ML20023A438

Forwards Info on Activities Re Task A-45,shutdown Decay Heat Removal Requirements.Activities Include Seismic Safety Margin Review Program,Fire Protection Program & Sys Reliability Studies

ML20023A438
Person / Time
Issue date:	04/08/1981
From:	Marchese A Office of Nuclear Reactor Regulation
To:	Murley T Office of Nuclear Reactor Regulation
Shared Package
ML20023A439	List: ML20023A438 ML20023A984
References
FOIA-82-412, REF-GTECI-A-45, REF-GTECI-DC NUDOCS 8104160723
Download: ML20023A438 (25)

Text

f.,.jpjti f

c.O u.a. -

o April 8,1981

,.s [' dkp d.(k jf'- 9 fH MEMORANDUM FOR: Thomas E. Murley, Dirutor i

Division of Safety Technology

~,

FROM:

A. R. Marchese

-()

' ' I I 684 Generic Issues Branch

• • %====

A Division of Safety Technology

(

^9

\\

THRU:

Karl Kniel, Chief Tb s

Generic Issues Branch Division of Safety Technology

SUBJECT:

ACTIVITIES RELATED TO TASK A-45, SHUTDOWN DECAY HEAT REMOVAL REQUIREMENTS Per your reouest, enclosed herewith is a discussion of those activities currently being pursued within the Comission and elsewhere (vendors, utilities, etc.) that are related to the Unresolved Safety Issue (USI) on Shutdown Decay Heat Removal Requirements. Task A-45. Alse. I have included a recommendation as to how we should handle each activity in terms of develcpment of a Task Action Plan. Table 1 at the end of the enclosure provides a specific reconnendation on how each activity should be handled. Generally, the recomendations fall into four cateoortes; nanely:

(1) suhtask under Task A-45 with the Generic Issues Branch (GIB) having the lead responsibilitly for resolution; (2) other Branch or Division has lead for resolution, but evaluation of the work by GIB to properly integrate the results needs to be a subtask under Task A-45; (3) other Branch or Division has lead for resolution, but GIB provides a strong interface to ensure that the results (including development of new requirenents) are integrated into Task A-45; and (4) GIB reviews the results of the activity when they become available and integrates them into Task A-45 efforts as appropriate.

As is obvious from the number of items in the enclosure, there are many activities going on related to decay heat renoval. Managing Task A-45 is going to represent a big coordination job to ensure the development of a consistent and comprehensive set of requirements. Wide distribution of this memorandwi will serve to alert all participants of the need for coordination and cooperation if MRC is to produce consistent and compatible licensing cuidance on this subject. I believe that I have touched base with almost everyone in the Cosmission that is working on activities that in one way or another are related to decay heat renoval, so in a

(

sense. preliminary interfaces have been established.

p.

By copy of this memorandun, those staff members involved in activities j ',I,i~

related to decay heat renoval should report back to me any discrepancies reported herein, anything that I missed, or any problems with ry suggested

(,

j

.16S N

T

~

~

g

)

s J-

=

?

l l

Thonas E. Murley '

l J

way to handle the activity as reflected in the enclosed table.

In addition, these individuals should keep me posted on any new developments in their areas that are relevant-to Task A-45.

If I missed any person i

that should be on distribution for this memorandum. I request that you pass same on to said individual.

need for a policy decision at the beginning en the sco>e,the possib10 I would also like to take this opportunity to highlight of effort under Task A-45. In order to define requirements for accepta >1e shutdown decay heat removal systems (DHRS). I presently envision that the scope of l

effort under Task A-45 include at least four major inter-related activities.

These are:

(1) development of criteria to judy the acceptability of lants;.(2) development of mesas for D! IRS in existing and proposed p(3) sssessment of DHRS. for specific plants;,

improvement of existing DHRS; or groups of similar plants to identify those for which DHRS impmvements are required; and (4) development of mcomendations of design altamatives j

for each plant, or gmups of plants,'in accordance with the criteria for acceptability (Item 4 will include a study of a separate, dedicated DMRS). A fifth major activity, or even an entirely separate approach, to Task A-45 could be based on a deterministic evaluation to determine what has to' be done to make all current operating LWRs meet current criteria. In part, this is being done for eleven plants under the Systeratic Evaluation Program (SEP).

With respect to performing the four major tasks outlined above I believe that we probably have to take a Probabilistic Risk Analysis j

(PRA) based approach, rather than one based strictly on either reliability of the DHRS or contribution to core melt probability. The need for this t t approach stems from two considerations.

First, the significance of the i

reliability of a given DHRS will depend upon its relative contribution, in terms of accident scenarios, to the overall risk derived from all scenarios, involving other safety systems and site considerations.

Second, risk reduction, rather than core melt probability or DHRS reliability, is likely to be the criterion which the industry (vendors, utilities, etc.) would advocate and also use to present its own assessment.

As previously indicated, a major parallel approach based solely on deterministic methods may also have to be pursued, which in itself is a considerable undertaking.

In this connection, a good suggestion (by R.

Lobel) was made to design a questionnaire that would result in licensees forming owner groups to determine what needs to be done to their plants in order to met current criteria.as reflected in the curmnt Standard i

Review Plans. Notwithstanding the above, I have been led to believe that some members of the ACRS will be dissatisfied if we recomend anything less than a dedicated, bunkered decay heat renoval system.

Because of the considerable variability in DHRS designs of LWks currently in operation, I am not optimistic about providing a sfogle technical resolution to Task A-45 that will be applicable to all plants. Hopefully, we can separate all existing plants into a limited num6er of grotas with

.mL to Ge3,. 4 the pleeti 6 a3 s -,, ffort under Ta sk. A-45...in..

14 the be L a L4 h e sin 11er manner. This should be a very early e I

4

..)

order to possibly 1,imit the amourtt of work req $1 red, t

'I j

h I

j _.

~

ihm e i.thiley April 8. l'al A' poli,- decision or, the scope of cf fo'rt becomes even snore significant if i.e tale a PRA bascd approach which unuld involve a substantial ef fort.

if TEls are not available for specific plants or groups of plants, llouever, we may be able to use the PRA infonnation that will be available in the near future, together with infoniution on DilRS design dit ferences to decide on grouping of plants.

In any event, we are going to have to inte our uope of ef fort under Task A-45 consistent with our available resources (internal staff,' technical assistance funding) and the required l

tiine to resolve this U51.

~

For tho'.c individuals on distribution, I would appreciate some, written early feedback on the contents of this meterandum before we get too far down the line in tenns of developnent of a Task Action Plan on Task A-45.

'.g fl 3 n,./f c-

/4:.

e &.2 C_

A. R. P.archese, Task Fanager Generic Is!.urs Branch. DST

/

.Inclosures:

As St.sted s:/ enclosures cc:

See next page N

s

\\

e N

e e

h a

9 e

er e

e e

\\-

. - n n

t i

i hvusi Murley April 8, 1981 w/cn. losures ct.

K. Kniel C. Graves F. Schroeder L. Shao T. Spels G. Bagchi B. Sheron S. Fabic V. Panciera L. Sullivan R. Lobel R. D1 Salvo R. Baer G. McPherson I. McKenna J. Richardson

0. Edison V. Benaroya.,..

t

.s :

R. Ferguson R. Mattson K. llerring H. Denton C. Nelson E. Case G. Ik)lohan

• A. Marchese (10)

F. Rosa R. Fraley ACRS (16)

P. 01Benedetto A. Thadant S. Israel.

.W.flefave N. Anderson Al-Szukiewicz.

J. Meyer D. Ross P.' Check.

~.

• .0. Parr.k, f%.c.!.M b:-.

21 D. Eisenhut E. ',

.c

.S'.Hanaudrjp.2..-U.%'N[t.

$.[.W,'C R. V611mer..

.'. x'.%

l i

G. Knighton g

J. Watt i.

. M. Taylor _.. 7

. y i

P. Baranowsky

..f 1, ~

f. Rowsome J. Stolz.

~

Z. Rosztoczy,-a.,.I

['

i J.'Conran g, Ernst'

'O

~

\\

R. Bernero

• s' "T

T., Scirt orougn'"

. :..' 3~ hy.,.

t

. -G.' Arlotto.

..f 5f. '

?

L... Rubenstein.

'h.

W. Morrison G.

J. Norberg S. Woo. key h

- W..An.d. e'cwn

~. '4' c

,y, w..,

s

,(,.

Por.,r

,w... o iw ye

&'.... v

. Nova >

s R.

Tedesco

(... t a( N '.,,i.

.:}'* H ' '-

..).'R6G 6MY ' " '...

' ' U.: :y.U.

c j

'W. Mil \\tead 4'#

J P..

.a'.

r a rumpaw w e. m.e r,.v

=

e._.-.~..

i s

iNCLO5tWE,

' ~

~

s

. 34 ty L-s-

ACilV111L',* RELATED TO 1 ASL A 452

[-

$HUIDOWN DICAY lif Al REMOVAL REQl!!RIMENTS.

L..

9

_;,..(

j m

'l.

Regulatory Guide 1.139 "routdanceifor Residual Heat Removal" (June 1980) 7

~

'This. propose'd Reg.tGuide'.'is considerably. more comprehensive than the

,;.previousve[sionl(Maflg8)or_,SRP15.4.7."ResidualHeatRemoval(RHR) fSystem.". 'IS,hou,1jl poin't ~out th'at, N,UitfG.0649 lists Task A-31. Residual

.>. ?.a 1

+,.; v > ~ =..

Heat.Rer' oval Requirem6ntias';an U5hthatihas.been resolved by.the issuance 2,

."

• v pt'.

.. -a:v.. : '....

• ' of the 4tay'~1978 'versforilof". Req.IGi$' n T ::i139.The..issub.,in A-31 was the derl

(

requ i rement to' g,..~e e :nj..+ e.c

"- ' n '...o from' hotto. cold. shutdown with'ut the.'use of offsite t

S o

power a'n'd ltt e'fiay [1kHNe( 15 kh gujdgprovidhs,a ' resolution of this. issue acceptable to tile staff. 'After the THI accident, it became

~

apparent'that the May.1978': version 3f.th'6,6 guide did not adequa'tely

"=

~

?

e

..c t ~ '

w ; 7 3.-

address.the need to cooldown with primar 7,gy coolant contariinated with c.,.

fission products (a k.1-actually experienced at. TMI),

.s.

m.

.'s....

'p., '.'.t

• . ' ~

' -Q.

' 4 ; 3 f;~.~&.g (A.'.[ Q,. C **'r'C W f

~9*

. ~

.The present. proposed versio'n!off.theluide does provide ' requirements for the RHR ' system to opera [Eiwith primary coofant tTiat. has been contaminated with activity resultingfrom.// degraded 'corb' accident '(f.~e., to encompass

., ~

. :.~ ;, :.

s.-

..8 theTMI-2~a(ctdent)'.

Tusk II+E.3.5.^" Regulatory Guide.".of'the TMI Action plan (NURLG-066bbtaththat dedraded core cond'itions w'ill not w<

...e,

be treated'in the latest'revifion to:Rs Q',,1;139., Howe'ver. SD (T.

s r

. n..;

v.

~

,3 Scarbrough)' claims',tha't'.dE yibEv'e Sno(" wasyi typoiraphicalferror; RSB n

y; -c w n.. ~..

.e

}

(T. Speis);plaims that,it was.not an : err'o'rQ, but 'that there was a cons'cious 1

and Eieliber, ate decision by NRR,,manaqcment.not to treat degraded core

- e..g

.s.

. r:s..

.;x v..

v-rond i t ions in 't h1Wev iTioII.9(A,the leu.: Gu ide bec'au'se 'shch.ef f ec ts will be treated in the context of the degraded core rulemaking.

I requested that V. Vniel resolve this apparent discrepancy between NRR and SD 1.

ria na gervn t.'

l i

A lot of wort has qnne into the preparatinn of this quide, including a nuut.er of i ter a t inna, annne

',D, NPP and ACR'..

In April 1980, the ACRS re< orrrnded t hat t he proposed quide he issued for publir rocrmint;-

• ;..... t. v i t i s.". J.i r.Eno,t l i' t ed in. aily ne di*i' of.3 ;, irt ans e or f rinri t v 7

.r o

~

7 y

this was af ter an extensive review by NRR.

In Jun'e 1980. 50 requested.

NPR's concurrence' to issue the guide, for public coment. The concurrence l

-package is presently being held by ths RSCB/ DST;' nothing has been done t

to' issue the guide since June 1980.-

I have reviewed this proposed guide and discussed 'it with key NRR and.50 staff nenbers; Notwithstanding the above discrepancy, I believe that it is consistent with present NRR staff thinking, represents a considerable improvement over the previous version and SRP 5.4.7 and should be issued for trial use and coninent. We cannot tolerate the above inaction for nine months. This guide is very relevant and important to resolution of SDHR requirements; therefore, I recommend that its continued development, including a consistent set of RHR requirements, be included as a subtask under Task A-45, and we should take the lead in the further development of Reg. Guide 1.139.

J 4-2.

Plants Without Seismically Qualified Auxiliary Feedwater System '

.(

(AFWS)

Ten PWRs have been identified as having an AFWS that does not meet our current seismic requirements. Based on a simplified probabilistic risk analysis provided in an August 8,1980 memo from R. J. Mattson to D. G.

Eisenhut, DST concluded that continued operation of these plants for the next three years is acceptable. The bottom line of the analysis is that the estimated probability of 3 x 10-4 per year of loss of decay heat removal due to a seismic event is acceptable for the next several years.

However, DST also concluded that a more detailed study of this question should be undertaken in the next several months and that plant operations could be permitted during the next several years unless the more detailed study shows otherwise. An ACRS letter of October 15, 1980 recomended that high priorily be given to resolution of this matter.

In a memo dated October 71, 1980, f. Schroeder expressed concern about the slower l

pace of the detailed study and suggested expeditious provision of a i

qualified dec ay heat reroval system, either an AIWS or a HPI feed and bleed ',y'.ttu.

In a rwrn, dated February 70, 1981, from D. G. Eisenhut

^

i to H. P. De-nton. D1 prov oh 4 a Mutliplant Att ion Plan C-14:

Seismic s

~

a 4

~ '

This action pian outlines an 18-month study Qualification of AFWS."

[

~ (starl' August 1980 - complete January 1982) to provide for a systematic l

upgrading of non-seismically' qualified AFW system's_ to a level of resistance I

for the'SSE consistent with that specified'for,other safety grade systems for operating reactors. This work is relevant to Task A-45 in the sense of generating requirements for seismic qualification of existing DHRSs.

Since DL has the lead on this work, I will plan on interfacing with them to ensure that any new requirements are integrated into Task A-45.

3.

Seismic Safety Margin Review Program (SSMRP)

The SSMRP is a five-year proaram intended to define the safety margin I

resulting from implementation of the NRC regulations in the seismic area. This study is also intended to define areas where additional j

efforts are necessary. The plant chosen fcr the study is the Zion plant which is typical of recent PWRs with fully qualified (i.e., confoming to Regulatory Guides 1.26, 1.29 and the Standard Review Plan) auxiliary feedwater systems. However, the SSMRP ar.6ytical methods and sensi.tivity studies can provide useful infonnation in the review of older plants as well. It is these aspects of the SSMRP which will be called upon in the action plan, C-14 as discussed in Item 2 above. To that end, the Office of Nuclear Regulatory Research has been requested to apply the SSMRP analytical methods to a study of the San Onofre 1 auxiliary feedwater system. Lawrence Livermore National Laboratory (LLNL) will perform sensitivity studies and an analysis of the failure probability of that system. The auxiliary feedwater system sensitivity studies for the Zion plant are complete and will be available for use in DL's review. OL

~

expects to use information developed by the SSMRP in their review of the licensee surveys, their review of the walk-down and tie-down survey results and their review of the ifcensee sei<mic upgrading programs.

This study, as well as the one discut. sed under item 2 above, will address the requirements for seismic qualification of decay heat removal systems and will also a'ddrese. (in a eense) the desirability of requiring a

'd dic ated decay heat rmoval *.rstem with seismic qualification for all e

4

,4, s

PWRs. Therefore, as with Item 2 above, I will plan on providing a f'

strong interfge with DL and RES to' ensure that the results of this program are integrated into Task A-45..

s 4.

Fire Protection Program The Comission published it's final rule in the Federal Register (Vol.,

45, No. 225, p. 76602) on November 19, 1980; the effective date of the e

rule was February 19, 1981. This rule will replace Appendix R. " Fire Protection Program for Nuclear Power Facilities Operating Prior to

~

January 1,1979," to 10 CFR Part 50.

In general, the rule sets forth fire protection features required to satisfy Criterion 3 of Appendix A

^

to 10 CFR.Part 50. Besides specific fire protection requirements, the rule provides for alternative or dedicated shutdown capability.

s Apparently, because of the pressures of TMI-2 and the NRR reorganization, the prov.ision for alternative or dedicated shutdown capability was not

[

reviewed by those NRR branches who nonnally develop such requirements.

The Chemical Engineering Branch (DE) could not get any other NRR branches to review their shutdcen requirements, so they went forward on their.

~

own.

The rule states that "..'.in areas where fire protection features cannot ensure safe shutdown capability in the event o'f a fire in that area, alternative or dedicated safe shutdown capability shall be provided."

The shutdown system shall be capable of achieving cold shutdown conditions within 72 hours8.333333e-4 days <br />0.02 hours <br />1.190476e-4 weeks <br />2.7396e-5 months <br /> and maintain cold shutdown conditions thereafter. The shutdownsystemneednotbeyesignedtomeetseismicCategoryIcriteria, single failure criteria, or other design basis accident criteria except where required for other reasons. Therefore, this dedicated system is essentially a minimum capability safe shutdown train. Apparently, for a potential fire in the cable spreading room of some plants, it is not possible or practicable to prot'ect redundant safe shutdown systems

'~

=

.-<m,,

-m m

S

'5

~

against adverse ef fects of the fire only through the use of fire protection features because the redundant safe shutdown systems in a given fire area are too clost to each other. Some plants (Oconce, McGuire) have already comitted to installing a new dedicated' shutdown cooling system l

4 f

just to meet our current fire protection requirements.

~

We will need to integrate the fire Protection Rule requirements into our study of shutdown decay heat removal requirements. There may be incon-sistencies (particularly in the time to achieve cold shutdown) between the fire protection rule and Reg. Guide 1.139 SRP 5.4.7 and BTP RSB 5-1.

Once inconsistencies are identified, we can decide on the best course of action for resolution.

It is anticipated that the decay heat removal requirements which result from the completion of Task A-45 will I

be more stringent than those steming from fire protection considerations; therefore, any alternative decay heat rqmoval system resulting from Task A-45 should satisfy the current fire protection requirements.

i 1

5.

Staff Position With Regard to Near-Term Construction Permit l

{

R_equtrements With Respect to Degraded Core Rulemaking -

februar1, l'M1 1

6 I

This position which is being considered as a final rule would (among other things) require licensees for NTCPs and M.s to perform a site / plant-specific probabilistic risk assessment (PRA) and to incorporate the results of'the assessment into the design of the facility. A prevention feature that must be considered is an additional decay heat removal system whose functional requirements and criteria would be derived from the PRA study. As these PRA studies become available, we will need to review and incorporate the results into Task A-45 as appropriate.

I' -

Ndtural, f.ircula,tinn Cooldown Itaseil on an evaluation of the generic aspects of the St. tucie natural

< nnvet t ion roolifown event in whic h a void forwd in the upper head of a

4 1

~

d the reactor vessel, the staff has identified (Ref., Memo, D. F. Ross to D. G. Eisenhut, "FollowJp Action on Natural Circulation Cooldown," dated

~

November 20,1980) additional efforts that must be taken by the industry to ensure that a similar event would not lead to more serious consequences.

r i

The staff suggests that a lower cooldown rate coupled with holdfr:g the plant at intermediate condit ions to allow the fluid in the upper vessel

. to equilibriate with the rest of the primary system, would preclude vessel voiding under natural convection cooldown. However, the lower cooldown rates would increase the time required.to achieve low pressure shutdown cooling system entry conditions and thus increase the time auxiliary feedwater is depended upon to remove _ decay heat from the loss-

/

of-offsite power case. The staff will require each operating PWR,

~

licensee to (1) demonstrate that their natural circulation cooldown procedure will preclude vessel voiding; (2) identify how'their operating procedures have been modified to ' assure that the operator avoids vessel voiding under controll.ed conditions, and recognizes it if and when it occurs; and (3) demonstrate that their plant has sufficient condensate-grade auxiliary feedwater capacity to support their cooldown method.

Specific wording of questions to licensees that would clarify our intent-was: suggested in a memo from T.' E. Murley to H. R. Denton, "NRC positions and Reviews Concerning Shutdown Cooling of Light Water Reactors," dated February 13, 1981.

DL is in the process of sending these questions out.

Although D51 has the lead in this area, we will provide a strong interface during the evaluation of the licensee responses to the above questions, especially if any new shutdown cooling requirements appear to be developing.

7.

Station Blackou_t_(Task A_-44)

A tas6 at tion plan f or the IISI, " Station Blackout," Task A-44, was approvr d by a mmo f rom R. M. Bernero and R. J. Mattson to H. R. Denton da t ei1 Nlv 14, 19H0.

The purpose of this task is to evaluate the adequacy h

O %R PE

j 7 _.

s.

L s.

of current licenting design requirements to assure that' nuclear power j

plants do not pose an unacceptable risk for a station blackout accident,

• The principal focus of this task will be on evaluating the reliabilit~y f~

of the emerge,ncy onsite AC power supplies. These studies will include l

enmining the reliability of shutdown cooling systems g'iven a loss of AC power supplies, an evaluation of the hazards posed by extended blackouts, and reactor coolant inventory regirements'during station blackouts.

This infonnation will be'used to establish acceptable requirements for

^

l AC power supply reliability and decay heat removal capability for station s

blackout, The completion date for this program is estimated to be October 1982.

i To some extent the effectiveness of decay heat removal systems is linked to that of the on-site and off-site power supplies.

Consequently, the

^

scope of work required in Tasks A-45 is complementary to that of Task A-44 Coordination of work li. the two tasks will be through the section leader in the GIB with technical responsibility for both tasks. He will ensure that any new requirements from Task A-44 that are related to shutdown decay heat removal are integrated into Task A-45 evaluations.

8.

Sa_ndia Program on Alternate Decay Heat Removal Concepts As part of Task II.E.3.4, " Alternate Concepts Research," of the TMI Action Plan (NtiftEG-0660), RES has Sandia Laboratories under contract to perform a study of alternate decay heat removal concepts for LWRs, including examination of current systems and proposed new system options for PWRs and 141Rs. The two-year Sandia study i; expected to be completed in September 1981.

Interim conclusions reached by Sandia af ter one year of study are provided below.

The best'way to improve the reliability of pressurized water reactor l

(PWP) decay heat ren.n va l is first to focus on improving the reliability 4

9 w

Sob M

em A s_

L V

.y, '

~

of the auxiliary feedwater and high pressure injection systems to cope with certain loss of feedwater transients an'd small loss 'of coolant.

accidents and then to assess how well these systeras can handle, special emergencies (e.g., sabotage, earthquake, airplane crash). for boiling l

.s s.

water reactors (BWRs), it was concluded that emphasis should be placed first on improving the reliability of the residual heat' removal and ht'gh l

pressure service water systems to cope with a loss of suppression pool cooling following a loss.of feedwater transient and then to assess how well these systems.can handle special emergencies.

It was found that, for both Pn/Rs' and BWRs, a design objective for alternate decay heat

~

removal systems should be at least an order of magnitude reduction in core Nitdown probability. Several alternate decay heat removal conce' pts

.which appear to ineet this design objective are proposed for further 1

evaluation.

This RES' program will be a very important subtask of our T'sk Action.

~..[

~

a

' Plan in terms of evaluating alternative shutdown decay heat removal

~

t systems. We have already started to interface with RES (Mat Taylor) and will continue the dialogue during the development and execution of our Task Action Plan. It is anticipated that it will be necessary to fund

~

l Sandia through FY82 to further support our efforts.

l 9.

Existing and Planned Systems Reitability and Risk Assessment 5tudies r

e There are numerous existing and forthcoming systems ~ reliability and risk assessment studies that will include among other things, an assessment 1

of the DilRS An tems of overall plant safety. The following studies are espected to be relevant to Task A-45:

- lion and Indian Point risk studies by Pickard, Lowe and C,arrick

- Liwrick study by General f lectric o

a h

l

- T

r 9,.

~

+-

e

- RSSMAP (Reactor Safety Study Methodology Application Program) conducted. for Sequoyah, Grand Gulf, and Crystal River by

~

RES/NRC, n

.p f

- Oconee study' by NSAC and Duke Power a,

- NREP (National Reliability Evaluation Program) risk studies to be conducted by the utilities.

Z

- WASH-1400 risk study for the Surry and Peach Bottom Plants

~

- IREP (Interim Reliability Evaluation Program) which will examine the contribution of various accident sequences to core melt li probability for Crystal River, Calvert Cliffs. Browns Ferry.

ANO-1, and Millstone-1 1l

- Big Rock Point risk study by SAI qL 4

- Davis Besse AFWS reliability study

~

The review and evaluation of the results of these studies (and perhaps others not yet identified) with respect to decay heat removal will be a subtask under Task A-45. The results will be integrated into Task A-45 as appropriate. We will need to depend heavily on the Reliability and Risk Assessment Branch (RRAB) to assist us in the evaluation and utilization of the above studies in performing Task A-45.

In a separate memo, I will plan on defining specifically what we need frcm RRAB.

The above studies may be helpful.in a determination of the extent to which gmuping of plants for the purposes of the overall Task A-45 program is possible.

10.

Syster.aticIvaluationProgramjs_E,PJ 01 has underway a Systematic f valuation Program (SEP) to perforin an integrated saf ety assessmnt f or 11 current ly operating plants licenseo s

s I

2<

i

_[

prior to 1970,[idluding Palisades, Dresden-1 and 2, Oyster Creek,

~

u Millstone-1, Ginna,..Haddem Neck, San Onofre, Lacrosse, Big Rock Point, c

and Yankee Rowe., This assessment examines the extent to which each.

facility meets'c'urrent criteria used by the Regulatory staff for licensing new facilities. This program effectively began in 1978. At that time, 137 topics were approved for study. The program is expected to be conpleted in 1982. Recently, the staff has issued evaluations (for all 11 plants) of those topics relative to Safe Shutdown Systems, including

'l residual heat removal systems reliability (Topic V-10.8), requirements...

for isolation of high and low pressure systems (Topic V-ll.A), RHR interlock requirements (Topic V-11.8), systems required for safe shutdown (Topic V11-3), station service and cooling water systems (Topic IX-3),

and auxiliary feedwater system (Topic X). We will need to review the

~

results of this program relative to shutdown decay heat removal and

.l

'~

integrate the results into Task A-45 as appropriate.

i 11.

Power Systems Branch (PSB) and Instrumentation and Control Systems, Branch LICSB) Activities Relative to Shutdown Decay Heat Removal Task II.E.3.1, " Reliability of Power Supplies for Natural Circulation "

of the TMI Action Plan has been completed. The staff issued require-ments for (1)' upgrading the pressurizer heater power supp'y and associated motive and control power interfaces sufficient to establith and maintain natural circulation in hot standby conditions, and (2) estaolishing new procedures arid training for maintaining the reactor coolant system (RCS) at hot standby conditions with only onsite power available.

IE will inspect the resulting inpleme'ntation.

ICSB is starting to receive licensee responses to IE Bulletin No. 79-27,

" Loss of Non-Class I-E Instreentation and Control Power Systems Bus During Operation," issued November 30, 1979; however, they don't have I

the n.anpower to perform a detailed review of all responses. Supplement 1 to Bulletin 79-27 will not be issued. The bulletin primarily addresses

)-

n l

1

.(

- 'Y C ;;G c.

. n,s

+

.Z l

(

m1 ' ' ;-

- 11,

. $ 'f L.

the loss of power to instrument buses and not 'other power buses; the.

w a

I

~

. w.n.

..f ',

bulletin does not require equipment to be, safety grade.

e e

~

l

.,gg

'C

~. _

l PSB is assisting in the fire protection reviews of. near tern 01. plants.

-" y, "

In a February 20, 1981 letter to all power reactor licensees with plants N'lO

~

... am nm licensed prior to January 1.1979, Ot.',provided a staff position on safe

1.,

shutdown capability relative to the recent1y issue"d Fire Protection Rule

,'; I p (Appendix R to 10 CFR 50) and a request'for additional inforCice

. f ",

relative to meeting Appendix R including an identification of all equipment (mechanical, electrical) and related cables that are required by the alternative or dedicated method of achieving and maintaining hot

, / i g' t '.-

^

shutdown to aid in identifying possible losses due to fire.

In a'meno

-9.J ],Q ;,

from P. Check to V. Noonan, dated February 23, 1981 PSB provides supple-h2.[

mental questions relative to fire protection.

__..g

iQ:cs a

w a

We will have to interface with ICSB and PSB to ensure that the results of their review of the above area's are integrated into Task A-45 as

. %p Jl

. appropriate. We will also need their assistance in Task A-45 wort N

3 insofar as establishing an integrated set of electrical requirements for O

any alternative decay heat removal systems.

12. High Pressure Residual Heat Removal System There is currently no activity in the Consnission or elsewhere that has

~

been identified. A high pressure RIR system that is capable of operating at reactor coolant system pressures and temperatures soon after reactor shutdown offers certain distinct. advantages, especially in simplicity of systen operation. We should agressively pursue this system; therefore, it is reconnended that a study of high p: essure RHR fystems be a subtask of Task A-45.

l H

s i

p.

- e 3

o

, a '. s

'~

(

?

-l l f '

12 -

y

+;

y,

,u-t

13. Auxiliary Feedwater ' System (AFWS) Evaluations U.

N j.?

$l In accordance with the TMI Action Plan,Section II.E.1 the Auxiliary s

Systems Branch (ASB) is performing work related to improving the reliability

~;

of the AFWS. The staff required implementation of the short-term and

~

long-term recoamendations for improving AFWS reliability as reported in i

NUREG-0645, " Report of the Bulletins and Orders Task Force," dated January 1980. The staff required all PWR operating license applicants to (1) evaluate AFWS reliability. (2) provide a deterministic AFWS

[

evaluation, and (3) provide AFW flow design basis information for staff i

review. In addition, the staff required the installation of (1) a control-grade system for automatic initiation of the auxiliary feedwater system that meets the single-failure criterion, is testable, and is powered from the emergency buses; and (2) control-grade indication of f

.}

auxiliary feedwater flow to each steam generator that is powered from

~

u emergency buses, in accordance with short-tern lessons leamed recommendatfons b

2.1.7.a and 2.1.7.b in NUREG-0578. The staff has been receiving responses from the affected licensees relative to the above and is in the process of evaluating the infomation, including writing SERs where positions will be taken. The ASB also plans to update SRP Section 10.4.9, " Auxiliary Feedwater System (PWR)," to make it consistent with B&OTF reconnendations.

~

We will need to maintain a strong interface with the ASB work to ensure t

that their positions are integrated and consistent with Task A-45 work.

1 We will also need help from the ASS on Task A-45 work related to alternative, dedicated shutdown decay ~ heat removal systems.

~

14 Reactor Vessel Cracki'ng As A Result of Overcooling Transients Thit is a relatively recent staff concern in which overcooling transients could re', ult in reactnr vessel failure by low temperature brittle fracture.

Severe therrml transients, such as those resulting from MSt.Bs, small LOCAs, plus others.in which the vessel could be subjected to high pressure

. l

~

I

~

-(

. ~

~

. w N

during or following a. transient, can cause deep vessel cracks if pre--

transient small cracks exist. Reactor vessel failure cou'Id result af ter -

,. T a transient,if the vessel isl pressurized with relatively cold water.

The. likelihood of vessel failure depends on its materials, irradiation and severity of the cooldown transient. -The most vulnerable plants are

. th'e oldei PWRs where significant radiation over 10-15 years has occurred

,resulting in radiation sensitive welds. The RSB and the MTEB are performing systems analyses and vessel integrity analyses, respectively. Appropriate licensees will be requested to perform such evaluations. Potential remedial measures include (1) automatic actuation of PORY to limit pressure, and (2) operator action to limit pressure. We will need to

. interface with RSB and MTEB in this area to ensure that any new systems I

considerations (such as pressure / temperature limits for maintaining vessel integrity) are integrated ~1nto Task A-45 efforts as appropriate.

15. Feed and Bleed System

^ -. -

This is a means of decay heat removal in PWRs that involves bleeding off primary coolant through the PORVs and injecting fresh coolant into the primary circuit by means of the HPI pumps.

In a memo to D. Eisenhut,

(DL) from F. Schroeder (DST), " Plants Without Seismically Qualified AFWS." dated October 21, 1980, preliminary design criteria were proposed for a seismically and environmentally qualified feed and bleed system.

For plants with HPl.; pumps..having a sufficiently high cut'off head (most f4W,1/2 'W);" fluid can be injected '...into the. primary system when the pressure is at.the PORV or safe'ty vapesetpoint. For plants with HPI pumps having l lower pressure cutoff EeadE.'(112 W and.most CE), it is necessary to manually open the PORVs in order to reduce the system pressure to the operating range of the HPI pumps.

Feed and bleed requires power. so t6e reliat.ilyty of emeigency AC power supplies is critical.

In a cierio to Farl,rniel.(GIB)' dated March 31,1981, " Status of feed and hiced fnr Emergency Dec2y Heat RemoEal," B. Shero,n '(RSB). documented Jhe present status of feed and hieed systems., We"should include a

.yster>at'ic ' examination of 'fet d and bleed s'ystems as a Aubtask under Task i

' ' k,,_

~ Y.h 5

A-45.

~

' ; - &n,5M Q:";;

acd'M

,y l_- - -

u,

'~ :'

.g O

g-p a

_.. y. < -

[

.o.

[.

. 14 l

t

'..t

[

(

t

~

c flux Condensatfori4

'?

i 4

16.

e c.

. ; {&

n Without overall ~ natural fcNeNion,aiid with steam generation in the

'~

.n..-em core, it may be possible~ that.the steam generating unitr, could be used

~r.

as reflux condensers to transfer a significant amount of heat to the secondary side.provided that. the; inlet temperature of the feedwater were sufficiently ' low. The\\ steam generated in the core could be condensed in the "up"41eg of a 0'-tubefteam generator and the condensed liquid

'}

j water would flow back into the' reactor vessel via the outlet main (i.e.,

in counter-current floiwith the steam). One of the Kemeny Comission recomendations was to pursue'the' use of steam generators in a reflux j

t

. condensing mode. Further investigation would be' required to establish the feasibility of this method of 5echy~ heat removal.

It is recomended

~

that this evaluation be a subtask under. Task A-45.

J r

17. Environmental Qualification The basic issue is to assure that safety systems that perfom their intended functions to mitigate the consequences of postulated accidents 3

are capable 'of pe.rforming their functions in the most limiting environrient which occurs as a result of tha'. accident. Confirmation that those systems will rema'in functional.under.postdlated accident condition' constitutes environmental qualification.-

~

~ l r

~

A. '.,

'f Currently, the Comission has' underway'.a program to reevaluate the 1

. Qualification of. safety-reljited electrical. equipment,in all operating,

)

reac trirs. As.partopthiNhEb'riras.. definitive,,criteriaforenviron-rnental qualification of safety-related electrical equipment have been J

develooed by the staff.

The Division of Operating Reactors' " Guidelines

\\

for ivaluating Environmental Qualification of Class IE Electrical Equipment in Operating Reactors" (00R Guidelines) were completed in November 1979.

In additinn, for reactors presently under licensino review and for future plants. the staff has developed NUREG-05M, " Interim Staf f Position on Invironrental Qualification of Safety-Related Electrical Equipment."

c w

v

- La _

1

'O

..k s

~~

On May 23,1980, the Comission ordered that the above two documents fonn the requirements which licensees and applicants must meet in order' y

~

~. to satisfy those aspects of 10 CFR 50, Appendix A General Design Criteria

. (GDC)-4', which relates to environmental qualification of safety-related 4

z.,

.5 equipment. As s' result, the staff is evaluating the qualification of

,};(

t

'o all electrical safety equipment in operating plant pursuant to the s-Nidelines.

If probless arise, the intent is' to resolve.the problem j

using NUREG-0588 as a guide for the staff's _ judgment.' For NT0Ls and future

~

, y" 1

plants, the staff is evaluating the qualification programs using NURER-0588.

y N

l In addition to the above requirements, the staff is in the process of

,i..

rulemaking which may modify or change the abovementioned requirerients.

~.,

Accordingly, we will need to provide a strong interface with EQ8 to

",.2

~

ensure that existing and any new environmental qualification requirements are integrated into Task A-45 in a consistent manner.

~ N 1

18. Severe Accident Mitigation Features and Degraded Core Rulemaking

[

l DSI is performing a study for the Zion and Indian Point (ZIP) units to determine to what extent additional design features, such as Filtered-4 Vented Containment System. Hydrogen Control System, Core Retention-System, and Passive Containment Heat Removal System, mitigate the consequences of severe accidents beyond the current design basis. A draft " Interim

^

peport" on the severe accident mitigat.lon features for'the ZIP plants has been issued by the staff for internal review.

It is expected that this interim report will be published in the spring.of 1981. There is alsn a parallel activity in progress on a ZIP probabilistic risk analysis (PR4) which is being sponsored by the 71P utilities and being perfomed 0

T

. - -.. y

• ~

'+

g

~

C w }.

f 9

m

(,

i..

l' '

'4:

~

by Pickard Lowe and Garrick (PLG). DST /RRAS has the responsibility,0f -

~

reviewing the PLG risk analysis, which is expected to be published in

.a

~

~-

. the sunener of 1981. RRAB w'111' make the' determination of whether ZIP,

. represents undue risk. Only if the staff concludes that ZIP represents undue risk and that mitigation features can substantially reduce risk will-the staff recocinend to, th.e Commission that certain mitigation features 'be required'.. If it is determined that they do not represent undue risk, the ZIP action will be incorporated into the upcoming rulemaking on degraded core accidents fcr all LWRs. NRR will need to review the

~ '

results of these efforts, in particular the cost / benefit evaluations, as

~

part of developing information for a decision on whether its best to 1 v.

" prevent" or " mitigate" the consequences of severe accidents for a v.

particular plant.

r,

19. ACRS Quantitative Safety Goals _

J.;.

In NUREG-0799, dated October 1980, the ACRS published "An Approach to Quantitative Safety Goals for Nuclear Power Plants." This is the ACRS proposal for a trial approach to quantitative safety criteria. The safety criteria or decision rules are broken down as follows:

-Limits are placed on the frequency of occurrence of certain hazardous conditions (hazard states) within the reactor.

- Limits are placed on the risk to the individual of early death, or delayed death due to cancer arising from an accident.

- Limits are placed on the overall societal risk of early or delayed death.

/

_e=

l l,

p+

-9,

}

. k J-l

.v i

- An "as_ low as reasonably achievable" approach is applied with ~

a cost-effectiveness criterion that inciddes both economic costs

~

)

.,m o..

. : e.

and a' monetary value of preventing premature. death.

,a.

3-

,c y

7,p a.

l

- A small element of risk aversion is applied to infrequent accidents.

~ involving large numbers of early deaths] compared to's similar

[.g 4:

number of deaths caused by many accidents each involving one

~'.,

I or two deaths.

p. ll i

The quantitative values suggested for use in the proposed decision rules.,

M, g

are intended to be applicable for new light water power reactors and nay

-,g,,

be more stringent than is deemed appropr.iate for existing plants.

l

. ;4 t

L ag Fith respect to hazard state limits. the limit on the frequency of a

.. ;f large offsite release assuming that a fuel melt has occurred, places x

enphasis on mitigation as well as prevention of serious accidents. Such

'N a division between accident prevention and accident mitigation is believed i(;

to.be necessary because of the difficulty in demonstrating with a very 3~

high degree of confidence that a frequency of large scale fuel nelt much 3

less than the proposed goal of 10-4 per reactor-year can be achieved in a

view of the complexities introduced by consideration of matters such as sabotage, earthquakes. and other potential multiple failure scenarios.

The development of acceptance criteria. upon which the adeouacy of current shutdown decay heat removal systems will be judged, will form a ma.jor suhtask under Task A-45.

As part of this subtask, we plan to evaluate the ACRS criteria and integrate their proposal as deemed appropriate.

70. ' PAS Interim Ouantitative Action Criteria In a nma from R. Bernero to R. Mattson dated July 22. 1980 PAS proposed an inter im standard for corrective action related to the probability of e

e o

=

M sr" W

J 18 -

1 y

a serious accident. The pr'oposed interim standard is as follows:

r s

CD > lx10-2/yr:

[

P f'x in days

[]

i 2

lx107 /yr > PCD > lx10-3/yr:

fix in months J-

_;i lx10-3/yr '> PCD,> lx10-4/yr: fix in years

~

4 f

z lx10-4/yr > PCD > lx10-5/yr: consider fixing

~

lx10-5/yr > P

acceptable CD where P is the probability of severe core damage CD

u.

As with the ACRS criteria outlined above, we plan to evaluate the proposed e

PAS criteria and integrate their proposal into Task A-45 as deemed appropriate.

s 21 Systems Interactions The problem of potential systems interaction has been under study for several years as part of Unresolved Safety Issue A-17. The Phase I effort has been completed by the Sandia 1.aboratories and reported in NUREG/CR-1321 (April 1980). ~The Sandia study used fault,-tree methods to identify component failure combinations that could result in a loss of

(

safety function.

The TMI Action Plan (NUREG-0660/Section II.C.3) also proposes a systems interaction program to define and subsequently implement system interaction regulatory requirements and guidance on LWRs. The Systems Interaction Branch (518) is developing an independent methodology for identifying

.ind evaluating systems interactions.

In this vein. SIB has contracted for studies at BNL BCL and Lt.L; interim reports have been received from those labs; namely NURfG/CR-1901. (dated January 1981), NUREG/CR-1896 (dated January 1981) and NUREG/CR-1859 (dated January 1981), respectively.

~

m M

4

~

- 19 This program wilk focus attention on common cause failures. kttention, will.be concentrated on those common support systems that feed primary safety' systems, where a common cause failure could result in the loss of

~

~

a prinary safety ' function. SIB is finding that 10ss of non-safety power supplies is the biggest cause of common cause failures.

In short, SIB is trying to find and fix the weak links in systems in terns of connon cause failures, including decay heat removal systems. Walkthroughs of the plant have been identified'as a valuable tool in providing important j

information on potential systems interactions. A detailed systems interaction evaluation will be done _for Indian Point 3.

SIB was planning to issue interim Regulatory guidance by the end of this fiscal year that would spell out what industry should do in regard to systems interactions. However, SIB has been broken up due to the heavy casework load and this interim guidance will be delayed.

Apparently, there will be two individu'als that will continue to be involved in the systems interaction program. We should plan on following the Indian Point' effort very closely relative to decay heat removal systems and establish a strong interface with SIB to ensure that any' new requirements in the area of systems interactions are integrated into Task A-45.

/

s e

I

~

-J

- w._w.m

'is m

t

.r a

TABLE 1 RECOMMENDATIONS ON HANDLING ACTIVITIES RELATED TO TASK A 45 Subtask Under Task A-45/

GIB Does Not Have GIB Does Not Have GIB Does Not H>ve Lead for Resolution /

Lead'for Resolution /

Lead for Resolurton/

GIB Provides GIB Reviews Results Subtask Under GIB Evaluates &

Strong Interface /

When Available/

Task A-4S/

Integrates GIB Integrates ^

GIB Integrates.

GIB has Lead Results Results

-i

?Results as

. Activity for Resolution.

Into Task A-45 Into Task A-45

' Appropriate

~

,.o

r..

. q;. ~

.1' T

1.

Reg. Guide 1.139, X

/

~

" Guidance for Residual Heat Receval" (June 1980) 2.

Plants Without

'X Seismically Qualified ATWS 3.

Seismic Safety Margin X

Review Program (SSMRP)

4. > Fire Protection Program X

5.

Staff Position With X

Regard to Near-Term Contruction Permit Requirements With 1

~

Respect to Degraded Core Rulemaking'

~

^

6.

Natural Circulation

~

X Cooldown 7.

Station Blackout X

(Task A-44)*

8.

Sandia Program on X

Alternate Decay Heat 3

Removal Concepts

'L'TB has lead for resolution of Task A-44.

l

U

/

TABLE 1 (CONTINUED)

Subtask Under

. Task A-45/

GIB Does Not Have GIB.Does Not'Have GIB Does Not Have

-Leod for Resolution /J.

Lead for Resolution /l Subtask Under GIB Evaluates &

Strong Interface / -

,GIB Reviews'Results' ',.

Lead for Resolution /

GIB Provides

~

.When Available/

~

Task A-45/

Integrates GIB Integrates

.GIB Integrates GIB has Lead Results Results Results as Activity for Resolution Into Task A Into Task A-45 Appropriate 9.

Existing and Planned (Needs further X

Systems Reliability and definition)

Risk Assessment Studies

10. Systematic Evaluation e

X I

' Program

11. PSB & ICSB Activities X

Relative to Shutdown Decay Heat Removal

12. High Pressure Residual X

j._,,

Heat Removal System

13. AFWS Evaluations X

14. Reactor Vessel Cracking X.

as a Result of Over-cooling Transients

15. Feed & Bleed System X

16. Reflux Condensation X

s 17.' Environmental 0'ualification X

e e

.y4 3

g

(

_c gf#

p 3, ' f ;. n'l, _ f*

'f5 s-

_r.

e l_& _1_

' 1

e' t

'e TABLE 1 (CONTINUED)'

Subtask Under Task A-45/

GIB Does Not Have GIB Does Not Have GIB Does Not Have Lead for Resolution /

Lead for Resolution /

Lead for Resolution /

GIB Pmvides GI8' Reviews Results Subtask Under GIB Evaluates &

Strong Interface /

When Available/.

Task A-45/

Integrates

.GIB Integrates

(GIB Integrates

~

GIB has Leads Results Results Resultstas1 :.

Activity for Resolutilon Into Task A-45 Into Task A-45

~Aporopriate]

. ;Jg,

.g.;,.., ww.

7 :g;p;c t rJ.

,.4 f.y.

r

18. Severe Accident

.X Mitigation Features I

and Degraded Core Rulemaking

19. ACRS Quantitative I

X Safety Goals

20. PAS Interim Quanti-X tative Action Criteria

21. Systems Interactions X

I

-,.i

.i s. ~

.i -

. w:..

~r s

t s

y.'-

r

'k-

• g t.

=.

=

-.M.-

+

..a f? V f\\.,

Y'

(; ;,i '

L c

g2FS i*d:':ORM DUf TOR: ;R. J. '!!attson, Director Division of fyster.s Safety Office of 1uclear Reactor Reculatior.

rn:r;:

R. Earnero, Director P'robabilistic Ana. lysis Etcff i'

of fice of 1:uclear Rogulatory Research.

Stt?. JECT:

CRYSTAL P.IVER ITCP STUDY A*!D INTERI?i CRITErlJ.

~ ' '

TOR ACTIO*i

,.c you cra Owcro fro our recent discucrienc the C istel F.iv::

e 3,

27, ? c;; dy 1 nc aring co.: letion.

Thz interr.:1 peer ravise k:,'

P.,

.::': :t:f f r.:r>2rr end the plant cr.:ncr's representativer has cos:

3 c.;. :::.i.

a h:.va r.:.ny co. cats fro = the !*r::: staff cud c:;r:=.

at k

1 [7p r ::.y fre.- the os nar shortly.

Our host estir. ate is that it vill 1:%2 cnoth:;r 2 r.onths to obt.ain all the co.snts, deal eith thc=.,

7 cnd co:picts the report.

As we cpproach the put,11 cation of this

?.fi r:.- ore va cre conscious of the corr.it=:nt to provida gencrr.1 cnf

,G; ple..it nyecific recorr.cacctions b: sed on the kno.1odas ceined ir.

.i

t. is relitbility evaluatica of the plant.

. 3 41 T.::c6 on s hat to see so far in the Crystal River 3 study, r.ayc:

~ <

1:

41.-:.; c:rr.ctive..ctica does not appe:r to be rcruired.

T.u

rc :.r.: prediction of.the likelihood of covere cores Ct=ae:

., pro::i?:.tely 3 :: 10'/yr.

This doen.not sicr.ific ntly c:: c.*

ta: r: r;n of valuos for other L:TRs we have ext: ined.

1.*

believ:

the f.'.I

.ncly:is of Crystal niver 3 is conservativo in nsny re:c. acts, c1though potential accident sequences which vore not g

snr.lyzed r:y countsr-balanco this apparent conservatis:

I

hould also note that the societal risk of Crystal River 3 it hold dot n by the vary small population around the site..-

Dased on the IREP study we nake the following recomandationc:

1.

Ensure that the licensoe's voluntary action to eliminate tht.

AC power dependency in the steast-driven amorgency foeowater train is proporly impler. anted.

~

2.

Verify the existence of or add to the technical specifications a listiting condition for operation that requires prompt shutdown if the steam-driven esmergency feedwater ptanp train and the electric-motor-driven emergency feedwater pump train are both inoperativa.

' t GN

,i ;

f. *:A:.t 5_o!*

~,

(

3.

Vcrify the ad:quacy of the licen:ce's procedures regarding the cheching of chech valvo position for those valves whoss fciluro uould cause a LOCA that blows down outside containrent cnd require appropriate testing in the Technicel Specifications.-

4.

Tho co non DC pcreer deNndoney between one diesel =snd the ~

~

^

et erponey fccounter syston turbine adnission valve should-be cli:-ineted.

We note, Lhouover, that one of the sugaestidse

~

rsCe by our contractor (to pocor the adr.insion valve from both

'q DO tre. ins) r.sy no~t be desirable sinco it r.my coespromise DC.

o. ar rcCuncancy.

An Ers turbina steam adr.ission valve that fr.ils open upcn loss of DO powar ray ha appropriate. 7.i

.z 3

5.

1.0Citicasi investigation of the diesel-gonorator failure hirtory is rece: rended (see the r.srorandun on this subject from G. Edicen to D. Eisenhut cnd N. !!oscicy dated July 2, 1 R).

~

C.

rar:.:nd o;ur tor trr.ining end procedura revice baced on 7

tv. IE.? c:.qc:necc.

It is.our understcnding that L'is.is ;nor '

=::::. :.y.. Oaz nCoqur.cy of this trciniac cnd pro:scura revic.?

^

.: li b2 cresrt:ined.

7.s is true in rost..nu=1 oar pcr. cr 47)s 1:ne, he.-...-related errors cro an 1: :orte.nt contributor to

.t : ri:%.

7

,e 7.

7::= ( c:y he:t closed cyclo cooling vstor systen (MICCCSF har C

t.. "> tr. ins sf.iich are ccr pletsly rcdundant.

'Ihis syt'.:os pre-vida: corfenant cooling to several enginacred csfoty feitures.

t..u:, a single failure would disable not only one trein of J::.i hut also on.2 tr.in of rnultipic enginesroC s:fety'

i. terce.

It r.sy bs proic.nt to r. edify ths DECCC3 to include a

c.2 or r 3ro prc.>crly engineerse cross-over points to rodu=s t!ais cc:. on coupling of z.nitiple syston=.

O.

32vic.: t sc-st:cr line rupture ra trix circuitry for actuation or ft.iluro r. odes chich might dischle both traans of emergency A

fue.iunter.

I any of the Dst? plcnts have systens such as thi=

ones uc are shout to look at another in the Arkansa~s 1 12rF ctudy.

It nay be appropriate to conduct a risk tradeoff study of taese systecs to see if they do indeed reduce overall rie::.

9.

consider the possibility of further rodifications to the Emergency reedwater system.

The Crystal. River 3 plant has a tro pu=p ETS arrangement.

With action on items 1, 2, 4 and 8 above the Crystal River 3 ETS is not notably unreliable.

Ilowever, here, as well as in other Ers studies we find inherent lire.itations in the two pump configuration.

s h

em

w~,w;,~ mm;.a.as _

.. &xLm.r, s...

2:n /. -n.

---

w,n t :m n = -- -

n

n pp g 2 2 L6D I'

l It. "..- t t s o n

- h, Icplicit in our recor=sndations' is a ~ stan.:ard for correctiva

' action related to the probability of a sorsous accident.

The standard we hava in nind in cd:ing thmse crystal P.iver recc -

raendetions is:

P

-2 CD 1x10

/yr:

fix in day:

-2 1::10

/yr CD 1::10

/yrr fix in years F

~# yr

/

CD 1::10

/yr consider fixing 12:10 CD 1x10

/yr:

see:ptr.51: r:nrs CD is the probshility e

<~

of savare core cacage.

• hs attached r2: orr.ndu= fro.-- Trr.n':.s rsrs 'rik as cers pors,soctiva.

to this intarin strndcrd for the prcS:tility of c:v2re cara Carats.

/

It crovic sor's intercrtinc inright: r.:t cr.1.' en incit icu:1 f:ilurs

. 'I

. prc :.tilities but the carbirc.C pro *.-.bilitie r for th pe,sulttien of e::istin, rc'.ctors.

It c::c: inly cuy7::tc tb.'.c es rhouli. procs-

• cit's I. P cnd ! PCP prorptly.

I s;ould no: considar t*to savare. core d.r.sgo prohtbility str.r:Ccre k

r.:c. tion.

I Leliev:: it is useful es a priority guida until s.

( fins 3 cLova as a chsrply Cafin2d st:ndard rse.dy for fo:::1 l-h:va L::ctor davaloped stanCards rhich s.-ill rclate public hasit'-

s riz.:s 'b:=': to core dar.aga prohcbilitics,.

/

r.. T.orn:ro, Director Trchnhilirtic.*.n:1ysis Graff offics of :!u:1 car 1.cculatory F.cscarch Enclosures nero dtd 7/11/00 to R. Lernero fr T. Ro.rsor*

cc:

D. Eisonhut J. Olshi~nski T. Novak R.

Reid G. Vissing f:. Ernst S. Israel T. Rowsorne W. Vesaly G. Edison i

H. Cullingford n-nima

/

k 9

__m.__.______

.~

JUL 11 %

MMORANDUM FOR: Robert M. Bernere. Ofracter m..

Probabf11stic Analysts Staff-

~-

Orffce of Incienr hogulatory assearch

.,g,.

FRW:

Freak N. Assesse.

Director l

Probabf11stic Analysis ff Office of maclear begolatory assearch SMSJECT:

BAD 7IT BEADLIIES BASS W NTFOTNETICAL INTERN M TAaLE RISE mesERS P j

o Mypothests 1:

Me afght accept op to a 0.15 stemenfustt of a sfgeffisant accNest. 9.e.I ese of the earteessess of IIt! er erne.'.

fa the Seterve betuses the dfssovery_et geleerehtitty

~

to a short term fin, amether 4.15 samen elle me doctde 4

e eMrenge peller se hockfitstead a tided 0.35 T..

. e[I fler the rest er the servlee life of s'oelt. O q '.

.s

'1?; u w L 7q.3 y.

fifty of J.u Corollary 1.1: A usest sees aleet sold nose a Iffettee

.008 er 4.M eEnese of a sfgelflemet assWast u.

%" p:pq6 / w.

g, dissevery).-

~ ' ~ ~

s

< u e y;;h.

. a s.

...:,c.

Carollary 1.2: A papelattee of ISO merst case pleets useM toise a FM sheece of Itytas est their days witteet en accNest. '

.T

~.-

Carollary 1.3: ese ces deftee as alleushle delay ear a slurt term fla from the 0.15 crfteries:

1s t g.W11

• A

. I % +#I-

~

s s

1, = 10 /pr.

+ t,g gg,,,,,

3 1 =to /pr.

s 8%

1 7'*r 1, = to-r/yr

+t1 1 seeth s

u j

s R 7 0109 0154.

l j

l

^>

y

_3

.~

3 JtlL 11 G80

%r: '. :acnero..

s

\\

Cor'c11ary 1.4:*

A ten year period to settle upon and implement a long

. ate is 1,

• 10,j/ year. term retrofiJ pol cy implies the m

(

Caro 11ary 1.5; A new plant with a 40 year planned service life must':.

meet, during the 30 tyear period after the retrofit u

. l policy goes into effect, a long tens rate

-I A

g, "

'3.3X10 / year Observation 1.1:' We can be fairly confident that A actual % 10-2,,,,

.j 7

based upon actuarial data, although there may be a few outifers among the operating plants..

Observation 1.2: To meet the goals overall, one should calculate the allowable short tens exposure with the time-to-discovery 95 added to the time-to-fix,10~gce there probsbly are some Si

, then we can conclude that i.

outliers with A ' actual IREP is moving too slowly to meet 0.15 chance / plant goal.

sJ However, we are likely to be within a decade of the goal.

g

f f Hypothesis 2.:

The population of reactors had a failure rate for significant 4

accidents.1 actual, distributed as follows before TMI:

1A

1. Units J Unit Years 10-2 5

28.57

,~

,d

, Avg. 5.71 year / unit 3X10~3 10

~

57.14 fw W reactw yeen

~

'10~3 10 57.14 among 70 units

+j 3X10 15 85.71 10~4 15 85.71 3X10-5 15 85.71 Total:

70 Units 400 Unit Years Corollary 2.1:

The occurrence rate for serious accidents in the 70'-plant

,opulation is then 1 industry =jng A =.096 per year.

p q

This implies that the Mean Time Between Failures (MTBF) is

~

MTBF = 10.4 calendar years.

f u.

I 9

e o

e

w t

r JUL I1 580

'..r t :4. L. uro '

i

. ?.

The probability of'no accidents in 400 reactor years is

~*ind 400 =.57 e

n.

, -+ P lone or more accidents) =.43 in 400 r yr.

Otscrvation 2.1: This appears to be consistent with the TMI experience and the occurrence of several close calls.

Hyecthesis 3.:

Suppose the TMI experience and ratchets lowers the top I

two l's by X / each.

3 Corollary 3.1: What then is the industry 1 and MTBF7 1 ind =.043/yr -* MTBF = 23.2 years

. t.

Caro 11ary 3.2: What, then, would be the probabil.1,ty of going 3 calendar r

years after THI accident free? P (3 yrs) = e-1 ind 3 yr t

o t

conversely P (cae or more accidents in 3 years) =.12

~

Observation 3.1:

We can test the plausibility of hypothesis 3 by conside' ring the experience with precursors since TMI. In the 1.3 years since TMI we have had one close call (Browns Ferry ATWS) which may have come within a decade of TMI-level severity (one of ten such events might produce extensive core damage).

Other precursors, such as the February incident at CR-3 are -

I think - more like two decades or more from extensive core damage.

Hypothesis 4.: Suppose 1 (incl. precursors)= 10 X 1 (Ind core damage).

Together with Hypothesis 3 we would expect 1 incl. pre =.43/yr, MTBF pre = 2.33 yr.

9 e

1) l 1

..1;

~ ~.

M ' 1 G80 9 _rt ::. ?.. u ro ?

Corollary 4.1:

The litelihood of gettin'g one or e.cre precursor events of theseveritysuggestedin0biervation3.1andHypothesis4 in the 1.3 years since T'il ls' P(1 +) = 1 -e - Pre X 1.3 yr,,43 Observation 4.1:

This seems reasonably congruant with the experience.

Hypothesis 5.:

At exactly 4 calendar years after TMI, all plants go through a safety analysis (IREP) which " instantaneously" catches 90% of the 3X10 cases and-80". of the 10~3 cases'

~

and 70% of the 3X10~4 cases and - where identified -

reduces these to 10~4 cases, i.e.

Before - IREP +++++ After P =.1 1 actta1 = 3X10~3 A act. = 3X10~3 1 actual = 10~3 P =.2 P =.3 A actual = 3XI

^

P =.4 A actual = 10 s

s 8efore - IREP: - : After P =.2 la = 10'3

-3 1 act. = 10 4

P =.3 la = 3X10 la = 10 P =.5 P =.3 la = 3X10

-4 1 act. = 3110 P =.7 la = 10~4

-4 la _ 10 before are unchanged.

The model, as thus far developed, stands as follows:

SW

't

.;.. c

~

. J

.'., t ::. -

...)

- J.;:..c i;;I,'.'

2 i..?

~

II D

.t.":.1 EP_

1

. e 1::!,,

(1y:.cs) p of Units e A i of Units 0 A Ef ~ect.ive ! Uc.its, 3 L i

5 10 i

~

-2

-3

-3

0. 5' 3X10

-3 5

3X10 10 3x10

-3 10 5.0 10 10 10 3X1d

-4 15 3X10 12.0 15 3X10

-4 15 10 37.'5 M

c, 10

-5 15

-5

-5 15 3X10 15.0 3X10 15 3X10 70 70 70.0

~~

Aind. =.095/y:ar Aind. =.013/ year 11rJ. =.01.*3/y.ar '

LSF : 10.4 cal.yr.

MTBF = 23.2 cal. yr.

(110F = 70 cal. ; r.

729 Unit yr.

= 1624 Unit yr.

= 4D00 Uni. yr..

=

P (*23 n e n tor yec.rs P (4 calcr..'ar yr.vs P (10 crlto.2r t;/o acch'c.gi.s) t;/o accident) y6ars w/o

=.57

=.84 accid.ni)

=.87

~

P (14:ce) =.13 P (1+ acc) =.43 P (16 acc) =.16

!!h.'t, th:r., is tha probability of rc.:sining accib:rt-free 0

rvation 5.1 fe e 1 17.I ti.r.s;;h 4 y;ars to IR P and 10 ye:rs thereafter to the loi.3 tera

" ix"?

~

P (14 accid:nt free years) = 284 X.87 =.73 P ( 1 or more accid:nts in 14 years) =.27 according to this very-ut v :tain tod:1.

~

Frank H. Rouso::=, Deputy Director Probabilistic Ana1ysis Staff Office of Nuclear Regulatory Research e

O l

,G,f

.a./ 'l -

' UNITED shATES e

f' NUCLEAR REGULATORY COMMISSION a

c, g

j non= crow. o. c. roess

%w*...., ~,'

EAR 31 WI j

MEMORAf.DUM_ FOR: Karl Kniel, Chief Generic Issues Branch, DST FROM:

Brian W. Sheron, Section Leader Section A, Reactor Systems Branch, DSI THRU:

Themis P. Spets, Chie

)

Reactor Systems Branch, SI l

SUBJECT:

STATUS OF FEED AND BLEED FOR OtERGENCY DECAY HEAT REMOVAL

-I Per my discussions with A. Marchese of your staff I as providing you with a status summary of feed and bleed as an emergency means of decay 1

heat removal. This summary is provided in the enclosure and is intended for your use in developing the.overall action plan' for USI A-45, Decay Heat Removal Reliability. The status susuury considers the present 1

l-capability of all operating plants to remove decay heat by feed and r

bleed.

It also addresses the relative risk reduction potential associated T;

with a feed and bleed capability. The general conclusions reached are Ul

~

that:

e Feed and bleed, if performed, should be at a relatively low (P relief valve setpoints) pressure.

e Feed and bleed capability can be accomplished in all PWts if a suffletent capability to depressurize the plant is available. For some plants, this would probably require

^ additional PORV capacity.

a e The probability of loss of all feedwater due to loss of all ac power is an uncertain but finite fraction of the total 1

probability of losing all feedwater due to all causes. The ac power-dependence of feed and bleed makes the overall risk

.l reduction questionable. This is because the risk dominant sequences result.from a loss of all ac power. Thus, feed and bleed will not igrove risk dominant sequences. However, substantial igrovement in assuring core cooling might be realized with feed and bleed.

4 4

~

C G

qn pA

{

t t

--t

[

1 IdAk'e i SSi Karl Kniel s The costs associated with increased pressure relief capability i

may be acceptable.when compared to other risk reducing modi-

~

fications.

Further study is probably warranted.

If you have any further questions, please contact me.

v

%. s

=

L Brian W. Sheron, Section Leader I

Section A Reactor Systems Branch i

Division of Systems Integration l

Enclosure:

l Status Sumary of Feed and. Bleed Capability in PWRs cc: w/ enclosure D. Ross P. Check T. Murley

-l F. Schroeder

~9 P. Norlan A. Marchese 1

A. Thadant R. Bernero P. Baranowsky J. Ebersole ACRS M. Sender ACRS H. Etherington, ACRS y

J. Ray, ACRS M. Plessett, ACRS D. Okrent, ACRS

{

P. Boehnert ACRS R. DiSalvo, RES

.]

j P. North, EG&G i

G. Johnsen, EG4G l

M. Taylor L. Cave j

~

l e

I

_a_ammw=m =,wgwnm m.m.

'{

M Enclosure STATUS

SUMMARY

OF FEED AND, BLEED CAPABILITY IN PWRS

1.0 INTRODUCTION

The feed and bleed process refers to direct removal of decay heat from the primary system utilir.ing the high pressure injection system and the pressure relief system. The use of this process for decay heat removal is not a preferred method but rather an emergency method when the secondary heat removal path is not available (i.e., no main or auxiliary feedwater available).

~

The capability to successfully feed and bleed the primary system in order to remove decay heat varies among not only the PWR vendors, but among the various plants designed by the same ' vendor. This is described L,

in more detail in the following sections.

2.0 BASIC REQUIREMENTS e

Notwithstanding certain constraints and limitations of feed and bleed which wili be discussed later, the two basic requirements needed to feed and bleed are (1) availability of AC power and (2) the capability to establish a system pressure which will support feed and ~ bleed.

v The first requirement, availability of electric power, is an obvious requirement since pumped flow is required, and all Hp! pumps have electric drives.

For some plants, electric power to operate certain valves is also necessary. The capability to meet the second requirement, to establish a system pressure which will support feed and bleed, is plant-specific and requires further discussion regarding plant capabilities.

3.0 PLANT CAPABIt.! TIES s

3.1 B&W-Designed Plants B&W plants of -the 177FA lowered-loop design have high pressure injection pumps (which do " double-duty" as the charging pumps for h

s

. _ _~

~

.+---m

.n

/

~

-' ['.

1 -

..z

+

insentory control) with shutoff heads. between 2700 and 3000 psi.L One plant,' Davis Besse 1 Niith is of the raised-loop design,;has"?7.

-w _

separete charging andlHPI.ptaips. The HPI pumps have a shutoff : '.-

hsad of 4000 ft, or about'1500 psi, while the charging pumps h' ave "a shutoff head of 6500 ft, or 2600 psi.

~

All of the B&W plants have one PORY and two safety valves. The set I

pressure on the safety valves is 2500 psig and the set pressure on the PORVs is 2255 psig.

~

For all B&W plants, except Davis-Besse 1, the HPI pumps have the capability to inject coolant and discharge it through either the PORVs or the safety valves. An estimate of the flow requirements is about 7 gonVMWth (based on converting subcooled (@*F) water to steam at about 2500 psi). The HPI capability in B&W plants is around 250 to 300 gpm per pump at 2500 psi. Therefore, the Hpi peps can remove all of the core decay heat within a few minutes after shutdown.

3.2 CE-Designed Plants All CE plant P,).

Capability 3

to depressurize uncertain and needs further evaluation _.

CE No capability to feed and bleed at high pressure. Plants with PORVs have questionable capability to depressurize adequately. Plants without PORVs must rely on auxiliary pressurizer spray to depressurize. This capability presently unknown.

W No capability for extended high pressure feed and bleed. Pump damage potential. Plants with 3 PORVs capable of depressurizing. Plants with 2 PORVs need further evaluation.

e f

,)

, hi w

N g

.s, 3..,., *[

'<~

3 p,

'e

.. )

i 2.s i

3

/

.9 j

s; i

.~{

.c.,

e

.+

o s,

..s

+*

g e

. e

, ',s }

t i-l-

. c.

..., 4...

,y.

. p. 3.?~'

'>,. ).i.

~g.

,.~<:m

~

s?-h w.

s

Q [_ _',9

',,, l

' f:>.)

. g:

,13 a

,,, ~

v n ;.. :+.

2;i.

,3, u.

,,,p ammisesamu d=4

af

. ? ^-

wnserer emesau-Wsm E

E

'3

. :s

~

E

_3 3

3

-3 3

8' I

~ l sg

• 3 *5 *5 ~ *3 *E

• I. *I' W:l e

.o,t.

a b

sanspessoaeg3 ding mon ator

.oota$.I 3

3 3

3

- 3 3

g!

aanssead @)g w

3 3

3 3

~g 3

3 uj

.-y g

u.

u. l l l l IEIl I

+b~

- r. 4

• S f8

/wvmot wid E }"NE ![ El El El,Q Eg Q

~

w w

a o

.x

.. 1 I

I I

I 5

I is w5C

- yo.w->m, -

se se s

s 3,

Be Re 3 3

un A,

1

.a:

E

'(960-44 It ded

~

7 i

o pos pea og e)) isd tsusodges

• gt *g 1,*g i

l

/WevW4L 'A l#*3 Asad SR SN EN 3R Et

)

=

a s

gb1 is E

l a D 3.i A 3 i E

2-7 I

e e

a 6

e e

e f

.d

,g 1.

t_

t*

.l r

m

.gr t_

2.

!i i! al !I3 1 Ea s a

~

.t., : A

'_ M.

'**'g

• • e E*;

y

g 7.y i

ly! E! an !I3 I l i s V'

n g

1

-a g

15! 2! saa s 31 Hi s a

nas gg I

II! !! aus l@l Ei s s s

==>

Eli g

-sg 8

gill 2! sua sil E! s

+

eg II:

=

m i

oS 8

e s-s=s aa

~~

3 E E

• s

! a = c13,a n s s3f Es-s a

g

-r

=

ses

--s

=

n t li

~

t E

E skt R? s s

= Igi ism s i t'-

al' 1

s gj ee.n

,, 3 1

=

s Ei E~ 3=i }

3 3 * ?g 1-l*31Ac3321 I l 1 ] ]1 -} E a - - E.{ !

'l f I }

. l }3}

}

e 3

e 0

9

l,;*.4 j

l

.N endsupuegs mm a'

s m.m s

s.m

,y

-s

~

a e

=

i i

n-

,a g

a.

>~ 3 g

g-

=

y I.

I 9

8 n,a ay, g

8 i

s'5 a

$_-a"4 is a b b b b

si a

g

=

EwEi!! li! Ik ! B lin Ih 11

1st.1h Jg.

!. u g

• * 'm

.,=

4E I

3 3

3 3

3 5

==

8 h

g

".A7 23 23 2!

23 is 3:

yf s.!L, E5[i it. bi Ski Ski Ski hik 4

.g

.g

.g g

s e

1

-w.=

351 3

~3 }l,l o 3 si..

2. w

.s..iv,e m 5 el 2E 2E .k. 5 35 eg 3 -m a. -~ a w l %g ? $5 1 b 2 fj~ 5'i i --3~ .h e I.i 5 1]" E g3 2.1 l g 0.E 2.i lailil3glg a=*-lja jjjju!: 3 g= IEd 8: ' *a ggg = A g e e

p !E 5 !E E 5 55 EE Em 3

• f*,J2,

~,, s -[ gs sg g ~ e y 7 i ".".5 5 5 2 5 3.5 5 61 g ",*3 I I 5 5 5 5 5 5 5 g .-,m EE dN l!L5 If ~18 13 's 9 $3.o..,umd 5k 13 . ~ ..}% 5%s.k,! jhjsg3h-a .~n y s m ~ 3 5 5 5 5 5 3 ?! 23 73 ?! ji 8 ^= j l*E m i i -t -A i -s i y sji 2 x&aji ski sak 2j s s sig - >=e ~::M ce,~2 2B !! 2J-s! 69 J J il 3 N i E3 5. !i _I 2 ~ e y V y a a e a lii i " y~ j!! in !=i,i '!.,' m JI 1 ,j11-! s 3 e 9 4

4 i g F G e e a o l ~ dumd Sel8Jeq3 weemetesse I 8 l mgsgeed . um a g I S sd ws s a ~ 3 ,.m e 4 eiw g 3 b ue ns L i e sd *) sled 1*s n -3s 'Angw/wev a dog artet j gg j:ns a g 1 d

• smgea Apses

~ E b* 34*9 e 5 _E I I m i H g 3 -] ( -~ =.srpes as.: J_4 g s la n

4. a. =

1,, ah s,w wi F -W8/ Vel /J4/94 gm aunes non t -R* [3 tI .j 3r m 2 h' i ? !,,s1 lil f ~ 1 + E

N r = 1 r TABLE 5 Probability Estimates

• Loss of Offsite Power 0.2 Failure of Emergency AC 10-2,.10'#

5 Pecovery of Offsite Power 0.5 Failure of AFW w/o AC. 5x10 2i.- 10-3 Loss of MFW 0.3 Failure of AFW 10-3 5x10-5. Recovery of MFW (1/2-1 hr) 0.1 Sequence Probabilities * (Loss of Offsite Power) X (Loss of Emergency AC) X (Loss of AFW) = 5 x 10 10 (Loss of Main Feedwater) X (Loss of AFW) = 3 x 10 - 1.5 x 10-5 s.

• Informally provided by P Baranowsky, RES t

e . I ' Y' ^ }}