ML20054G565
| ML20054G565 | |
| Person / Time | |
|---|---|
| Site: | Clinch River |
| Issue date: | 04/30/1982 |
| From: | Ami Agrawal, Bari R, Buslik A BROOKHAVEN NATIONAL LABORATORY |
| To: | Morris W Office of Nuclear Reactor Regulation |
| References | |
| CON-FIN-A-3360 BNL-NUREG-31297, NUDOCS 8206220068 | |
| Download: ML20054G565 (97) | |
Text
_
L.NUREG
-31297 I
FORMAL REPORT Limited Distribution REVIEW 0F THE STATUS OF CRBR LICENSING TECHNICAL ISSUES RELATED TO HEAT REMOVAL SYSTEM AND SEVERE ACCIDENT ANALYSIS 7
DATE PUBLISHED - APRIL 1982 D[PARTME NT Of NUCLE AR ENERGY BROOKHAVEN NATIONAL LABORATORY UPTON NEW YORK 11973 J
N( '()Ii
('
(,.
s) k(
h' (fI hie h fl h h"[
V4' e < ' N.:. ea Rea Rey...p;0r 3r t
4
( If d ' I f ' f 4(]1 'I
',].
- b 0;O sy:n r,
PDR H [' '
8,'06/JDON1
..I.
i e'
45 e s 1
ni 4
e N OT I C I;
'I hn rep..ri w as piepared a. an acc ount of w..r k sponsored hs the t'mied States
( ;oset ninent Neither the t'n:ted States nor the Unitnl States Nutl+ ar Regulatory
(:onnnuwion, nor ans of their emplos res, nor any of their r oneractors. suleinera(tors, or tbrir t'm]do} re s Inahr%.tllY wari.In.s, r\\l>It w or I!Dplied, t,nr a%Utnri Jins lrgal hahihn or trsinnsibrhts for the au uran. completeness or uv fulne ss of any infortna-tion. a ppu at us, prmf u. t or prm ew dat lowd or re presents that its use w ould not mtr mer pris atch ow nr.1 ri.ghts i
h BNL-NUREG-31297
-INFORMAL REFORT LIMITED DISTRIBUTION REVIEW 0F THE STATUS OF CRBR LICENSING TECHNICAL ISSUES RELATED TO HEAT REMOVAL SYSTEM AND SEVERE ACCIDENT ANALYSIS o
. "5 8
R. A. Bari A. K. Agrawal A. J. Buslik R. D. Gasser T. Ginsberg J. G. Guppy H. Ludewig I. A. Papazoglou K. R. Perkins W. T. Pratt Department of Nuclear Energy BROOKHAVEN NATIONAL LABORATORY Upton, New York 11973 i
April 1982 I[
>V l
Prepared for l
U. S. Nuclear Regulatory Connission l
Washington, D. C.
20555 Under Interagency Aareenent DE-AC02-76CH00016 l
i O
9 O!
ABSTRACT A review is presented of the status of licensing technical issues re-lated to heat removal system and severe accident analysis for the Clinch River Breeder Reactor.
Discussions are presented on the following topics:
opera-tional transients; natural circulation heat removal and the overall capability of the shutdown heat removal system; containment thermal analysis associated with design basis sodium spills and with core-disruptive accidents; reliability and risk analysis; loss-of-heat-sink accident progression with scram; reactor physics issues related to the heterogeneous core design; transition phase of 1
the core-disruptive accident; structural analysis.
This work is part of the initial phase of the re-initiation of the licensing review of the Clinch River Breeder Reactor.
O r )
-lii-
.e TABLE OF CONTENTS
~
~
Pace q
ABSTRACT.................................
iii L IS T OF FI GUR E S...........v. '......'.............
vi i ---
~
viii LIST OF TABLES.............................
1.
INTRODUCTION
- 1..-. : g'........
1 1.1 Background...........................
1 1.2 The Role of Brookhaven National Laboratory in tie. Earlier Phase '
3 of the Licensing Review 1.3 Scope of this Report..........
3 S'
2.
OPERATIONAL TRANSIENTS 2.1 Objectives and Issues
...................J.5 2.2 Reactor Physics Analysi s....................
7 2.2.1 Control Rod Worth....................
7 2.2.2 Reactivity Feedback Coefficients 10
~
2.2.3 Reactivity Feedback Due to Core Compaction 11 2.3 Thermal-Hydraulic Analysi s.............. 1
'll 2.4 PPS/PCS Systems and Their Interaction 14 14 2.5 Rod Design Criteria 15 2.6 Sodium-Water Reaction.....................
16 2.7 Pipe Breaks..........'................
17 REFERENCES s
3.
NATURAL CIRCULATION AND SHUTDOWN HEAT REMOVAL............
18 3.1 Natural Circulation Data Base 18 3.1.1 FFTF Natural Circulation Data..............
18 3.1.2 In-Vessel Flow Redistribution.............
20 3.1.3 Cladding Temperature Limits for Natural Circulation...
21
-iv-
TABLE OF CONTENTS (CONT.)
Page 3.1.4 Applicability of FFTF Data to CRBR 21 3.2 Thermal-Hydraulic Characterization at Natural Circulation Condi-tions 22 3.3 Shutdown Heat Removal System Issues 24 3.3.1 Introduction 24 3.3.2 Shutdown Heat Removal Systems and Events 24
.3.3.3 Issues Related to Individual SHRS Components 31 3.3.4 Issues Related to Total Integrated Plant Response During SHRS Events..,.....................
35 3.3.5 Approach to Resolution of Issues 39 REFERENCES 41 4
CONTAINMENT ANALYSIS 44 4.1 Large Sodium Releases in the S/G Building 44
- 4. 2 Containment System Design 48 4.2.1 Design Basis Accidents 48 4.2.2 Accommodation of a Meltdown...............
48 4.2.3 Sodium-Concrete Reaction Rates 49 4.2.4 Availability of Containment Passive Heat Sinks 49 4.2.5 Debris Bed Coolability 50 4.2.6 Hydrogen Generation and Auto Ignition..........
50 4.2.7 Mitigating Features...................
51
-s 4.2.8 Potential for Basemat Penetration............
51 e
4.2.9 Model Validation 51 4.3 Radiological Consequences 52 REFERENCES 54
-v-m.maammm sar
TABLE OF CONTENTS (CONT.)
Page 55 5.
RELIABILITY ANALYSIS 55 5.1 Risk Assessment 5.2 Reliability Analyses of the Shutdown Systems..........
56 56 5.2.1
Background
5.3 Shutdown Heat Removal System Reliability............
59 64 REFERENCES...............................
65 6.
LOSS OF HEAT SINK 67 REFERENCES................................
71 7.
REACTOR PHYSICS..........................
8.
CDA ENERGETICS: TRANSITION PHASE LICENSING ISSUES 79 8.i NRC Staff Position......................
79 79 8.2 CRBR Project Position 8.3 Major Changeh;During Past Four Years.............
80 8.4 Is sue s to be Reso l ved..................... 80 82 REFERENCES...............................
83 9.
STRUCTURAL ANALYSIS 85 10
SUMMARY
87 ACKNOWLEDGEMENTS............................
b D
-vi-
LIST OF FIGURES Fiaure Title Page 2.1 CRBRP Homogeneous and Heterogeneous Core Designs.
8 2.2 Reactivity Feedback Calculational Scheme.
12
~3.3.1 Plant Conditions after Trip Versus Loss of Offsite Power.
28 3.3.2 Shutdown Heat Removal System Schematic (without DHRS
~
and Guard Vessels.
29 3.3.3 Schematic of the DHRS.
30 3.3.4 DHRS Flow Operation.
33 6.1 Contour of k =1.0.
68 e
6.2 Channel Assignments for Thermal-Hydraulic Calculation.
68 6.3 Flow Fraction.
69 6.4 Sodium INmperature.
70 7.1 Homogeneous Core Layout.
77 7.2 Heterogeneous Core Layout.
78 e
,j l
e
-vii-
LIST OF TABLES t
Table Title Page 2.1 Issues on Operational Transients 9
3.1 Factors Potentially Affecting Initial Assessment Results for CRBRP 23 3.3.1 Grouping of SHRS Events 25 3.3.2 Shutdown Heat Removal Systems and Components 26 3.3.3 SHRS Issues - Component Level 32-3.3.4 SHRS Issues - Integrated System Level 37 4.1 Summary of Outstanding Technical Issues Related to CRBR Containment Analysis 45 4.2 Summary of the Project's Position on the NRC Staff Review 47 5.1 Point Estimate, Median, and 90% Probability Band of the Failure Probability per Year under Various Assumptions for the Shutdown System of the CRBR 58 5.2 Failure Probability, per year, of the SHRS Due to LOSP Shutdown Initiating Event 62 5.3 Summary of Results for Failure of Short-Term Forced Ci rculation 63 7.1 Doppler Coefficient for B0C1 73 7.2 Sodium Void Reactivity ($)
74 l
7.3 Fast Reactor Physics Codes Available at BNL 75 l
9.1 Issues in the Structural Analysis Area 84 hv f
l I
i
-viii-
1.
INTRODUCTION
1.1 Background
The formal licensing review of the Clinch River Breeder Reactor (CRBR) be-gan in 1975, when the Preliminary Safety Analysis Report was submitted as part of the application for a construction permit to the Nuclear Regulatory Commis-sion (NRC). This application differed from those presented for commercial light water reactors in several ways, which were a direct result of the rather
~
different design, i.e., sodium coolant, highly enriched fuel, breeding blanket, etc.
However, perhaps the most interesting aspect of this application was the emphasis placed on reliability analysis in the development of the safety-related design considerations for the plant.
During the licensing review, several technical issues received much atten-tion by the applicant and the NRC.
These include, but are not limited to:
a) the impact of core disruptive accidents on the reactor vessel and on the con-tainment building, b) the redundancy and diversity of the shutdown heat removal system, c) the capability of the plant to remove decay and sensible heat by natural circulation, and d) the selection of a design basis accident for con-tainment.
In the Spring of 1977, President Carter put forth an energy and non-pro-liferation policy which did not include plans for completing the CRBR.
Shortly thereafter, the Energy Research and Development Agency (ERDA) requested of NRC an indefinite postponement of the hearings associated with Limited Work Author-ization.
As a result, NRC terminated its formal review of the application.
Notwithstanding the foregoing, funding for the CRBR Project continued.
During the past four years, analyses were performed, the design was modified (particularly the reactor core), and major components were fabricated.
The position of the present administration is to complete the CRBR.
In g
order to resume the licensing review of the CRBR, NRC and its consultants must participate in at least two processes:
- 1) refamiliarize themselves with the safety issues that were associated with the licensing review at the time that it was suspended, and 2) assess the impact of any technological advances of new safety issues that emerged during the past four years. )
The purpose of this report is to initiate these processes.
The general focus of this report is on the systems and phenomena related to decay heat re-moval and severe accident analysis.
The draft version of this report was sub-mitted to NRC in November 1981.
There have been no major changes in the con-tent of the final version of this report.
This report was written for interim use only and it should not be regarded as an evaluation of the current design of the CRBR.
W=
0 4
e 1.2 The Role of Brookhaven National Laboratory in the Earlier Phase of the Licensing Review Brookhaven National Laboratory provided technical assistance to the reg-ulatory staff since the inception of the CRBR licensing review in 1975.
The program contained a diverse range of technical topics which included: core disruptive accident analysis (loss-of-flow, transient overpower, transition phase, fuel pin failure dynamics, and loss of heat sink with scram); plant transient behavior (core thermal response to pipe ruptures, pump seizures, and natural circulation events); post accident heat removal phenomena (in-vessel fuel debris relocation, ex-vessel debris cooling, sodium fires, hydrogen com-bustion, and containment thermal response); reactor physics analysis of dis-rupted core configurations. Many of the activities in these areas were also carried out during the period 1977-1979 in support of regulatory staff's safeiv review related to the operation of the Fast Flux Test Facility.
Over the same period of years BNL was active in the development of the SSC code and in simulation experiments related to disrupted core accidents.
In addi-tion, analyses were performed on LNFBR piping integrity and on sodium materials behavior.
1.3 Scope of this Report This report identifies and discusses a broad range of technical issues related to the licensing of the CRBR.
It does not include (except for a brief i
discussion of the transition phase) a discussion of issues related to un-protected transients such as the loss-of-flow and the transient overpower sce-l narios. A discussion of these transients are the main focus of a parallel ef-l fort being conducted by Los Alamos National Laboratory.
Several of the issues
(
identified in this report do, however, relate directly to unprotected tran-sients.
e p
Section 2 contains a discussion of operational transients that are gen-erally considered in Chapter 15 of a Preliminary Safety Analysis Report.
Sec-tion 3 addresses the issues associated with natural circulation heat removal and the overall capabilities of the CRBR shutdown heat removal system.
Section 4 focuses on containment thermal analysis associated with design basis sodium spills and with core disruptive accidents.
Reliability and risk analysis is i
I - -
i a
discussed in Section 5.
In Section 6, a discussion of the loss-of-heat-sink l
accident, with scram, is provided.
Section 7 contains a discussion of the
]
reactor physics issues related to the new heterogeneous core design.
Section 8 l
adresses issues related to the transition phase of the core disruptive accident.
A discussion of structural analysis issues is presented in Section 9.
- Finally, i
a summary of this report in connection with the relevent technical issues for l
CRBR licensing are presented in Section 10.
.l 4
I i
h i
i 4
D l
1 I l'
2.
OPERATIONAL TRANSIENTS For the purpose of this report, the term " operational transients" is in-tended to include all events that are likely to occur at least once during the operating lifetime of the the CRBR plant.
These transients are considered in design basis accidents, and no specific quantification of their likelihood of occurrence is given.
In all such incidences, credit is taken for proper opera-
~
tion of the Plant Protection System (PPS).
Thus we are not concerned, in this section, with the core disruptive accident scenarios.
The likelihood of occur-rence of an incidence due to failure of a component or a group of components is -
discussed in Section 5.
This section provides an assessment of the safety analysis contained in Chapter 15 of the CRBRP Preliminary Safety Analysis Re-port (PSAR).
2.1 Objectives and Issues The objectives of this section are:
(a) to assess the safety analysis as reported in Chapter 15 of the CRBRP PSAR, (b) to ascertain that all significant accident scenarios are con-
- sidered, (c) to provide a list of subtask areas where work is needed, and (d) to ascertain that the classification of events and the design criteria used are consistent.
There are many issues which need to be resolved. Most of the concerns noted in this section stem from the following:
(a)
Change in Core Design - The current design of CRBRP reactor core uses heterogeneous core, as opposed to the earlier design involving homogeneous core (Refs. 2.1,2.2).
This change causes substantial variation in key reactivity feedback parameters.
Computations of these parameters are more complex and subject to larger variations / uncertainties than for their counterparts in homo-geneous core design.
(b)
Control Rod Worths - The number of control rods used in the current design is reduced from a total of 19 to 15.
The computation of the worth and margin of a control rod for the heterogeneous core is also more complex because such a computation must be done using space-time coupled method. __ _....
(c)
Thermal-Hydraulic Analysis - Most of the PSAR analyses for opera-tional transients are done by using either the DEMO-Rev4 (Ref. 2.3) or FORE-2fi (Ref. 2.4) computer codes, or both.
The DEMO-Rev4 code has been known to have several major deficiencies (Ref. 2.5).
Perhaps the most significant one, as it relates to the analysis of undercooling transients, is the usage of pure trans-port time-delay model in the piping (Ref. 2.6).
The CRBR Project has been aware of many of these shortcomings in the DEM0-Rev4 code -- they intend to use DEM0-Rev5 (Ref. 2.7).
The FORE-2M code (Ref. 2.4) is used for most reactivity insertion events.
A verification of this code is needed.
It is not clear, however, as to what extent new analyses will be performed.
In any case, there is a clear need to audit all thermal-hydraulic calculations.
(d) Multiple Fault Events - The CRBRP Chapter 15 analyses are based on the consideration of a single failure at a time.
An evaluation is then made to ascertain the consequences of each single failure.
In view of the THI-2 inci-dence, additional effort is needed to examine the consequences of secondary failures or complicatirg features so that the significance of multiple failures could be addressed.
(e) PPS/PCS Design - The change in the reactor core design, as noted
'above, can have significant impact on the Plant Protection System (PPS) and the Plant Control System (PCS) designs and their settings.
It is not clear whether the impact of the new core has been assessed on PPS and PCS.
Furthermore, any interdependency of these two systems needs to be assessed.
(f) Rod Design Criteria - The CRBRP rod (fuel and blanket) design crite-ria are based on evaluation of the Cumulative Damage Function (CDF), which is at variance with the temperature limits used in LWRs.
There is a lack of suf-ficient data base.
It should also be noted that even for an anticipated event of control assembly withdrawal at power, a significant amount of fuel melting in a hot fuel channel is noted in Section 15.2.1.2 of the CRBRP PSAR.
Since 3
the peak linear power in the hot blanket channel is more than 207 higher than the hot fuel channel, a significant amount of blanket fuel may also be molten for this anticipated transient.
The acceptability of molten fuel either in a fuel or blanket rod must be assessed.
The CDF criteria need close scrutiny.
(g)
Shutdown Heat Removal - The long term consequences of many of the op-erational transients place a reliance on the shutdown heat removal system.
The issues connected with the SHRs are discussed in Section 3. - - -
A brief list of major issues is given in Table 2.1.
Short discussions are provided in subsequent subsections.
2.2 Reactor Physics Analysis An essential element of the CRBRP performance evaluation is a verified data base for key reactor physics parameters such as the reactivity feedback coefficients, power coefficient, the control rod worths and the available con-trol rod margins.
Because of the sensitivity of these parameters to burnup of fissile fuel and breeding of fertile fuel, this data base is needed for the entire fuel cycle from the beginning-of-cycle (B0C), to the equilibrium cycle,
~
to the end-of-cycle (EOC).
The need for this data base has increased because of the change from the homogeneous core to the heterogeneous core design in the CRBRP. Another complicating factor is that the criticality calculations for the heterogeneous core are more complex (and subject to larger variations be-tween the computed and measured ratio) than those for the homogeneous core.
The reactor physics data base is also needed for thermal-hydraulic perfor-mance evaluation for essentially intact reactor core as well as damaged or de-graded core conditions.
In this section, required parameters for the intact core conditions are discussed.
Reactor physics analysis for the damaged core conditions is discussed in Sections 6 and 7.
2.2.1 Control Rod Worth The CRBRP heterogeneous core differs from the previous homogeneous core in many ways.
Figure 2.1 is a schenatic of these two layouts (Refs. 2.1,2.2).
The heterogeneous core uses 32.87. onriched fuel (Pu/(U+Pu)), as opposed to 17.4 and 25.17. enrichment in the homogeneous core.
The fissile inventory is also significantly larger (10 to 257.). At the same time, the CRBR Project stated that the control rod worth requirement is considerably lower for the hetero-geneous core ($17.44 primary for BOC4 as opposed to $26.53 primary for the equilibrium cycle of the homogeneous core).
The computation of the control rod worth and the available margin is a complex problem, as it requires two-space dimensions coupled with time kinetics computer codes (see Section 7). An inde-pendent assessment is needed to assure that the total available worth with one worst rod stuck will exceed, at all times, the required control rod reactivity.
4 CRDRP HETEROGENEOUS CRBRl9 HOMOGENEOUS CORE DESIGN m
f CORE DESIGN l
2
~
.~
9
%. ', y' e.;
s g
i o
m x
n y
l-a
(
,h"v_,
I"
),
"4
,,,i 04 h
,,\\
g l'
(
D H
n
, \\. <y
~
4,. /"'
,, j i
g'
~
t,<
F g-qy j'
6" q
\\
C g
g f
h' 1/'
h'
)
~A I
( y
" b<
, f
's
,y m
1' N
p l j }$;M}H N
H i
)
X 4
p n-H H..
)
y
_e Q
s
.,i x
,fi
~
o,,
t i
L
-q
/
)'
t" j
H
. s L'J g
/
v). : -
I. L k
,(."
ih o
8
'L m
. g..
f.
m
+
' i 6
l 5
Q H,. 4 g G,
j I
n
,y
.f ' l.
. f " n.,
s m
..s rn 9
5 g
CYCLES ODD EVEN
~
,m
~
i 158 162 h FUEL ASSEMBLIES 108 INf1ER CORE al.
=
h FUEL ASSEMBLIES SO OUTER COllE O auNxEr ASSEMaty 300 isa RADIAL SHIELD ASSEMBLIES O n'anxeT asSrMat'Es PRIMARY CONTROL ASSEMBLIES S
324 Q RADIALSillELD ASSEMBLIES 15 Q PRIMARY CONTROL ASSEMDLIES
@ ALTERNATE F,UEL18LANKET 5
l 4
h SECONDARY CONTROL ASSEMBLIES s
ria la F j
4
- (
TABLE 2.1 Issues on Operational Transients l
i Area Issues and Comments i
Re' actor Physics Criticality calculations Feedback reactivity, including Doppler and sodium temperature coefficients Control rod worth - margin and requirements Reactivity effect due to bowing of fuel and blanket assemblies Thermal-Hydraulic Undercooling events Analysis Reactivity insertion events Local fault events Combination of different faults l
PPS/PCS Systems Design specification of PPS/PCS systems Interaction between PPS and PCS systems Sensitivity of settings due to variation in core physics parameters Rod Design Criteria Cumulative damage function vs. temperature limits Data base Comparison with LWR design criteria Steam Generator Sodium-water reaction Integrity of IHTS and IHX
_g_
2.2.2 Reactivity Feedback Coefficients In going from a " cold" (just critical) to " hot" (poser operation) condi-tion the temperature of the core rises, which causes changes in the atom densi-ties as well as microscopic reaction cross-sections.
The effects of these changes are expressed in terms of five temperature coefficients of reactivity:
a) the Doppler coefficient, b) the sodium temperature coefficient, c) the fuel expansion temperature coefficient, d) the fuel-element bowing coefficient, and e) the power coefficient. All of these coefficients should be verified, par-ticularly the Doppler coefficient which acts very promptly.
The CRBR Project has computed this value to be -0.0084 This value for other LMFBRs (with homo-geneous core) ranges from -0.0032 to -0.0060.
A careful assessment of this prompt negative coefficient of reactivity is essential, as it also impacts on the Plant Protection System (PPS) and the Plant Control System (PCS).
The Dop-pler coefficient is the key feedback reactivity contributor in determining the course of most operational transients.
The reactivity effect associated with the fuel rod bowing can be either positive or negative depending upon the assembly support structure and the ra-dial temperature gradient.
This effect was first observed in the EBR-I reac-tor.
Interestingly enough, the accompanying positive reactivity was at first associated with the Doppler effect (which can be positive for highly enriched fuel), but later was confirmed to be due to inward thermal bowing of the fuel assemblies (Ref. 2.6).
In the CRBRP the assemblies are held at the top and bottom of the core.
Depending upon the radial temperature gradient, fuel as-semblies can bow either toward the center of the core, resulting in a positive i
reactivity contribution, or away from the core, thus causing a negative reac-tivity contribution. This effect can result in a substantial net positive re-activity.
It should be evaluated for different power-to-flow conditions and also for different interassembly gap sizes (nominal gaps to reduced gaps). For CRBRP, the project-computed bowing reactivity contribution for the homogeneous core was as high as +654 For the heterogeneous core, this value should be I
smaller.
The reactivity coefficients associated with sodium temperature and with fuel expansions also need to be computed.
l l
Once the reactor attains its operating power and flow, subsequent tran-sient analyses require essentially similar reactivity feedback coefficients.
A l..
generalized representation of the reactivity feedback on reactor power is shown (Ref. 2.6) in Figure 2.2.
Any deviation in reactor power causes deviation in the fuel and blanket temperatures, and the coolant temperature.
The reactivity effects associated with the change in fuel temperature are predominantly the Doppler and, to some extent, fuel thermal expansion. A deviation in the sodium temperature gives rise to a reactivity contribution due to the resulting change in sodium density. Furthermore, a change in the coolant temperature also re-sults in a change in the duct wall temperature, thereby resulting in an addi-tional reactivity contribution from bowing of fuel and blanket assemblies.
The sum of all these individual contributions with the applied reactivity, such as ~
control rod movement, is the total reactivity.
The neutron kinetics equations, usually the point-kinetics equations, may now be solved to get a new value for the reactor power.
2.2.3 Reactivity Feedback Due to Core Compaction A sudden core compaction, due to seismic loading, will result in a posi-tive reactivity. The magnitude of this reactivity change has been estimated, by the CRBR Applicant, to be a 30( step for the Operating Basis Earthquake (0BE) and a 604 step for the Safe Shutdown Earthquake (SSE).
The PSAR formerly considered, in addition, a 90t step reactivity insertion.
The magnitude of the reactivity insertions, due to core compaction for the OBE and the SSE must be verified for the new heterogeneous core.
2.3 Thermal-Hydraulic Analysis There are a large number of transients for which thermal and hydraulic performance evaluations must be made.
These calculations should be made by us-ing first the primary scram function, and then the secondary scram function to allow for situations where the primary scram signal failed to scram the re-actor.
For example, in the event of a reactivity insertion of, say, 24/sec, the primary scram function is the Flux-Pressure and, if this function is as-sumed to be inoperative, the secondary scram function is the Flux-Total Flow.
The response of the plant for this event (2d ramp) must be evaluated for both the primary and secondary scram functions.
For the purpose of thermal-hydraulic analysis, various transients can be grouped into three categories:
undercooling, reactivity insertion, and local fault events. A partial list of the undercooling events is:
i -
m-.
e 8 ]
NEUTRON DEVIATION IN
- e
_ I KINETICS POWER
]
DOPPLER EFFECT DEVIATION IN pg FUELS BLANKET
- FUEL EXPAN$iCN EFFECT i
I D EVI ATIO N IN SC0lVM DENS 6TY 5001UM e-EFFECT T E M PE R ATU R E i
AssEusty ec v) ucr EFFECT
[
T E MPER ATURE t
Figure 2.2 Reactivity Feedback Calculational Scheme J.. _..
Loss of electrical power Primary pump trip Primary pump seizure Coolant leaks Intermediate pump trip Intermediate pump seizure Closure of the evaporator Closure of the superheater Loss of main feedwater Loss of auxiliary feedwater Steam line break Failure of the steam bypass system Intermediate heat transport system leak Check valve failure Sodium-water reaction Pipe breaks Steam-tube failure Loss of heat sink Loss of normal shutdown heat removal A partial list of the reactivity insertion events is:
Control assembly withdrawal during startup Control assembly withdrawal at power Seismic reactivity insertion - step reactivity Gas bubble passage through the core Movement of fuel and blanket assemblies Cold sodium insertion Control assembly insertion Mal-operation of PCS Some of the important local fault events are:
Blockage at the assembly inlet Blockage within the assembly Misloading of assembly in wrong inlet module Interchanging of fuel and blanket assemblies Side-way compaction of fuel assemblies Propagation of pin-to-pin failure Propagation of assembly-to-assembly failure The approach to be used here will include single faults as well as possi-ble resulting faults.
For example, if it turns out that misloading of a fuel assembly (wrong flow inlet module or interchanging with a blanket assembly) goes unnoticed (which is possible), then the consequence of a reactivity inser-tion for this misloaded assembly must also be investigated. - --
9 2.4 PPS/PCS Systems and Their Interaction The Plant Control System (PCS) is designed to take corrective measures in order to avoid spurious scrams or shutdowns.
On the other hand, the Plant Pro-tection System (PPS) is designed to initiate reactor scrams if the perturbation exceeds preset limits.
While in many cases the two functions can be complemen-tary, there may be situations where the action by one could counteract the other.
For example, the FFTF design initially contained a flow dependent limit within the automatic flux controller power set point circuitry such that the automatic flux controller would reduce power in response to a reduction of pri,
mary loop flow. This control feature was included so that a constant flux / flow ratio could be maintained automatically during normal operations.
However, if a rapid loss of primary flow event were to occur while operating under auto-matic flux control, the resultant power decrease delayed the time to trip ini-2 tiation for both the flux /oressure and flux / flow protective functions such that hot channel cladding temperatures exceeded allowable limits.
On the basis of these studies, this control feature was deleted from the FFTF design. While this specific feature is not in the CRBRP design, the interaction of the PPS and PCS systems must be studied.
The settings for various PPS functions are based on the neutronic and thermal-hydraulic performance of the reactor core. Any uncertainty or revi-sion in key neutronic or thermal-hydraulic parameters can have significant impact on the acceptability of PPS settings.
For example, if the Doppler coef-ficient of reactivity were to be one-half of the value used in developing PPS settings, a significant amount of fuel could conceivably melt.
It is also pos-sible that, in this case, a different PPS function will initiate scram.
In other words, the influence of the new core design, with associated feedback coefficients on the system response, must be evaluated in order to verify the efficacy of the PPS.
2.5 Rod Desian Criteria The CRBRP fuel management scheme involves batch replacement of the fuel and blanket assemblies (Ref. 2.2).
All fuel and inner blanket assemblies are replaced at two year intervals.
The first row of radial blankets is replaced after four years of operation, and the second row of radial blankets is re-placed after five years of operation.
The safe operation of these rods over l
i their stated residence time is being quantified in terms of the Cumulative Dam-age Function (CDF). For emergency events, design criteria calls for the CDF to be less than 1.0 and the accumulated plastic and thermal creep strain to be less than 0.2%.
For the faulted events, the criterion used is that the clad-ding temperature not exceed 2475'F (its melting temperature), and this is con-
!o servatively translated into a limit to prevent boiling.
The CRBRP PSAR reports that even for an anticipated transient, like the control rod withdrawal at power, a substantial amount of fuel in the hot fuel channel can be molten.
The same should be true for the hot blanket channel (which has roughly 20% more power than the hot fuel channel). The acceptabil-ity of molten fuel, even for an anticipated transient, therefore, must be ex-ami ned.
It is conceivable that the molten fuel could ooze through the pellet gaps and contact the cladding tube, thereby resulting in rod failures.
The adequacy of the rod design criteria for such an anticipated transient must be assessed.
j The CRBRP design criteria appear to have been developed in lieu of a tem-perature limit for emergency events.
This is in variance to the approach used in the light water reactor (LWR) plants. An extensive review and assessment of the CDF criteria, therefore, must be undertaken.
Another major comment here is related to the implementation of the design criteria.
If the maximum cladding temperature were to be used as a design criterion, implementation can be more direct than the CDF criterion.
One would still have to rely on analytical ca-pabilities to substantiate the temperature limits.
It appears, however, that the CDF criteria require additional reliance on analysis.
2.6 Sodium-Water Reaction l
The materials performan.ce in LMFBRs has been discussed and reviewed in a special issue of Nuclear Technology (Ref. 2.8).
It is seen that the steam gen-erators have had a history of tube failures because of a number of factors such j
as weldment problems, stress corrosion cracking, water chemistry, temperature transients, etc.
In LMFBRs (including CRBR) such a failure results in a chemi-cal reaction between sodium and water.
Depending upon the number and extent of l
tube failures (and hence the amount of water available for reaction with sodi-um), the resulting energy release may cause further propagation of tube fail-ures.
On the other hand, the consequences may be controlled if detected early. _ _
1 Instrumentation is provided to arrest the consequence of a tube failure in the steam generator and the reaction products are removed by a cleanup system.
On the other hand, it is quite possible not to detect small leaks until they prop-agate to involve many tubes.
An example is the BR-5 steam generator explosion in 1973 in the USSR.
The problem here has several aspects:
a) to determine conservatively the extent of tube failures before corrective action may be taken, b) to compute the hydrogen generation rate and the pressure source term, c) to perform hydro-dynamic analysis to ascertain the integrity of the intermediate piping system, and finally, d) to ascertain the integrity of the intermediate heat exchanger.
2.7 Pipe Breaks Even though the NRC staff has accepted the Project's po'sition of leak before break in the primary cold leg, certain design problems remain to be resolved; the main one being the adequacy of the leak detection system.
Addi-tionally, the question of leak before break in the primary hot leg is still an open issue.
s '
v- - _ _ _
REFERENCES 2.1 Clinch River Breeder Reactor Plant - Prelininary Safety Analysis Report, Project Management Corporation (1975), including Amendment 64, dated January 2982.
2.2 P. W. Dickson, "Briefino on Heterogeneous Core Characteristics", Bethesda, Maryland, October 15, 1981.
2.3 W. H. Alliston et al., " Clinch River Breeder Reactor Plant; LMFBR Demo Plant Sinulation Model (DEM0), Topical Report", CRBRP-ARD-0005, February 1978 (available from US/00E Technical Information Center).
2.4 J. V. Miller and R. D. Cof field, " FORE-2M: A Modified Version of the FORE-II Computer Program for the Analysis of LMFBR Transients",
CRBRP-ARD-0142, November 1976 (available from US/00E Technical Information Center).
2.5 R. Pyare and J. G. Guopy, " Transient Analysis of LMFBR Plant with SSC-L and Comparison with DEM0 Code", Brookhaven National Laboratory Report, BNL-NUREG-29313 (1981).
2.6 A. K. Agrawal and M. Khatih-Rahhar, "Dynsmic Sinulation of LMFBR Systems",
Atomic Energy Review, 18, 329 (1980).
2.7 W. J. Severson, " Briefing on CRBRP Shutdown Heat Removal - Performance Evaluation", Bethesda, Maryland, November 2, 1981.
2.8 Materials Performance in Nuclear Steam Generators, Proceedings published in Nucl. Technology, Sj! (1981).
e 3.
NATURAL CIRCULATION AND SHUTDOWN HEAT REMOVAL During the earlier phase of the CRBR licensing review, it was a regula-tory staff position that no credit would be given for shutdown (decay and sen-sible) heat removal by natural circulation through the heat transport trains of the plant.
This position was based on the lack of a data base to support the potential natural circulation capability of the CRBR design.
Additionally, there are numerous concerns regarding the adequate simulation / prediction of the thermal / hydraulic characterization of the system operating under natural circulation conditions.
In this section, the issue of natural circulation capability is discussed in light of information that has become available during the past four years.
Subsequently, a discussion of the issues related to the performance capability of the shutdown heat removal system components for various operating regimes is provided.
3.1 Natural Circulation Data Base To adequately portray system operation under natural circulation condi-tions, there are many inter-related factors which must be represented.
The overall driving force for the natural circulation flow is a balance between competing pressure losses and gains throughout the reactor vessel and primary system. The net driving head at these conditions is only a few kPa (1 psi =
6.9 kPa).
Some test results (most notably from FFTF) are now obtainable, or will be shortly.
These must be assessed closely to determine their true applicability in leading to a reduction of uncertainties in the analytical models and codes.
Most of the LMFBR operating experience (Ref. 3.1) has been in pool-type reactors such as PFR and PHENIX, or in very small experimental loops such as EBR-I, EBR-II, SEFOR, KNK-2 and J0Y0.
The only large scale loop-type reactors outside of the USSR are Fermi-1, FFTF, SNR-300 and MONJU.
Natural circulation tests were not performed in Fermi before the project was canceled, but consider-able data were obtained for operational transients.
Neither SNR-300 nor MONJU have been completed, so the large scale loop data base is limited to FFTF.
3.1.1 FFTF Natural Circulation Data The FFTF design relies on passive natural circulation in the heat trans-port system as a backup to forced circulation as a means of decay heat removal. --
As such, the project staff included an extensive array of natural circulation tests (Ref. 3.2) as part of the test program.
In the Safety Evaluation Report (Ref. 3.3), the NRC recommends that only limited operation be approved until natural circulation capability has been verified, including verification of the IANUS and FLODISC codes used for Chapter 15 analyses in the FSAR.
The natural circulation test program has been completed, but the test data have not been formally released and do not appear to have been analyzed extensively.
How-ever, some conclusions can be made based on the non-nuclear natural circulation tests and preliminary results (Ref. 3.5) of the nuclear tests.
Specifically, (1) Heat exchanger performance (both the Intermediate Heat Exchanger (IHX) and the Dump Heat Exchanger (DHX)) was not well predicted at natural cir.-
culation conditions.
In the case of the DHX this appears to be due to poor modeling (modeling the DHX as a counter-flow heat exchanger 'ather than a r
cross-flow heat exchanger).
But in the case of the IHX, no such phenomenologi-cal basis appears to be present and the project has reduced the heat transfer coefficient in the IHX by a factor of 10 to achieve agreement with the data.
The lack of heat exchanger performance data at natural circulation flows was identified by Perkins and Bari (Ref. 3.6) as a problem area, and it apparently remains so.
(2) Pump performance data have been obtained for the FFTF pumps but coastdown times and stopped rotor losses are very design-dependent.
Low flow frictional torque and pressure loss data from FFTF could provide a useful com-parison to CRBR pump description data, but better data should be available from the CRBR component verification program.
Informal discussions with CRBR proj-ect staff indicate that water tests have been performed for a CRBR prototype pump, bt t the data have not yet been submitted.
(3) Upper plenum mixing data in FFTF are very limited.
The FFTF project staff interprets (Ref. 3.5) the data as supporting a two-zone mixing model (as i
a function of momentum and buoyant head) for pony motor flow rates, but they acknowledge severe stratification at natural circulation flows.
Even at full flow rates the upper plenum mixing appears to be incomplete and a new phenome-non of bleed flow is identified (Ref. 3.5) to account for discrepancies between the upper plenum temperature and the hot leg temperature.
Previous CRBR natu-ral circulation calculations (Refs. 3.7,3.8) assumed stratification of the upper plenum (thereby reducing the buoyant head term), but the present proj-ect's verification program (Ref. 3.9) uses a multi-region upper plenum model which is difficult to relate to the previous assumptions.
(4) Possible stratification in the piping at natural circulation flows was identified as a problem area for both CRBR (Ref. 3.7) and FFTF (Ref. 3.10).
From a licensing standpoint, stratification is important in that the effect may be detrimental to natural circulation and is not modeled in the one-dimensional codes (Refs. 3.11,3.12) used in licensing analyses.
However, stratification effects can be accounted for even in a one-dimensional code (Ref. 3.13) if the _
magnitude of the effect is known a priori.
While there is no direct measure-I ment of temperature distributions in the FFTF piping, the FFTF project now ac-knowledges (Ref. 3.5) that the limited data indicate that piping stratifica-tion occurred in the primary hot leg during the natural circulation tests.
(5) Check-valve pressure drop is also expected (Ref. 3.14) to be an im-portant contributor to natural circulation performance.
FFTF has character-ized their check valves in the component testing program but, as with the pumps, the pressure loss characteristics are very design-dependent.
The CRBR project staff has reduced their estimate of pressure losses at the check valve i
l (Ref. 3.9), but they have not yet submitted supporting data.
3.1.2 In-Vessel Flow Redistribution l
1 Most of the experimental data and analytical calculations have concen-trated on determining flow and temperature patterns at full flow and full power conditions.
However, recognition of uncertainties in local flow rates has gen-erated considerable interest in analytical (Refs. 3.15, 3.16, 3.17, 3.18, 3.19) and experimental (Refs. 3.2,3.20,3.21) investigations of flow redistribution at natural circulation conditions.
The original CRBR project calculations (Ref. 3.8) for natural circulation assumed no flow redistribution within the core, but BNL pointed out (Ref. 3.7) that this procedure underestimated the av-erage blanket temperature difference by about 25"., and boiling would be pre-I dicted in the hot channel for the natural circulation event (using the hot I
channel factors developed by the project).
Both the FFTF and the CRBR projects have relied heavily on the FFTF natural circulation tests to verify their flow redistribution models.
However, instrument tree bypass flows make the FFTF data difficult to interpret. Only the Fuels Open Test Assembly (F0TA) data _
l appear (Ref. 3.5) to be accurate at natural circulation flow rates.
As pre-viot. sly indicated (Ref. 3.4), the relative difference between F0TAs is not a sensitive measure of flow redistribution.
In-vessel flow redistribution, including recirculation, was shown (Ref.
3.22) to be an important mitigating effect for the loss-of-heat sink scenarios (see Section 6).
However, subsequent analyses indicate that recirculation would not be sufficient to delay boiling for FFTF (Ref. 3.23) and that inter-mediate powered assemblies could be expected to stagnate and boil early in the loss-of-heat-sink transient for CRBR (Ref. 3.24).
3.1.3 Cladding Temperature Limits for Natural Circulation There has been considerable interest in whether sodium boiling is a via--
ble means of decay heat removal.
However, both analyses (Refs. 3.25, 3.26) and l
experiments (Refs. 3.27,3.28) tend to be somewhat contradictory.
While the debate can be expected to continue, the CRBR project has taken the position that they will use a temperature limit for design events which, conservatively, precludes boiling.
In this regard, it should be noted that the sodium boiling point at atmospheric pressure is 1620 F.
The boiling point can, of course, be higher in the core when the gravitational head is considered.
Neglecting the pressure drop due to forced flow (assuming low-flow natural circulation condi-tions), the gravitational head of sodium above the core exit region brings the boiling point up to 1700 F.
3.1.4 Applicability of FFTF Data to CRBR While the FFTF natural circulation test program provides considerable sup-port for the viability of natural circulation in FFTF, the degree of conserva-tism in the licensing calculations (Ref. 3.3) has yet to be assessed.
At the present time, apparent inconsistencies (Ref. 3.5) in the FFTF data make inter-pretation difficult, but once the data are released for detailed evaluation, some of the inconsistencies may be resolved.
The released data should also help to establish the validity of the plant transient codes such as SSC (Ref.
3.30) and DEMO (Ref. 3.11).
In any case, the importance of component specific data appears to make it necessary to reevaluate CRBR natural circulation per-formance in the light of new CRBR component test data and FFTF loop performance data. _...
3.2 Thermal-Hydraulic Characterization at Natural Circulation Conditions There are numerous concerns regarding the adequate simulation / prediction of the thermal / hydraulic characterization of reactor system operation under natural circulation conditions. The applicant has an intensive on-going effort in this area aimed at the demonstration of natural circulation capability in the CRBRP.
The CRBR Natural Circulation Verification Program (to be submitted) is intended to validate DEM0-REVS, COBRA-WC and F0%2A.
However, none of these codes were used in the natural circulation analysis (Ref. 3.8) and DEM0-REV5 is not yet available.
In the previous review (Ref. 3.7), the blanket thermal performance was found to be a dominant concern and could approach boiling even with one set of pony motors on.
The previous analysis appears to have little relevance to the present verification effort due to a number of important changes in system characteristics. Most notably, the impact of the new heterogeneous core configuration on natural circulation capability was not evaluated in the previous review.
The influence of this new design must be carefully considered on a consistent basis.
The CRBR Project's briefing on decay heat removal (Ref. 3.9) indicated (see Table 3.1) the major changes in their analysis which affect core thermal performance, but they provided little basis to assess the importance of the changes.
The summary only raised additional concerns.
In particular, (1) The pump stop time and stopped rotor loss estimates are important changes from the previous submission and are based on prototype data not yet submi tted.
(2) The estimated check-valve pressure drop has also been changed sig-nificantly and the supporting data should be submitted.
(3) The effects of core flow and heat redistribution to be included in the CRBR project analysis are expected (Ref. 3.7) to raise estimates of blanket temperatures rather than lower them, as implied by the project.
(4) The importance of a new upper plenum model is difficult to assess without having the model available.
(5) The decrease in estimated decay power is said to be due to a decrease in uncertainty estimates, but such a decrease in uncertainty appears to con-flict with present NRC requirements (Ref. 3.31). -
e Table 3.1 FACTORS POTENTIALLY AFFECTING INITIAL ASSESSMENT RESULTS FOR CRBRP 1976 Revised Effect On Predictions Predictions Peak Temperatures
- 1. Pump stop time 56 sec
> 110 sec.
Lower
- 2. Core flow and heat Neglected To he included Lower redistribution (fixed flow fractions)
- 3. Reactor upper Neglected Ul3 r.1odel included Raise '
internals
- 4. CKV AP
.07 psi @
.05 psi O Lower 3% flow 3% flow w
- 5. Pump stopped rotor
.08 psi @
.11 psi @
Raise AP 3 % llo w 3 % llo w (Based on water data)
- 6. Reactor AP
.145 psi O
.178 psi 0 Raise 3% flow 3% flow
- 7. Piping / plena heat Neglected included Will smooth temps, capacity effects small effect on core temps
- 8. Total decay power 4.15% @ 100 sec.
3.65 0 100 sec.
Lower 3.4 % @ 300 sec.
2.9 @ 300 sec. 6054 304 e
1 i
s The new heterogeneous core, along~with over-cooled inner blankets at be-ginning of cycle conditions, would appear-to make thermal-striping (and associ-ated thermal stresses) a potential' problem for the new core, particularly at be gi nni ng-o f-li fe.
3.3 Shutdown Heat Removal System Issues 3.3.1 Introduction The general issue to be addressed in this section is the capability of the various shutdown heat removal systems (SHRS) proposed for the current CRBRP design to remove the decay and sensible heat for all postulated operating con- ~
ditions.
In the aftefbath of Three Mile Island, decay. heat removal systems are receiving increased attention to ensure that their capabilities are properly verified under all potential modes of operation.
The breakdown of this general issue into individual issues is approached by:
- 1) addressing the issues re-lated to the individual components of the various SHRS, and 2) addressing the issues relating to the operation of the total integrated system.
Additionally, general issues relating to the appropriate design limits to be applied to the various events, as well as the identification of the significant events, are discussed.
Some of these issues are probabilistic in nature, while others are mechan-istic. The intent here is to address in more detail those issues of a thernal/
hydraulic nature, while just highlighting those in the probabilistic or relia-bility areas, as they are discussed more fully in Section 5.
In this section, the categorization of potential shutdown heat removal events is first discussed, along with a brief description of the various SHRS, to indicate the scope and complexity of these systems.
Next, specific issues identified at this time regarding each of the systems and their integrated operations are presented.
Finally, an approach to the resolution of the issues and questions is proposed.
3.3.2 Shutdown Heat Removal Systems and Events The discussion of events including operation of the various shutdown heat removal systems focuses on the categories of scenarios, as summarized in Table 3.3.1.
A list of the various SHRS and subsystems is given in Table 3.3.2.
-?4-
/
Table 3.3.1 Grouping of SHRS Events Group' Assumed Power SHRS Components Sources Available Operating (1)
Normal Shutdown All off-site and Main heat transport loops, on-site turbine bypass and main feedwater (2)
Upset On-site only PHTS, IHTS, evaporators under natural circulation, auxiliary feedwater and PACC, steam drums (3)
Eme rgency Batteries Only Loop Three PHTS and IHTS only Loop 3 PACC and turbine-driven auxiliary feedwater pump (4)
Faulted At least diesels One or more PHTS. loops and and batteries DHRS f
l
\\
Table 3.3.2 Shutdown Heaf' Removal Systems and Components System Components Main Heat Primary Heat Transport System (PHTS)
Transport System including IHX and pony motor Intermediate Heat Transport System (IHTS), including pony motor Steam Generator System Evaporator Module Superheater Module Recirculation Pump Main Feedwater System Protected Air Cooled Condensor (PACC)
Turbine Bypass s
Main Condensor Steam Generator Auxiliary Feedwater System (AFWS)
Auxiliary Heat One turbine-driven pump Removal System Two motor-driven pumps (SGAHRS)
Protected water storage tank (PWST)
Protected Air Cooled Condensors (PACC)
Superheater Isolation Values Evaporator and Superheater Vent Valves Di rect Heat Removal Vessel Overflow Tank Service (DHRS)
Two sodium electromagnetic pumps Overflow heat exchanger (Na/NaK)
Two NaK pumps Two cirblast heat exchangers (ABHX)
(NaK/ Air) l
(,
I (1) For normal plant shutdown (i.e., assuming no loss of off-site or on-site power), the CRBRP design provides for the removal of decay and sensible heat via forced convection through all main primary heat transport systems (PHTS) and steam generator systems (SGS), see Figure 3.3.la.
All PHTS and IHTS pony motors are operating; the main feedwater pumps and recirculation pumps are operating, and the residual heat is ultimately removed via the main condensor.
Thus, no reliance is placed on natural circulation.
However, the actions of a complex control system scheme and the subsequent manipulations of pumps and valves are required to assure the smooth operation of this procedure.
(2) For loss of off-site power events (assuming availability of diesel generators), heat is still removed via forced convection utilizing pony motors in the PHTS and IHTS.
However, in the SGS, power is not available to the main feedwater and recirculation pumps. Thus, the water side of all evaporators and superheaters must rely on natural circulation.
Additionally, operation of the two main subsystems of the steam generator auxiliary heat removal system (SGAHRS), specifically the auxiliary feedwater system (AFWS) and the protected air-cooled condensor (PACC), are automatically initiated (see Figures 3.3.lb and 3.3.2).
The SGAHRS requires the corrplex automatic operation of various control systems, pumps and valves, as well as a water supply from the protected water storage tank (PWST) or main condensate storage tank to assure its mission success. The PACC, which provide the ultimate long term coolability, rely solely on natural circulation on the water side for heat removal.
(3) Under loss of all off-site and on-site power events, heat vill be-removed by natural circulation within the primary and secondary sonun loops.
The SGAHRS will use the steam dump valves to remove heat until the PACC can accept the load with natural circulation of air.
Auxiliary feedwater is provided from the PWST by the turbine-driven pump only, which is driven by steam from the steam drum (s).
(4) For shutdown heat removal events in which loss-of-heat-sink (LOHS) through the SGS and/or IHTS is assumed, decay heat is to be removed through the direct heat removal service (DHRS) provided via the reactor vessel sodium over-flow system (see Figure 3.3.3).
The operation of this system requires the manual realignment of six valves and the startup of various pumps. Addition-ally, for the present CRBRP design, the DHRS a) cannot operate in a natural circulation mode and thus must have at least the diesel generators availabl ',
Figure 3.3.1 PLANT CONDITIONS AFTER TRIP VERSUS LOSS OF OFFSITE POWER Bypass to
-J L
Condenser g
- h P
(a)
Trip j
V n
~10%
PACC
~10%
tp Main 5
~100%
y To SGAHRS 1_'
_j L
Vents (b)
Loss of dIN Offsite Power y
F Q
G4 Ydg,
PACC
^ "
- H ' ^ 'Y
~3%
_I rw Natural Circillation 0054 256
Figure 3.3.2 SHUTDOWN HEAT REMOVAL SYSTEM SCHEMATIC i
CWITHOUT DHRS AND GUARD VESSELSD FROM OTHER D "D
CONTAINMENT g Au
,Rv ik di NE GENERATOR c::9'Q
=
7
'/'
V AL VING SUPERHf ATER
[ DESUPERHE ATER h WN(T N"A f d g
HO fwlit l
C u
PRIMARY l
iP rN Qq di SODIUM PUMP INT E R ME DI AT E
""";, "' ""' " '?i""'5
~
e:,
T;,e; R
~
~
R
""l"
_J,_
Qf ATM MAIN f f ED PUMPS T
go
?
""'-* L - s p
m M
$RY d
PUMPS r,
m m
AFW TURBINE PROF f CTf D WATER RfACTOR l
s, i _'_ _.r -g-----, TM. A,7, --,
DRUM j
svoaAce rANR l
~
s r
' aoM
-oRAmA-s
.VA, ORATOR.
ij t' ! a WA,iR l
AL, a y,- *-q) 4 a; I
~
r-a j
p
,R,c,Rc,UM,
, t,or,,yt - - + -- ---- -,;
a
'a c
l;f;" RUM, 4 i
ti i
e i
o.s.,
m l
O **
6 FROMothtR r,
s ruur c g]
toors t - -- - - - - - a i
V TYPICAL OF 3 100% REDUNDANT LOOPS I
l
Figure 3.3.3 THE DIRECT HEAT REMOVAL SERVICE VALVE LINEUP PERMITS THE USE OF THE EVST NAK AIRBLAST HEAT EXCHANGERS AS THE SINK AIR BLAST EXPANSION HEAT EXPANSION TANK 1 EXCHANGERS TANK 2 m
n h
JL 7
v v
NaK NaK OVERFLOW PUMP 1 1
2 PUMP 2 HEAT EXCHANGER h
N' W"
v a
i, a
3 a
^
h COLO i
m TRAPS
)MJ REACTOR h
, r n
U VESSEL
^
5 h*
-o<: :=
c
- >o--
JL h
w5(
>-T0 EVSTw y
OVERFLOW MAKEUP VESSEL PUMPS 7 77 P0514-9 9
b) requires approximately one-half hour to accomplish the manual realignment, c) assumes the integrity of all PHTS, and d) requires that forced convection be continuously provided via all three PHTS pony motors.
(5) The consequences of LOHS scenarios which could ultimately lead to core fuel / cladding melt and loss of coolable geometry are addressed in a sepa-rate section (see Section 6).
3.3.3 Issues Related to Individual SHRS Components a
Certain issues exist which must be addressed for the various components and subsystems of the shutdown heat removal systems to assure that they will
~
successfully fulfill their respective functions.
Those issues currently iden-tified are discussed in the following subsections and highlighted in Table 3.3.3.
Direct Heat Removal Service (1) Effect of Sodium Level:
In the current design, the inlet to the DHRS (refer to Figure 3.3.4) is located behind the thermal liner (i.e., the inlet does not penetrate through the thermal liner).
Sodium into the thermal liner originates from the bottom (which will essentially be cold sodium at the inlet plenum temperature) and from the top through thermal liner ports which are lo-cated only about 2.5 inches below the nominal sodium operating level.
Conse-quently, the only source of upper plenum sodium to the DHRS will be through these highly elevated ports.
Thus, the sodium level under required DHRS opera-tion must be closely assessed.
Additionally, the DHRS will offer no heat re-moval capability for any serious PHTS pipe break events.
(2) Upper Plenum Effects: The applicant currently assumes that the upper plenum sodium is well mixed, given that all PHTS pony motors are providing forced convection during DHRS operations.
Current evidence (see also Section 3.1) indicates that this may not be the case.
Since the only source of hot so-dium received by the DHRS comes from the uppermost portion of the upper plenum (refer to Figure 3.3.4 and also to previous discussion under this subsection),
the impact of stratification and subsequent heat removal capability must be investigated.
Additionally, the outlet from the DHRS exits directly into the upper plenum through the thermal liner (refer to Figure 3.3.4) at a location.
r Table 3.3.3 SHRS Issues - Component Level SHRS Component Identified Issues DHRS e Effect of sodium level e Upper Plenum effects e Thermal stress limits e Single valve failure e Capability of instrumentation and control PACC e Verification of natural circulation capability e Verification from prototype testing e Capability of instrumentation and control Main Evaporator e Verification of natural circulation capability Modules e
Verification from prototypic testing g
Figure 3.3.4 DHRS Flow Operation r
I'
{
-((
c=,f 5
s 1
i c
's "0"
ELEVATI0t1 E8 rd
{f F I) llh
-i I40MIt1AL OPERATIllG
=C
^ ~ ~
~ x - THERMAL LIllER PORTS l V (1.25"0D) (,-89.625in.
LEVEL OF SODIUM -87 Ill.
)
L
/
(\\z
[
z
( -148 Itt. -Gj
\\
g j-/
t% =
{ -156 in.
j [
e '/ 's DHRS MAKEUP tl0ZZLE
= -
DHRS OVERFLOW li
\\
l10ZZLE ll
\\
=
/
a k
f-f M.
T (l'/__h l
)ll1
\\
\\ !g l
V
' (_ -282.15 i n.
l i
f F
[f 1
i
,/
.FWl[l EK ll
/3N b
l b)
. a
?f w;
hiss b
ws
-i------
almost opposite from the inlet.
The inlet stream velocity is around 6-10 feet /
second.
The applicant is claiming the usage of a conservative estimate, based on experimental evidence (see also Section 3.4) to provide an upper bound on the amount of cold sodium from the DHRS outlet which may potentially communi-cate directly with the DHRS inlet.
This direct cross flow will limit the effectiveness of the DHRS and must be verified. Also, the admissible usage of data derived from water tests as applied to a sodium system under these circum-stances must be addressed.
(3) Adequacy of DHRS Design to Thermal Stress: Under the proposed oper ating conditions, the DHRS will be exposed to sodium temperatures in excess of 1100 F (865 K) for extended periods of time.
Since the ex-vessel storage tank components, which comprise the NaK portion of the DHRS, are normally exposed to operating temperatures in the 400-600 F ( ~ 475-590*K) range, the structural in-tegrity of the DHRS must be verified. Also, for the design basis event under which DHRS operation is necessary, the sodium entering the upper plenum from the DHRS will be at approximately 550-600*F (assuming 11 MW heat removal) while the upper plenum bulk sodium and structural temperature may be above 1100 F.
(4) Single Valve Failure:
In the required manual realignment of the sodium overflow system and the ex-vessel storage tank system to form the DHRS, operation of a single, non-redundant valve is required.
The reliability and unavailability of this valve must be determined.
(5)
Instrumentation and Control:
The capability of the instrumentation and control information available must be verified to ensure that the operator can correctly ascertain the true plant condition and implement any required DHRS action.
Protected Air-Cooled Condensor (PACC)
(1)
Verification of Natural Circulation Capability:
The PACC subsystem is designed to operate solely by natu al circulation on the wa,ter side.
On the air side, the preferred operating mode is via forced convection; however, nat-ural circulation capability is claimed.
The applicant has as yet provided no substantive basis to verify the natural circulation capability of such a unit for these applications.
In view of the heightened importance placed on the PACC, particularly for loss of all on-site and off-site power events, the.
assuran:e of natural circulation capability must be verified.
Also, in those areas wnere operation under abnormal conditions (e.g., asymmetric transients or single failure cannot be tested (see next subsection)), verified analytical tools must provide supportive analyses.
(2 ) Assessment of Prototype Testing:
The applicant currently plans to test a full scale prototype of the PACC subsystem. However, due to experi-mental test facility limitations, full power operation is currently question-able.
Also, it is essential that the testing provide a sufficient basis to e
ensure that any analytical tool used to predict PACC response to potential transie7ts outside the scope of the experiment can be adequately validated.
(3 )
Instrumentation and Control:
The capability of the instrumentation and control information available must be verified to ensure that the plant operate r can correctly ascertain and implement any required PACC action.
Main St eam Generator Evaporators (1)
Verification of Natural Circulation Capability:
Whenever off-site i
power i 3 unavailable, the evaporator modules will be required to operate under natural circulation on the water side, with no effective operator intervention possibl e.
Short term steam venting will be accomplished through steam d, ump valves downstream of the superheaters.
Subsequently, the superheaters will be-come ef #ectively isolated and heat removal will be accomplished directly from the evasorator to the steam drum to the PACC, via natural circulation.
Due to the importance of the evaporators to operate under natural circulation whenever required, it is. essential that tnis capability be either experimentally verifie d or analytically calculated by validated predictive tools.
(2 ) Assessment of Prototype Testing:
The applicant presently plans to test a 'ull scale steam generator module (evaporator and superheater designs are ide,tical).
This testing must be closely followed to ensure that it pro-vides a sufficient data base either to experimentally verify a,ll modes of re-quired,ormal and abnormal operating conditions or to validate analytical tools used tc predict operating conditions not covered during the planned testing.
3.3.4
- ssues Related to Total Integrated Plant Response During SHRS Events The capability to remove sensible and decay heat under all postulated ac-cident :onditions involves the complex interaction of a number of main plant components and SHRS components.
Two identified issues in this integrated plant response area are general in nature, while others may be grouped under the var-ious categories of events discussed in Section 3.3.2 and noted in Table 3.3.1.
The following subsections delineate these concerns, which are also highlighted in Table 3.3.4.
1)
Specification of design limits There must be a clear definition of the design limits within which the plant response must fall for specific events. The classifications designated by the project include normal, upset, emergency and faulted events.
For those events involving SHRS response, the appropriate limits for the various cate-gories of events must be clarified.
Thus, the capability of the system res-ponse can be quantitatively assessed.
2)
Identification of significant SHRS events The scenarios involving operation of the SHRS are many and varied. A con-sistent means must be used to identify the significant transients of interest and rank them according to their relative severity so that the appropriate de-sign limits discussed in the previous subsection can be applied.
- 3) Plant response to SHRS events assuming availability of off-site and on-site power Identified issues of concern here include:
a) Effects to individual components due to single failures such as; in-advertent operation of the turbine driven auxiliary feedwater pump; stuck open relief valve; failure in the turbine bypass system; leak in the main feedwater header common to all SG loops; inadvertent opera-tion of a PACC.
b) Effects to the remainder of plant due to asymmetric operation result-ing from failures.
c) Capability of instrumentation and control available to the operator to ascertain the true plant condition and implement corrective action if requi red.
d)
Impact of aforementioned failures under any proposed plant operations with one loop out of service.
4 a
Table 3.3.4 SHRS Issues - Integrated System Level l
e Specification of design limits, e Identifica-ica of significant events.
o Plant response to single failures for nomal, upset, eme gency and faulted events.
e Effects due to asymmetric operation.
e Capability of instrumentation and control.
4 i
37_
- 4) Plant response to SHRS events given loss of off-site power supplies Here the effects due to single failures and resulting asymmetric opera-tions are more pronounced.
Identified issues of concern include:
a) Loss of a diesel generator and resulting unavailability of dependent components such as pumps, valves and controls.
Delayed loss of a die-sel may be potentially worse with regards to a transition to natural circulation in the affected loop (s) due to degradation of thermal head.
b) Failure of the protected water storage tank auxiliary feedwater supply ~
which supplies all SG loops.
c) Failure of an auxiliary feedwater pump or its associated controls, particularly the full-sized turbine driven pump, d) Failure in the turbine bypass valve system.
e) Capability of instrumentation and control available to the operator to ascertain the true plant condition and implement corrective action if required.
f)
Impact of aforementioned failures under any proposed plant operations with one loop out of service.
- 5) Plant response to SHRS events given loss of all off-site and on-site power supplies The plant is required to operate with natural circulation under blackout conditions as discussed in Sections 3.1 and 3.2.
Identified issues of concern include:
a) Failure of turbine driven auxiliary feedwater pump.
b) Failure of eithcr superheater or steam drum vent valve in loop 3.
c) Failure of protected water storage tank system.
d) Effect of upper plenum stratification.
Stratification may be severe under these very low flow conditions (see also Section 3.1).
e) Capability of instrumentation and control available to the operator to ascertain the true plant condition and implement corrective action if required.
- 6) Plant response to LOHS events when the DHRS is operatino The applicant position for these scenarios appears to be that all PHTS must be intact and power to all primary loop pony motors must be available.
Under these circumstances, a single additional failure could lead to a loss of coolable geometry.
Such single failures and additional outstanding issues include:
a) Failure of one pony motor (instantaneously or delayed) b) Failure of the single non-redundant valve in the DHRS (see Section
~
3.3.3) c) Failure of any DHRS pump d) Failure of Na/NaK heat exchanger e) Failures of either Nak/ air heat exchanger f)
In-vessel sodium level drops below thermal liner ports (see Section 3.3.3) g) Stratificacion in upper plenum (see Section 3.3.3)
'3.3.5 Approach to Resolution of Issues The general approach necessary to resolve the issues and questions re-garding the capability of the shutdown heat removal systems can be summarized by grouping the previous discussions into four main areas:
- 1) Define clearly the appropriate safety and design limits for the vari-ous types of SHRS events.
- 2) Use a consistent methodology to identify all potentially significant SHRS transients.
The CRBRP design must be carefully reviewed, with particular emphasis on recent changes, to ensure that all important transients are identi-fied.
Here a combined effort involving probabilistic and realjability analy-sis, engineering judgment and supportive parametric-type deterministic analysis will be required.
i a
- 3) Conduct a close assessment of all completed and proposed experiments which provide substantiation or expansion of the SHRS data base or which lead to the reduction of uncertainties.
- 4) Apply validated predictive tools to analyze potential SHRS events and to provide any required independent assessment of results supplied by the applicant.
o I
J e
i 4
I J
f I
s l
l l
l.
REFERENCES 3.1 D. Majumdar and R. J. Cerbone, " Survey of Available LMFBR Facilities for Testing Safety-Related Thermal-Hydraulic Computer System Codes",
BNL-NUREG-23429 (October 1977).
3.2 G. D. Bouchey, S. L. Additon and E. A. Parziale, " Planned Natural Cir-culation Testing in the Fast Flux Test Facility", ANS Transactions, 21,
- p. 522 (June 1978).
3.3 U. S. Nuclear Regulatory Commission, "FFTF Safety Evaluation Report",
USNRC Report NUREG-0358 (August 1978).
3.4 K. R. Perkins, R. A. Bari and L. C. Chen, "An Evaluation of Natural Cir-culation Tests in FFTF", BNL Informal Report, BNL-NUREG-29348 (March 1981).
3.5 R. L. Stover et al., "FFTF Natural Circulation Seminar, HEDL, Richland, Washington (September 30,1981).
3.6 K. R. Perkins and R. A. Bari, " Preliminary Review and Evaluation of Natural Circulation Decay Heat Removal Capability in CRBR and FFTF", BNL Informal Report, BNL-NUREG-21913 (October 1976).
3.7 K. R. Perkins, R. A. Bari and D. C. Albright, " Uncertainties in the Calculated Response of the Clinch River Breeder Reactor During :latural Circulation Decay Heat Removal", BNL Informal Report, BNL-NUREG-22715 (April 1977).
3.8 R. R. Lowrie and W. J. Severson, "A Preliminary Evaluation of the CRBRP Natural Circulation Decay Heat Removal Capability", WARD-D-0132 (March 1976).
3.9 J. R. Longenecker et al., " Briefing on CRBRP Shutdown Heat Removal",
Bethesda, Maryland (November 2,1981).
3.10 K. R. Perkins, R. A. Bari, L. C. Chen and D. C. Albright, " Analysis of FFTF System Transients", BNL Informal Report, BNL-NUREG-25560 (January 1979).
3.11 Westinghouse ARD, "LMFBR Demo Plant Simulation Model (DB10)",
WARD-D-0005, Rev. 4 (January 1976).
3.12 C. F. Wolfe, "IANUS - Application Information", WARD-FPC '1253, Rev.1 (May 1976) (Proprietary).
3.13 M. Khatib-Rahbar, I. K. Madni and A. K. Agrawal, " Impact of Thermal Buoy-ancy in LMFBR Piping Systems", Proc. of Specialists' Meeting on Decay Heat Removal and Natural Convection in FBRs, Brookhaven National Labora-tory (February 28-29, 1980).
REFERENCES (CONT.)
3.14 I. K. Madni and A. K. Agrawal, "LMFBR System Analysis: Impact of Heat Transport System on Core Thermal-Hydraulics", Specialists Issue of Nucle-ar Engineerino & Design, Nucl. Eng. Design 62[, 199 (1980).
3.15 A. K. Agrawal et al., " Transient Simulation of LMFBR Systems", presented at the Mathematics and Computation Division Topical Meeting on Improved Methods for Analysis of Nuclear Systems, Tucson, Arizona (March 28-30, 1977); Nucl. Sci. Eng., 64, 480 (1977).
3.16 A. K. Agrawal, J. G. Guppy, I. K. Madni and W. L. Weaver, " Dynamic Simu-lation of LMFBR Plant Under Natural Circulation", ASME Paper No. 79-HT-6, presented at the 18th ASiE National Heat Transfer Conf., San Diego, Cal-ifornia (August 5-8, 1979); J. Heat Transfer, 103, 312 (1981).
3.17 J. Muraoka et al., "FL0 DISC - A Dynamic Core Flow Distribution Code:
Evaluation of the Total Loss of Electric Power Event", HEDL-TC-874 (May 1977).
3.18 J. V. Miller and R. D. Coffield, " FORE-2M: A Modified Version of the FORE-II Computer Program for the Analysis of LMFBR Transients", WARD-D-0142 (May 1976).
3.19 K. R. Perkins.and R. A. Bari, "Interassembly Flow Redistribution at Natural Circulation Conditions in the Fast Flux Test Facility", Trans.
Am. Nucl. Soc., 30, 413 (November 1978).
3.20 M. L. Millburg, J. A. Hassberger and C. J. Boasso, " Natural Circulation Heat Transfer Testing with a Simulated Full-Scale LMFBR 217-Pin Elec-trically Heated Fuel Assembly", HEDL-TME 77-3 (June 1977).
3.21 R. M. Singer et al., " Studies Related to Emergency Decay Heat Removal in EBR-II", Proc. of the International Meeting on Fast Reactor Safety Tech-nology, p.1590, Seattle (1979).
3.22 C. K. Chan, T. K. Min and D. Okrent, "A Look at Alternative Core Disrup-tion Accidents in UiFBRs", UCLA-ENG-7720 (February 1977).
3.23 K. R. Perkins, W. T. Pratt and R. A. Bari, " Evaluation of In-Vessel Natural Circulation During a Hypothetical Loss-of-Heat-Sink Accident in the Fast Flux Test Facility", BNL Informal Report, BNL-NUREG-26565 (August 1979).
3.24 M. J. Khatib-Rahbar, J. G. Guppy and A. K. Agrawal, " Hypothetical Loss-of-Heat-Sink and In-Vessel Natural Convection: Homogeneous and Hetero-geneous Core Designs", Decay Heat Removal and Natural Convection in Fast Breeder Reactors, Hemisphere Puolishing Co., New York, N.
Y.
(1981).
3.25 F. E. Dunn, " Severe FFTF Natural Circulation Transients with Boiling",
ANL/ RAS 76-26 (September 1976). -
REFERENCES (CONT.)
3.26 K. R. Perkins and R. A. Bari, "SAS-3D Evaluation of Boiling at Decay-Heat Levels in FFTF", Trans. Am. Nucl. Soc., 33, p. 516 (1979).
3.27 P. W. Garrison, R. H. Morris and B. H. Montgomery, i' Natural Convection Boiling of Sodium in a Simulated FBR Fuel Assembly Subchannel", Proc. of the International Meeting on Fast Reactor Safety Technology, Vol. IV, Seattle (August 1979).
3.28 A. Kaiser, W. Peppler and M. Straka, " Decay Heat Removal from a Pin Bun-die", CONF-761001, IV, p.1578 (October 1976).
3.29 Preliminary Safety Analysis Report, Clinch River Breeder Reactor Plant, Project Management Corporation, docketed June 1975.
3.30 A. K. Agrawal et al., "An Advanced Thermohydraulic Simulation Code for Trar.sients in LMFBRs (SSC-L Code)", Brookhaven National Laboratory, SNL-NUREG-50773 (February 1978).
4.
CONTAlfNENT ANALYSIS Technical issues related to the adequacy of the CRBR containment were sum-marized in Reference 4.1 at the termination of NRC licensing of the Project.
In Table 4.1, we summarize these issues and provide a brief description of how much additional review was considered necessary at that time. The Project responded (Ref. 4.2) to the NRC staff review and in Table 4.2 we summarize their assess-ment of the resolvability of each of the items noted in Table 4.1.
In this section, we follow the format indicated in Table 4.1.
Initially, we discuss large sodium releases in the Steam Generator (S/G) building in Sec- -
tion 4.1.
We then discuss containment system design in Section 4.2.
- Finally, in Section 4.3 we discuss the impact of the new core design on radiological consequences associated with accidents beyond the design basis.
The discussion on radiological consequences was not included in Reference 4.1 so it does not appear in Table 4.1.
4.1 Large Sodium Releases in the S/G Building In Table 4.1, we summarize the major issues remaining (in the view of the NRC staff) at the suspension of CRBR licensing.
Further evaluation of the models and assumptions used by the project in their analysis of sodium fires was considered necessary.
The adequacy of the S/G building design pressure and temperature was not considered to be established.
Additional information on the analyses of sodium fires was requested (Q001.703).
Finally, further assessment of the adequacy of the proposed fire suppression and nitrogen flooding systems for sodium fires was also considered necessary.
The CRBR Project's response to the above issues is briefly summarized in Table 4.2.
The Project staff indi-cated their intention to update the PSAR to address the issues of model assump-tions, S/G building design and the request for further information.
- However, with regard to the fire suppression systems the project staff considers that sufficient information was provided in the PSAR for the NRC evaluation.
A supplement to the PSAR was planned by the Project staff for further clarifica-tion in this area.
Clearly, a review of all of this additional information is required to assess the adequacy of the Project's analysis of large sodium re-leases in the S/G building.
Issues discussed in Section 4.2 related to the choice of the DBA and integrity of the cell liners also impact the issues in this section.
l l
Table 4.1 Summary of outstanding technical issues related to CRBR containment analysis
- Technical Issue Comments large Sodium Releases o Understand and evaluate models used a
in S/G Building in PSAR.
o Adequacy of S/G building design pressure and temperature.
o Additional information requested on sodium fire analysis.
o Adequacy of proposed fire suppression and N2 flooding systems.
Containment System Design Large Sodium Releases o DBA selection, and DBAs o Preliminary review of containment isolation system-further review needed.
o Control the accumulation of H '
2 o Review of autocatalytic recombination test results needed.
o Capability of TMBDB systems to function in hostile environments.
o Pipe break spectrum.
o Cell liners as ESF.
o SOFIRE, SPRAY, and CACEC0 verification.
o Analysis of spectrum of sodium spills and behavior of cell liners.
- Letter, W. P. Gammill to L. W. Caffey, dated November 9,1978; Reference 4.1.
Table 4.1 Summary of outstanding technical issues related to CRBR containment analysis (Cont.)
Technical Issue Comments Containment System Design (cont.)
Accommodation of o Containment integrity must be provided -
meltdown for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> following CDA.
o Substantiation of the assumptions regard ing core debris-material interactions.
o Evolution and control of H2 - auto-ignition claim requires further sub-stantiation.
o Concrete structural analyses.
o Dose mitigation features.
o Interaction of TMBDB features with ESF features.
o Revised TMBDB report review.
=
. 1
Table 4.2 Summary of the Project's Position on the NRC staff review
Containment System Design Large sodium releases o Project considers selected DBA and DBAs to be conservative.
o Features have been included to ensure protection against DBA.
o Commitment to mitigate hypothetical accidents beyond design basis, o Equipment survival in hostile environ-ments to be in TMBDB report.
Accommodation of o NRC concerns will be addressed in meltdown TMBDB report.
- Letter to H. R. Denton, Project Evaluation of the NRC Staff Review of CRBRP," dated March 3,1979.
4.2 Containment System Design Issues related to Containment System Design were subdivided into Large Sodium Releases, Design Basis Accidents (DBAs), and the Accommodation of a Core-meltdown (refer to Table 4.1).
There are a number of issues which are of concern under all of the above classifications.
For example, the evolution and control of H2 could be of concern in both DBA analysis and in the accommoda-tion of accidents beyond DBA.
Also the performance of the cell liners is cru-cial for DBAs and accidents beyond DBA.
Indeed the final selection of a DBA will influence the extent to which some of the issues included in Table 4.1 need to be considered. Therefore, we have selected major issues in Table 4.1, which we consider to be still unresolved at this stage.
We discuss these issues in the following sections.
4.2.1 Design Basis Accidents l
At the termination of CRBR licensing, the adequacy of the DBA analysis was in question (Ref. 4.1).
It remains for the NRC staff to identify an adequate l
DBA.
A BNL analysis (Ref. 4.3) of the DBA proposed before licensing termination indicated some divergence from the CRBR Project's analysis.
This was mostly due to assumed sodium-concrete reaction rates and to differences in heat transfer rates and choices of thermal models (thermal equilibrium model).
These differences should be addressed and factored into the analysis of the selected DBA.
4.2.2 Accommodation of a Meltdown All of the outstanding issues (refer to Table 4.1) associated with accom-modation of a meltdown are addressed by the CRBR Project in their TMBDB report (Ref. 4.4).
In order to establish the adequacy of thermal margins beyond the design base, a careful review of Reference 4.4 is necessary.
At the suspension of the CRBR licensing effort, an assessment had been performed and a report (Ref. 4.3) written by BNL regarding the DBA analysis in CRBR., An effort was initiated to analyze the containment response to a core meltdown accident but the work was terminated.
It is, therefore, recommended that this effort be resumed.
We have had only a limited review of Reference 4.4 at BNL, but it appears that issues still remain, which require further resolution.
In the following sections we discuss some of these issues, which we found during our preliminary review of Reference 4.4.
More issues may be found as part of the detailed evaluation of Reference 4.4 4.2.3 Sodium-Concrete Reaction Rates This issue has not been resolved in a consistent or satisfactory manner.
The question of penetration rate as well as the ultimate depth of penetration are still open.
Sandia experimental results tend to favor more rapid penetra-tion rates (i.e., 4 to 6 in/hr) with (under certain conditions) no reaction product inhibition or termination of the reaction front propagation. These results differ from those obtained in the HEDL experiments which imply a reac-tion rate of 1/2-1 in/hr and a maximum penetration of 12 inches. We consider that the parametric study performed in the TMBDB report (Ref. 4.4) should as a minimum be consistent with the penetration rates used for the analysis of FFTF core meltdown accidents (Ref. 5) (6 in/hr for 12 in was used for FFTF HEDL-TC-1175).
The sodium concrete reaction rate strongly impacts the time of incipient sodium boiling in the reactor cavity and this, in turn, is the primary contain-ment pressurization mechanism.
Since the energy supplied to the sodium pool by the sodium concrete reaction rate may be comparable to the decay heat, it is a crucial parameter for predicting whether or not the 24 Lopr non-venting cri-terion can be met.
4.2.4 Availability of Containment Passive Heat Sinks The CRBR Project's analysis of the TMBDB (Ref. 4.4) relies heavily upon the availability of the passive heat sinks in the containment to meet the 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> venting criterion. Transfer of heat to these existing structures is critical if full credit is to be taken for their heat absorbing capacity.
The presence, in the containment building atmosphere of enormous quantities of aerosols (mostly Na20 and NaOH) subsequent to incipient sodium pool boiling, and,the accompany-ing plate-out of these aerosols on all the available surfaces in the RCB will certainly inhibit the flow of heat by the normal transfer mechanism to the existing passive heat sinks.
Thick layers of plated materials having both high porosity and low heat transfer properties will effectively insulate and delay the transfer of energy to the available structures.
This mechanism is appar-ently not considered in the TMBDB analysis (Ref. 4.4).
An assessment should 1 l
l
.~
~
i; therefore be made of the degree to which this mechanism may negatively effect the capacity to contain the post-melt-through environment for the prescribed period.
4.2.5 Debris Bed Coolability The redesigned CRBR core is sufficiently different to warrant not only a recalculation of the coolability of sodium flooded debris beds using an average decay heat level, but one which accounts for the heterogeneous distribution of the enrichment zones.
A scenario can be visualized in which selective melting of the highly enriched zones of the core occurs preferentially, resulting in a
~
concentration of decay heat in a smaller quantity of debris and perhaps in a more rapid penetration of the vetsai head.
New estimates should also be made of the amount of upward (primary piping and upper plenum) and downward fuel relocation in terms of the new heterogeneous core enrichment configuration.
Possible stratification of enrichment zones in the debris may affect the coolability of debris beds both in-vessel and ex-vessel, but the largest effect would probably be in-vessel.
In a BNL assessment (Reference 4.10) of thermal margins beyond DBA for FFTF we assumed that the core debris did not form a sodium flooded debris bed.
We therefore assumed that the core debris could directly attack the reactor cavity concrete even with a large quantity of sodium also in the cavity. This assump-tion was an important consideration for FFTF because of the presence of the unlined subcavity beneath the reactor cavity.
We found that the integrity of the floor between the cavity and the subcavity was crucial to our analysis of containment pressurization.
The noncoolable debris bed assumption may not be such an important consideration for CRBR (no unlined subcavity) but the con-sequences of such an assumption should be addressed.
No such assumption is at present made in the TMBDB report (Reference 4.4), which is again not consistent with the analysis (Reference 4.5) that has already been performed for FFTF.
4.2.6 Hydrogen Generation and Auto Ignition At the point when CRBR licensing was suspended, the applicability of the experimental data (Ref. 4.11) and extrapolation of that data had been called into question (Ref. 4.1).
In particular, at lower temperatures, entrainment of large quantities of sodium droplet aerosols were required to assure a continuous
hydrogen combustion and avoid an accumulation of H2 in the containment building.
Whether this can be guaranteed has not been established.
4.2.7 Mitigating Features Independent review of the containment response to mitigating features such as vented-filtered systems has not yet been performed.
In view of the possibil-ity of more rapid sodium-concrete penetration rates and the subsequent increased density of aerosols in the RCB, an independent assessment should be made of the design and capability of the proposed scrubbing system to adequately remove the particulate and volatile radioactive materials. Also, an assessment is required -
to determine the efficiency of the annular cooling system in the presence of inhibiting aerosol plate-out.
In addition, the question of the, feasibility of an ex-vessel core catcher and/or Na K cooling system may again emerge.
4.2.8 Potential for Basemat Penetration The analysis performed in the TMBDB report (Reference 4.4) does not reflect the latest experimental and analytical models available (Reference 4.12) to pre-dict the potential for basemat penetration by thermal attack of the core debris.
In Section 4.2.5 we note the potential for a noncoolable debris bed to form and concreta penetration to occur even in the presence of sodium.
The assumption in the TMBDB report (Reference 4.4) that no penetration of the core debris into the concrete can occur until all the sodium is boiled dry underestimates the poten-tial threat to basemat integrity. Also, the model used (TRUMP) cannot model the potential for a solid core debris to penetrate concrete.
SANDIA experiments have shown that molten core debris rapidly forms crusts, which prevent dilution of the core debris (and hence dilution of the volumetric heat source) but allows concrete penetration.
The analysis of basemat penetration and the TMBDB report (Reference 4.4) is inadequate and should be improved to include currently avail-able models.
4.2.9 Model Validation Although the CRBR Project claims that the CACEC0 code has been fully validated, we suggested that the code be reviewed by an independent (from the writers of the code) organization. A code with the degree of complexity con-tained in CACECO requires more than experimental validation of individual models. An overall evaluation of the code's energy and mass balances and the way in which the individual models mesh with the containment response models are extremely important.
In addition to the obvious checks against simply hand cal-culations, the code results should be compared to those obtained by alternate codes su'ch as the BNL developed CONAN code'(Ref. 4.6).
Care must also be taken in accepting, as experimentally validated, models which were obtaiiied by extra-polation of non-prototypical experiments. These types of issues should be con-sidered in any " independent" assessment.
A number of model deficiencies were detected and reported by BNL (Ref. 4.7) before the CRBR licensing review was suspended.
Some of these_ problems may still remain, e.g., the assumption of thermal equilibrium between the contain-ment atmosphere and the sodium pool in the reactor cavity.
A careful in'quiry should be made into the areas that have previously been identified as problem-atical as well as a thorough re-evaluation of the latest version of the code.
Validation of the sodium fire codes has been undertaken at BNL. The NACOM' code (Reference 4.8) was developed at BNL to analyze sodium spray fires.
NACOM was compared with the SPRAY code (Reference 4.9) (used by the Project in their analysis of sodium spray fires) and reasonable agreement found.
However, extra-
.polation of both of the above models, which are based on single droplet burning models, to toe turbulent conditions that occur within the spray core still re-quire resolution.
Comparison of the predictions of the above models with the more advanced spray model (SOMIX) is needed.
4.3 Radiological Consecuences The radiological consequences analysis for the revised CRBR is based on the homogeneous : ore configuration. The claim is made in this analysis that, even with the higrer plutonium inventory in the heterogeneous core, its isotopic makeup is sucn that the radiological impact due to plutonium aerosols is lower.
In view of this implied sensitivity to plutonium isotopic content, a compara-tively more severe situation might arise if the heterogeneous core was fueled with homogeneous core plutonium.
Table 4-6 of the CRBRP-3 (Re'f. 4.4) seems to indicate that the homogeneous core is fueled with LWR gra'de plutonium, while the heterogeneous core is fueled with FFTF grade plutonium.
Since both grades were l
considered for the homogeneous core, it is not unreasonable to considir both grades for the heterogeneous core.
A further point, of lesser importance, but which should be considered is the different burnup expected for the two cores. 1
~.
o
'r t
u -
At equilibrium'the burnup in the heterogeneous core is expected to be approxi-mately 25% lowar than.in the homogeneous core. This change in burnup will
~ ' affect t.he fission product inventory which in turn will have an effect on the y
radiological consequence.
o'
]
i WD s
+
l i
I i
i, i
4 4
i
' j i
f
R[FERENCES 4.1.
Letter, W. P. Ga'mmill to L. W. Caf fey, dated November 9,1978.
4.C.
Letter to H. R. Denton, "Projrct Evaluation of the NRC Staff Review of CRBRP," dated March 5, 1979.
~
4.3.
R'. D. Gasser, S. S. Tsai, D. C. Albright and W. T. Pratt, " Containment Design ~ Basis Accident Analys:s for the Clinch River Breeder Reactor,"
BNL-NUREG-25561, January 197!.
4.4.
"CRBRP-3, Hypothetical Core Ilisruptive Accident Considerations in CRBRP, Volume 2, Assessment of Thenial Margin Beyond the Design Base," Issued by -
the CRBRP Project Office, Ma ch 1980.
4.5.
R. J. Jensen et al., "Furthe ' Studies of Containment Margins in FFTF for Postulated Failure of In-Ves ;el Post-Accident Heat Removal," HEDL-TC-1175, July 1978.
- 4. 6.
R. D. Gasser, "CONAN:
An LM BR Containment Response Computer Code,"
BNL-NUREG-CR-1355, November 1979.
- 4. 7.
S. S. Tsai, R. D. Gasser and W. T. Pratt, " Containment Design Basis Ac-cident for LMFBRs: Review of Methods," BNL-NUREG-23221 September 1976.
4.8.
S. S. Tsai, "The NACOM Code for Analysis of Postulated Sodium Spray Fires in LMFBRs," BNL-NUREG-51180, NUREG-CR-1405, March 1980.
- 4. 9.
P. R. Shire, " SPRAY Code User's Report," HEDL-TME 76-94, March 1977.
4.10.
R. D. Gasser and W. T. Pratt, " Containment Response to Postulated Core Meltdown Accidents in the FFTF," BNL-NUREG-24141-R, August 1978.
4.11.
R. W. Wierman and R. K. Hilliard, " Experimental Study of Hydrogen Forma-tion and Recombination under Postulated LMFBR Accident Conditions," Han-ford Engineering Development Laboratory, December 1976.
4.12.
W. T. Pratt, R. D. Gasser and R. A. Bari, " Potential Influence of Core /
Concrete Interactions on PWR Containment Pressurization," Trans. Am.
Nucl. Soc. 39,609(1981).
9
5.
RELIABILITY ANALYSIS This section reports on issues related to reliability analysis of the CRBR.
Section 5.1 provides some comments on the risk study conducted by the CRBR Proj-ect.
Sections 5.2 and 5.3 contain updates of the issues related to the reli-ability of the shutdown system and shutdown heat removal system, respectively.
5.1 Risk Assessment In March 1977, the CRBRP risk assessment report was issued (Ref. 5.1).
The report states that its objective was to provide a realistic evaluation of the risk to the public from the CRBR plant.
A methodology similar to the one used in the Reactor Safety Study (Ref. 5.2) was used.
Since this risk study was issued near the end of CRBR licensing activities, it has not been reviewed and evaluated in depth by BNL.
However, BNL has performed extensive reliability analyses on the systems supporting the two main safety functions, namely, the reactor shutdown function and the shutdown heat removal function.
The major issues that these studies have identified are presented in the corresponding sections of this report (See Sections 5.2 and 5.3).
This section contains general comments on the overall risk assessment methodology and results.
The methodology for the assessment of the CRBRP risk can be divided into the following steps.
1.
Accident Sequence Definition and Quantification.
2.
Accident Analysis and Evaluation.
3.
Consequence Modeling.
4.
Risk Evaluation.
The following issues are pertinent to the completeness and validity of the analysis.
o The risk study has been performed for the 1976 CRBRP design. Before the results of this study can be extrapolated to the latest CRBRP design, a careful examination of the potential impact of the design changes in the various parts of the analysis should be performed.
o The handling of the potential dependences among systems should be re-viewed and evaluated. Of particular interest are interactions between key systems that can contribute (through their failure) to the frequency of accident initiators or increase the failure probability of a safety function.
Both a qualitative analysis should be performed (having as its objective the identification of existing dependences) and a quan-titative assessment of the impact of the potential dependences on the probabilities of the various accident sequences.
o The coupling of the reactor shutdown and the shutdown heat removal event _
trees with the containment event tree through the core response and pres-sure vessel failure model is highly subjective.
A more detailed model-ing of the dependences between the failures of systems, the response of the core, the pressure vessel failure modes and the containment failure modes (Ref. 5.3) is required.
o The numerical results are point estimates obtained by inputting the median values of the various parameters. The use of median values in the input does not yield the median value of the final result.
Yet, the results of this study are compared with the median results of WASH-1400.
Furthermore, even if the exact median of the CRBRP risk were presented, the comparison of two median is not adequate since the median does not contain any measure of the spread of the results.
Hence, an uncertainty analysis should be performed and the total range of the CRBRP risk should be estimated and presented.
5.2 Reliability Analyses of the Shutdown Systems 5.2.1
Background
l BNL has performed an extensive analysis of the reliability of the shutdown systems of the CRBRP, as well as, their contribution to the probability of Loss of Core Coolable Geometry (Ref. 5.3).
The BNL study has employed the most advanced methodology for analyzing the 1
reliability of large systems, namely, the modeling of the stochastic behavior of the systems by Markov processes.
This approach allows for a more realistic modeling of the system that avoids unnecessary conservative assumptions. l 1
The BNL analysis includes dependences between the components of the elec-trical subsystem of the primary and secondary systems as well as between the components of the mechanical subsystems.
Interdependences between the unavail-ability of the systems and the occurrence of the transients have been included, resulting in the removal of unnecessary conservatisms.
Thus, whether the shut-down system responds successfully to a challenge depends on its state and on the type of transient, e.g., for limited response transients only the mechanical subsystem is required to respond.
Successful challenges reveal partial failures of the system (e.g., one channel down out of three) and result in system renew-al. Finally, the time-dependent nature of the model accurately calculates the -
probability of loss of core coolable gecmetry by avoiding the double counting introduced by the approximation:
failure probability = (frequency of challenge) x (system unavailability)
The BNL model allows for inspection and maintenance procedures that depend on the state of the system and include the possibility of human errors.
Inspec-tion of the protective function networks of the primary and secondary systems are staggered to minimize the effect of human errors.
Two inspection policies are possible depending on whether there are tripped channels in one or both shutdown systems.
Uncertainties are expressed by assuming the failure rates, the repair rates and all other input variables as random variables, distributed according to given probability density functions. The distributions are such that the upper 90% percentile of the input parameter is one order of magnitude higher than the lower 90% percentile.
To assess the sensitivity of the probability of loss of core coolable geometry to interdependences and human errors, four special cases were examined:
(a) the failure rates of the components are completely independent and the in-spection is perfect, i.e., every four weeks the electrical subsystems are com-pletely renewed; (b) the failure rates and the states of the components are completely independent, and the inspection is perfect; (c) the failure rates of the components are completely independent, but inspection is imperfect, i.e.,
human errors are possible; and (d) interdependences exist, and the inspection is imperfect.
For these cases, medians, 90% probability bands, and point estimates of the RSS tailure probability per year are listed in Table 5.1.
Table 5.1 s
Table 5.1 Point Estimate, Median, and 90% Probability Band of the Failure Probability per Year under Various Assumptions for the Shutdown System of the CRBR (Point estimate is the value of the failure probability obtained when the input variables are assumed to be fixed at their means.)
Failure Probability Case Assumption Point Estimate Median 0.05 Percentile 0.95 Percentile a
fio dependences, perfect inspection 8 x 10-10 1 x 10-9 4 x 10-11 2 x 10-8 b
Dependences, perfect inspection 2 x 10-7 8 x 10-8 1 x 10-9 6 x 10-6 Qg c
flo dependences, imperfect 4 x 10-7 2 x 10-7 1 x 10-8 5 x 10-6
,inspection d
Dependences, imperfect 2 x 10-6 2 x 10-6 2 x 10-7 2.x 10-5 inspection O
e e
5.3 Shutdown Heat Removal System Reliability Work on the reliability of the Shutdown Heat Renoval System (SHRS) was performed at BNL prior to cessation of licensing activities for the CRBe..
This work, as pertains to the most recent design of the SHRS in existence at the time of the cessation of licensing activities, is summarized in Reference 6.4 The design considered was capable (under certain circumstances) of using the Direct Heat Removal Service (DHRS) to successfully remove decay heat even if heat transport through the main heat transport system is cut off at the intermediate heat exchangers. Consideration was given only to the so-called baseline design.
case in which operation on less than three loops was not permitted.
Pony motors in one primary loop and its associated intermediate loop were assumed to be ca-pable of being powered by a battery with a two hour capacity.
The analysis given in Reference 5.4 confined itself to the initiator considered most impor-tant, the loss of offsite power initiator.
It was assumed in Reference 5.4 that natural circulation in the steam generator loop is sufficient for SHRS mission success, or else that the recirculation pump in the steam generator loop 2 may be operated by a battery.
It is really not clear whether one requires both motor-driven pumps in the auxiliary feedwater system (AFWS) or only one, for AFWS mission success.
The CRBRP risk assessment report (see Ref. 5.1, p.10-24, section 10.3.5) assumed one motor-driven pump was sufficient.
However, according to the CRBR PSAR (Ref.
5.5, p. 5.6-1C, Amendment 43 dated January 1978) "each motor-driven pump is half-size, such that the combination can supply the required flow to all three loops." This statement could be interpreted as meaning that both motor-driven pumps are required for AFWS mission success.
This point must be clarified.
Reference 5.4 presents results for both cases.
Results were presented a Reference 5.4 for two different sets of diesel generator failure probabilities.
One of these was the Reactor Safety Study (Ref. 5.2) diesel generator data, where the probability of one-diesel generator failing to start (and take on load) was.03 per demand, and the probability that both fail to start (or take up load) is.01.
The second set of diesel generator failures probabilities had a.01 per demand probability of failure for one die-sel generator, and a.001 per demand probability for the common mode failure of both diesel generators.
The Reactor Safety Study data for common mode failure of the diesel generators is probably somewhat pessimistic, but the second set of diesel generator failure probabilities are optimistic, when compared to present experience.
Reference 5.4 considered the SHRS reliability for two different probabili-ties for restoration of offsite power within two hours:
.1 and.01.
The signi-
=
ficance of the two hours is that the batteries on the pump motors will be drained in two hours.
(Consideration should also be given to the effects of loss of D.C. power for control and instrumentation for an extended loss of all.
AC electric power).
The probability of.1 for restoration of offsite power in two hours was taken from the Reactor Safety Study (see Figure III.6.4 in Appen-dix III of Reference 5.2) and came from Bonnsville power grid data.
- Actually, even the probability of.1 for restoration of offsite power in two hours may be optimistic.
The reason is that the Bonnsville power data was for restoration of single line outages, and the distribution function for the time to repair of a single line may depend on whether the outages considered are those which cause loss of only a single line or those which cause total loss of offsite power.
There are indications in the data collected by Ray Scholl (Ref. 5.6) that this is the case.
The frequency of loss of offsite power used in Reference 5.4 was
.2/ year; TVA grid specific values could be better.
The results obtained in Reference 5.4 are summarized in Table 5.1 of that reference and are reproduced here in Table 5.2.
Since the work perforned in Reference 5.4, additional work has been per-formed on the SHRS system reliability by Jamali, as part of his thesis.
Thi s work is summarized in Reference 5.7, by Jamali and Kerr.
The results obtained by Jamali (see Table 5.3) give much higher unavailabilities for the SHRS than were obtained in Reference 5.4 Some of the reasons for this are:
1.
Reference 5.7 uses, for a loss of main feedwater frequency 2/ year, which is reasonable.
However, no credit is given for timely' restoration of main feedwater.
The Reactor Safety Study assumed only 1% of all loss of main feedwater transients could not be repaired in 1/2 to I hour (see Appendix V, p. 37, of Reference 5.2). - -
2.
Reference 5.7 gave no credit for the DHRS as an alternative path for decay heat removal, independent of the AFWS.
3.
Reference 5.7 assumed an AC dependency of the steam-turbine driven pumps in the AFWS, because of its dependence on the ventilation system. The ability of the steam-turbine pump in the AFWS to operate in the absence of AC power should be confirmed.
4 Reference 5.7 included many types of common mode failures by a variant of 6
the 6-factor method (Ref. 5.8) where factors 02/1 for the probability of failure of a second component given failure of the first, and 03/2 for the probability of failure of the third component given failure of the first two, are obtained. However, many maintenance-caused failures of, e.g., valves are assumed coupled, not only valves performing similar functions in redundant trains of a system. This very likely introduces a conservative bias in the results, since the probability a second valve fails given the first loss depends on how similar the valves are in func-tion and in location, and in the maintenance they are subjected to.
The precise manner of incorporating the somewhat weaker dependence of valves performing dissimilar functions is of interest, however.
Jamali has commented on some nonconservative aspects of his analysis, in-cluding neglect of certain types of electrical common mode failures, neglect of some coupled human failures, and fault-tree truncation.
4 Table 5.2 Failure probability, per year, of the SHRS due to LOSP shutdown initiating event.
Failure Probability, Per Year, of SHRS W
(Due to LOSP)
Case 6 1 a2 Case 62 a1 Case 6 2 a2 Case 6 1 at 4
(a) No credit for natural circulation in sodium loops after two hours from shut-down; p2 =.1 3 x 10-4 /yr 2 x 10-4/yr 2 x 10-5/yr 2 x 10-5/yr (b) Like (a),
but p2 =.01 9 x 10- 5 /yr 5 x 10-5/yr 6 x 10-6/yr 6 x 10-6 /yr (c) Credit for natural circulation in sodium loops after two hours from shut-down, or negligible probability that off-site power will not be re-stored in two hours 7 x 10-5/yr 3 x 10-5/yr 4 x 10-6/yr 4 x 10-6 /yr
= Probability of f-site power is not restored by two hours af ter p2 shutdown.
Key to Cases:
- Reactor Safety Study diesel generator failure data.
The prob
=
7 6 i ability one diesel generator f ails to start is.03; the proba-bility both fail to start is
.01.
The probability one diesel generator fails to start is.01; the 6 2 : probability both fail to start is.001.
One motor-driven pump loop in the AFWS is insufficient for AFWS at : mission success.
One motor-driven pump loop in the AFWS is sufficient for AFWS a2 :
mission success. -.-
8
- t k
~
V Table 5.3 Summary of results for failure of short-term forced circulation.
!!edian Failure In it ia to'r Unavailability Proba bility Major Sequences Critical Component Failures per year Normal 2.1 x 10-4 3 x 10-3 e Pony Motors Dependent Failures e
Shutd own of Pony Motors e Loss of all AC Power e Faults in Hydrogen Detection System e Dependent Pony Motor e Turbine Driven Failures Combined Auxiliary Feed-with a Maintenance water Pump Fault Dependent Failures e
of Valves b
Loss of 4.4 x 10-2 9 x 10-3 e Turbine Driven Dependent Failures e
Y Offsite of Diesel Generators -
Auxiliary Feed-Power (LOSP) water pump e Loss of Chilled Water Sys.
e Diesel Generators e De pendent Failures e Normal and Emer-of Valves gency Chilled Water Systems Rupture of Main
'7.2 x 10-3 1.4 x 10-2 i
e Turbine Driven Delendent Failures e
Feedwater llcader of Valves in Auxiliary Auxiliary Feed-(I.oss of Main Feedwater System water pump Feedwater Systen) e !!aintenance Fault Dependent Failures of e
Diesel Generators Combined on Circuit with Failure of Normal Breakensand Gas Chilled Water System Coole rs e Loss of DC Power System.
e Normal Chilled A or B Combined with Water System Failure of Normal Chilled Water Systen
l REFERENCES 5.1 CRBRP Risk Assessment Report, CRBRP-1, March (1977).
5.2 Reactor Safety Study, "An Assessment of Accident Risks in U.S. Commercial Nuclear Power Plants," WASH-1400 (NUREG-75/014), U.S. Nuclear Regulatory Commission, October (1975).*
5.3 I. A. Papazoglou and S.E.P. Gyftopoulos, "Markovian Reliability Analysis i
Under Uncertainty with an Application on the Shutdown System of the Clinch River Breeder Reactor," NUREG/CR-0405, BNL-NUREG-50804, September (1978).--
5.4 A. J. Buslik, I. A. Papazoglou, and R. A. Bari, " Reliability of the CRBRP Shutdown Heat Removal System," Paper V.2 in the Proc. of the ANS/ ENS /0 ECD Topical Meeting on Probabilistic Analysis of Nuclear Reactor Safety, May 8-10, 1978, Los Angeles, California.
5.5 Clinch River Breeder Reactor Plant, Preliminary Safety Analysis Report, Project Management Corporation.
5.6
" Loss of Offsite Power, Survey Status Report, Revision 3," prepared by Raymond F. Scholl, Jr., enclosure to a Memo from D. Crutchfield, Chief, Operating Reactors Branch #5, Division of Licensing, USNRC to D.
Crutchfield, Acting Chief, Systematic Evaluation Program Branch, Division of Licensing, dated September 25, (1980).
5.7 K. Jamali and W. Kerr, "Probabilistic Methods for Accident Progression Analysis," paper presented at the International ANS/ ENS. Topical Meeting on Probabilistic Risk Assessment, September 20-24, 1981, at Port Chester, New York (proceedings in print).
5.8 K. N. Fleming and P. H. Raabe, "A Comparison of Three Methods for the Quantitative Analysis of Common Cause Failures," Paper X.3 in the Proc. of the ANS/ ENS /0 ECD Topical Meeting on Probabilistic Analysis of Nuclear Reactor Safety, May 8-10, 1978, Los Angeles, California.
- 5. 9 WARD-D-0118, " Reliability Assessment of CRBR Reactor Shutdown System,"
Westinghouse Electric Corporation, November (1975).
4 6.
LOSS OF HEAT SINK This accident scenario differs from other core disruptive accident sce-narios in that it is assumed that the reactor is scrammed and thus, initially the only heat source is the decay heat of the fuel.
Furthermore, it is assumed that no heat sinks exist and thus eventially the coolant boils, cladding and subassembly walls melt and relocate forming blockages and eventually the fuel compacts. This fuel compaction may lead to a critical configuration in a fast reactor. At this point the heating is not only due to decay heat but also due to fission heating.
In an analysis of such a scenario for the homogeneous core-CRBR it was found (Ref. 6.1) that all relocation patterns of the fuel and con-trol material led to the formation of a critical configuration.
Figure 6.1 shows a contour plot of active core height and control subassembly height for multiplication factor (ke)=1.
It is seen that a large area of the map is ex-cluded due to compaction limits.
Furthermore, it is seen that no compaction could reach the compaction limit without first passing through the k =1 e
contour.
In the analysis mentioned above, the natural circulation of the coolant was not determined in great detail.
It is clear from an analysis carried out subse-quently that considerable recirculation flow loops are set up (Ref. 6.2) (up hot subassemblies and down cooler subassemblies) in cores experiencing such an acci-dent scenario.
This phenomena is illustrated on Figures 6.2, 6.3 and 6.4 which show the channel assignments, flow fraction and maximum sodium temperatures of a heterogeneous core following a LOHS immediately after scram (power is assumed to be 6%).
From Figure 6.3 it can be seen that there is substantial reverse flow in channel 19 (control subassembly) and a low flow in channels 1, 2 and 9.
Fig-ure 6.4 shows that the sodium reaches boiling temperatures in channels 1 and 8 approxinately 90 seconds after the accident is initiated.
These channels do not represent the hottest or coolest subassemblies, but some intermediate subassem-bly, whose flow rate is not great enough to provide sufficient. cooling.
The spatial incoherency in the onset of coolant boiling implies an incoherency in the location of core damage, (steel and fuel melting and relocating) which af-fects the progression of the accident in both space and time.
If, for example, a highly inccherent situation arises, the possibility of a recriticality could be avoided if the fuel has a chance to drain away in selected position before -
the remainder of the core compacts.
Furthennore, the amount and timing of internal blanket material relocating could affect the progression of the acci-238U and would thus tend to retard the dent, these being nostly composed of formation of a critical mass.
In the event that the fuel drains out of the core region, thus reducing the probability of assembling a critical mass in this region, the possibility of a critical mass forming in the inlet plenum region has to be addressed.
Studies (Ref. 6.3) of this problem for the homogeneous core have indicated that if the fuel and steel are separated and collect in the bottom of the inlet plenum a determined for critical nass results.
In addition, most of the values of ke various configuration in this study have been determined to be close to.9 or l a rge r.
In view of the uncertainty (Ref. 6.4) connected with the determination of k for compacted cores, a value of ke 2,.9 should be studied with great e
Care.
In detennining the formation of the blockages due to steel relocation, care should be taken to allow for the smaller pitch / diameter of 1.072 in the internal blankets as compared to a value of 1.24 in the driver fuel region. Clearly the formation of blockages in the internal blankets will be enhanced relative to the fuel regions.
Finally, due to the incoherent nature of the boiling of coolant, melting and relocation of steel and fuel the neutronic calculations would re-quire a space-time kinetic model with appropriate feedback coefficients (Doppler and volumetric change).
Finally, the mechanical work that such an accident might imply would have to be detenained based on some terminating scenario, i.e., upper axial blanket collapse, pressure driven compaction etc.
n
REFERENCES 6.1 R. A. Bari, H. Ludewig, W. T. Pratt, Y. H. Sun, " Accident Progression for a Loss-of-Heat-Sink with Scram in a Liquid Metal Cooled Fast Breeder Reac-tor," Nucl. Tech., 44, 357 (1979).
6.2 M. Khatib-Rahbar, J. G. Guppy, A. K. Agrawal, " Hypothetical Loss-of-Heat Sink and In-Vessel Natural Convection: Homogeneous and Heterogeneous Core Designs, Proc. Decay Heat Removal and Natural Convection in Fast Breeder.
Reactors," Hemisphere Publishing Corp. (1981).
6.3 A. Busiik, "Recriticality Potential at the CRBR Reactor Vessel Bottom,"
BNL/FRS 75-7,(1975).
6.4 D. hade, " Critical Experiments on Severely Damaged Cores," Presentation to ACRS, September 27,(1978).
l P.
e 1.0 0.9
- k. < t.O O.8 E
~
s 0.7 I
O G O6 I
k, > l.O
- O.5 O
LIMITING CCRE 0.4 - H EIGHT
~
w 3
4 g 0.3 O.2 LIMITING CONTROL O.1 SU9 ASSEMBLY HEIGHT 1
f f
1 f
I f
I O
O.2 0.4 0.6 0.8 1.0 CONTROL SUSASSEMBLY HEIGHT (m)
Figure 6.1 Contour Of k =1.0.
e (308 12 52 3 s.3 lit [\\
2 a x +0.e*FsC3%6 20%f is 18 1
E sw SSEveLY tuwete 307
\\ iF it il XX CMam%El, hWMSte 30 44 302 30 Sa i 3ol [\\.,s / io \\ ir /..30S [ no \\ 47 i4 3:0 So \\ 14 IS S
si 4
S. 2 /,3, 304 2:0
\\ i. / 3. \\ i. / go,/;o \\ ;ir / ;,
\\ ir / i, \\
3 3 \\ i / ; \\ i4
'2
[46 \\ [ as io \\ 88 II
'4 4
42 2
205 283 44
\\4 [4F \\ i3
,N S
7 2
S le it IS i'
206 214
- 2. 7 102 /ac,t \\ ii / e.
2 1
S ir i
t 26 207 37
[ 6. k4
[ 36 t
2 16 9
17 42 6
S t
10 9,6
\\3 /
1 2 04
\\ iS 20.s 0 4 io 2
0 es j c, i at 25 202
/,'. \\ " / AW
,i, io / i \\ i3j io /
2 i
6 eO IS )
/ 4 \\
.0
- c. ; io f rect 33 io -
cx S
\\S / 2 \\.
s';
i
,2
/
S 1
2
/3,\\','
r S
12S 2 /
3 r '
( S.S
.t.
.e, 2
\\'.'/
( 41
/
3 Figure 6.2 Channel assignments for thermal-hydraulic calculation.
.~
0.02 i
i i
i i
i i
HETEROGENEOUS CORE DES 0.01
< 9; b
2 p
O. O o
<C T
-0.Ol l
y ' li.
3
~
o
-0.02
_I u_
<- 19
-0.03
-004 O
20 40 60 80 900 12 0 14 0 160 180 TIME (s)
Figure 6.3 Flow fraction.
s
1400 i
i i
i i
i i
i HETEROGE'NEOUS CORE DESIGN i
1300 BOILING d
z l
e W
1200 mo H
8 c <t II00
? CCu CL s 1000 w
900 800 I
l 0
20 40 60 80 10.0 12 0 14 0 16 0 18 0 TIME (s)
Figure 6.4 Sodium temperature.
+
l O
O V
g
7.
REACTOR PHYSICS The CRBR core design has been revised and the new layout is outlined in Amendment 51. Briefly, the core layout has been changed from a homogeneous core with two enrichment zones (inner core N 17%, outer core N 27%) to a heterogen-eous core with a single driver fuel enrichment (32.8%) and approximately two internal blanket zones.
These two designs are shown as Figures 7.1 and 7.2, which illustrate the two enrichment zones of the homogeneous core and internal blanket zones of the heterogeneous core, respectively.
From a safety point of view the revised core design implies different reactivity feedback coefficients.
and thus different response to operational transients and behavior under acci-dent conditions.
Table 7.1 compares the Doppler coefficient for the two cores, ignoring the radial blanket and the upper and lower axial blankets at BOC.
From the table it can be seen that the major contribution to the Doppler coefficient in the heter-ogeneous core comes from the internal blanket while in the homogeneous core the major contribution comes from the inner core zone.
Thus, although the total co-efficient is higher in the case of the heterogeneous core the fuel region has a much lower coefficient than the homogeneous core which is all fuel. At BOC, this difference could be significant, since the internal blanket zones generate a relatively small fraction of the power and any power rise due to a transient will first occur in the fuel.
At EOC the situation is not as clear since the power fraction generated by the internal blanket has increased and at this time some combination of the coefficients (fuel and internal blanket) will play a role.
Table 2 shows the sodium void reactivity for the two cores at B0C1 and E0C4.
This table shows that the total sodium void reactivity is higher for the heterogeneous core compared to the homogeneous core.
However, in the case of the heterogeneous core the fuel and internal blankets contribute approximately equally to the total, while in the case of the homogeneous core the inner core plays a dominant role.
It is seen that the sodium void reactivity associated with the fuel region in the heterogeneous core is much lower than that asso-ciated with the inner core region of the homogeneous core.
In the event of an accident leading to sodium voiding, the heterogeneous core is expected to start voiding in the fuel zones at BOC, since the blankets are overcooled at this time.
In the homogeneous core, the void generally starts in the inner core. -
Thus, at BOC the initially added reactivity due to sodium voiding is much larger in the homogeneous core than in the heterogeneous core.
At E0C the situation is not as clear as voiding could start in the internal blankets of the heteroge-neous core (depending on the orificing) and then move to the fuel region thus l
changing the space-time dynamics of the power shape.
The spatial voiding pat-tern is very important in determining the accident sequence.
The above discussion on feedback coefficients implies that a redetermina-tion of these together with steady state power shapes, and a value of Seff a
characteristic of this core is required.
Furthermore, a determination of the.
(
steady state condition of the core would be required as a starting point for the analysis of transients and accident scenarios (scrammed or unscrammed).
The possibility of a re-criticality during an accident in a fast reactor requires a steady state configuration even in a scrammed reactor.
The calculational capability to carry out such an analysis is in place at BNL.
Table 7.3 is a tabulation of fast reactor physics codes available at BNL and their status as of November 1980.
However, in order to have confidence in the adequacy of the analysis methods, they should be benchmarked against selec-ted heterogeneous core critical experiments (ZPPR-7 and ZPPR-8).
k J
l Table 7.1 Doppler coefficient for B0C1 (-T $$ x 10-4).
dT e
Heteroceneous Core s
Fuel Internal Blanket Total 25.8 44.0 69.8 Flooded 16.6 35.4 52.0 Voided Homogeneous Core Inner Core Outer Core Total 44.3 13.7 58.0 Flooded 23.1 8.1 31.2 Voided i
Table 7.2 Sodium void reactivity ($).
Heteroceneous Core 3
Fuel Internal Blanket Total BOC1 1.51 1.4 2.9 E0C4 2.31 1.64 3.95 Homogeneous Core Inner Core Outer Core Total l
80C1 2.71
- 0.83 1.88 l
E004 3.55
- 0.21 3.34 e
I 4.
Table 7.3 Fast reactor physics codes available at BNL.
Multio_rouc Cell Code's 2
Operational - Multigroup cross-section preparation code.
MC -11 Operational - Multigroup fast reactor cell code (slabs and
~
50X cylinders)
Operational - One-dimensional cell calculations for resonance RABBLE i
range
~
Operational - Cross-section preparation code MJOY Transoort Theory Operational - One-dimensional Sn code.
AMISM Operational - Two-dimensional Sg code TWOTRAN 00T 4.2 Operational - Two-dimensional Sn code
& 3.5 MonteCahlo Operational (needs modifications) -' Monte Carlo code for reac-MORSE 1
tor analysis, using multigroup library Diffusion Theory Operational - One-dimensional code used for crcss-section 1-0X preparation Operaticnal - Similar to 1-DX, more flexible SPHINX Operational - One-dimensional burnup code (limited to 13 SIZILE
~
groups) 203 Operational - Two-dimensional code with burnup option; several revisions. available allowing for upscattaring and multi-chain fission product models.
S Table 7.3 (c::nt'd) 31F-30 Operational - Three-cimensional ccde 30B Operational - Three-dimensional code PERT-V Operational - Perturbation theory code, interfaces with 2-03
~
Core Dynamics with Feedback V
FX-2 Running - needs checking; two-dimensional space-time kinetics code t..
Fission Product Codes CI ! DER Operational - Study fission prcduct models TCAFE'd Operational - Collapses multigroup fission product library for use in cit! DER,1-0X or 2-DB l
i l
l l
t
. i 1
', j
__7_
~
~
~
~
>- l
,/,,'O ',
. y f
^
\\ y /,,,,,
/, !
gj,' /,,' '// s P
\\\\
P
'//
/j 7p l=
- ' / b Ns\\ f IET _-
S
\\
/'!
s \\
,,, //.
m,,
\\\\
y/,,
3 s
fj,
'm
. -7/,
\\\\
//,^'
P '\\
\\\\
/)
s
' [ y //
'\\
\\
\\
' g \\Y 4
' I,1
/,
3 s
f
\\\\
///, ~
' \\'(s \\\\P,("/f;-,
H
~/'//
\\\\
\\
\\\\
\\
' ~ "///
\\\\
\\\\
S
\\\\
s
//t ' //
s \\' '
//
\\
y
,,,n,gi;!_//,'p/s ; o
~'
h INNER CORE h RADIAL SHIELD 324 10 8 h OUTER CORE h PRIM ARY CONTROL. 'I5 l
90
@ RADI AL BLANKET
@ SECONDARY CONTROL 150 4
-l Figure 7.1 Homogeneous core layout.
/,
,,/
//
fx
'/
(f "d/'
/
./
,D-
/g
/
9
'/
, //
/
/
,,,l/
'l/
gA
, //
/
gg
/
/
'./
,, 'g.
,,, N/ '/:' /, y 4,'4 o' /,, o '// / l, /,, J' 4 'E // C y s. Af/ h l / / h 6 ALTERNATE FUEL BLANKET 0156 FUEL ASSEMBLIES Q'/f ASSEMBLIES A\\ Q 76 INNER BLANKET ASSEMBLIES @ 15 CONTROL ASSEMBLIES Q132 RADIAL BLANKET ASSEis1BLIES30G RADIAL SHIELD ASSEMBLIES 4 Figure -'.2 Heterogeneous core layout..
8. CDA ENERGETICS: TRANSITION PHASE LICENSING ISSUES This section provides a discussion of issues related to the transition phase of core disruptive accidents. BNL has an NRC-sponsored experimental program which has produced key information related to phenomenon involved in the transition phase. 8.1 NRC Staff Position The NRC staff position with respect to the transition phase was outlined g in NUREG-0122 (Ref. 8.1). The core meltdown phase of the CDA was created by. attempting to identify recriticality scenarios during the meltdown sequence. The objective of this approach was to determine the potential for recriti-cality events. It was recognized that a mechanistic treatment of this portion of the CDA was not possible due to a lack of analytical tools. On the basis of the analysis it.was concluded that "... although there is potential for large reactivity insertion during meltdown, there does not appear to be suffi-cient driving forces to cause a sustained prompt-critical excursion which is not enveloped by the present staff position." (The staff position specified that CRBR should accomodate a "... work-energy release of 1200 MW-sec based on fuel vapor as the working fluid and expansion to one atmosphere.") 8.2 CRBR Project Position The current position of the CRBR project (Refs. 8.2, 8.3) is that the transition phase is the most likely progression for the LOF-CDA-core meltdown. It is argued, that this leads to a non-energetic accident termination. Recriticalities are possible but incoherencies prevent them from becoming " sustained" and energetic. Large-scale molten pool development is unlikely since fuel losses away from the core region are likely to occur prior to development of such pools, b 2 The CRBR Project position is based upon a combination of separate neu-tronic and thermal-hydraulic computations. The latter computations are based upon fuel relocation and volumetric boiling molten pool models. Details of the calculational methods have not yet been supplied.. - - _ _ _ -. - - - _ - --
8.3 Major Chances Durina Past Fou-Years Two major factors have been irtroduced since 1977 which could affect the approach to transition phase issues involved in the licensing process. Fi rst, the design has gone from the homogeqeous to the heterogeneous core. This, it is claimed in CRBR-3, leads to a greater likelihood that the transition phase path would be the path to CDA termination rather than early disassembly. This y argues that closer scrutiny should be placed on analysis of the transition phase se-quence for the heterogeneous CRBR. The second factor is that the SIMMER (Ref. 8.4) code is now available for analysis of transient phase se-quences. This will allow a more mechanistic treatment of the transition phase than was possible earlier. 8.4 Issues to be Resolved The technical issues with respect to transition phase analysis have not changed over the past four years. Uncertainties remain.is such key areas as: mechanism of fuel disruption, molten material relocation and solidification, behavior of molten boiling masses of fuel and steel, etc. These uncertainties will not be resolved on a short time scale. While SIMMER is available, it is ~ Some out-of-pile simulant experi-recognized that it is largely unverified. I ments have been performed to help guide application of SIMMER in transition phase calculations. However, there have been no in-pile prototypic fuel as-sembly meltdown experiments performed which would provide a meaningful level of verification of the code for licensing application. The technological uncertainties and shortcomings described above suggest t that the question of accident energetics will not be resolved on the short term. Analysis of the transition phase progression for the heterogeneous CRBR is, however, required to evaluate the project position and to provide support for development of a NRC licensing position. The approach should be two-fold: (i) A phenomenological analysis of the transition phase for the heter-ogeneous CRBR should be performed, which parallels the effort in the BNL transition phase assessment report (done for the homogeneous core) (Ref. 8.5). This approach applies phenomenological arguments to trace possible accident proaression paths. Tine scales
,f. / for important events are-computed and potential paths to recriti-cality are identified. L (ii) Accident sequence. calculations should be performed using the SIMMER code. These calculations should be coupled to initiation phase l calculations, and should be guided by available experimen_tal' data relevant to transition phase phenomenon. g; . g,. 8 l 1 O J .c REFERENCES 8.1. J. F. Meyer et al., "An Analysis and Evaluation' of the Clinch River 9s Breeder Reactor Core Disruptive Accident Energetics," NUREG-0122, March
- 1 1977.
8.2. R. M. Stark, " Summary of Hypothetical Core Disruptive Accident Con-y siderations," Presentation to NRC, Bethesda, Md., October 1981. N 8.3. CRBRP, " Hypothetical Core Disruptive Accident Considerations in CRBRP," Volume 1: Energetics and Structural Margin Beyond the Design Base," CRBRP-3, January 1979. "O 8.4. C. Bell et al., " SIMMER-I: An S Implicit, Multifield, Multi-n Component, Eulerian, Recriticality Code for LMFBR Disrupted Core An-alysis," LA-NUREG-646741S, January 1977. 8.5. G. A. Greene et al., " Assessment of the Thermal-Hydraulic Technology of the Transition Phase of a Core-Disruptive Accident in a LMFBR," BNL-NUREG-27366, February 1980. t l
4 9. STRUCTURAL ANALYSIS The primary objectives of this section relate to assessing the structural adequacy of the Primary Heat Transport System (PHTS), the reactor vessel, the closure head as well as the Direct Heat Removal Service (DHRS). Such an as-sessment should be made for all accident conditions including seismic loadings i (the safe shutdown earthquake), and core disruptive accident loads. These loads should be combined with the loads from the duty cycle. An adequate as-sessment is particularly needed for the plant towards the end of its useful 6 cycl e. The issues that need resolution are: (a) Integrity of the PHTS piping network for a number of loadings inclu-ding thermal / pressure (static) and seismic disturbances.(dynamic), (b) Integrity of the DHRS piping network with its convection to the Re-actor Vessel for loadings noted above, j (c) DHRS system does not use the guard vessel or sleeves around the prim-ary coolant (radioactive). Adequacy of this concept needs clear as-sessment. k C l, j
Table 9.1 Issues in the Structural Analysis Area Item Issues and Comments Primary Piping System Piping network analysis for a combination of loads, y particularly towards the end of its useful life (de-signed plant life-30-years) ~ DHRS DHRS system does not have guard vessel or sleeves around pipings that carry radioactive sodium. Sub-stantial quantity of radioactive sodium can leak in the containment. Recent CRBRP analysis for thermal margin beyond design basis (TMBDB) accident did not make any assessment of large leaks through DHRS early in the core disruptive accident scenario. Need to assess integrity of the DHRS inlet and outlet nozzles for CDA loadings. Containment Performance Need to assess performance of the reactor containment building subsequence to sodium fire and also subsequent to core-concrete-sodium interactions. l 4
S 10.
SUMMARY
This report has identified several technical issues that will require resolution during the course of the forthcoming licensing review of the CRBR. It is clear that the new amended PSAR will need an evaluation. For operational transients, a careful review will be needed of the selec-I tion of initiating events, of the crit?-ia put forth for acceptance of the con-sequences of events, and of the method. and tools used in the evaluation of the consequences of postulated events. Particular attention should be given to the 1 impact of the new core design on these technical issues related to operational transients. An evaluation of the shutdown heat removal system capability is needed. Particular attention should be given to redundancy and diversity of heat re-moval paths and to the capability of the systems to remove shutdown heat under a spectrum of transient conditions. The potential capability of the heat re-moval system to remove shutdown heat under natural circulation conditions should be reviewed. The available results from the natural circulation test program for the Fast Flux Test Facility should be evaluated and assessed for their implications for the CRBR configuration. In the area of containment heat removal, a review must be performed of the CRBR Project's approach to design basis sodium spills. Additional analyses, if needed, should be identified as a result of this review process. The analysis and philosophy associated with Thermal Margins Beyond Design Basis (TMB08) must be evaluated. The heat removal capability of the containment to a spectrum of severe core damage conditions should be evaluated. An assessment should be performed of the interactions of sodium with concrete to establish reaction and penetration rates. The risk assessment report that was prepared by the CRBR Project in 1977 has not yet been evaluated. In addition, the reliability of th,e shutdown sys-tem and of the shutdown heat removal system have not been evaluated for the new design changes. Review activities on these topics should be commensurate with the role that reliability and risk assessment will play in fiRC's licensing re-view of the CRBR.
The heterogenous core design will need review and evaluation for reactor physics concerns as well as for more general core disruptive accident issues. As the licensing review of the CRBR goes forward, it is expected that some resolution will be achieved on several of the aforementioned technical issues, t t O k' t i 1 't ,v, ,,,.,.,n --s nw w ,wv <--e-,,w,-,-- m-r e,-
t ACKNOWLEDGEMENTS The authors are grateful to R. J. Cerbone and M. Khatib-Rahbar for helpful remarks on this report. 4 e 9 DISTRIBUTION LIST BNL Distribution CRBR Task Force (11) Division Heads / Associate Chairmen (6)- Deputy Chairman (1) Chairman (1) Nuclear Safety. Library (2) NRC Distribution W. Morris (6) P. Check (1) J. Long (1) S. Sands (1) H. Holz (1). J. Hanson (1) C. Allen (1) R. Becker (1) L r i l l l
~ INTERIM REPORT Alk$ (30te ank or kas ocak SS/Sfeneet fW - i Accession No. Contract Program or Project
Title:
Case Review'of CRBR Heat Transport System Design e Subject of this Document: Review of the Status of CRBR Licensing Technical Issues Related to Heat Removal System and Severe . Accident Analysis Type of Document: Infonnal Report Author (s): R. A. Bari, et al. Date of Document: April 1982 Responsible NRC Individual and NRC Office or Division: William Morris CRBR Program Office Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Coimiission 1 Washington, D. C. 20555~ 1 This document was prepared primarily for preliminary or internal use. It has not received full review and approval. Since there may be substantive changes, this document should not be considered final. j Brookhaven National Laboratory Up ton, New York 11973 Associated Universities, Inc. for the U.S. Department of Energy l Prepared for U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Under Interagency Agreement DE-AC02-76CH00016 FIN-3360 m}}