ML20101B469
| ML20101B469 | |
| Person / Time | |
|---|---|
| Site: | Arkansas Nuclear  |
| Issue date: | 05/29/1992 |
| From: | James Fisicaro ENTERGY OPERATIONS, INC. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| REF-GTECI-A-47, REF-GTECI-SY, TASK-A-47, TASK-OR GL-89-19, TAC-M74906, TAC-M74907, NUDOCS 9206040084 | |
| Download: ML20101B469 (16) | |
Text
.
s 4-Entergy
[al m r aac-o Operations
- s w 31:o
+ + w-a May 29, 1992 OCAN059207 U. S. Nuclear Regulatory Commission Document Control Desk Mail Station P1-137 l
Washington, DC 20555 i
Subj ect: Arkansas Nuclear One - Units, and 2 Docket Nos. 50-313 & 50-368 License Nos. DPR-51 & NPF-b Resolution of Unresolved Safety Issue A-47,
" Safety Implication of Control Systems in LWR Nuclear Power Plants" Ceneric Letter 89-19 (TAC Nos. M74906 and M74907) i Gentlemen:
Generic: Letter 89-19 was issued September 20, 1989 (OCNA098921),
re.ues ting action related to Unresolved Safety Issue A-47 concerning cuatrol systems in LWFs.
The staff concluded that all PWR plants should provide automatic steam generator overfill protection and establish plant procedures and technical specifications to assure the system remains available.
The licensee was to provide a response to the NRC outlining the : Intended compliance or justification for not implementing the recommendations.
On March 19, 1990 'atergy Operation responded-in letters ICAN039001 and 2CANO39001 that taete were several issues that needed to be considered prior to installation of an automatic overfill protection system at Arkansas Nuclear One, Units 1-and 2 (ANO-1 & 2).
Specifically, assumptions and information utilized in the NRC evaluation were outdated or unsupported, and an evaluation assessing the negative impact on safety of the proposed modification was not performed.
Due to the potential negative safety impact associated with the modification,'Entergy Operations proposed to assess the issue in the Inlividual Plant Examination (IPE) process.
The NRC responded in letters dated October 1, 1590 (1CNA109002 and 2CNA109002), that Entergy Operations' proposal was unacceptable and that implementation of Generic Letter 89-19's recomr andations should progress.
gQ(\\'
9206040084 920529 PDR ADOCK 05000313 p
\\\\
4 t
N. S. NRC May 29. 1992
-Page 2 The Combustion Engineering owners Group (CEOG), which ANO-2 is a member, then contacted the NRC and arranged for a-meeting to discuss the generic letter. Entergy Operations notified the NRC in letter dated November 19 ~
1990 (0CAN119003), that a revised Aesponse to the generic letter for both units would be developed based on the outcome of the meeting.
based on the common position taken on the generic letter, for both ANO-1&2 and the generic nature of'the CEOG presentation, Entergy Operations believed tl.at the results of the meeting would be applicable to both ANO units..
On November 20, 1990, the NRC and representatives of the CEOG met.
The CEOG presented their evaluation and demonstrated that steam generator overfill was not a significant safety concern and that the installation of an automatic steam generator overfill protection system could, in fact, degrade safety and _was not cost justified.
- be group requested that the NRC staf f consider the CEOG position as c basis for not implementing the generic letter's recommendations and eccept the analysis as the justification for resolution.
The 5xc staff committed to assess the information and stated that implementation was to be delayed until the staff evaluation was completed.
Based upon this NRC response, implementation of Gh 89-19 for ANO-1&2 was dt layed unt11' KRC review of the CEOG data was complete.
Based on a recent discussion with the ANO-2 NRR Project Manager, ANO has been informed that the NRC Staf f has completed their. review of the CEOG data and are planning to close this issue for the CEOG in correspondence soon to be issued. During this discussion,_the Project Manager requested that Entergy Operaticns provide a discussion of the appl.icability'of the CEOG presentation to ANO-1.
Attached is the requested information.
Entergy Operations believes that this submittal provides suf ficient information for the Staff to close this lasue for ANO-1.
thould you have any questions regarding this issue, please contact me.
Very truly yours,
/7 u,h.
C c
a
/ fb James J. Fisicaro
//
Director, Licensing JJF/RWC/sjf j
Attachment cc:
Mr. Robert Martin U. S. Nuclear Regalatory Commission Region IV 611 Ryan Plaza Drive, Suite 400 Arlington, TX 76011-8064 4
1
-U.
S. NRC May la '1992 Page 3 NRC Senior Resident In:spector Arkansas Nucleat One - ANO-1 & 2 Number 1, Nuclear Plsnt Road Russellville, AR 72801 s
Mr. Thomas W. Alexion NRR Project Managet, Region IV/ANO-1 U, S. Nuclear Regulatory Commission NRR Mail Stop 13-11-3 One White-Flint North 11555 Rockville Pike Rockville, Maryland 20852 Ms. Sheri Peterson NRR Project Manager, Region IV/ANO-2
.f U. S. Nuclear Regulatory Commission 5
NRR Mail Stop 13-11-3 One White Flint North 11555 Pockv111e Pike Rockville, Maryland 20852 id i
'S
. ~ - -
.i DISCUSSION OF THE APPLICABILITY OF THE CEOG PRESENTATION TO ANO-1 WITH RESPECT TO STEAM GENERATOR OVERFILL PROTECTION BACKGROUND Generic Letter (GL) 89-19 recommends that the Babcock and Wilcox (B&W) -
designed PWR plants have an automatic steam generator overfill protection to mitigate main feedwater overfeed events.
In addition, plant procedures and technical specifications are to include provisions to periodically verify the operab111ty of the overfill protection. A third recommendation, which is not applicable to ANO-1, involves the installation of automatic protection tc orevent steam generator dryout on loss of control system power. These recommendations are based in the probabilistic risk analysis (PRA) performed by_ Pacific Northwest Laboratory (PNL) and documented in NUREG/CR-4386 (Reference 1).
A review of this and related documents in conjunction with the CEOG position presented to the NRC in November of 1990, reveals similar concerns and ccaclusions for ANO-1.
Specifically, assumptions and information utilired in the NRC PP.A evaluation are outdated or unsupported, and an
-v evo. tion assessing the negative impact on safety of the proposed modification was not performed. A review of the major-core Camage scenario, the assumptions utilized in the NRC PRA evaluation, and the safety benefit /value impact analysis demonstrates that a steam generator overfill prote >2.ou system should not be installed at ANO-1.
DISCUSSION OF THE PUBLIC RISK ASSESSMENT The major overfill transient scenario assessed in Reference 1 at Oconee dictated that there must be failures that would initiate a main feedwater (MFW) overfeed, a failure of the MFW trip signal, and.a failure of the operator to isolate feedwater flow. As the steam generator overfills, water spills into the main steam line, eventually resulting in a main steam line break (MSLB) du, to the static and dynamic water loads on the piping. The steam generator experiences a pressure transient upon blowdown of the secondary side following the postulated MSLB. The
'3 pressure differential across the steam genarator tubes is then postulated to induce one or more steam generator tube ruptures (SbTR).
High pressure injection into the primary system continues to maintain core cooling as long as a water source (reactor au11 ding sump-or borated water storage tank (BWST)) is available.
If the MSLB location is outside the Reactor Building but upstream of the main steam isolation valve (MSIV),
sufficient primary water is lost through the ruptured tubes to eventually exhaust the BWST inventory, at which point core damage is postulated to occur. The public risk due to this scenario as described in Reference 1 dominates the total risk associated with control system failure scenarios.
As noted earlier the major core damage scenario considered in Reference 1
.is a steam generator overfill with a-resulting SGTR.
NUREG/CR-4386 a;uo discusses overfill scanarios in conjunction with two other accident sequences.
In one sequence an overfill is experienced and a transient shutdown is effected with the power conversion system unavailable because of degr ading conditions in the secondary ride. This scenario is 1
considered to account for-the potential of a main turbine trip before the point of spillover in the main stenm line.
The other scenario cesumed a MSLB after overfill led to core damage and included the sequence in the-repert to be conservative. These accident sequences are not considered in this discussion since their contribution to the_public risk has acen detarmined by Entergy Operations and the_NRC to be acceptably low.
It should be noted that. the CEOG concerns are also applicable to these other cases considered in Reference 1, although only the averfill_with a MSLB and SGTR will be explicit 1y considerad here, o
The public risk is assessed by first analyzing the frequency of an overfeeding event and the probability that-the operator fails to terminate the overfeed to establish an initiating event (IE) frequency.
This is then combined with the probabilities of a MSLB given overfill into the steam lines, break location (i.e., inboard of the MSIVs), SGTR given a MSLB, and core-melt givea 1,cimary-system injection sources are-exhausted out the ruptured SG tube (s).
Each of these probabilities directly contribute to the resulting assessment of public risk. The assessment provided in the Reference 1 NUREG, however, made incorrect-assumptions in deriving these probabilities which artificially increased the public risk calculation (Table 1).
A.
STEAM GENERATOR CVERFILL INITIATING EVENT FREQUENCY One area that was investigated by the CEOG was the assumed initiating event frequency in relation to the probability of an operator failing to termitate an overfill scenario.
The probability of an operator failure to terminate the cverfill was estimated at 4
0.7 (recognized by PNL as an upper bound estimate of operator error) for the B&W units.
In plant-specific PRI.s, such overfill scenarios would be assigned an operator failure proiability an order of magnitude lower, resulting in an associated order of magnitude further reduction in public risk.
It should be noted that NUREG/CR-4386 recognized that the A-47 issue deals with control systems routinely under operator coctrol, and therefore interaction of the operator with failure diagnosis and recovery is an appropriate consideration; and also recognized that the average failure probability would be lower in plants with simulator programs stressing proper diagnosis of failures.
For ANO-1, as well as other B&W plants, MFW overfeed due to control system malfunctions receives special attentica in operator training due tc the smaller secondary volume of the B&W once-throagh steam generator (OTSG) and its associated responsiveness. As a result, the probability that an operator fails to terminate an overfeed event can be readily reduced to 0.07 which produces an initiating event frequency of 0.0006/yr (0.006/yr
- 0.07/0.7).
2
- _ _ _ _ _ _ _ _ = _ _ - _ - _ _ -.
V B.
PROBABILITY OF A MSLB Tho'pcobability of a MSLB given a steam generator overfill event utilized in the assessment was also unrealistic and inconsistent with the NRC's own analysis. -Reference 1,-the NRC PRA evaluation for B&W plants, acknowledged that although several spillover events have occurred in U.S. commercial plants resulting in support damagt, no steam line failures have occurred and that this was an crea of likely conservatism. The basis for the probability cited an Reference-1 accounted for uncertainties associated with dynamic.
loading and waterhammer offects.
NUREG/CR-3958 (Reference.2).
provides a discussion-on the studies of overfill in which the MSLB probability was based and this was cited as part of the CEOG.
position. One study indicated a low probability of MSL failure (1 x 10-*) when static forces caused by the deadweight of water in filled steam linea were considered. Deadweight loading was also addressed in the draft version of NUREG-0844 which assigned a 1 x 10 ' probability.
This NUkEG, however, was silent in relation-to d:
alc loading and the MSLB probability.
As a result, the NRC
,uation of the B&W units in Reference 1 incorporated a 0.95 es vaiae.
When the final version of NUREG-0844 (Reference 3) was issued there were savers-changes that effected the overfill evaluation.
Although the draft NUREG, which was utilized as an input to Reference 1, did not address dynamic loading, the final report considered the issue. NUREG-0844 was updated as follows (Section 3.4.1, page 3-10):
The staff has assessed the change in the probability of failure of the main steam line from the increased stress levels associated with the deadweight of water in the team lines.
Analyses have been performed of the increase in stress levels that would result from filling the steam lines in several plants.
Information extracted from analyses on the Ginna, Zion 1, Waterford 3, and Oconee 3 plants indicates that, although in some cases the spring hangers may be loaded slightly beyond a
their operating range, they will not fail and that the stress levels in the main steam line will in all cases remain within the limits allowed by che ASME Code.
In addition to the analyses available, the steam lines were inspected af ter overfilling events at Oconee and Ginna and no indications of failures or incipient failures were found. Therefore, the staff concludes that the probability of failure of the main steam lir.e is not increased by the deadweight loading. Nor is there considered to be a significant potential for failure from waterhammer since the water in the steam lines will be essentially saturated. Accordingly, the estimates of risk in this report for event sequences that consider. failure of the main steam lines are bas 3d on a conservatively determined conditional probabtlity of main steam line failure of 1 x 10-'/ overfill event.
3 A
Yhe ef fect of overfill on main steam line integrity continued to___be examined under GI-135.
On October 5, 1989, the NRC presented the results of the GI-135 program to the ACRS. Task 4 of GI-135 (Steam Generator Overf111) addressed stean line integrity concerns due to the steam generator being overfed or otherwise filled with water.
The presentation showed that the NRC had resolved the steam generator overfill issue due to the small risk invc1ved.
This conclusion was' based on analyses which indicate that some spring hangers may be loaded beyond specification due to deadweight loading, but they will not fall.
In addition, because the water in the steam lines is at saturation temperature and. pressure, the potential for failure due to condensation induced waterhammer is small.
Overfills that have occurred under similar conditions have resulted in little or no damage to steam line piping. Therefore, based on the.results of the NUREG-0844 Final Report and GI-135, a reduction in the probability-of a MELB due to an overfill from 0.95 to 1 x 10-8 is appropriate and justified.
C.
PROBABILITY OF A MSLB LOCATED OUTSIDE THE REACTOR DUILDING AND INBOARD OF THE MSIVs GIVEN A MSLB The third probability that was assessed in the CEOG position is the HSLB location.
In the dominant scenario, core scit occurs as a result of a loss of RCS inventory through an unisolable steam line break (in conjunction with tube ruptures) which eventually exhausts the borated water storage tank. The steam line break locai on probability is based on the assumption that a MSLB has an equal probability of occurring upstream or downstream of the main steam isolation valve,-- A break upstream is asaumed to result in core melt since all water exiting the break would be lost outside of the g
Reactor Building.
In reality, the MSIV is located relatively close to the outside Reactor Building wall, which results in the majority of piping upstream of the MSIV beitig located. inside the Reactor Building.
If the MSLB occurred inside_the Reactor Building, water lost through the break would be collected in the Reactor Building sump and be available for recirculation.
Thus, core melt would not occur without additional failures.
Reference 1 assumes a probability of 1.0 since Oconee has no MSIVa, although it acknowledges that valves are present in the general population of B&W PWRs.
For these plants a 50 percent probability is to be i
utilized.
The maximum probmotlity, however, cannot exceed the product of 0.5 times the ratio of the main steam line piping length outside the Reactor Building to the MSIV to the total main steam line piping' length-to the MSIV.
For ANO-1, this probability is 1.55 x 10-8 ar.d should be utilized for its assessment.
)
4
4 D.
PROBABILITY OF A SGTR GIVEN A MSLB-i As noted earlier
- o final version of NUREG-0844 differed from the draft version whit,was utilized to prepare the overfill evaluation for the B&W units.
The probability of SGTRs after a steam line break was taket from the draf t report and adopted unchanged in the Reference 1 analysis (Section 2.3r page 2.8).. The early draft of NUREG-0844 established the prcbability of tube rupture due to a MSLB as 0.034.
This probability was broken down as follows:
p (1 SGTR)
= 0.014
= 0.017 p (2-10 SG1Rs) p (>10 SGTRs)
= 0.003 In preparing the final report for NUREG-0844, the NRC changed certain assumptions and spproaches in cciculcting the probabilities-associated with single and multiple SGTRs. The total probability of tube rupture due to an MSLB was revised to-0.0505.
This probability was broken down as follows (Section 3. 4.6, Sequences 8A, 8B, and BC, of Reference 3):
p (1 SGTR)
= 0.025
- o p (2-10 SGTRs)
= 0.025 n
p (>10 SCTRs)
= 0.0005 Although the overall SGTR probability was increased..the probability of rupturing greater than 10 tubes was decreased by nearly an order-of magnitude. The NUREG/CR-4386 analysis was particularly sensitive to the value assumed for p(>10 SGTRs) in the calculation of the core melt probability.
L.
PROBABILITY OF C0kE MELT GIVEN MSLB INBOARD OF THE MSIV Utilizing the SGTR probability values from NUREG 0844 Final Report changes and reduces the probebility of_ core melt given a MSLB inboard of the MSIV. Table 2 illustrates the differences by s
reproducing the table from Reference 1 page 2.10 which uses the drafc NUREG 0844 probabils tles and a revised table incorporating the appropriate final vers'. values.
By examining this table, it can be seen that the reduction in probability of greater than 16 SCTRs by more than na order of mrEnitude results in a notable change in the total core melt value.
By merely substituting the final report values and making no other changes, the core melt probability'givea a MSLB inboard of the MSIV is modified to 5,25 x 10-6 from an original value of 1.66 x 10-8 Further reduction in the core melt probabil!ty can be justified by _
?
reassessing the probability of loss of BWST before a RCS depressurization for ruptures
'f more than 10 tubes. This prob &bility dominates the core _relt calculation (Table 2).
It should be noted that the loss of tha BWST supply probebility utilized in Reference 1 was based unon L.e unrealistic estimate of t
time of I hour to empty the tank for Wessinghouse plants with 20 ruptures, the assumptions of runout flow for low pressure safety 5
9 injection, and no cperator action to depressurize the RCS and stop leak ficw.
As noted earlier, opnrator training alone can rednen this value significantly.
This is an area, however, that cannot be easily quantified and the 0.5 value utilized in Reierence I will be included and considered conservatisn. Therefore, tha probability of core melt given a MSLB ir. board of the MSIV is 5.25 x 10-" which is not only appropriate but is conservative.
F.
PUBLIC RISK CALCULATION A new public risk can be calculated utilizing the probabilities previously discussed. Table 1 illustrates the assessment using the Reference 1 analysis values and t.he new probabilities.
NURE",/CR-4386 delineates a public risk of AS.4 tan-rem /yr and the new calculation yields a value cf 2.3* x 10-'
,.an. rem /y r.
This is a sign!ficant reductica in the public risk.
This public risk, however, cannot be evalu tted without considering the negative impact that the proposed steam generator ovm fill protection system could have on safety.
The automatic feedwater pump trip function receramunded by GL 89-19 can itself cause a loss of feedwater accident due to spurious actuation or testing failures.
Adverse consequences can also result from spurious actuation during other events. Unfortunately, the analysis in NUREG/CR-4386 and the value/ impact analysis do not address the negative impact to safety due to installation of an overfill p~otection system. Using the same approach as the PNL study incluuing highly conservative failure assumptions, multiple failures, a high probability of operator failure to restore feedwater, etc., the public risk due to installation of the feedwater pun..
trip could be significant. At a m in i.nu m, -it must be con,;dered and will result in a reduction in the benefit attributed to such a system.
The CEOG presented the results of a scoping calculation using generic data and the noted con" '.mtive approach to estimate the core damage probability due to testing of the proposed system.
The calculation result was a core damage probability of 1.4E-06/yr.
This corresponds to a core melt prchability from the poter.tial overfeed of 4.88E-11/yr.
Conbining these probabilities indicates that the proposed modification would degrade sa fety (4.88E-11/yr -
1.4E-06/yr = - 1.4 E-06/yr).
G.
COST BENFFIT OF GL 89-19 PROPOSED MODIFICATIONS The final item that warrants consideration for ANO-1 is the safety benefit and value impact of the ecommended steam generator overfill protection system. Th is, however, is dif ficult to assess since the assumptions made in the regulatory analysis were not applicable to ANO-1.
The bases for the recommendations in GL 89-19 are discussed in NUREG-1218 (Reference 4) whi b used the calculations of NUREG/CR-4386 (Reference 1) to tocte the safety benefit and value 6
impact of various proposed upgrades.
The feedwater control system at ANO-1 is significantly dif ferent from Oconee, and as a result the values for both costs and henciits of the proposed upgrades which were ustd in the NRC s regulatory analysis do not apply to ANO-1.
For example, ANO-1 hus made mejor improvements in the FFW control system, and in the lntegrated Control System (ICS) over the past several years which make the actus1 probability of a MFW overfeed due to control system failures s4gnificantly lower than that assumed for Oconee.
A further examination of the factors discussed above should Icad to an estimated risk reduction for the applicable control system failure scenarios well below the point at which the NRC's value/ impact guidelines would conclude that hardware changes are a viable option.
.re significantly, when plant specific factors are taken into account, the actual risk reduction due to an overfi)1 protection system may actually be less than the risk increase due to spurious operation of the system.
Based on the above concerns, Entergy Operetions believes that, for ANO-1, the actual risk due to overfill scenarios is substantially lower than estimated in the basis NUREGs for GL 89-19.
It should be noted that NUREG-1218 incurrectly assumed that all B&W plants other than Oconee either had in place or had committed to modify their designs to include a safety grade overfill protection system. The Emergency Feedwater Initiation and Control (EFIC) system at ANO-1, a safety grade system, was originally designed with the capability for MFW overfill protection.
Doe te the concerns related to adverse consequences resulting from snurious cperation, questionable cost / benefit (cost estimated
,a excess of $1 million dollars),
ad expected (which have subsequently been implemented) improvements in the MFV and ICS control systems, Entergy Operations determined that overfill protcetion implementation was not appropriate. The MFW overfill issue was specifically addressed by Entergy Operations as part of the B&WOG Safety & Per-formance Improvement Program (SPIP).
Reference 4 specified a value of less than $200,000 for the installation of an automatic overtill protection system for the CE FWR plants which was used by the CEOG in their ^ valuation.
Since ANO-1 cannot be assessed 1n relation to the alternatives discussed for the B&W PWR plants, this evaluatien will utilize the low value of $200,000 as the cost to install the proposed system and the value which makes the option viable (Table 1). It should be reiterated that Entergy Operations estimated than an overfill systen, would cost in excess of one million dollars to install, Using $1000/ man-rem reduction in public risk yields a seven dollar cost benefit over 30 years without considering the negative impact of the system on safety. A calculation such as the CEOG presented would yield a negati/e cost benefit to the plant (Table 1).
If the seven dollar cost benefit, however, is compared to the NRC estimated $200,000 modification cost, the overfill protection modification is not warraated.
7
-n
CONCLUSION In conclusion, the insta11r.tlon of an automatic steam generator overfill protection c,yst.em for ANO-I is not justified from a public health and safety or test benefit standpoint.
Reviewing the data and input assumptions related to ANO-1, a B&W PWR plant, the same generic concerns pretented to the NRC by the CEOG on November 20, 1990, for CE PWR plents can be applied.
It has been shown that the public risk value can be reduced to 2.3 x 10-4 man-rem /yr from 45.4 man-rem /yr which is a difference of over five orders of magnitude.
Deliberation of these values and a seven dollar cost benefit vith a low value of $200,000 installation cost demonstrates that the generic letter recommendations are not justified for implementation at ANO-1, 1
8 l
l
'R'y.( '
,l. 'g. ~ f ~, I ^
.4<,;*,.>.
.),,
l REFERENCES 1.
NUREG/CR-4386, Ef fects of Control System Failures on Transients, Accidents, and Core Melt Frequencies at a Babcock and Wilcox Pressurized Water Reactor, Pacific Northwest Laboratory, Dec.cmber 1985.
2.
NUREG/CR-2958, Ef fects of Control System Failures on Transients, Ac.cidents and Core-Melt Frequencies at a Combustion Engineering Pressurined Water Reactor, Pacific Northwest. Laboratory, March 1986.
3.
NUREG-0844, NRC Integrated Program for the Resolution of Unresolved Safety Issues 4-3, A-4, and A-5 Regarding Steam Generator Tube Integrity, U. S. Nuclear Regulatory Commission, September 1988.
4 NUREG
~1, Regulatory Analysis for Resolution of USI A-47, U. S.
Nuclear Regulatory Comminsior., July 1989.
5.
CEOG-90-330, CEOG/NRC Meeting on S/G Overfill Protection (GL 89-19);
November 20, Forwarding of Summary, November 27, 1990.
6.
CEOG-91-143, NRC Summary of November 20, 1990 ~r0G/NRC Meeting on Generic Letter 89-19, March 11, 1991.
s 4
9 m
n J
.W
TABLE 1 CALCULATION OF PUBLIC RISK 1.
Definitfor. of Terms F0E:
Frequency of an Overfeeding Event P0F:
Probability of an Opc ator Falling to terminate the event PEBWST: Probability of Loss of BWST befcre RCS Depressurization (Table 2)
IE:
Steam Generator In! lating Event Frequency PMSLB:
Probability of a MSLB given overfill and spillover l
l PMSLbL: Probability of a MSLB Located outside the Reactor Builuing and inboard of the MSIVs given a MSLB PSGTR:
Probability of a SGTR given MSLB PCM:
Probability of Corn Melt give,n MSLB inboard of the MSIV and outside the React.r Building II.
Pormula for Public Risk (PR) l IE TOE
- P0F
=
i PCM = E (PSGTk
- PEBWST )
See Table 2 g
f i=1 IE
- PMSLB
- PMSLBL
- PSTGR
- PCM/PSGTR
- 4.8x10' man-rem PR
=
core melt III. Public Risk Utilizing Values from NUREG/CR-4386 P0F
= 0.7 IE
= 0.006/yr PMSLB = 0.95 PMSi,BL = 1.0 (Oconee does not have MSIVs)
PSGTR = 0.034 PCM
= 1.66 x 10-'
PR = 0.006
- 0.95
- i,0
- 0.034
- 1_.66 x_1_01' cure melt.
- 4, 8 *r 1b' man-rem yr 0 0J4 come melt
= 4.54 y 10' man-rem
'; t Total Public Risk for 30 years = 1360 man-rem 10
TABLE 1 (Continued)
CALCULATION OF PUBLIC RISK IV.
Public Risk Utilizing the New Values Applicable for ANO-1 P0F = 0.07 IE = 0.0006/yr PM SLB = 1 x 10 - '
PMSLBL = 0.155 PSGTR = 0.0505 PCM = 5.25 x 10-*
PR = 0.0006
- 1 x 10-3
- C.155
- 0.0505
- 5.25 x 10-6 core melt
- 4.8 x 10' man-rem yr 0.0505 core melt
= 2.34 x 10-" man-rem yr Total Public Risk for 30 years = 7.03 x 10-' man-rom V.
Cost Beneff t of Overfill Protection Assumption:
$1000/ man-rem benefit and $200,000+ modification cost.
Given: Total Public Rirk for 30 yea s = 7.03 x 10-8 man-rem Conclusion; Benefit of $7 over 30 years opposed to a $200,000 cost that was assumed appropriate provides justification for not implementing the modification on the basis of the cost benefit.
4 L
'NvREG 1218 delineates cost projections between $100,000 and $1,100,000 to upgrade stready existing overfill protection.
In additica, the CEOG utilities admitted during the November 20, 1990 meeting that a system could n_ot be installed for
$200,000 Therefore, this value is not representative of ANO-1 since it would cost far mo: 9 to install the automatic overfill p otection system.
1 11 l
4
TABLE 2 CORE MELT PROBABILITIES GIVEN MSLB CONSIDERING SGTTi NUREG/CR-4386 fREFERENCE 1) P. 2.10 Utilizes draf t NUREG 0844 Case 1: Rupture of Aain Steam Line Inboard of the MSIV Probability of Prob. of Net Number Prob. of Loss of RWST before Failure to Core-Felt of SGTRs Rupture RCS Depressurization Isolate SG Prob.
i 1
0.017 1E-03 1
1.7E-05 2 to 10 0.014 1E-02 1
1.4E-04
>10 0.003 0.5 1
1.5E-03 Total Probability of Core Melt Given MSLB Inboard of MSIV 1.66E-03 Case 2: Rupture of Main Steam Line Downstream of the MSIV Probability of Prob. of Net Number Prob. of Loss of RWST before Failure to Core-Melt of SGTRs Rupture
'a.~_Dqpressurization Isolate SG Prob.
1 0.017 1E-04 1E-03 1.7E-09 2 to 10 0.014 1E-03 lE-03 1.4E-08
>10 0.003 1E-03 1E-03 3.0E-09
~
l Total Probability of Core Melt Given MSLB Downstream of MSIV 1.87E-08 4
12 l
i
_m.m
-w w
W g
TABLE 2 (Continued)
CORE MELT PROBASILITIES GIVEN MSLB CONSIDERING SGTR REVISED PROBABILITIES Utilizes NUREG 0844 Finst Report (Reference 3)
Case 1: Rupture of MSL Inboard of the MSIV Probability of Prob. of Not i
Numb r Prob. of Loss of RWST before Failure to Core-Molt of SGTRs Rupture RCS Depressurization Isolate SG Prob.
1 0.025 IE-03 1
.2.5E-05
?. to 10 0.025 1E-02 1
2.5E-04
>10 0.0005
-5E-01 1
2.5E-04 Total Probability of Core Melt Given -MSLB Inboard of MSIV 5.25E-04 Case 2: Rupture of MSL Downstream of MSIV Probability of
' Prob. of Not Number Prob. of Loss of RWST before Failure to Core-Melt of SGTRs Rupture RCS Depressurization Isolate SG Prob.
1 0.025 1E-04 1E-03 2.5E-09 2 to 10 0.025 1E-03 1E-03 2.5E-08
>10 0.0005 1E-03 1E-03 5.9E-10
)
Total: Probability of Cote Mel,t Given MSLB Downstream of MSIV-2.8E,
c 13 A
__