ML20247F608

Reg Guide 1.157,Task Rs 701-4, Best-Estimate Calculations of ECCS Performance

ML20247F608
Person / Time
Issue date:	05/31/1989
From:	NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES)
To:
References
TASK-RE, TASK-RS-701-4 FACA, REGGD-01.157, REGGD-1.157, NUDOCS 8907270196
Download: ML20247F608 (19)

Text

-

U.S. NUCLEAR REGULATORY COMMISSION M:y 1989 Lc@[%(\\)

REGULATORY GUIDE w***

OFFICE OF NUCLEAR REGULATORY RESEARCH 1

i REGULATORY GUIDE 1.157

)

}

(Task RS 701-4)

BEST-ESTIMATE CALCULATIONS OF EMERGENCY CORE COOLING SYSTEM PERFORMANCE A. INTRODUCTION use either ~ Appendix K features or a realistic 1

)

evaluation model. These realistic evaluation modelsa q

Section 50.46, " Acceptance Criteria for Emer-must include sufficient supporting justification to i

gency Core Cooling Systems for Light-Water Nuclear demonstrate that the analytic techniques employed 1

Power Reactors," of 10 CFR Part 50, " Domestic Li-realistically describe the behavior of the reactor censing of Production and Utilization Facilities," re-system during a postulated loss-of-coolant accident.

quires that light-water nuclear reactors fueled with Paragraph 50.46(a)(1) also requires that the uranium oxide pellets within cylindrical zircaloy clad-uncertainty in the realistic evaluation mode! be ding be provided with emergency core cooling systems quantified and considered when comparing the results (ECCS) that are designrd in such a way that their of the calculations with the applicable limits in calculated core cooling performance after a postu-paragraph 50.46(b) so that there h a high probability 1

lated loss-of-coolant accident (LOCA) conforms to that the criteria will not be exceeded, certain criteria specified in paragraph 50.46(b). Para-graph 50.46(b)(1) req,.ses that the calculated maxi-This regulatory guide describes models.8 correla-O mum temperature of fuel element cladding not be tions,* data, model evaluation procedures, and meth-ods that are acceptable.to the NRC staff for meeting

/ greater than 2200 F.

In addition, paragraphs 50.46(b)(2) through (b)(5), which contain required the requirements for a realistic or best-estimate calcu-hmits for calculated maximum cladding oxidauon and lation of ECCS performance during a loss-of-coolant accident and for estimating the uncertainty in that maximum hydrogen generation, require that calcu-lated changes in core geometry remain amenable to cooling and that long-term decay heat removal be

'For the purpose of this guide, the terms"bestrstimate"and"realis-tic" haw the same meaning. Both terms are used to indicate that the provided.

techniquesatterrpt topredictrealistiereactorsystemthermal-hydraulic response. Best-estimate is not used in a statistical sense in this guide.

On September 16,1988, the NRC staff amended

%e term " evaluation model" refers to a nuclear plant system com-puter c de r any ther analysis tool designed to predict the aggregate the re4uirements of $ 50.46 and Appendix K, behavior of a reactor durms a loss-of-coolant acudent.1: can be either "ECCS Evaluation Models" (53 FR 35996), so that best-estimate o< consern tive and may contain many correlations or these regulations reflect the improved understand-

" del 5-ing of ECCS performance during reactor transients

%e terremodel" refer. to 4 set of ecrations derived from funh-that was obtained through the extensive research mental physicallawfhat is designed an pudict the detaMs of a speci6e

,' nenomenon.

performed since t! e promulgation af the :.riginal

• The term treWion" refm to an equation having empirically de-requirements in January 1974.

Paragraph termined eonnanis such thal it can predict some details of a specific phe-50.46(a)(1) now permits iicensees or applicants to nomenon for a limited range of conditions.

UsNRC REGUIAToRY PUIDEs The guides are Jesued tr the followl ten broad divisions:

negulatory Guidus are issuor! ta describe sad make availabie to the pute lac rriethods acceptable to the NRc staff of implemranting specHic parts of it's Commission's. regulations, to delineate tethniques used by the

,I,, Power Rerctors

6. Products

. Researc4 and Test Reactors

7. Transportation stW al evahaating specific problems or perulated accidents, or to pro.

3'. Furts and f/ata ta!s Facitfies

8. Occupational Health wir guidanne to applicants. Regulatory %Jides are net s4 stitutes er

1. tulations, and compliance with them (e not raqu:rded. IAthodt.:xhd

4. EnWonmental and Siting O. AntWust and Financial Review e4 lutions different frorn those set out in the guides will be acceptabe If

6. Ha aririt a.nd Plant Protection

10. General ti sy prodde a basis fts the findings reQuisits to tne issuance or continu-at e of a permit or Itcense by the Commiss,on.

Com d WW ps % b MW h h hnM PdmW

/

Thl6 oulde was issued after considera0cn of comments received from office at the current GPo price. Information on current GPO prices may t

the puw a.

ants and suggest ons for improvements in these be obtained by contacting the Suponntendent of Documents, U.S.

r guides are encou, aged at all times, and guides will be revised, as ap.

Government Printing office, Post office Box 37082. Washington, oC propriate, to accommodate comments and to reflect new information or 20013-7082, telephone (202)275-2060 or (202)276-2171.

experlence.

Written comments may be submitted to the Regulatory Publicatbns issued guides may also be purchased from the National Technical infor-Branch. OFIPS, AHfLt. U. S. Nuclear Regulatory Commisson, Washing-rnation Serv 6ce on a standing order basis Detalls on this service may be ton. oC 20555.

obtained by writing NTIS, 6285 Port Royal Road, Springfield, VA 22161.

8907270196 890531 PDR REGGD 01.157 R PDR

calculation. Methods for including the uncertainty in vided information that allows for quantification of the comparisons of the cabulational results to the cri-that conservatism. The results of experiments, com-teria of paragraph 50.46(b), in order to meet the re-puter code development, and code assessment allow quirement that there be a high probability that the more accurate calculations, along with reasonable es-criteria would not be exceeded, are also described in timates of uncertainty, of ECCS performance during this regulatory guide. Paragraph 50.46(a) also per-a postulated loss-of-coolant accident than is possible mits licensees to use evaluation models developed in using the Appendix K procedures.

conformance with Appendix K.

It was also found that some plants were being re-Other models, data, model evaluation proce-stricted in operating flexibility by limits resulting from dures, and methods will be considered if they are conservative Appendix K requirements.. Based on the supported by appropriate experimental data and research performed, it was determined that these re-technical justification. Any models, data, model stri:tions could be relaxed through the use of more evaluation procedures, and methods listed as accept-realistic calculations without adversely affecting able in this regulatory guide are acceptable in a ge-safety. The Appendix K requirements tended to di-neric sense only and would still have to be justified to vert both NRC and industry resources from matters the NRC staff as being appropriately applied and ap-that are relevant to reactor safety to analyses with as-plicable for particular plant applications.

sumptions known to be nonphysical.

The regulatory position in this regulatory guide in recognition of the known conservatism in Ap-lists models, correlations, data, and model evaluation pendix K, the NRC adopted an interim approach in procedures that the NRC staff considers acceptable 1983, described in SECY-83-472,6 to accommodate for realistic calculations cf ECCS performance. It industry requests for improved evaluation models for

(

also provides a description of the acceptable features the purpose of reducing reactor operating restrictions.

of best-estimate computer codes and acceptable This interim approach was a step in the direction of methods for determining the uncertainty in the calcu-basing licensing decisions on realistic calculations of lations.

plant behavior. Although the approach permits many j

"best-estimate" methods and models to be used for The Advisory Committee on Reactor Safeguards licensee submittals, it retains those features of Ap-has been consulted concerning this guide and has concurred in the regulatory position.

Pendix K that are legal requirements.

Any information collection activities discussed in The current revision of 9 50.46 permits ECCS this regulatory guide are contained as requirements in evaluation models to be fully "best-estimate" and re-10 CFR Part 50, which provides the regulatory basis moves the arbitrary conservatism contained in the for this guide. The information collection require.

required features of Appendix K for those licensees ments in 10 CFR Part 50 have been cleared under wishing to use these improved methods. Safety is best served when decisions concerning the limits within OMB Clearance No. 3150-0011.

which nuclear reactors are permitted to operate are based upon realistic calculations. This approach is B. DISCUSSION currently being used in the resolution of almost all reactor safety issues (e.g., anticipated transients with-The criteria set forth in 9 50.46, " Acceptance cat scram, pressurized thermal shock, and operator Criteria for Emergency Core Cooling Systems for guidelines) and is now available for one of the last Light-Water Nuclear Power Reactors," and the cal-remaining major issues still treated in a prescriptive culational toethods specified m Appendix K were manner, the loss-of-coolant accident promulgated in January 1974 after extensive rulemaking hearings and were based on the under.

The NRC staff amended 9 50.46 of 10 CFR Part 50 to allow realistic methods to be taed for the ECCS standing o.t ECCS performance available at that time.

In the years following the promulgation of tSose Performance calculations in p' tace of the evaluation rules, the NRC, the nuclear industry, and several for, models that use the requirtJ Appendix K features.

eign institutions have conducted an extensive prc,.

This rule change also requires analysis of the uncer-pam oi research that has greatly improved the un.

taint) of the best-estimate calculations and requires that this uncertainty be considered when compning derstr iding of ECC5 ;wrfonnance during a postiated loss-of-coolant accident. The methods the results of the calculations to the limits of para-specified in Appendix K were found to be highly con.

graph 50.46(b) so that frste is a high probabihty that servative; that is, the fuel claddmg temperatures ex-pected during a loss-of-coolant accident would be information Report from Wmiam J. Dncks to the Commissk>ners.

much lower than the temperatures calculated using dated November 17,1983.* Emergency Core Coohng System Analy-sisMethods." SITY-83-472. Available for mspectionor copying for Appendix h, methods. In addition to showing that a fee in the NRC Pubbe Document Room. 2120 L Street NW.,

Appendix K is conservative, the ECCS research pro-Washington, DC.

l 1.157-2 l

' the criteria will not be exceeded, in this manner, the phenomenon. The model should be compared more realistic calculations are available for regtPatory with applicable experimental data and should predict l-decisions, yet an appropriate degree of conservdsm the mean of the data, rather than providing a bound n

would be maintained.

to the data. The effects of all important variables

(

Many of the methods and models needed for a should be considered. If it is not possible or practical to consider a particular phenomenon, the effect of I

best-estimate calculation are the same as those used ign ring this phenomenon should not normally be previously for evaluation model analyses. Although licensees and applicants are well acquainted with treated by fncluding a bias m the analysis directly, but should be included as part of the model uncertainty.

them, explicit guidance on acceptable methods and The importance of neglecting a particular phenome-models (based on NRC experience with its own best-n n should be considered withm the overall calcula-estimate advanced codes such as TRAC-PWR, tional uncertainty.

TRAC-BWR, RELAPS, COBRA, and FRAP) would be useful. Further, the NRC has not previously pub.

Careful consideration should be given to the J

11shed acceptable methods for uncertainty analyses, range of applicability of a model when used in a best-Therefore, guidance on methods acceptable to the estimate code. When comparing the model to data, NRC staff for calculating ECCS performance and for judgments on the applicability of the data to the situ-estimating the uncertainty are provided in the follow.

ation that would actually occur in a reactor should be I

ing Regulatory Position.

made. Correlations ger erally should not be extrapo-

~

lated beyond the range over which they were devel-C. REGULATORY POSITION beyond the conditions for which valid data compari-1.

BEST-ESTIMATE CALCULATIONS l

to the effect of this extrapolation and the effect A best-estimate calculation uses modeling that should be accounted for in the uncertainty evalu-l attempts to realistically describe the physical proc-ation. The fundamental laws of physics, well-esses occurring in a nuclear reactor. There is no established data bases (e.g., steam tables), and sensi-unique approach to the extremely complex modeling tivity studies should be used to assist in estimating the of these processes. The NRC has developed and a'-

uncertainty that results from extrapolation.

sessed several best-estimate advanced thero!-

A best-estimate code contains all the models nec-hydraulic transient codes. These include TRAC-essary to predict the important phenomena that might PWR, TRAC-BWR,.RELAP5, COBRA, and the occur during a loss-of-coolant accident. Best-estimate FRAP series of codes (References 1 through 6)'

code calculations should be compared with applicable These codes reasonably predict the major phenom' experimental data (e.g., separate-effects tests and in-ena observed over a broad range of thermal-tegral simulations of loss-of-coolant accidents) to de-hydrauhc and fuel tests. Licensees and applicants termine the overall uncertainty and biases of the cal-may use, but are not limited to, these codes and the culation. In addition to providing input to the specific models within them to perform best-estimate uncertainty evaluation, integral simulation data com-calculations of emergency core cooling system parisons should be used to ensure that important phe-(ECCS) performance. Since the NRC staff has not nomena that are expected to occur duting a loss-of-performed the plant-specific uncertainty analysis re-coolant accident are adequately predicted. This is an quired by the revised 9 50.46 of 10 CFR Part 50, the idealizect characterization of a best-estimate code. In licensee must demonstrate that the code and models practice, best-estimate codes may contain certain used are acceptable and applicable to the specific fa-models thu are simplified or th4t contain conserva-cility over the intended operating range and must tism to some degree. This commatism may be intro-quantify the uncertainty m the specific apphcation.

duced for the following reason.5:

General attributes expected in a best-estimate calcu-lation are described here in Regulatory Position 1; 1.

The model simplification or conservatism has special considerations for thermal-hydraulic best.

little effect on the result, and therefore the estimate codes are presented in Regulatory Position development of a better model is not 2; and specific examples of features that are consid, justified ered acceptable best-estimate models are given in 2.

Tne uncertainty a a pzrticular modelis diffi-Regulatory Position 3. Other models or correlations cult to determ6r, and only an upper bosmo will be considered acceptable if their technical basis is can be determind demonstrated with appropriate data and analysis.

3.

The particular application does not require a A best-estimate model should provide a realistic totally best-estimate calculation, so a bias in calculation of the important parameters associated the calculation is acceptable.

(

wPh a particular phenomenon to the degree practical The introduction of conservative bias or simplifi-with the currently available data and knowledge of cation in otherwise best-estimate codes should not, 1.157-3

however, result in calculations that are unrealistic, or nodes to represent the system. Sensitivity studies that do not include important phenomena, or that and evaluations of the uncertainty introduced by contain bias and uncertainty that cannot be bounded.

noding should be performed. Numerical methods i

Therefore, any calculational procedure determined treat time in a discrete manner, and the effect of to be a best-estimate ccde in the context of this gt-e time-step size should also be investigated.

or for use under paragraph 50.46(a)(i) should 2.1.2 Computational Models compared with applicable experirnental data to en.

sure that the calculation of important phenomena is A best-estimate code typically contains equations realistic.

for conservation of mass, energy, and momentum of the reactor, coolant and noncondensible gases, if im-2.

CONSIDERATIONS FOR TIIERMAL-portant (e.g.. air, nitrogen). Energy equations are IlYDRAULIC DEST-ESTIMATE CODES also used to calculate the temperature distribution in Some features that are acceptable for use in reactor system structures and in the fuel rods. The l

best-estimate codes are described in the following required complexity of these equations will vary de-paragraphs. Models that address these features may pending on the phenomena that are to be calculated be used with the basic proviso that a specific modelis and the re. quired accuracy of the calculation. NRC acceptable if it has been compared with applicable staff experience with its own best-estimate computer experimental data and shown to provide reasonable codes has indicated that separate flow fields for dif-predictions. Reference 7, " Compendium of ECCS ferent fluid phases, or types, and calculation of non-Research for Realistic LOCA Analysis" equilibrium between phases may be required to calcu-late some important phenomena (e.g.,

(NUREG 1230), provides a summary of the large ex.

perimental data base available, upon which best.

countercurrent flow, reflood heat transfer) to an ac-estimate models may be based. While inclusion in ceptable accuracy. The NRC staff has also deter-Reference 7 does not guarantee that the data or mined that certain phenomena require that the equa-model will be acceptable, the report describes and tions be solved in multiple dimensions.11owever, one-dimensional approximations to three-erences a large body of data generally applicable to best-estimate models. NUREG-1230 also provides dimensional phencmena will be considered accept-documentation of NRC studies of the effect of reac.

able if those approximations are properly justified.

tor power increase on risk, background information Other basic code features include equations of state on the ECCS rule, and a description of the methodol, and other material properties. Sensitivity studies and ogy developed by NRC for estimating thermal.

comparisons to data should be performed to deter-hydraulic transient code uncertainty.

mine the importance of the simplifications used.

1 For any models or correlations used in a best-3.

BEST-ESTIMATE CODE FEATURES estimate code, sufficient justification must be pro-3.1 Initial and. Boundary Conditions and vided to substantiate that the code performs adc Equipment Availability quately for the classes of transients to which it The heat generated by the fuel during a loss-of-applied. In general, the features of best-estimate c lant accident depends on the power level of the thermal-hydraulic transient codes have uncertainties, reactor at the time of the loss-of-coolant accident and associated with their use for predicting reactor syster n the history of operation. The most limiung mitial response, These uncertaintiu should be considered c nditions expected over the life of the plant should as part of the cuerall uncertamt) analysis described in be based on sensitivity studies. It is not necessary to Regulatory I'osition 4.

assume iritial conditions that could not occur in com-21 Basic Structure of Cohs bink.on. For example, beginning-of life peaking fac-

ors together with end-of-life decay heat dc not re-2.1.1 Numerical Methods q' tire considerat'on. Given the assumed initial conditions, relevant factors such as the actual total A best-estimate code mes a numerical scheme for solvirg the equatms used to predict the thermal-poner, actual peaking factors, and actual fuel condi-ti ns should be calculated in a best-estimate manner.

bydraube behavior of the reactor. The numerical scherr;e is, in itself, a corWex prccess that can play The calcuMions pewrmed should be represen-an important role in the ovuall cdculation. Careful tatise of the spectrum of possible t reak sizes from the numerical modeling, wnsbvity studies, and evahi-fual double-ended break of the largest pipe to a sim ations of numerical error should be performed to en-small enough that it can be shown %. smaller breaks sure that the results of the calculations are represen-are of less consequence en Aose already consid-tative of the models used in the code. Numerical ered. The analyses should also include the effects of simulations of complex problems, such as those con-longitudinal splits in the largest pipes, with the split sidered here, treat the geometry of the reactor in an area equal to twice the cross-sectional area of the approximate manner, making use of discrete volumes pipe. The range of break sizes considered should be 1.157-4

.sufficiently brcad that the system response as a fanc-creep strain in cladding during steady-state operati m, Lion of break size is well enough defined so that inter-reducing the gap between the fuel pellet and clad-pelations between ' calculations, without considering ding. Ciadding creep is a function of fast neutron (p unexpected behavior between the break sizes, mayflux (>l MeV), cladding temperature, hoop stress,

be made confidently.

and material. Cladding materials may be cold-Other boundary and initial conditions and equip-

1. Y "".8".ess-relieved or fully recrystallized, and there is a significant difference m the magnitude of ment availability should be based on plant technical specification limits. These other conditions include, er epdown betwepn these materials. During pellet-ca ng mechanical interaction, cladding experi-but may not be limited to, availability and perform-mces e rmauon imn: tensk cmep, dch is sig-(

ante of equipment, automatic controls, and operator

)

• " ' " Y C""*

D C *E'***'**

)

actions. Appendix A to 10 CFR Part 50 requires that creep. An acceptable model for cladding tensile a single failure be considered when analyting safety

'" P P

system performance and that the analysis consider d

the effect of using only onsite power and only offsite power.

Best estimate fuel models will be considered ac-ceptable provided the models include essential phe-3.2 Sources of IIcat During a Loss-of-Coolant nomena identified above and provided their technical Accident basis is demonstrated with appropriate data and Models should account for the sources of heat analyses.

l discussed below r,nd the disutbution of h at

3. 2.1.1 Model Evaluation Procedure for l

production.

Stored Energy and lleat Transfer in Fuel Rods. A l

model to be used in ECCS evaluations to calculate l

3.2.1 Initial Stored Energy of the Fuel internal fuel rod heat transfer should:

The steady-state temperature distribution and a,

Be checked against several sets of relevant stored energy in the fuel before the postulated acci-data, and dent should be calculated in a best-estimate manner for the assumed initial conditions, fuel conditions, b.

Recognize the effects of fuel burnup, fuel and operating history. To accomplish this, the ther-pellet cracking and relocation, cladding S mal conductivity of the fuel pellets and the thermal creep, and gas mixture conductivity.

conductance of the gap between the fuel pellet and The model described by Lanning (Ref. 8) com-the cladding should be evaluated. Therrnal conduc-pared well to in-pile fuel temperature data. Best-tivit) of fuel is a function of temperature and is de.

estimate models will be considered acceptable graded by the presence of gases in crack voids be-provided their technical basis is demonstrated with j

tween fuel fragments. An acceptable model for appropriate data and analyses.

j thermal conductivity should be developed from the 3.2.1.2 Experimental Data for Stored in-pile test results for fuel centerline and off-center Energy in Fuel Rods and IIcat Transfer. The I

temperatures, taking into accourt the conductivity of correlations and data of Reference 9 are acceptable gases in crack voids, for calculating the initial stored enugy of the fuel and Thermal conductance of the fuel-cladding gap is subsequent heat transfer.

d a strong functico of hot gap size and of the composi-3.2.2 Fission IIcat tion and pressure of the gases in the fuel rod. The Fission heat should be included in the calculation calculation of hot gap size should take into account and should be calculated using best-estimate reactiv-UO or c.ixed-oxide fuel swelling, densificatien*

ity and reactor kinetics calculations. Shutdown reac-2 creep, thermal expansion and fragment relocation, tivities resulting from temperatures and voids should

.and cladding creep. Fuel swelling is a function of also be calculated in a best-estimate manner. The temperature and burnup. Fuel densification is a func-po;nt kinetics formulation is considered an accept-tien of burnup, temperature, and initial density. Den-able best-estimate method for determining fission sification can remit from hy&ostatic stresses imposed heat in safety calculations for loss-of-coolant acci-on fuel during pellet-chdding mechanicalinteraction dents. Other best-estimate models will be considered and Aould be considered. Fuel creep is a function of acceptable providea their technical basis is demon-time, temperature, grain size, density, fission rate, strated wi h appropriate data and analyset Control t

oxygen-to-metal ratio. and external stress. Fuel ther-rod assembly insertion may be assumed if it is ex-mal expansion represents dimensional changes in pected to occur.

unirradiated fuel pellets caused by changes in tem-perature. An acceptable model for the above fuel pa-3.2.3 Decay of Actinides

\\

rameters should be based on in-pile and out-of-pile The heat from radioactive decay of actinides, in-test data. Cladding creep introduces compressive cluding neptunium and plutonium generated during 1.157-5

operation as well as isotopes of uranium, should be 3.2.6 IIcat Transfer from Reactor calculated in accordance with fuel cycle history and Internals known radioactive properties. The actinide decay Heat transfer from piping, vessel walls, and inter-heat chosen should be appropriate for the facility's nal hardware should be included in the calculation operating history. Best-estimate models will be con-and should be calculated in a best-estimate manne.

sidered acceptable provided their technical basis is Heat transfer to channel boxes, control rods, guide demonstrated with appropriate data and analyses.

tubes, and other in-core hardware should also be considered. Models will be considered acceptable 3 2.4 Fission l'roduct Decay Heat provided their technical basis is demonstrated with 1

appropriate data and analyses.

i The heat generation rates from radioactive decay

(

of fission products, including the effects of neutron 3.2.7 Primary to Secondary Heat Transfer capture, should be included in the calculation and (Not Applicable to Boiling Water should be calculated in a best-estimate manner. The Reactors) energy release per fission (Q value) should also be Heat transferred between the primary and secon-calculated in a best-estimate manner. Best-estimate dary systems through the steam generators should be methods will be considered acceptable provided their considered in the calculation and should be calcu-techrucal basis is demonstrated with appropriate data lated in a best-estimate manner. Models will be con-and analyses. The model m Reference 10 is consid-sidered acceptable provided their technical basis is ered acceptable for calculating fission product decay demonstrated with appropriate data and analyses.

heat.

3.3 Reactor Core Thermal / Physical Parameters 3.2.4.1 Model Evaluation Procedure for 3.3.1 Thermal Parameters for Swelling Fission Product Decay Heu. The values of mean energy per fission (Q) and the models for actinide and Rupture of the Cladding and Fuel Rods decay heat should be checked against a set of relevant data.

A calculation of tt}e swelling and rupture of the cladding resulting ?.om the temperature distribution 3.2.5 Metal-Water Reaction Rate in the cladding nid from the pressure difference be-tween the inside and outside of the cladding, both as The rate of energy release, hydrogen generation, a function of time, should be included in the analysis and cladding oxidation from the reaction of the zir-and should be performed in a best-estimate manner.

caloy cladding with stearn should be calculated in a The degree of swelling and rupture should be taken best-estimate manner. Best-estimate models will be into account in the calculation of gap conductance, considered acceptable provided their technical basis cladding oxidation and embrittlement, hydrogen gen-is demonstrated with appropriate data and analyses.

eratior., and heat transfer and fluid flow outside of For rods calculated to mpture their cladding during the cladding. The calculation of fuel and cladding the loss-of-coolant accident, the oxidation of the in-temperatures as a function of time should use valuer, side of the cladding should be calculated in a best-of gap conductance and other thermal parameters as estinate manner.

functions of temperature and time. Best-estimate methods to calculate the swelling of the cladding 3.2.5.1 Model Evaluation Procedure for should take into account spatially varying cladding M etal-Water Reaction Rate. Correlations to be used temperatures, heating rates, anisotropic material to calculate metal-water reaction rates at less than or properties, asymmetric deformation of cladding, and ergual to 1700*F should:

fuel rod thermal and mechanical parameters. Best-estimate methods will be considered acceptable pro-a.

Be checked against a set of rclevant data, vided their technical basis is demonstrated with ap-and propriate data.and analyses.

3.3.2 Other Core Thermal Parameters b.

Recognize the effects of steam pressure, pre-As necessary and appropriate, physical and c,xidauon el the cladding, deformation dur-ing oxidat*on, and internal oxidation from chemical changes in in-core materials (e.g., eutectic both steam anel 002 fuel.

formation, phase change, or other phenomena

( aused by material interaction) should be accounted The data of Reference 11 are considered accept-for in the reactor core thermai analysis. Best-estimate able for calculating the rates of energy release, hydro-models will be considered acceptable provided their gen 1,eneration, and cladding oxidation for cladding technical basis is demonstrated with appropriate data temperatures greater than 1900*F.

and analyses.

1.157-6

3.4 Blowdown Phenomena For critical flow from small breaks under strati-fied conditions, currently acceptable test data for as-3.4.1 Break Characteristics and Flow sessing models and codes include those reported by:

( ]j dents, a spectrum of possible break sizes should be

/

In analyses of hypothetical loss-of-coolant acci-Anderson and Owca (Ref. 21) e considered, as indicated in Regulatory Position 3.1.

Re;mann and Khan (Ref. 22) e The discharge flow rate should be calculated with a critical flow rate model that considers the fluid Schrock et al. (Refs. 23 and 24) e conditions at the break locatien, upstream and i o-stream pressures, and break geo. metry. The u 3.4.2 ECC Bypass Bow model should be justified by comparisor> to ap-The best-es:imate code should contain a calcula-j plicable experimental data over a range of conditions tion of the amount of injected cooling water that by-for which the modelis applied. The model should be passes the vessel during the blowdown phase of the a best-estimate calculation, with uncertainty in the loss-of-coolant accident. The calculation of ECC by-critical fkiw rate included as part of the uncertainty pass should be a best-estimate calculation using y

evaluation. Best-estimate models will be considered analyses and comparisons with applicable experimen-i acceptable provided their technical basis is demon-tal data. Although it is clear that the dominant proc-l strated with appropriate data and analyses, esses governing ECC bypass are multidimerssional, single-dimensional approximations justified through 3.4.1.1 Model Evaluation Procedure for Discharge Flow Rate. Critical flow models to be sufficiern analysis and data may be acceptable. Best-employed m ECCS evaluations should:

estimate methods will be considered acceptable pro-vided the.ir technical basis is demonstrated with ap-a.

De checked against an acceptable set of rele.

propriate data and analyses. Cooling water that is not varc data, expelled, but remains in piping or is stored in parts of l

the vessel, should be calculated in a best-estimate i

b Recognize thermal nonequilibrium conditions manner based on applicable experimental data.

j when the fluid is st,bcooled, and j

3.4.2.1 Model Evaluation Procedure for 1

I-Provide a means of transition f im nonequi-ECC Bypass. A correlation or model to be used to c.

librium to equilibrium conc. dons.

n loate ECC bypass should:

The uncertainties and bias of a correlation or a,

Be checked against an acceptable set of rele-model used to calculate critical flow should be stated, vant data, and as well as their range of applicability, b.

Recognize the effects of pressure, liquid sub-The mechanistic thermal nonequilibrium and slip cooling, fluid conditions, hot walls, and sys-model of Richter (Ref.12) compares well to small-tem geometry, and large-scale test data (Ref.13).

Uncertainties and bias in the correlations or models used to calculate ECC bypass should be 3.4.1.2 Experimental Data for Discharge Flow Rate. An acceptable set of relevant critical flow stated, as well as the range of their applicability.

data should cover the fluid conditions, geometries, For scaled-down PWR downcomers, correlations and types of breaks pertinent to light-water reactor by Beckner and Reyes (Ref. 25) compared well to loss-of-coolant accidents. The following tests should the bypass data of References 26 and 27. Correla-be considered in establishing an acceptable set of tions of Sun (Ref. 28) and Jones (Ref. 29) compare relevant data:

well to counter-current flow limiting (CCFL) test data of interest to BWRs.

Marviken tests (Ref.14) e 3.4.2.2 Experin, ental Data for ECC Bypass.

Moby Dick experiments (Ref.15)

The followmg tests shoWd tw considered in establishing a set of data for scaled-down PWR Brookhaven craical flashing flaws in nozzles downcomers:

e (Ref.16)

BatteHe Columbus test (Ref. 26) e l

Sozzi-Sutherland tests (Ref.17)

=

Creare test (Refs. 26 and 27) e Edwards experiments (Ref.18)

For a full scale PWR vessel, ECC bypass data will e

I L-}

become available from the forthcoming upper g

Super Moby Dick experiments (Refs.19 and plenum test facility (UPTF) experiments performed 20) as part of the.2D/3D program sponsored by the 1.157-7

o Horizontal tubes l

Federal Republic of Germany, Japan, and the United GE tests (Refs. 33 and 36)

States.

e Rod bundles

/

For BWRs, the following test should be consid-ered in establishing r.n acceptable set of relevant GE tests (Ref. 37) l data:

4 3.7 Momentum Equation i

SSTF test data (Refs. 30 through 32)

The following effects should be taken into e

account in the two-phase conservation of momentum 3.5 Noding Near the Break and ECCS Injection equation:

(1) temporal change in niomentum, Point (2) momentum convection, (3) area change mornen-tum flux (4) momentum change due to compressibil-The break location and ECCS injection point are areas of high fluid velocity and complex fluid flow ity, (5) pressure loss resulting from wall friction, and contain phenomena that are often difficult to cal.

(6) pressure loss resulting from crea change, and culate. The results of these calculations are often (7) gravitational acceleration. Best-estimate modeis will be considered acceptable provided their technical highly dependent on the noding. Sufficient sensitivity basis is demonstrated with appropriate data and studies should be performed on the noding and other important parameters to ensure that the calculations analyses.

provide realistic r:sults.

3.8 Critical lleat Flux Best-estimate models developed from appropri-i 3.6 Frictional Pressure Drop ate steady-state or transient experimental data should The frictional losses in pipes and other compo-be used in calculating critical heat flux (CHF) during nents should be calculated using models that include loss-of-coolant accidents. The codes in which these variation of friction factor with Reynolds number and models are used should contain suitable checks tc, en-account for two-phase flow effects on friction. Best-sure that the range of conditions over which these estimate models will be considered acceptable pro-correlations are used are within those intended. Re-vided therr technical basis is demonstrated with ap-search has shown that CHF is highly dependent on propriate data and analyses.

the fuel rod geometry, local heat flux, and fluid con-ditions. After CrlF is predicted at an axial fuel rod 3.6.1 Model Evaluation Procedure for location, the calculation may use nucleate boiling Fractional Pressure Drop W

sb correlations if the calculated local fluid A model for frictional pressure drop to be used and surface conditioni justify the reestablishment of in ECCS evaluation should:

nucleate boiling. Best-estimate models will be consid-ered acceptable provided their technical basis is dem-a.

Be checked againe a set of relevant data, onstrated with appropriate data and analyses.

and 3.9 Post-CIIF Blowdown Heat Transfer b.

Be consistent with models used for calculat-Models of heat transfer from the fuel to the sur-ing gravitational and acceleration pressure rounding fluid in the post-CHF regimes of transition drops. If void fraction models or correlations and film boiling should be best-estimate models based used to calculate the three components of the on comparison to applicable steady-state or transient t

total pressure drop differ one from another' data. Any model should be evaluated to demonstrate a quantitative justification must be provided.

that it provides acceptable results over the applicable i

Uncertainties and bias of a correlation or model ranges. Best-estimate models will be considerert ac-should be stated as well as the rege of applicability.

ceptable provided their technical basis is demon-strated with appropriate data and analyses.

3.6.2 Experimental Data for Frictional 3.9.1 Model Evaluation Procedure for Pressure Drop Post-CHF lleat Transfer An acceptable set of relevant data rhould cover, A model to be used in ECCS evaluation to calcu-as far as possbie, the ranges of parameters (mass late post CHF heat transfer from rod bundies should:

flux, quality, pressure, Puid physical ;,ropertMs, roughness, and geometries) that are found in actual plant applications. Tne following tests should be con-u.

Be checked against an acceptable set cf rele-sidered m establishing an acceptable set of relevant vant data, and data:

b.

Recognize effects of liquid entrainment, ther-l e

Vertical tubes mal radiation, thermal nonequilibrium, low CISE test (Refs. 33 and 34) and high mass flow rates, low and high power 1.157-8

. densities, and saturated and subcooled inlet variables. Pr represents the Prandtl number, and Nu conditions.

is the Nusselt number. The physical properties may The uncertainties and bias of models or correla, be defined as wall, film, or vapor values.

[]

tions used to calculate post-CHF heat transfer should A distinction from, and transition to, laminar i

/

be stated as well as the range of their applicability, convection (i.e., Re <2000) should be made, with a value of the laminar heat transfer for rod bundles 3.9.2 Experimental Data for Post-CHF that is t.ppropciate for the applicable bundle geometry Heat Transfer and flow conditions.

The acceptable set of relevant data should cover Other forms and values, depending on the bun-power densities, mass flow rates, fluid conditions, die Ecometry and flow conditions, are also appropri-7 and rod bundle geometries pertinent to light-water re-ate.

actor designs and applications. The following tests should be considered in establishing an acceptable set 3.9.3.2 Experimental Data for Heat of relevara data:

Transfer from Uncovered Rod B undles. An j

l acceptable set of relevant data for post-CHF heat t

ORNL tests (Refs. 38 and 39) transfer from uncovered rod bundles should cover j

e power densities, fluid conditions, and rod bundle J

FLECHT-SEASET tests (Ref. 40) geometries pertinent to light-water reactor design and e

application. The following tests should be considered i

INEL tests (Ref. 41) in establishing an acceptable set of relevant data:

e ORNL data base (Ref. 4?)

ORNL-THTF tests (Refs. 43 and 44) e e

FLECHT-SEASET tests (Refs. 45 and 46) i e

3.9.3 Post-CHF Heat Transfer from

)

Uncovered Bundles ORNL data base (Ref. 42)

During some time periods of small-break loss-of-3.10 Pump Modeling coolant accidents and during portions of large breaks prior to reflood, partial or complete core uncovering The characteristics of rotating primary system may be calculated to occur. Under these circum.

pumps should be derived from a best-estimate dy-

)

namic model that includes momentum transfer be-e\\ stances, special considerations for calculating heat

[

transfer are necessary.

tween the iluid and the rotating member, with vari-()

able pump speed as a function of time. The pump 3.9.3.1 Model Evaluation Procedures for model resistance and other empirical terms should be

)

Heat Transfer from Uncovered Rod Bundles. A

,iustified through comparisons with applicable data.

correlation to be used in ECCS evaluations to The pump model for the two-phase region should be calculate heat transfer from uncovered rod bundles verified by comparison to applicable two-phase per-should:

formance data. Pump coastdown following loss of power should be treated in a best-estimate manner. A a.

Be checked against an acceptable set of rele-locked rotor following a large-break loss-of-coolant vant data, and accident need not be assumed unless it is calculated b.

Recognize the effects of radiation and of to occur. Best-estimate models will be considered ac-laminar, transition, and turbulent flows.

ceptable provided their technical basis is demon-strated with appropriate data and analyses.

Uncertainties and bias in the models and correla-tions used to calculate post-CHF heat transfer should 3.11 Core Flow Distribution During Blowdown be stated, as should the range of their applicability.

The core flow through the hottest region of the The correlation derived should include a stated c re during the blowdown should be calculated as a procedure for correcting for radiative heat transfer function of time. For the purpose of the;e calcula-and for estimating the vapor temperatures. The Hot-ti ns, the hottest region of the core should not be tel procedure cited in Reference 43 is a satisfactory greater than the size of one fuel assembly. Calcula-example.

Li ns of the flow in the het re; ion should take into account any cross-flow between regions end ae) flow The ttirbulent correlation may be of the generel blockage calculated to occur during the blowdown as form:

a result of cladding swelling or rupture. The numeri-W=ARFP, cal scheme should enstue that unrealistic oscillations p

of the calculated flow do not result. Best-estimate

(

l for higher Reynolds numbers (Re), where the coeffi-rnodels will be considered acceptable provided their (j cients A, m, and n are modifications from the basic technical basis is demonstrated with appropriate data Dittus-Boelter form and may be functions of ether and analyses.

1.157-9

3.13 Post-Blowdown Phenomena effects of the compressed gas in the accumulator fol-lowing accumulator water discharge should be in-3.12.1 Containment Pressure cluded in the calculation. Any model or code used The containment pressure used for evaluating for this calculation should be assessed against applica-cooling effectiveness during the post-blowdown phase ble experimental data. Reference 7 describes a large of a loss-of-coolant accident should be calculated in a body of refill /reflood thermal-hydraulic data obtained best-estim te manner and chould include the effects from the 2D/3D program that is appropriate for of containment heat sinks. The calculation should in-consideration.

clude the effects of operation of all pressure-reducing equipment assumed to be available.

hestimate 3.12.2.2 Experimental Data for Post-Blowdown Thermal Hydraulics. The (pilowing tests I

models will be considered acceptable provided their should be considered when establishing an acceptable technical basis is demonstrated with appropriate data set f relevant data:

and analyses.

e s M 4 and W e

3.12.2 Calculation of Post-Blowdown Thermal Hydraulics for Pressurized ORNL tests (Refs. 43 and 49)

Water Reactors FLECHT-SEASET test (Ref. 45)

The refilling of the reactor vessel and the ulti-mate reflooding of the core should be calculated by a THETIS tests (Ref. 50) best-estimate model that takes irato consideration the o

thermal and hydraulic characteristics of the core, the 3.12.3 Steam Inte - don with Emergency emergency core cooling systems, and the primary and Core Coohng Mater m Pressurized secondary reactor systems. The model should be ca-a r acmrs pable of calculating the two-phase levelin the reactor during the postulated transient. Best-estimate models The thermal-hydraulic interaction between the will be considered acceptable provided their technical steam or two-phase fluid and the emergency core basis is demonstrated with appropriate data and cooling water should be taken into account in calcu-lating the core thermal hydraulics and the steam flow analyses.

through the reactor coolant pipes during the time the 3.12.2.1 Model Evaluation Procedures for accumulators are discharging water. Best-estimate Post-Blowdown Thermal Hydraulics. A correlation models will be considered acceptable provided their or model to be used in ECCS evaluation to calculate technical basis is demonstrated with appropriate data level swell should be checked against an acceptable and analyses.

set of relevant data and should recognize the effects of depressurization, boil-off, power level, fluid 3.12.4 Post-Blowdown Heat Transfer for conditions, and system geometry.

Pressurized Water Reactors The correlation proposed by Chexal, Horowitz, During refilling of the reactor vessel and ultimate and Lellouche (Ref. 47) provides acceptable results reflooding of the core, the heat transfer calculations when compared to experimental data reported in should be based on a best-estimate calculation of the Referencet 43, 48, 49, and 50.

fluid flow through the core, accounting for unique Uncerta'nties and bias of a correlation or model emergmy g te c ling systems. The calculations should also include the effects of any flow blockage used to calculate level swell should be stated, as "E ** *

"E #

should the range of applicability.

j rupture. Heat transfer calculations that account for j

The primary coolant pumps should be assumed two-phase conditions in the core during refilling of j

to be operating in the expected manc.er, based on the the reactor sessel should be justified through com-l assumptions of Regulatory Position 3.1, when calcu-parisons with experimental data. Best-estimate mod-lating the resistance effered by the pumps to fluid els will be considered acceptable provided their tech-flow. Models will be considered accept @% pravided nical basis is demonstrated through comparison with their technical basis is demonstrated through com-appropriate data and analyses.

prison with appropriate data and analyses.

The FLECHT-SEASET tests (Refs. 40, 45, and The total fluid flow leaving the core exit (car-

46) should be considered when establishing an ac-ryover) should be calculated using a best estimate ceptable set (,f relevant data. Reference 7 contains model that includes the effect of cross-flow on car-extensive information regarding a large amount of ex-ryover and core fluid distribution. Thermal-hydraulic perimental reflood heat transfer data. This informa-phenomena associated with unique emergency core tion should also be considered when aeveloping and cooling systems, such as upper plenum injection and assessing models. The results from the 2D/3D pro-upper head injection, should be accounted for. The gram are particularly relevant.

1.157-10

3.13 Convective Ileat Transfer Coefficients for different from those phenomena that would occur-

)

' Boiling Water Reactor Rods Under Spray during a large-break lowof-coolant accident. The Cooling distribution of liquid throughout the reactor system, Models will be considered acceptable providec; in addition to the total liquid inventory, is of in-9 their technical bases can be justified with appropriate creased importance for the small-break loss-of-data hnd analyses. These models should contain the coolant accident. A number of special factors must following:

be given increased consideration in small-break loss-of-coolant accident calculations to correctly predict 1.

Following the blowdown period, convective phenomena influenced by the h:iuid inventory heat transfer coefficients should be deter-distribution.

mined based on the calculated fluid condi-Break flow may be greatly influenced by the loca-tions and heat transfer modes withm the bun-tion and specific geometry of the break. For a break die and on the calculated rod temperatures.

in a horizontal pipe containing stratified flow, the 2.

During the period following the flashing of quality of the break flow will be a strong function of the lower plenum fluid, but prior to ECCS the assumed location of the break on the pipe (e.g.,

initiation, heat transfer models should in-top or bottom). Small-break loss-of-coolant accident dude cooling by steam flow or by a two-calculations should, therefore, include various as-phase mixture, if calculated to occur, sumed break locations in the spectrum of breaks ana-lyzed. The assumed operating state of the reactor 3.

Following initiation of ECCS flow, but prior coolant pump will also influence the distribution of to reflooding, heat transfer should be based liquid throughout the system and the amount of liquid on the actual calculated bundle fluid condi-lost through the break.

tions and best-estimate heat transfer models that take into account rod to-rod variations The pump operation assumptions used in the cal-m heat transfer.

culations should be the most likely, based on operat-ing procedures, with appropriate consideration of the j

4.

After the two-phase reflood level reaches the uncertainty of the pump operation during an actual level under consideration, a best-estimate event. Level depression in the core region and r,ubse-heat transfer model should be used. This quent core heatup may be influenced by liquid mociel should include the effects of any flow holdup in the steam generator tubes, manometric ef-O blockage calculated to occur as a result of fects of liquid in the piping and loop seal region, and cladding swelling or rupture.

liquid levels relative to vent paths for steam through upper plenum bypass flow paths and vent valves.

5. Thermal-hydraulic models that do not calcu-late multiple channel effects should be com-Steam generator heat transfer under " reflux" or j

pared with applicable experimental data or

" boiler-condensor" modes of operation may also more detailed calculations to ensure that all strongly influence core inventory through level de-important phenomena are adequately Pression and the effect on total system pressure and, calculated, thus, on ECCS flow. These phenomena should be carefully considered in the calculation. Sensitivity 3,14 Boiling Water Reactor Channel Box Under studies of the importance of these effects should be i

Spray Cooling performed for use in the uncertainty evaluation.

Following the blowdown period, heat transfer lleat transfer from an uncovered core under from the channel box and wetting of the channel box high-pressure conditions typical during a small-break should be based on the r.;culated fluid conditions on loss-of-coolant accident may include contributions both sides of the channel box and should make use from both convective and radiation heat transfer to of best-estimate heat transfer and rewetting models the steam. Models will be considered acceptable pro-that have been compared with applicable experimen-vided their technical basis is demonstrated through tal data.

comparison with appropriate data and analyses. Spe-C

""C"E" "E ""'"*

3.15 Special Considerations for a Small-Break fer is given in Regulatory Position 3.9.3.

Loss-of-Coolant Accident in Pressurized Water Reactors 3.16 Other Features of Best-Estimate Codes The slower small-break loss-of-coolant accident No list of best-estimate code features could be leaos L r.uio conditions characterized by separation all-inclusive, because the important features of a of the fluid phases versus the more homogeneous best-estimate code may vary depending on the tran-fluid conditions that would result from rapid large-sient to be calculated and the required accuracy of 9 break loss-of-coolant accident trawients. Phenomena the calculation. Because of this, no attempt has been that would occur in a PWR during a small-break loss-made to construct an exhaustive list of best-estimate of-coolant accident would, therefore, be significantly code features. Rather, features that were identified as 1.157-11

important for inclusion in Appendix K were used es a tainty to ensure that meaningful comparisons are be-basis for the above list. These features are not neces-ing niade.

sarily any more or less important than other code fea-tures, t ut were highlighted because it is necessary to 4.

ESTIMATION OF OVERALL give specific examples of how current best-estimate CALCULATIONAL UNCERTAINTY models may vary from methods used traditionally in 4.1 General evaluation model codes using the various Appendix K conservatism. In addition, models have not been The term " uncertainty," when applied to best-estimate thermal-hydraulic transient codes, is used at meluded for areas in which the best model would be 1

wo levels. At the lower or more detailed level, the highly dependent on the specif c plant design or the specific transient under consideranon.

term refers to the degree to which ar. individual model, correlation, or method used within the code The NRC staff believes that good examples of represents the physical phenomenon it addresses.

best-estimate thermal-hydraulic transient codes are These individual uncertainties, when taken together, those deseloped by the NRC (e.g., TRAC-PWR, comprise the " code uncertainty."

TRAC-BWR, RELAP5, COBRA, and FRAP). Al-The combined uncertainty associated with indi-though these codes are subject to further improve-vidual models (i.e., code uncertainty) within the best-ment, based on their ongoing use and assessment, estimate codes does not account for all of the uncer-they currently proside reasonable best-estimate cal-tainty associated with the model's use. In addition to culations of the loss-of-coolant accident in a full-scale the code uncertainty, various other sources of uncer-light-water reactor. This is substantiated through the tainty are introduced when attempting to use best-code development and assessment literature gener' estimate codes to predict full-scale plant ated by the NRC and its contractors over the past thermal-hydraulic response. These sources include several years.

uncertainty associated with the experimental data It is possible, however, to describe in general used in the code assessment process (including appli-how other features of best-estimate codes should be cability of the data to full-scale reactors), the input constructed. Two basic criteria should be applied, boundary and initial conditions, and the fuel behav-completeness and comparisons to experimental data.

ior. Additional sources of uncertainty stem from the use of simplifying assumptions and approximations. A 3.16.1 Completeness careful statement of these assumptions and approxi-Best-estimate codes should contain models in niad ns should be made, and the uncertainty assoc -

ted with them should be taken into account. There-sufficient detail to predict phenomena that are impor-f re, the "overall calculational uncertainty" is tant to demonstrate compliance with the acceptance criteria specified in paragraph 50.46(b) of 10 CFR defined as the uncertainty arrived at when all the Part 50 (e.g., peak cladding temperature). Simplifi-c ntributions from the sources identified above, m-cations are acceptable as lopg as code uncertainties ciuding the code uncertainty, are tak.en into account.

or biases do not become so large that they cast doubt A 95% probability level is considered acceptable on the actual behavior that would occur or on the to the NRC staff for comparison of best-estimate pre-l true effect of assumed initial and boundary condi-dictions to the applicable limits of paragraph tions (e.g., equipment sizing, safety system settings).

50.46(b) of 10 CFR Part 50 to meet the requirement Comparisons of the overall calculations to integral ex.

of paragraph 50.46(a)(1)(i) to show that there is a periments should be performed to ensure that impor-high probability that the criteria will not be exceeded.

tant phenomena can be predicted and to help in The basis for selecting the 95% probability level is l

making judgments on the effect of code simplifica-

p. 'mari:y for consistency with standard engineering I

tions. Consideration should also be given to the un-m tice in regulatory matters involving thermal hy-certainty and validity of the experiment to ensure that (1..ulics. Many parameters, most notably the depar-meaningful comparisons are being made.

ture from nucleate boiling ratio (DNBR), have been found acceptable by the NRC staff in the past at the 3.16.2 Data Comparisons 95% probability level.

Individual best-estimate models should be com-This 95% probability level would also be applied pared to applicable experimental data to ensure that to small-break loss-of-coolant accidents, which have reahstic behavior is predicted and that mlevant ex-a higher probability than large breaks. The dominant perimental variables are included. Uncertamty analy-factors influencing risk from small-break loss-of-ses are required to ensure that a major bias does not coolant accidents includ; equipment availability and i

exist in the models and that the model uncertainty is operator actions. Calculational uncertainties are small enough to provide a realistic estimate of the ef-much less important than factors such as operator feet of important exper!.aental variables. Uncertainty recognition of the event, the availability of equip-analyses should also consider experimental uncer-ment, and the correct use of this equipment. The use 1

1 1.157-12 l

l L____________

I of a best-estimate calcult. tion with reasonable and best-estimate capability of the code. For large-break quantifiable uncertainty is expected to provide a re-loss-of-coolant accidents, the most important key pa-duction in the overall risk from a small-break loss-of-rameter is peak cladding temperature, which is ad-

[ )g coolant accident by providing more realistic calcula-dressed by one of the criteria of paagraph 50.46(b)

("

tions with which to evaluate operator guidelines and and has a direct influence on the other criteria. In determine the true effect of equipment availability.

addition, a code uncertainty evaluation should be performed for other important parameters for the Regulatory Position 3 provides a description of transient of interest to evaluate compensating errors.

(

the features that should be included in the overall For small-break loss-of-coolant accidents, the clad-code uncertainty evaluation that is called for in para-ding temperature response is the most important graph 50.46(a)(1). This uncertainty evaluation parameter; however, the ability of the codes to pre-should make use of probabilistic and statistical meth-dict overall system mass and reactor vesselinventory ods to determine the code uncertainty. For a calcula-distribution should also be statistically examined.

)

tion of this complexity, a completely rigorous mathe-matical treatment is neither practical nor required. in In evaluating the code uncertainty, it will be nec-

]

many cases, approximations and asnmptions may be essary to evaluate the code's predictive ability over made to make the overall calculational uncertainty several time intervals, since different processes and evaluttion possible. A careful statement of these as-phenomena occur at different intervals. For example, sumptent a7d appruimations should be made so in large-break loss-of-coolant accident evaluations, that the NRC staff may make a judgment as to the separate code uncertainties may be required for the validity of the uncertainty evaluation. The purpose of Peak cladding tempercure during the blowdown and the uncertainty evaluation is to provide assurance p st-blowdown phases. Justification for treating these that for postulated loss-of-coolant accidents a given uncertainties individually or methods for combining plant will not, with a probability of 95% or more, ex-them should be provided.

ceed the applicable limits specified in paragraph The experimental information used to determine 50.46(b).

code uncertainty will usually be obtained from facili-ties that are much smaller than nuclear power reac-4.2 Code Uncertainty tors. Applicability of these results should be justified f r I rger scales. The effects of scale can be assessed This regulatory guide makes a distinction be-3 through comparisons to available large-scale separate-tween the terms " code uncertainty" and "overall cal-effects tests and through comparison to mtegral tests I

culational uncertainty." The latter term is defined fr m vari us sized facilities. If there are scaling prob-above and includes the contributions to the uncer-lems, pasulady W pWons are nonconsmak, tainty described in Regulatory Positions 4.2 and 4.3.

the code should be improved for large-scale plants The components of the code uncertainty fi.e., the

• " nuclear reactors). Codes not having scaling ca-contribution to the overall uncertainty from the mod-PaMa.y M not be acceptable if their predictions are els and numerical methods used) are described in n nc mmau,ve.

this section.

4.3 Other Sources of Uncertainty

]

Code uncertainty should be evaluated through di-rect data compariron with relevant integral systems When a best-estimate methodology is used to

]

and separate-effects experiments at different scales.

predict reactor transients, sources of uncertainty In this ma mer, an estimate of the uncertainty attrib-other than the limitations in the individual models l

utable to the combined effect of the models and cor.

and numerical methods (i.e., code uncertainty) are l

relations within the code can be obtained for all introduced. The following contributors to the overall i

scales and for different phenomena. Comparison to a calculational uncertYnty should also be considered in sufficient number of integral systems experiments, the uncertainty analysis, from different test facilities and different scales, 4.3.1 Initial and Boundary Conditions should be made to ensure that a reasonable estimate and Equipment Availability of code uncertainty and bias has been obtained.

When necessary, separate-effects experiments should When a plant input model is prepared, certain be used to establish code uncertainty for specific phe, relationships describing the plant boundary and initid nomena (e.g., comparisons to Cylindrical Core Test c nditions and the availability and performance of Facility data to ascertain code uncertainty in model-equipment are defined. These include factors such as ing upper plenum injection performance). Code com-initial power level, pump pu6amance, valve activa-tion times, nd control systems functioning. Uncer-parisons should account for limitations of the meas-urements and calibration errots, tainties associated with the boundary and initial con-f ditions and the characterization and performance of

(

There comparisons should be performed for im-equipment should be accounted for in the uncertainty portant key parameters to demonstrate the overall evaluation. It is also acceptable to limit the variables 1.157-13

to be considered by setting their values to conserva-and justification of the statistical distribution and in tive bounds.

the estimation of its statistical parameters. If a normal distribution is selected and justified, the probability 4.3.2 Fuel llehavior limit can be conservatively calculated using two Variability of the results of plant transient calcu-standard deviations. The added conservatisrr. of the lations can result from uncertainties associated with two standard deviations compared to the 95th fuel behavior, which are not included in the compari-percentile is used to account for uncertainty in the sons of code resuas with integral experiments since probability distribution. Other techniques that most integral tests use electrically heated rods. This account for the uncertainty in a more detailed i

uncertainty includes many effects such as fuel con-manner may be used. These techniques may require ductivity, gap width, gap conductivity, and peaking the use of confidence levels, which are not required factors. These uncertainties should be quantified and by the above approach.

used in the determination of the overall calculational The evaluation of the peak cladding temperature uncertainty.

at the 95% probability level need only be performed 4.3.3 Other Variables r N worspcase break identified by the break spec-trum analysis in order to demonstrate conformance There may be individual models within the best-with paragraph 50.46(b). However, in order to use estimate code whose effect may not have been evalu-this approach, j istification must be provided that ated by the comparison to the integral systems data.

demonstrates that the overall calculational uncer-For example, since most integral systems experiments tainty for the worst case bounds the uncertainty for use electrically heated rods, uncertainties associated other breaks within the spectrum. It may be neces-with the prediction of core decay heat and cladding sary to perform separate uncertainty evaluations for metal / water reaction have not been evaluated. In ad-large-and small-break loss-of-coolant accidents be-dition, to demonstrate the overall adequacy of the cause of the substantial difference in system thermal-predictive ability of the best-estimate code, it may be hydraulic behavior.

necessary to use empirically arrived at break-discharge coefficients to obtain a reasonable break The revised paragraph 50.46(a)(1)(i) requires flow. The uncertainties in the individual models that that it be shown with a high probability that none of have not been evaluated by comparison to integral the criteria of paragraph 50.46(b) will be exceeded, systems data should be quantified and used in the and is not limited to the peak cladding temperature criterion. Ilowever, since the other criteria are determination of overall code uncertainty.

strongly dependent on peak cladding temperature, 4.4 Statistical Treatment of Overan exphcit consideration of the probability of exceeding Calculational Uncertainty the other criteria may not be required if it can be The methodology used to obtain an estimate of demonstrated that meeting the temperature criterion the overall calculational uncertainty at the 95% prob-at the 95% probability level ensures with an equal or ability limit should be provided and justified. If linear greater probability that the other criteria will not be exceeded.

independence is assumed, suitable justification should be provided. The influence of the individual 4.5 NRC Approach to LOCA Uncertainty parameters on code uncertainty should be examined Evaluation by making comparisons to relevant experimental Chapter 4 of the " Compendium of ECCS data. Justification should be provided for the as-Research for Realistic LOCA Analysis" (Ref. 7) sumed distnbution of the parameter and the range considered.

presents a methodology that has been used for

}

evaluating the overall calculational uncertainty in In reality, the true statistical distribution for the peak cladding temperature predictions for key parameters (e.g., peak cladding temperature) is best-estimate thermal-hydraulic transient codes that unknown. The choice of a statistical distribution the NRC has developed.

should be verified using applicable engineering data and information. The statistical parameters appropri-D. IMPLEMENTATION ate for that distribution should be estimated using available data and results of engineering analyses.

The purpose of this section is to provide informa-Supnorting documentation should be provided for tion to applicants and licensees regarding the NRC this selection process. These esdmated values are as-staff's plans for using this regulatory guide, sumed to be the true values of the statistical param-Licensees and applicants may propose means eters c>f the distribution. With these assumptions, an other than those specified by the provisions of the upper one-sided probability limit can be calculated at Regulatory Position of this guide for meeting applica-the 95% level. As the probabihty limit approaches ble regulations. This guide has been approved for use 2200*F, more care must be taken in the selection by the NRC staff as an acceptable means of 1.157-14

complying with the Commission's regulations and for allow the use of realistic models as an alter-evaluating submittals in the following categories:

n tive to the features of Appendix K of a

10 CFR Pan 50.

s n

1.. Construct. ion permit applicants that choose to 77%

hi

?

make use of the provisions of 9 50,46 that

1. f allow the use of realistic models as'an alter.

3.

Operating reactor licensees will not be native to the features of Appendix K of evaluated against the provisions of this guide 10 CFR Part 50.

except for new submittals that make use of the provisions of 9 50.46 that allow the use '

2.

Operating license applicants that choose to of realistic models as an alternative to the make use of the provisions of 9 50,46 that teatures of Appendix K of 10 CFR Part 50, n

fA_

4 i

6 s\\

l i

1 1.157-15 l

j

REFERENCES 1.

Los Alarnos National Leboratory, " TRAC-NUREG-17, August 1977. (Available from PF1/ MOD 1: An Advanced Best-Estimate Com-NTIS.)

puter Program for Pressunzed Water Reactor Thermal-llydraulic Analysis,"

NUREG/

12. H. J. Richter, " Separated Twe-Phase Flow CR-3858 (LA-10157-MS), July 1986.

Model:

Application to Critical Two-Phase Flow," EPRI Report NP-1800, Electric Power 2.

Idaho National Engineering Laboratory, Research Institute, Palo Alto, CA April 1981.

" TRAC-BD1/ MOD 1: An Advanced Best Esti-13.

D. Abdollahian et al., " Critical Flow Data Re-mate Computer Program for Boiling Water Re.

actor Transient Analysis," NUREG/CR-3633, 4 view and Analysis," Report NP-2192, Electric Powur Research Institute, Palo Alto, CA, Janu-Vols. (EGG-2294), April 1984.

l ary 4982.

3.

Idaho National Engineenng Laboratory,

14. USNRC, "The Marviken Full Scale Critical "RELAP5/ MOD 2 Code Manual," Vols.1 & 2, Ph.w Tests Summary Report," (Joint Reactor NUREG/CR-4312, August 1985. (Available in Safeq Experiments in the Marviken Power Sta-the NRC Public Document Room.)

l tion, Swden), NUREG/CR-2671, May 1982.

4.

Pacific Northwest Laboratory, " COBRA / TRAC

15. M. Reccreux,,, contribution to the Study of m

- A Thermal-Hydraulics Code for Transient Two-Phase Steam-her Cnucal Flow," Ph.D Analysis of Nuclear Reactor Vessels and Pri-Thesis, L'Umversite Scientifique Medicale de mary Coolant Systems," NUREG/CR-3046, 5 Grenoble 1974. (English translation available Vols. (PNL-4385), March 1983.

from NTIS, LIB /Trans-576.)

5.

L. J. Siefken et al., "FRAP-T6: A Computer

16. N. Abuaf, G. A. Zimmer, B. J. C. Wu, " A Code for the Transient Analysis of Oxide Fuel Study of Nonequilibrium Flashing of Water in a Rods," NUREG/CR-2148 (EG&G, EGG-2104),

Converging-Diverging Nozzle,"

NUREG/

May 1981.

CR-1864, Vols. 1-2 (Brookhaven National Laboratory, BNL-NUREG-51317),

March 6.

G, A. Berna et al., "FRAPCON-2: A Com-

39g2, puter Code for the Calculation of Steady State Thermal-Mechanical Behavior of Oxide Fuel

17. G. L. Sozzi and W. A. Sutherland, " Critical Rods," NUREG/CR-1845, January 1981-Flow of Saturated and Subcooled Water at High Pressure " General Electric Company, GE Re-7.

" Compendium of ECCS Research for Realistic port NEDO-13418, 1975. (Available in the LOCA Analysis," NUREG-1230 December NRC Public Document Room.)

1988.

18. R. A. Edwards and T. P. O'Brien, " Studies of 8.

D. Lanning and M. Cunnmgham, " Trends in Phenomena Connected with the Depressuriza-Thermal Calculations for Light Water Reactor tion of Water Reactors," Nuclear Energy (Jour-Fuel (1971-1981)," in Ninth Water Reactor nal of the British Nuclear Energy Society), Vol.

Safety Research Information Meeting, 'USNRC, 9, No. 2, April 1970.

NUREG/CP-0024, Vol. 3, March 1982.

19. Co,maissariat a L'Energie Atomique, C. Jeen-9.

Idaho National Engineering Laboratory, dey et al., " Auto vaporization d'ecoulements

• M ATPRO Version 11 (Revision 2): A Har.d-eau /vapeur," Report Tl, No. 163, Centre book of Materials Properties for Use in the d' Etudes Nucleaires de Grenoble, Dept. des Analysis of Light-Water Reactor Fuel Rod Be-Reacteurs a Eau, Service des Transferts Thei-havior," NUREG/CR-0497, Rev.

2, August miques, Grenoble France, July 1981. (Copies 198L may be obtained from Maurice uomolinski, CEA, B.P. No. 6, 92260 fontenay-aux-Roses

10. American Nu' lear Society,

" American Na-Cedex, France.)

tional Standard for Decay Heat Power in Light Water Reactors," ANSI /ANS-5.1-19V9, August

20. C. Aaredey and L. Gros d'Aillon, " Critical 1979. (ANS,555 North Kensington Avenue, La Flows in a Short Super Moby Dick Pipe," Rap-Grange Parx, ilhnois 60525.)

port TT/SETRE/71, Centre d' Etudes Nucleaires de Grenoble, Grenoble, France, September L'

11. J. V. Cathcart et al., " Zirconium Metal-Water 1983. NRC Translation 1401 available from the Oxidation Kinetics: IV Reaction Rate Studies,"

NRC Public Document Room (52 FR '6334), ac-Oak Ridge National Laboratory, ORNL /

eession number 8704060298.

1.157-16

21. J. L. Anderson and W. A. Owca, " Data Report

31. D. G. Schumacher et al., "BWR Refill-Reflood for the TPFL Tee /Critica! Flow Experiments,"

Program Task 4.4 - CCFL/ Refill System Ef-NUREG/CR-4164 (EG&G

Idaho, Inc.,

fects Tests (30 Sector). SSTF Systems Re-EGG-2377), November 1985.

sponse Test Resuks," NUREG/CR-2568 (Gen-i 9

22. J. Reimann and M. Khan, " Flow Through a eral Electric Company, GEAP-22046, EPRI l

NP-2374), April 1983.

Small Break at the Bottom of a Large Pipe with Stratified Flow," Nuclear Science and Enginecr-

32. J. A. Findlay, "BWR Refill-Reflood Program ing, Vol. 88, pp. 297-310, November 1984.

Task 4.4 - CCFL/ Refill System Effects Tests (30* Sector). Evaluation of ECCS Mixing Phe-

23. V. E. Schrock et al., " Steam-Water Critical n mena," NUREG/CR-2786 (General Electric Flow Through Smal! Pipes from Stratified Up.

Company, GEAP-22150, EPRI NP-2542), May stream Regions," in Heat Transfer 1986; C. L.

1983.

Tien, V. P. Carey, and J. K. Ferrell, Editors;

\\ 01. 5, pp. 2307-2311; Hemisphere Publishing

33. G. P. Gaspari, C. Lombardi, G. Peterlongo,

. Pressure Drops m Steam-Water Mixtures.

Corp., 242 Cherry St.,

Philadelphia, PA 19106, 1986' Round Tubes Vertical Upflow," Centro Infor-mazioni Studi Esperienze,

Milan, Italy,

24. V. E. Schrock et al., "Small Break Critical Dis-CISE-R83,1964. (Available from NTIS.)

charge - Roles of Vapor and Liquid Entrain-

34. A.

Alessadrini, G.

Peterlongo, R.

Ravetta, ment in Stratified Two-Phase Region Upstream "Large Scale Experiments on Heat Transfer and of the Break," NUREG/CR-4761 (Lawrence Hydrodynamics with Steam-Water Mixtures.

Berkeley Laboratory, LBL-22024), December Critical Heat Flux and Pressure Drop Measure-1986.

ments in Round Vertictl Tubes at the Pressure of 51 kg/cm2 abs," Centro Informazioni Studi

25. W. D. Beckner and J. N. Reyes, Research in-Esperienze, Milan, Italy, CISE-R86, 1963.

formation Letter No.128, "PWR Lower Plenum fAvailable from NTIS.)

Refill Research Results," USNRC, December 8, 1981. (Available in the NRC Public Document 35.

E. Janssen and J. A. Kervinen "Two-Phase Room.)

Pressure Drop Across Contractions and Expan-sions: Water-Steam Mixtures at 600 to 1400 9

26. W. D. Beckner, J. N. Reyes, R. Anderson, psia," AEC R&D Report GEAP-4622,1964.

" Analysis of ECC Bypass Data " U.S. Nuclear (Available in the NRC Public Document Room.)

Regulatory Commission, NUREG-0573, July 1979.

36. E. Janssen and J. A. Kervinen, "Two-Phase

27. C. J. Crowley et al., "1/5-Scale Countercurrent Pressure Drop in Straight Pipes and Channels:

Flow Data Presentation and Discussion,"

Water-Steam Mixtures at 600 to 1400 psia,"

NUREG/CR-2106 (Creare Incorporated, Creare AEC R&D Report GEAP-4616,1964. (Avail-TN-333), November 1981.

able in the NRC Public Document Room.)

37. R T. Lahey, B. S. Shiralkar, J. W. Radcliffe,

28. K. H. Sun, " Flooding Correlations for BWR

\\

"Two-Phase Flow and Heat Transfer m Mulu-Bundle Upper Tieplate and Bottom Side-Entry r d Geometries:

Subchannel and Pressure Orifices," in Multi-Those Transport: Funda-mentals, Reactor Safety, Applications, Vol.1, Drop Measurements in a Nine-Rod Bundle for T. N. Veziroglu, Editor, tiemisphere Publishing Diabatic and Adiabatic Conditions," AEC R&D Cor p.,

242 Cherry St.,

Philadelphia, PA Report GEAP-13049, General Electnc Com-19106, 1979.

Pany, March 1970.

38. G. L. Yoder et al., " Dispersed Flow Film Boil-

29. D. D. Jones, "Subcooled Counter-Current Flow ing in Rod Bundle Geometry - Steady State Lirr.iting Characteristics of the Upper Region of Heat Transfer Data and Correlation Compari-a BWR Fuel Bundle," General Electric Com-sons," NUREG/CR-2435 (Oak Ridge National

pany, NEDG-NUREG-23549, July 1977.

Laborator), ORNL-5822), April 1982.

(Available m the NRC Public Document Room.)

39.

D. G. Morris et al, " Dispersed Flow Film Boil-ing of High Pressure Water in a Rod Bundle,"

30. J. A. Findlay, "BWR Refill-Reflood Program NUREG/CR-2183 (Oak Ridge National Labora-Task 4.4 - CCFL/ Refill System Effects Tests tory, ORNL/TM-7864), September 1982.

9 (30 Sector). Evaluation of Parallel Channel Phenomena," NUREG/CR-2566 (General E!cc-

40. N. Lee et al., "PWR FLECHT-SEASET Un-tric Company, CEAP-22044, EPRI NP-2373),

blocked Bundle, Forced and Gravity Reflood November 1982.

Task Data Evaluation and Analysis Report,"

1.157-17 l

l 1

l

NUREG/CR-2256 (Westinghouse Electric Cor-search Institute, Palo Aho, CA, December poration, WCAP-9891. EPRI NP-2013), No-1986.

vember 1981.

48. D. S. Seely and R. Muralidharan, "BWR Low

41. R. C. Gottula et al., " Forced Convective, Non-Flow Bundle Uncovery Tests and Analysis,"

equilibrium, Post-CHF Heat Transfer Experi-NUREG/CR-2231 (General Electric Company, ment Data and Correlation Comparison Re-GEAP-24964 EPRI NP-1781), April 1982.

port." NUREG/CR-3193 (EG&G Idaho, Inc.,

EGG-2245), April 1985.

49. T. M. Anklam, "ORNL Small-Break LOCA Heat Transfer Series 1: Two-Phase Mixture

42. G. L. Yoder, " Rod Bundle Film Boiling and Level Swell Results," NUREG/CR-2115 (Oak Steam Cooling Data Base and Correlation Ridge National Laboratory, ORNL/NUREG/

t Evaluation," NUREG/CR-4394 (Oak Ridge Na-TM-447), September 1981.

tional Laboratory, ORNL/TM-9628), August 1986.

50. D. Jowitt, "A New Voidage Correlation for Level Swel! Conditions,"

Winfrith UK,

43. T. M. Anklam et al., " Experimental Investiga-AEEW-R-1488, December 1981. (Available in tions of Uncovered-Bundle Heat Transfer and the NRC Public Document Room.)

Two-Phase Mixture-Level Swell Under High-Pressure Low Heat-Flux Conditions," NUREG/

51. J. A. Findlay, "BWR Refill-Reflood Program CR-2456 (Oak Ridge National Laboratory, Task 4.8 - Model Qualification Task Plan,"

ORNL-5848), April 1982.

NUREG/CR-1899 (General Electric Company, GEAP-24898, EPRI NP-1527), August 1981.

44. G. L. Yoder et al., "High Dr;out Quality Film Boiling and Steam Cooling Heat Transfer Data ADDRESSES from a Rod Bundle," NOREG/CR-3502 (Oak Ridge National Laboratory, ORNL/TM-8794),

NUREG-and NUREG/CR-series documents are January 1984.

available from the Government Printing Office

45. S. Wong and L. E. Hochreiter, " Analysis of the (GPO) and the National Technical Information FLECHT-SEASET Unblocked Bundle Steam Service (NTIS).

Cooling and Boiloff Tests," NUREG/CR-1533 U.S. Government Printing Office (Westinghouse Electric Corporation, Post Office Box 37082 WCAP-9729, EPRI NP-1460), March 1981.

We.shington, DC 200134 082

46. M. J. Loftus et al., "PWR FLECHT SEASET National Technical Information Service 21-Rod Bundle Flow Blockage Test Data and Springfield, VA 22161 Analysis Report," NUREG/CR-2444, Vol.1-2 (Westinghouse Electric Corporation.

Documents that are in the NRC Public Document WCAP-9992, EPRI NP-2014), September Room are available for inspection or copying for a 1982.

fee.

47. B. J. Chexal J. Horowitz, G. Lellouche, " An USNRC Public Document Room Assessment of Eight Void Fraction Models ior 2120 L Street NW.

I Vertical Flows," NSAC-107 Electric Power Re-Washirgtrin, DC l

l

)

i 1.157-18

REGULATORY ANALYSIS l

A separate regu'.atory analysis has not been pre-Revision of the ECCS Rule and Supporting Regula-

.\\

pared in support of this regulatory guide. The regula-tory Guide " is available in the NRC Public Docu-

' tory analysis that supports the rulemaking effort also ment Room, 2120 L Street NW., Washington, DC, covers this regulatory guide. " Regulatory A.nalysis for under Regulatory Guide 1.157 (52 FR 6334).

I i

?-

E 9

i s

)

i

)

l i

,y-8 4

' d l

I 1

I i

l e

-(

1.157-19

UNITED STATES nost etas, ung

. NUCLEAR REGULATORY COMMISSION Po87Aar

• Fe<8 PAio WASHINGTON, D.C. 20555 PERMIT ho. G $7 OFFICIAL BUSINESS PENALTY FOR PRIVATE USE,4300 k

O s

l l

@1