB exemption requests for Byron Units 1 and 2 and Braldwood Units 1 and 2 (November 1984). l S86-044 04/01/86 m-
DRAFT Reva0
'A
- 14. Duke Power Company, AIB exemption request for Catawba Units 1 and 2 (November 1983).
- 15. Texas Utuities Generating Company, AIB exemption request for Comanche Peak Units 1 and 2 (December 1985).
- 16. Public Service of New Hampshire, AIB exemption request for See. brook Unit 1 (December 1985).
- 17. Carolina Power & Light Company, AIB exemption request for Shearon Harris Unit 1 (February 1985).
- 18. Houston Lighting & Power, AIB exemption request for South Texas Project Units 1 and 2 (August 1984).
- 19. Georgia Power Company, AIB exemption request for Vogtle Units 1 and 2 (April 1985).
- 20. Washington Public Power Supply System, AIB exemption request for Washington Nuclear Project Unit 1 (October 1985).
- 21. Illinois Power Company, AIB exemption request for Clinton Unit 1 (Apr01985).
- 22. Public Service Electric and Gas Company, AIB exemption request for Hope Creek !
Station (December 1985).
- 23. Commonwealth Edison Company, AIB exemption request for La Salle Units 1 and 2 (April 1985).
- 24. U.S. Nuclear Regulato,y Commission, Reactor Safety Study, Report WASH-1400, NUREG-75/014 (October 1975).
- 25. Probability of Pipe Pallure in the Reactor Coolant Loops of Westinghouse PWR Plants, Lawrence Livermore N ational Laboratory, Report UCID-19988, ,
N UREG/CR-3660 (1984k i VoL 1: Summary Report j Vct 2: Pipe Failure Induced by Crack Growth :
Vol 3: Guillotine F reak Indirectly Induced by Earthquakes !
t Vol 4 Pipe FaDure Induced by Crack Growth, West Coast Plants
- 26. Probability of Pipe Failure in the Reactor Coolant Loops of Combustion ,
Engineering PWR Plants, Lawrence Livermore National Laboratory, Report U CRL-53500, N UREG/CR-3663 (1984).
Vol. It Summary Report Vol2: Pipe Failure Induced by Crack Growth {
Vol. 3: Guillotine Break Indirectly Induced by Earthquakes l
S86-044 04/01/86
I'. '. DRAFT Rcy, 0
- 27. Probability of Pipe Failure in the Reactor Coolant Loops of Babcock & Wilcox PWR Plants, Lawrence Livermore National Laboratory, Report UCRL-53644, N UREO/CR-4260 (1985).
Vol.1: Summary Report Vol. 2: Guillotine Break Indirectly Induced by Earthquakes
- 28. Probability of Failure in the Reactor Coolant Pling of BWR Pl' ants, Lawrence Livermore National Laboratory, report in preparat..on. See also G. Holman, "Pipe l Ruptures in BWR Plants," Lawrence Livermore National Laboratory, Report UCRL-93181 (January 1986).
- 29. W. Andrews, et al., Guidelines for Nuclear Power Plant SafetyIssue Prioritization Information Development, Pacific Northwest Laboratory, Report PN L-4297, N UREG/CR-2800 (1983b
- 30. D. Strip, Estimates of the Financial Consequences of Nuclear Power Reactor Accidents, Sandia N ational Laboratories, Report N UREG/CR-2373 (1982).
1 i
- 31. Letter from E. Rodabaugh (BCR Associates) to S. Hou (U.S. NRC) on ths subject 4 of "BTP MEB 3-1, Relation Between Usage Factor and Allowable Stresses," dated
, September 9,1985. .,
.I .
I 1
J 4
i i
i t:
l S86-044 04/01/86 ,
k
! b
- DRAFT Dev, 0
.~
DISCLAIMER This document was prepared as an account of work sponsored by an agency of the United l States Government. Neither the United States Government nor the University of California nor any of their employees, makes any warranty, express or impued, or assumes any legal liability or responsibuity for the aceuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights, Reference herein to any speelfie commercial products, process, or service by trade name, trademark, manufacturer, or otherwise, does not necessarny constitute or imply its endorsement, recommendation, or favoring by the United States Government or the University of California. The views and opinions of authors expressed herein do not necessarDy state or reflect those of the United States Government or the University of California, and shall not be used for advertising or product endorsement purposes.
t 04/01/88 SsG-044 i
, s R;v, 0
,- . DRAFT d
APPENDtX A
~
Proposed Revisions to SRP 3.4.1 f
1 s
l
[
f l
E f
t S86-044 A-1 04/01/86
,- s DRAFT Rcy.O ia
't 4...
+. .
.t 4 . t 4 s t
- l, i4 i
. t ,
l- '
.. -l, k ?
I I
L Y
' ?
6 i AFFEllDEX3 i
9 i
, A t -
3
! l 1
i 1 t p
1
}
)
a i
) =.
. i e
l 2 1 J r J
l
- t 1 !
i i
e
. 1 k
t j-I e iA f 1
4 i . .
.}
F i t i f 846-044 B-1 04/ol/ss l
I i i
- j.
- _ ., n ,- ,,- . ...-- - -. . . . .-
i s
g* '
DRAFT Rev.O !
i ^* i Table B.I. Westinghouse Plants Operating Plants Beaver Valley 1 Surry 1 l
- Byron 1 Surry 2 Callaway 1 Trojan !
Catawba 1 Turkey Point 3 '
Comanche Peak 1 Turkey Point 4 Diablo Canyon 1 Watts Bar 1 ;
Diablo Canyon 2 Wolf Creek .
Donald C. Cook 1 Yankee Rowe Donald C. Cook 2 Zion 1 !
Zion 2 l 3
Joseph M. Farley 1 1 Joseph M. Parley 2
^
I
, Robert E. Ginna '
( Haddam Neck Construction Plants t Indian Point 2 . (
i Indian Point 3 t Kewaunee Beaver Valley 2 f McGuire 1 Braidwood 1 '
McGuire 2 Braldwood 2 I
- North Anna 1 Byron 2 i
North Anna 2 Catawba 2 Comanche Peak 2 !
Point Beach 1 Point Beach 2 Shearon Harris 1 Prairie Island 1 MR1 stone 3 .
Seabrook 1 l Prairie Island 2 i H.B. Robinson 2 Scuth Texas 1 j Salem 1 SouthTexas 2 i l
Salem 2 Vogtle 1 San Onofre 1 Vogtle 2 :
i( SequoyahI Watts Bar 2 i Virgt! C. Summer 1 l lt r 3- Total operating units 41 1299 Total Average remainingremainingIlfe life(py)
(yr ) 31.7 !
i Total construetwn unita 14 i Total forward fit (pyh 20.1 Average forward fit (yr) 1.4 560 1 Total Average remaining remaininglife life (py(h yr) 40.0 l Total unita 55 1859 Total Average remainingremainingIlfe life(py)
(yr ) 33.8 l l
I 586-044 B-2 04/01/86 l l t
! l
)
i i
-,___-e.. ._,.-..r_. .,,-_,_,.._____.-.._._._.._~___.,___.___y_ -
' Rev. 9
' - DRAFT o
s Table B.2. Combustion Engineering Plants Operating Plants Construction Plants Arkansas 2 Palo Verde 2 Calvert Cliffs 1 Palo Verde 3 ,
Calvert Cliffs 2 Fort Calhoun ,
Maine Yankee Millstone 2 !
Palisades Palo Verde 1 j St.Lucie1 St. Lucie 2 San Onofre 2 3
San Onofre 3 Waterford 3 ,
i Total operating unita 13 Total remaining life (py) 430 Average femalning life (yr) 33.1 4
Total construetion unita 2 Total forward fit (py) 2 Average forward fit (yr) 1.0 Total remaining life (py) 80 Average remaining life (yr) 40.0 Total unita 15 i
Total remaining life (py) 510 Average remaining lite (yr) 34.0 l i !
i l
I f
B-3 04/01/86 886-044
a 4
- -DRAFT Rev. 0
.o e
Table B.3. Babcock & WUcox Plants Operating Plants Construction Plants Arkansas 1 Bellefonte 1 Crystal River 3 Bellefonte 2 Davis-Bease I l
Oconee 1 Oconee 2
- Oconee 3 Rancho Seco Three Mile Island 1 1
i u Total oper ating unita . 8
. Total remaining life (py) 240 Average remaining life (yr) 30.0 1 Total construction unita 2 Total forward fit (pyh 8.7 g Average forward fit (yr) 4.4 y 80 Total Averageremalning remaininglife life(py)
(yr ) 40.0
)
Total unita 10
{. Total remaining life (pyh 320
$ Average remaining IWe (ys 32.0 4
I L
i 886-044 B-4 04/01/86
f, i Cev. 0 j' '
DBAFT
,4
. t
i Table B.4 Summary for AllPWR Plants 4
Total operating units 62 .
Total remaining life (py) 1969
- Average remaining life (yr) 31.8 .
' Total construction unita 18 Total forward fit (py) 30.8 Average forward fit (yrh 1.7 Total remaining life (py) 720 Avorage remaining Ilfe (yr) 40.0
, Total units 80 Total remaining life (pyh 2689 Average remaining life (yr) 33.6 I .
t l
i
?
d I
9 r
B-5 04/01/86
- } $86-044 I
1
P-
- COV.0 DEAFT
,1 a'
L Table B.5 General Electrie @WR) Plants t
Operating Plants
}
Duane Arnold Peach Bottom 3 Browds Ferry 1 Perry 1 :;
l Brown's Ferry 2 PDgrim 1 -
Browds Ferry 3 Quad Cities 1 Brunswick 1 Quad Cities 2 Brunswick 2 River Bend 1 Cooper Shoreham Susquehanna 1 U Dresden 2 Dresden 3 Susquehanna 2 [
, Fermt2 Vermont Yankee FitaPatrick WNP 2 Orand Gulf i ,
Hatch I lt Construction Plants O Hatch 2 La Salle 1 ll 1i La Salle 2 I Limerlek 1 Clinton MR1 stone 1 Hope Creek 1 ,
Limerick 2 i>
Mentleello Peach Bottom 2 Nine Mlle Point 2 l
Total operating units 31 r
997
- l. Total Average remaining remaininglife lif e(py)
(yr ) 32.2 k Total construction unita 4 ,.
? Total forward fit (py) 5.2 -
1 Average forward fit (yr) 1.3 160 i
- Total remaining life (py()
t Average remainingIlfe yr) 40.0 [
Total units 35 1157 '
l Total Average remaining life (py()
remaininglife yr ) 33.1 1
i l S86-044 B-6 04/01/86 I
i ll
. - . - - - - - - . ---,.,,c.---.--.--n --- - ,, =- ~-,-, -,, ,. - , - -- --- --