ML20040G056
| ML20040G056 | |
| Person / Time | |
|---|---|
| Site: | LaSalle  |
| Issue date: | 02/09/1982 |
| From: | Sargent C COMMONWEALTH EDISON CO. |
| To: | Schwencer A Office of Nuclear Reactor Regulation |
| References | |
| RTR-NUREG-0737, RTR-NUREG-737, TASK-2.F.2, TASK-TM NUDOCS 8202110189 | |
| Download: ML20040G056 (25) | |
Text
, ;
. Commonwealth Edison O' ona Fust Nition11 Nta, Chicago, lihnois C ] Addrsss Riply to: Post Office Box 767 g Chicago, Illinois 60690 Februa ry 9, 1982 a g 3
s? " sN
/?
j te6':%&O n.
.g Mr. A. Schwencer, Chief -
g CO --
Licensing Branch #2 ;94 ,t/h' lg '
Division of Licensing Le /
U. S. Nuclear Regulatory Commission Washington, DC 20555 {f p 4 w
Subject:
LaSalle County Station Units 1 and 2 Inadequate Core Cooling NUREG 0737 Item II.F.2 NRC Docket Nos. 50-373 and 50-374
Dear Mr. Schwencer:
The purpose of this letter is to provide Commonwealth Edison Company 's position relative to the Inadequate Core Cooling (Incore Thermocouple) issue as it applies to LaSalle County Station.
Mr. R. J. Mattson requested the LaSalle County Station input on this subject during a meeting with the BWR-Owners Group on January 27, 1982.
As a member of the BWR-Owners Group, Commonwealth Edison not only participates in the evaluation o f Inadequate Core Cooling (ICC) but also commits to endorse any practical and cost-ef fective resolution to the ICC questions as approved by the NRC staff.
Until the final resolution is implemented LaSalle County Station will address the problem of inaccurate water level indica-tion due to boil-off by emergency procedures and operator training.
As part of the development o f the procedures, the applicable recommendations of G.E. SIL 299 Supplement 1 were incorporated.
Enclosed for your information is a revised edition of Appendix L.31 of the LaSalle County Station FSAR. In this revision an attempt was made to organize the material into four separate positions on ICC in contrast to the chronological study positions previously reported. This document is currently undergoing internal-review and is expected to be incorporated in an Amendment to the FSAR in early March.
hO s
II C202110189 820209 PDR ADOCK 05000573 A PDR
e A. Schwencer February 9, 1982 If there are any further questions in this regard, please contact this office.
Very truly yours, bb C. E. Sargent Nuclear Licensing Administrator 1m Enclosure cc: NRC Resident Inspector - LSCS (w/o Attachment) l 3420N
~ '
5 -
LSCS-FSAR AMENDMENT 56 HAY 1981 L.31 INADE00 ATE CORE COOLING INSTRUMENTS III.F.2)
FUEL LOAD AND LOW POWER TEST REOUIREMENT:
Licensees shall provide a description of any additional instru-mentation controls (primary or backup) proposed for the plant to supplement existing instrumentation (including primary coolant saturation monitors) in order to provide an unambiguous, easy-to-interpret indication of inadequate core cooling (ICC). A description of the functional design requirements for the system shall also be included. A description of the procedures to be used with the proposed equipment, the analysis used in developing these procedures, and a schedule for installing the cquipment shall be provided.
Chances to Previous Requirements and Guidance .
(1) Specify the " Design and Qualification Criteria" for the final ICC monitoring system in section, " Clarification" (items 7, 8, and 9), Attachment 1, and Appendix A.
(2) Specify complete documentation package to allow NRC evaluation of the final ICC monitoring systems to begin
, on January 1, 1981.
f3) No preimplementation review is required but post-implementation review of installation and ;
preimplementation review before use as a basis for operator decisions are required. .
(4) Installation of additional instrumentation is now required by January 1, 1982. .
(5) Clarification of item (6) has been expanded to provide licensees / applicants with more flexibility and diversity in meeting the requirements for determining liquid level ,
indication by providing possible examples of alternative methods.
Previous guidance on the design and qualification criteria for upgrading of existing instrumentation was based on Regulatory Guide 1.97, which is still being developed. Detailed design requirements for incore thermocouples and additional instrumentation were not specified. The pertinent portions of draft Regulatory Guide 1.97 have now been included as Appendix A.
Design requirements for incore thermocoup..es used 1 in theonly).
(PWR's ICC monitoring system are specified in Attachment The only significant change in design requirements involves a relaxation of qualification requirements for display systems This facilitates procurement of amenable to computer processing.
computer systems ar.d makes feasible the use of cathode ray tube (CRT) displays that may be aeeded for proper interpretation This of some reactor-water-level systems under development.
L.31-1
I
~
AMENDMENT 56
- LSCS-FSAR HAY 1981 ,
relaxation can be accomplished without compromise of ICC monitoring reliability by requiring 99% availability for the display systems, by requiring postaccident maintenance and by accessibility for noncedundantICC portions of the that monitoring system, include relying on diverse methods of com'pletely qualified display systems.
The staff has concluded that1981 the previous installationfor additional instrumentatio
- requirement of January 1, unrealistic for most licensees, due to procurement a development problems associated with proposed measurementthe staff methods.' Further, acceptable for use until development programs have been completed. .
Clarification Design of new instrumentation should provide an (1) unambiguous indication of ICC. This may require new measurements or a synthesis of exisiting measurements which meet design criteria (item 7).
(2) The evaluation is to include reactor-water-level -
. indication.
k3)
Licensees and applicants are required to provide the necessary design analysis to support the proposed final instrumentation system for inadequate core cooling and to evaluate water theand level merits of various to monitor 'nstruments other parameters toindicative monitor of core-cooling conditions.
(4)
The indication of ICC must be unambiguous in that it should have the following properties:
(a) It must indicate the existence of inadequate corehigh-void cooling' caused by various phenomena (i.e.,
fraction-pumped flow as well as stagnant boil-off);
and, (b) It must not erroneously indicate ,ICC because of the presence of an unrelated phenomenon.
(5)
The indication must give advanced warning of the approach of ICC.
(6)
The indication must cover the full range from For normalwater-example, operation level to complete core uncovery. instrumentation may be chosen to pr warning cf two-phase level drop to the top of the core and could be supplemented by other indicators such as incore and core-exit thermocouples provided that the indicated temperatures can-be correlated to provide L.31-2
. . LSCS-FSAR AMENDHENT 56 MAY 1981 .
indication of the-existence of ICC and to infer the extent of core uncovery. Alternatively, full-range level
- instrumentation to the bottom of the core may be employed in conjunction with other diverse indicators such as core-exit thermocouples to preclude misinterpretation due to any inherent deficiencies or inaccuracies in the measurement system selected.
(7) All instrumentation in the final ICC system must be -
. evaluated for conformance to Appendix A, " Design and
- Qualification Criteria for Accident Monitoring Instrumentation," as clarified or modified by the provisions of items 8 and 9 that follow. This is a new requirement.
(8) If a computer is provided to process liquid-level signals for display, seismic qualification is not required for the computer and associated hardware beyond the isolator or input buffer at a location accessible for maintenance following an accident. The single-failure criteria of item 2, Appendix A, need not apply to the channel beyond the isolation device if it is designed to provide 99%
availability with respect to functional capability for
- liquid-level display. The display and associated . .
hardware beyond the isolation device need not be Class.IE, but should be energized from a high-reliability power source which is battery backed. The quality assurance provisions cited in Appendix A, item 5, need not apply to this portion of the instrumentation system.
This is a new requirement.
(9) Incore thermocouples located at the core exit or at discrete axial levels of the ICC monitoring system and which are part of the monitoring system should be evaluated for conformity with Attachment 1, " Design and
. Qualification Criteria for PWR Incore Thermocouples,'
which is a new requirement.
(10) The types and locations of displays and alarms should be determined by performing a human-factors analysis taking
,into consideration:
(a) the use of this information by an operator during both normal and abnormal plant conditions, j (b) integration into emergency procedures, (c) integration into operator training, and (d)_other alarms during emergency and need for prioritization of alarms.
i 9
L.31-3 i
AMENDMENT 60
- .. - LSCS-FSAR MAY 1981 DATED REOUIREMENT:
This requirement appliesThis to all operatingmust requirement reactors and applicants be implemented by for operating license.
January 1, 1982.
A postimplementation review will be' performed for installation, and a preimplementation review will be performed prior to use.
(Ref. 34).
See NUREG-0737 Section II.F.2 .
POSITION: ,
Initial Response:
Operational Appraisal PART A.
The includes provision of postaccident fuel level indication in t e LSCS BW installed water level instrumentation for the h
These wide range instruments measure water level control room.
above and below the top of the active(with fuel no(vessel jet elevatio 366 inches). -
conditions at zero psig in the vessel and drywellThe full height o pump flow).aone instruments which read out on indicating recorders.
instruments used for operational control The normal water level inches above the top of the active fuel are indexed some 161.5 hence a continuous precursor
.(at vessel height 527.5 inches); fuel.
above the top of active fuel to the bottom of the act .
These water level instruments they are notareproposed part of the or standa'rd additive. BWR/5 d level instrumentation;Indeed, based on the BWR Owners' Group eval i for the in Subsection 2.3.2 of NEDO-24708A,is needed for doperational co approach tonoinadequate cooling, norprovides other instrumentation for detection as accuratelevel an -
For a'BWR, ready indicationchannels of adequate cooling as does the waterUpon decreasin of data in BWR/5 s). safety response and instruments (11 water level (below Level 2), the automati: i itiate the operator's procedural response is alwayslow-pressure the same - nBecaus .
ECCS make-up to the core. no ambiguity supplied directly to the core via high-pressure or sprays or via core flood-up from the bottom (LPCI), tinent~
~
exists as to whether the level instruments relay perActually, with w cooling information.no excessive temperature rise is possible.
flooding it, ingful Ad'ditional core temperature information would not be mean because: any core d
(1) With core sprays actively cooling the core, thermocouple readout could not represant fuel or clacoul temperature; hence, ac,tions.
L.31-4
AMENDMENT 00 LSCS-FSAR ,
MAY 1981 of steam and vapor
' (2) With core flood-up, even the presenced fuel element would make a which cools a partially submergetemperature measu i
r manual initiation (3) Any corrective -
response as a follow up. t.ould requ retaken by the by the operator, ltaneous SSE and loss Even'for the unrealistic postulateRCIC including of simu and ADS /SRV which hav of all ECCS equipment, ns whenever level external energy requirements,same -provide water to the co indications signal the need.
' Group pro-The following initial evaluation by the BWR tOwners This information vides the basis for the foregoing statemenis. h mocouples for was'provided to ACRS as comment on core exGuide t t er 1.9 7.
BWR's per Draft 2 of Revision 2 to Regulatory
- 1. Technical Backoround 1.97 for requiring core exit nitorino is to indicate The reason cited in Regulatory ce of fuel cladding t they desire Guidetemocrature to b, reach.
identify m the potential for or actual occurren damage; they have The local NRC staffand hasthealso indicated oftha hot areas propagation ouples core should be utilized.
be sufficient to derectre spray (o suggested that approximately.50 thermoc I This quantity is felt by the staff of cooling toblockage conditions of 5-10%
at a high confidence level and witQuadrential differentiationth I attrition.
was thought possible with about 50 exit temperatures.
(Time Period.
1.2 Detection of Hich Core Temoerature Number 1) a [ Prior to Core Uncovery s with very strong i
vessel. The The BWR operates under saturated condit sureonphase (steam) is eous natural circulation inside the reactor prestemper -
in equilibrium (identical) "
" Addit.ional InformationBoiling Water b The GE Generic Report NEDO-24708A, 1979, concluded that, ling is assured. Required fo Therefore, dated Augustcovered with water, firct be acore adequate challenge coo threatening l ther e reactor mus t water level is the key para- Water for a cladding breach, core. Thus, d operator actions are based.
I to uncover the f accomplishment of the core The BWR/5 cool-provides l
meter level ison which also both automatic the primary measure ot an situations.
During this time period, l
multiple and (See Response to redundant 0031.287).. water leveing safety fu I
- poses j L.31-5
- MAY 1981 core exit'thermocouples would indicate, at most, the saturation temperature corresponding to the reactor vessel pressure. Core exit thermocouple readings would probably be erratically indicating lower temperatures due to the subcooling effect of ECCS (ct;9 sprays and LPCI). The use of core. exit thermocouples would not provide useful additional information for the plant operator and the erratic readings would of necessity be disregarded because saturated temperature data is of no use in assuring core coverage. With water in.the core or impingina on the core, there is no way for fuel cladding or for the fuel l
^
itself to become overh~eated.
- b. During Fuel Heat-up Following Core Uncovery (Second Time Period)
The second time period when knowledge of core exit temperature was thought to be useful was during fuel heat-up following core <
uncovery from water starvation. It is during this time that the
. potential for cladding breach exists and, depending on the duration and amount of core uncovery, the potential may exist for creating local flow blockage as a result of core damage. Reactor vessel water level provides the ability to detect core uncovery and,.thus, by itself, indicates the potential for cladding perforations. - Prior to core uncovery, automatic and operator l manual actions would already be underway to restore water level to re-cover the core. Continued monitoring of reactor water level and water makeup system performance parameters provide the capability to monitor this critical safety function.thePre-l vention of core deterioration via inventory make-up is .
priority action, rather than analysis of extent of core damage which will not alter the priority to cover the core with water.
There are many parameters available to the operator that are more reliable indicators of actual fuel clad breach than would be
-shown by core thermocouples. These include high steamline radiation, high off-gas radiation Icvels, high area radiation levels in the containment, high hydrogen concentration in the containment, and high radioactivity in reactor water or Details of these current provisions are suppression pool water.
discussea in Section 1.3 below.
Core exit temperature measurement will not provide an unambiguous -
indication of either the potential for or actual clad damage during core uncovery. This results because the BWR's multiple, safety-grade core spray systems continue to supply water spray over the top of the core even though the core may be uncovered in a bulk sense. Even if there is only one core spray system functioning (our of four provided), the core exit temperature, whether measured locally or in bulk, will not be superheated but will be at the saturation temperature for the vessel. The core sprays or RCIC need only provide 300 gallons per minute of their total typical design flow rate of 12,000 gpm to remove any superheating in the steam. In the BWR/5 and 6 designs, the low pressure coolant L.31-6
LSCS-FSAR AMENDMENT 60 MAY 1921 ,
injection (LPCI) sys tem directly floods the core bypass region, providing further subcooling. The Staff contends that these ECCS functions not be considered when determining the merits of core exit temperature measurement; that contention is unreasonable.
During fuel heat-up following core uncovery, there was initially thought to be one condition for the BWR where a core temperature measurement might* provide unambiguous and definitive information useful to the operator. This was thought to occur in the highly
-unlikely event that, following a loss of water inventory and _
absolutely no normal, emergency, or alternate water makeup systems were available to replenish coolant inventory to the pressure vessel while the reactor sensible heat level remains high. During this situation, the core is cooled by water and steam flow for a considerable period of time until the water in the core region is boiled of f. The fuel temperature begins to rise only af ter the water level has decreased to where approximately one-third of the core remains uncovered. Under such conditions, measurement of steam superheat anywhere above the core region was thought to indicate core heatup with an accompanying low water level. This postulate was the original basis for the staff requirement for incore thermocouples, the inference being that such thermocouple read-out would result in operator action.
Should this condition occur, the operator would already be taking
'all appropriate actions to restore water level above the core based only on knowledge that water level is low and/or no injec-tion has occurred. Indication of elevated core temperature, inferred later from the superheated steam indication, would not affect or change the operator 's prior response to the situation.
Propogation of core damage requires the creation of uncoolable geometry which in the BWR is constrained by the channelized fuel assemblies. The postulate of absolute water starvation simul-taneously across all channels, such that no core flow exists during boil-off, leads to the conclusion that core thermocouples could. not ' provide useful information because a core melt is postulated; however, additional information is provided below concerning core damage. propagation.
~
- c. During the Core R'ecovery Phase (Third Time Period) -
The third time period, called the recovery phase, covers the interval after the operator has restored the water level in the core region. If there were no significant core damage, core exit temperature measurement would not provide any relevant information. The possibility of thermocouples providing useful information for operator actions has been raised by the Staff for
- Subsequently, evaluated in detail as reported in Part C of this response and found not to be possible for even this case.
(See p L. 31-15) .
L !.11-7
e LSCS-FSAR AMENDMENT 60
. MAY 1981 k
. the situation when 5-10% of the core is damaged. The Staff contends that high core exit temperature readings would indicate localized propagating core damage and guide the operator in long-term decision making.
This position is not correct because: (a) once water level is restored in the core, core damage will not propagate to the rest of the core from the postulated 5-10% damaged core, and j (b) temperature readings would not provide relevant information.* -
A detailed discussion of both these points follows.
Core damage propagation, when the core is covered, has b.een i
c'iscussed in a Licensing Topical Report NEDO-10174, " Consequences of a Postulated Flow Blockage Incident in a Boiling Water Reactor," October 1974. Because each fuel bundle in the BWR core is surrounded by a flow channel, cross-flow between bundles is eliminated and any thermal-hydraulic effects of localized core f damage (within a bundle) remain localized. Each channel forms'an y
essentially independent flow path connecting the upper and lower To assure no damage to an plena and the core bypass region.
undamaged fuel assembly, the channel flow coolout requirment is
[
! less than one gallon of coolant per minute. Because there are three independent flow paths into each fuel assembly (the top and
- bottom.of the fuel bundle, and the flow paths between the bundle and bypass region around the bundle), any core damage propagation requires almost complete blockage of all three flow paths.
Calculations have shown that all three paths have to be greater than 99% blocked for any damage to result. Even if almostthis total flow blockage of the bundles were postulated momentarily, j
situation would not persist very long. Localized heating of the cladding would result in molten cladding coming into contact with I the channel wall. Such localized heating of the channel would eventually form a hole in the chann'1, thus opening another tiow path for the coolant from the bypass Calculations wereflow region also to enter performed forand the cool the fuel rods. _
situation with a 5-10.% core damage and with a postulated uncoolable geometry to determine if superheated steam can be detected in the region around the damaged portion of the core.
The calculations were done assuming the available instruments (thermocouples) were those directly adjacent to the bundles in the damaged core region. The analyses show that the heat generation rate (decay heat'and heat from the metal water reaction) in the post-recovery phase are so low that, under all
' analyzed situations, nucleate boiling would be maintained and no superheat would be measured in the bypass region surrounding the damaged core.
if a temperature sen-It has been suggesEcd by the NRC staf f that,to the assumed local blockage and if it sor was located adjacent the operator were postulated that it could indicate some superheat,This would then l response would be to restart recirculation pumps. However, force coolant through the partially blocked flow paths.
as indicated above, superheat would not be observed and the operator would have no knowledge that this action is desirable.
In addition, because of the strong inherent n,atural circulation L.31-8
l LSCS-FSAR AMENDMENT 60 MAY 1981 in the BWR, this action would be likely to be helpful for only a very limited situation where greater than 99% but less than 100%
of all available flow paths were blocked. Therefore, operator actions would be no difterent: the principal emphasis would l
i still be to maintain reactor water inventory. The addition of 50 thermocouple data readouts may, indeed, add to operator confusion such that the reliability of operator action is decreased.
- d. Mandated Incore Thermocouples for Detection of High Core Temperatures The most practical location to install mandated thermocouples in a l BWR is in the in-core power range monitor (PRM) ins trument assemblics.
All other locations (see Section 2) woulti require additional penetrations and major redesign of the vessel internals and/or the fuel bundles. A review of the temperature response of a thermocouple in the PRM assembly indicates that it would provide only a gross indication of core discharge superheat conditions, and then only for the unlikely event that no water makeup systems were operating for an extended period. But, for such a situation as discussed above, a single thermocouple anywhere above the core would provide comparably useful information as to the existence of a bulk superheat condition. Figure L.31-1 shows the response of the variables already available to the operator to quide his actions during a core uncovery. event. It also shows the expected temperature response of thermocouples in the PRM tubes when there is nc normal, emergency or alternate water makeup systems of any kind in operation. The comparisons show that the operator already has multiple and unambiguous indications to guide his -
actions during the core heatup time period.
Incore thermocouples provide no useful information for improved operator action to control water level for damage prevention in the BWR. Their diagnostic or post-event analysis function is discussed next.
l.3 Detection of Procacatino Core Damace For the worst-case assumptions related to the second time period (i.e., uncovered core and no water make-up), for which the NRC staff proposes that thermocouple indication would be useful, alternate means are available to provide trend information relating to the possible propagation of core damage. Those means which were previously available or are presently required by NUREG-0578 and which provide direct indication of propagation of core damage, with or without ECCS functional, incl.ude :
(1) reactor and suppression pool water / containment air sampling and analysis for radioactive material, (2) containment gross gamma monitoring, and (3) containment hydrogen monitoring.
- a. Analysis of reactor water samples would measure fission product activity and the concentration of dissolved hydroge.
in the reactor water. The fission product activity from the gap / plenum would be released within several minutes after the onset of. fuel clad perforations. The reactor water sampling L.31-9
l LSCS-FSAR AMENDMENT 60 !
. MAY 1981 .
i system is sufficiently sensitive to de:er: :ne hydrogen I concentration resulting f rom the reactic: :f as little as four pounds of zirconium. This is equiva;ent to a metal-water reaction involving about 3% of tr.e :: adding of a single l fuel bundle.
For the dry-core case, vessel _ depressurica:icn is expected.
It will occur naturally if the event is in::iated by a primary system break of sufficient size. :: will occur by automatic or manual actuation for the no ~ :reak or small-break case via safety-relief valvo actuation. Thus, for the entire spectrum of initiating events, indicat. ion f core .d'amage is provided by various instruments in the ccn:ainment. These include the suppression pool water / containment air sampler system, containment gross gamma monitor, and the containment hydrogen monitor. The gross gamma moniter can detect fuel clad gap / plenum activity release within several minutes from the onset of clad perforation.
Radioactivity due to noble gases alone sh: Id provide suf ficient indication of propagation of c:re damage. For the relatively straightforward case involving blockage of a single fuel assembly during normal plant :eration, analysis 1
(NEDO-10174 previously cited) shows that within 9 seconds, fuel l element melting would be detected by the steamline radiation monitor; scram and steamline closure would follow within 4 seconds. The of f-gas radiation monitor wcu'_d alarm within 2 minutes.
The more complex case involving main stgar isolation valve (MSIV) closure for reasons other than highFor steamline that case, the radiation has also been investigated.
saf ety relief valves open within seconds to relieve vessel pressure, and noble gases are transported Jia the saf ety/ relief valve discharge piping to the suppression pool water, thereby being released to the containment wetwell.
The results of this analysis are illustrated in Figure L.31-1 for the situation in which all reactor water rakeup systems (normal, emergency, and alternate) are postulated to remain inoperative forl an extended period. Eventually the water level is reduced such For the that the readings on all thermocouples would increase.
situation in which the bulk water level has been significantly _
reduced there would be little or no correlation between In thermocouple readings and core-area cross sectional heating.
this case, the insufficient reactor water inventory would affect i
all fuel assemblies independent of whether or not local blockage
! exi'sts. The extent to which actual fuel failures progress could only be assessed by monitoring fission gas release to the primary f
' system or to the containment. Gross gamma monitoring snculd I
L.31-10
~
LSCS-PSAR RMENDMENT 60
- MAY 1981 provide a more rapid indication of progressive core damage. It is noteworthy, however, that any such information does not result in a different operator response; water injection for core recovery is the only corrective action. Continuing indica-tions of the rate of progressive core damage can be obtained from the suppression power water / containment air sampler systems as well as the containment hydrogen monitor. The containment hydrogen monitor is expected to be sufficiently sensitive to detect a core damage threshold as low as 2% to 3% of a core-wide metal-water reaction per day.
Again, this information is not operational data but post-event analytical data.
- 2. Design Considerations for Thermocouple Placement Three possible locations for thermocouples within a BWR were evaluated. These are: within or on the fuel assembly; on the shroud head with leads projecting downward to near the fuel assembly discharge; and in the PRM assemblies. While detailed design investigations have not been performed, the first con-cep t is considered unacceptable because it would create local- l
. ized flow disturbances and cladding stress concentrations with an accompanying potential for initiating fuel damage. The first and second options are both considered unacceptable due to the interaction created between the thermocouple lead supports and the ECCS functions--specifically core spray.
Multiple penetrations are required to route the thermocouple leads into the core. These options also significantly impact I the duration of each refueling outage. For both, the required number of thermocouples would be large because the BWR utilizes a channeled fuel design which as previously discussed, localizes core damage and prevents propagation across the whole core.
Only placement in the PRM assemblies is conceptually feasible I without ext.cnsive-plant redesign.
19 the BWR/5 design, the PRM assemblies are secured to the top grid within the vessel. The top of the PRM latches approx-imately 10 inches below the top of the channel of the fuel assembly. The PRM latching mecahnism design precludes locating the thermocuoples higher than approximately 13 inches below the top of the fuel channel. A detailed evaluation of the time response use of thermocouples in the PRM assemblics indicated an unacceptable time lap and spacial dependency.
1 Conclusion It has been shown that core exit thermocouples in a BWR can-not provide operationally meaningful data regarding inadequate core cooling, not only because such thermocouples cannot indicate true fuel conditions in the BWR saturated steam environment with L.31-ll
LSCS-FSAR AMENDMENT 60 MAY 1981 db or without the ECCS core sprays, but also because corrective operator action requires inventory make-up via ECCS initiation for every degraded vessel inventory condition including the extreme instance should total loss of vessel water level indi-cation occur.
The mandated addition of quadrentially located thermocouples to the BWR/5 reactor would result in extensive instrument modifications, addition of needless RPV and primary containment penetrations to provide reasonable core coverage for relative comparisons (quadrential) of core cooling in a transient degraded situation, and would yield ambiguous temperature indications on which prudent operational decisions could not be based. Such indications should not be fed into the control ,
room to confuse the operators. Moreover, even it constrained to a non-operational, i.e., analytical, purpose, their use-fulness is questionable.
No realistic postulate exists for a combination of events where core outlet thermocouples can yield operationally significant data for BWR's. As an analytical aid, core exit
,thermocouples in a BWR core can contribute only a gross indi-
- cation of an after-the-fact condition for which existing safety equipment provides more direct information on which correct operator response is based and which actions assure adequate public health protection. For the above reason, CECO does not recommend the installation of incore thermocouples at La Salle.
L.31-12
PART B. SECOND RESPONSE: VALUE-BENEFIT & RISK ASSESSMENT OVERVIEW This report provides an evaluation of the value/ benefit of incore thermocouples in the LaSalle reactors.
The regulatory staff via NUREG-0737 Item II.F.2 on Inadequate Core Cooling has suggested that the addition of incore thermocouples might improve post-accident monitoring of BWR core conditions. Part A of this response evaluated the operational significance of core thermocouples in BWRs and concluded that incore thermocoup,les are not justified in the LaSa'ie reactors because:
- 1) They do not provide useful additional inf ormation on w hich credible operator ctions can be based; indeed, automatic system responses e..d/or execution of operator emergency procedures obviate any operational need for thermocouples in a BWR, and
- 2) The read-out from the in-core thermocouples in a BWR cannot indicate fuel integrity status under normal or operational scenerios nor in accident / transient situations; alternate information from existing containment monitoring systems and water / air sampling systems can provide timely indication of core degradation thresholds.
This report concludes that incorporating thermocouples in the LaSalle reactors entails an unwarranted increase in occupational exposure and an unnecessary cost is encountered with no resultant decrease in public health risk.
1.0 Introduction As a result of the Three Mile Island - 2 accident, the NRC perceived a need for all reactor types to use in-core thermocouples l
to provide an unambiguous, easy-to-interpret indication of inadequate core cooling. The basis for this new requirement was that in-core thermocouples were of some value in determining core conditions following the accident. This position was promulgated j as a requirement via Item II.F.2 of NUREG-0737.
Almost concurrently with this NI'3EG-0737 requirement, Regulatory
- Guide 1.97 Revision 2, Draf t 2 was issued. This draft also required in-core thermocouples for BWR's. The cited reason for requiring them for post-accident monitoring was to indicate the potential for, or actual occurrence of, fuel cladding breach. The i
NRC staff also indicated that they desired to identify local hot l areas and the propagation of core damage with incore thermocouples. They suggested that approximately 50 thermocouples should be utilized. This quantity was judged by the NRC to be sufficient to detect block age of 5 -10% of the core w ith no spray (or other ECCS) at a high confidence level and with a sufficient allow ance f or attrition. No analysis nor engineering justification l
l to acknowledge the difference between the PWR involved in the accident and other types of reactors was cited f or including BNR's I
L.31-14
in the same requirement. Part A of this section addressed the NRC-proposed 50 thermocouples plan and reported that installation of such a number of thermocouples in power range monitors was impractical. The LaSalle presentation to ACRS specifically enumerated why incore thermocouples were of no significance to the operational decisions made by a BWR operator and concluded that the NRC's postulated-case of thermocouples indication of a superheat in a f uel assembly was applicable only f or a static post-accident scenerio for which all appropriate operator actions would have already taken place, with or without the thermocouple indications.
GE and the BWR Owner's Group had previously commented and discussed Revision 1 of Regulatory Guide 1.97; this eventually led to Revision 2 of Regulatory Guide 1.97. In this version, the NRC staff cited its reasons for now requiring 16 in-core thermocouples for BWR's to monitor core cooling and to provide a diverse indication of w ater level.
This staff position is recorded in LSCS-SER Supplement 1 (June 81) along with analytical predictions of thermcouple response for that very limited case where no core spray exists and water level is near mid-plane of the f uel zone with no make-up w ater available from any source and system pressure at ambient conditions.
(Additional simplifying assumptions used in the NRC analysis included no radiative, conductive or convective heat transfer from the fuel bundle to any supporting structure or cooling media).
It is Edison's concern that the installation of in-core thermocouples in the BWR will not satisfy either NRC purpose.
The installation of instrumentation is expensive in dollars and significantly costly in man-rem dosage during installation and maintenance. Installatien of in-core thermocouples in BWR's is inconsistent with the ALARA and cost-benefit concepts.
Edison believes the solution to achieving adequate post accident monitoring of core cooling and reactor water level is through present BWR w ater level instrumentation, and other instrumentation already incorporated into BWR plant designs.
2.0 Installation and Maintenance of Thermocouples in BWR-5 Reactors
(
The placement of thermocouples w as discussed in Part A.2 except f or the potential f or radiological exposure which was evaluated subsequent to that w rite-up. For the LaSalle reactors, the minimum annual increased exposure to plant personnel due to PRM-installed thermocouples would be in the range of 2 to 15 man-rem per year due to maintenance activities.
Installation of thermocouples requires tw 3 containment penetrations fue redundancy of separated instrument divisions corresponding to the electrical divisions which provide instrument power. Such penetratio.as are not currently available at LaSalle. Exposure of installation w orkers f or these penetrations is estimated at a minimum of 100 man-rem for an installation job of 2000 man-hours at LaSalle.
L.31-15
With installation in the PRM assemblies, considerable difficulty is expected both during installation and maintenance because of increased complexity in CRD removal in a restricted space where the added thermocouple leads would have to be located. Increased cable damage and connector damage can be expected during maintenance activities in this restricted space.
The detailed cost estimate for a 16-thermocouple installation is over $2,000,000 per unit with material costs approximately one-half and manpower costs the remainder. This is a late 1981 estimate; for conditions at the second outage, additional cost escalation should be applied.
Note that the application of the single-failure criterion of Table 1, Item 2 of Reg. Guide 1.97 would essentially reduce a 16-thermocouple installation to eight by loss of one division.
Further loss by the accident consequence criteria would eliminate readings from another 50%, thus leaving only four indicating thermocouples. Thus, it is concluded that an installation of 16-thermocouples does meet the single-failure criterion as cultines, but only magically could any conclusion be made that the l four surviving thermocouples would exist with one in each reactor l quadrant. Realistically, the presumed loss of thermocouples is of no consequence because, as previously discussed the surviving thermocouples are of no use in detecting timely ' Scal fuel temperatures anyw ay.
3.0 Risk Assessment GE evaluated the usefulness of in-core thermocouples to decrease radiological risk to the public f or a BWR-4-MK 2 power plant under the assumption that thermocouple output was not only available (timely) but also that it correctly indicated that reactor water level w as below the top of the active fuel. The NRC-Staff postulated condition of partial core-uncovery without a w ater-makeup capability was included in the grouping of accidents and transients used to analytically represent potential radioactive releases.
It w as assumed that incore-thermocouples installed in a BWR core would allow a few minutes early warning of complete core uncovery and that the sequential partial core overheated condition (termed partial melt for convenience) occurs with a freauency of 10-4 per reactor year. The hypothesis of a complete core melt reduces the
~
usefulness of core thermocouples to zero anyday. The risk from core melt sequences is not dominant nor even mildly influenced by an operator's early know ledge of incipient core damage; however, in this evaluation the operator was allowed to restart available ESF equipment. ,
This PRA utilized the methodology and techniques of WASH-1400.
Event sequences and their probabilities were determined for typical groupings of accidents and transients, each with detailed event i
trees and fault tree analyses. System f ailure probabilities were amalgumated from component and operator f aults which contribute to L.31-16
i .
L failure of the systems to function as required. An estimation was l
j made of the magnitude of radioactive release for each event sequence where f ailures indiated potential core overtemperatures due to lack of coolant. This estimate included the typical distribution of involved radionuclides and their respective release i times. Unique release fractions for significant dose contributing isotopes were then applied for the surrounding system equipment, containment, and plant arrangement with typical non-conservative credit .for various radiological f ractionation processes which occur during an accident with a w ater cooled power plant.
f With the resulting releases (source terms) from the plant and from l analyses of dispersion, transpurt, and receptor exposure via NRC approved models, the public risk was categorized into early fatalities or latent cancer fatalities.
The overall consequence was then expressed as the sum of the products of each potential public risk due to a particular accident / transient times the probability of occurrence of that accident / transient. The methodology yields a relative rank ing of
- contributing accident sequences which when analyzed further can disclose the significance of having or not having thermocouple i indication of in-core conditions, under the assumption of course l that operator action is quantified to take credit for improved performance due to thermocouple indication. For this evaluation, i the quantification of changes in operator response was based on the Handbook on Human Reliability (NUREG/CR-1278) plus practical l engineering experience in cases where task perf ormance times were needed.
J j The human error probability due to contradictory or confusing signals was reduced f rom the original PRA value of 1.0 X 10-4 to 2.0 X 10-4 based on the assumption of non-ambiguity previously stated. The operator-inattention human error probability was likew ise reduced f rom 5.0 X 10-2 to 1.0 X 10-3 on the assumption that thermocouples would annunciate impending core over temperature and that operators would be trained to correctly respond to that annunciaton. The human error probability due to failure to follow procedures (respond correctly) was therefore lef t
, unchanged at j 5 X 10-2 These changes essentially reduced the operator's failure probability for manual initiation from 9.8 X 10-2 to i 5.1 X 10-2 for those event sequences where an automatic _
initiation was not relied upon. In all cases concerning ECCS -
injection systems, the net effect of these human response factors on the systems-level accident sequence was negligible.
The role of the LaSalle operator in the manual initiation of the automatic depressurization system ( A05) as a back-up to its automatic initiation is somewhat diff erent as relates to quantizing potential human error. This is due to the routine nature of manual ADS initation imparted to the operator through training and procedures. For this action the original BWR PRA ascribed a probability of operator failure to manually initiate ADS at i 2X 10-3 w hereas f or LaSalle this w as improved to 1 X 10-3 on L.31-17 4
l
the basis that the thermocouples correctly identify the need for ADS and the operator responds accordingly with a lesser unreliability. The net affect of this operator improvement is to decrease the overall ADS failure to depressurize from 1.05 X 10-3 to 1.03 X 10-3 The final core overheat probability freauency changed from 9.3 x 10-6 to 9.5 x 10-6 This produces no consequence on early fatalities and a 2 percent' decrease in latent fatality expectation.
As can be seen f rom their minimal effect on a typical BWR probability risk assessment, in-core thermocouples do not significantly af f ect public risk. This conclusion correlates well with the basic BWR design concept that the operator is considered a back-up to the automatic equipment for safety functions. In any f ailure setting the hardsare f ailures tend to dominate the risk assessment results because operator effectiveness is classically derated to an arbitrarily pessamistic level. Requiring thermocouples for operator reference in the post accident situation does not decrease event probabilities not can it affect safety consequences because, realistically speaking, the thermocouple output does not affect operator response. This PRA shows that even if such output were assumed to affect operator response, the net effect on safety consequence is trivial. Another way of expressing the same conclusion is that in-core thermocouples do not provide operationally significant information.
If the standard of tiUREG-0739 were used f or cost benefit, comparisons at five million dollars per averted early fatality and one million dollars for every averted latent death, the benefit f rom incore thermocouples would be 0.01 X 106 or $10,000.00. The comparison to an installation cost of $2,000,000 indicates a benefit to cost ratio of less than 1 to 100,. Clearly, the economics of this evaluation cannot justify the installation of in-core thermocouples on a public risk reduction basis.
The logical reason for this PRA result is that thermocouples have been suggested on the premise that they provide the operator with a diverse indication of water level below TAF. In commerical BWR's, the operator has always been considered a back-up to automatic safety equipment for safety functions. In a failure setting, mechanical or hardware failures dominate the risk because operator effectiveness is arbitrarily derated to a very pessimistic level.
Since TMI-2, training requirements and symptom-oriented emergency procedures and control-room human-factor improvements have no doubt increased operator reliability, but the primary BWR safety contributors are still the engineered safety features of the equipment. To require thermocouples for operator reference in a post accident situation cannot decrease accident event probabilities nor can it affect safety consecuences because the thermocouple outputs do not affect operator responses in the real world. From this, then comes the conclusion that thermocouple usefulness is constrained to post-accident use as an analytical tool for a very specific case which has a vanishingly small probability of occurrence.
L.31-18
4.0 Operational Evaluation of Sixteen Thermocouples in BWR's Monitoring of core cooling and providing diverse indication of water level are postulated reasons for incore thermocouples in BWR systems. These functions are not currently performed by incore thermocouples, but are covered by the water-level measuring equipment in LaSalle reactors.
Monitor Core Cooling In assessing possible plant safety improvement resulting from core exit temperature measurements, applicable time periods during the course of an event were evaluated in Part A of this report. That evaluation is applicable t'o sixteen thermocouples in ARM locations within the core as well as for the fif ty thermocouples originally addressed as core exit thermocouples.
Diverse Indication of Water Level .
For LaSalle reactors, water level is the primary measure of core cooling during accident situations because it represents the safety status of the reactor better than any other plant variable.
Section L.67 and the response to Q031.287 provide a description of the LaSalle level measurement systems which are suf ficient and necessary to reliably monitor water level during all inventory threatening events. For any condition, the BWR has a designed water level indicating system - normal operations, refueling, post accident, upset conditions. Also unique symptom-based procedures exist to direct operator actions whether an indication of water l level exists or not. With existing level measuring equipment, measurement response time is not a problem. By training and current practice, the BWR operator equates reactor cooling with water level. He is not trained in terms of fuel temperatures response nor core thermal status. With the existing multiple -
redundant water level monitoring systems and with the emergency operating guidelines, the operator does not need incore thermocouple information. Assimilation of such new information may only conf use the operator by adding meaningless parameters which detract f rom restoring water level.
5.0 Conclusion Placement of thermocouples in LaSalle Reactors does not reduce public risk ; the projected radiological consequences remain essentially unchanged. Occupational radiation exposure would increase at LaSalle if thermocouples are mandated because of the original installation exposure and continued maintenance and replacement tasks. A commensurate improvement in plant safety cannot be identified.
L.31-19 j
4 PART C. THIRD RESPONSE: THERMOCOUPLE RESPONSE TIMES OVERVIEW The LaSalle SER Supplement No. I records the staff belief that in-core thermocouples (in ARM thimbles) would be responsive indicators of fuel temperatures and thus indicate the approach to or existence of inadequate core cooling. Preliminary calculations were cited to show a time responsiveness and temperature sensitivity that was acceptable. From these preliminary calculations, with conservative heat transfer assumptions, the conclusion was made that in-core thermocouples w ill f unction independent of location in the in-core assemblies to provide operationally meaningful data on ICC after all ECCS's have failed.
This third response addresses that aspect of the ICC issue.
1.0 Introduction A more detailed analysis performed by S. Levy f or the BWR Owner's Group was just completed to verify the preliminary BWR analysis and to determine the best location for in vessel thermocouples. The analysis included natural convection in a model having more axial nodes and more radial nodes with radial radiation accounted f or.
The ef fect of level swell f rom SRV opening and closing (8 to 14 f t of vessel level variation) shows that in-core thimbles and the top of the active fuel are periodically covered and uncovered such that in-core thermocouples would intermittently indicate some temperature above Tsat only during the period when the SRV was closed. In the meantime, fuel temperatures would be progressively increasing. A higher peak-clad-temperature would be reached for a more rapid variation in vessel level. Within any given bundle the fuel rods would experience approximately the same temperature excursion radially at the peak plane; axially the maximum temperature above Tsat occurs at approximately 0.8 of the bundle height. The core exit temperature did not correlate directly w ith peak fuel temperature but depended upon level-drop-rate in the vessel. The time lag w as at least 10 minutes. Basically, thermocouple response was degraded by the time rate of change of pressure in the vessel.
The conclusions are that in-core thermocouples are a poor indication of fuel temperature in a realistic analysis of a reactor vessel w here w ater level changes (sw ell) and pressure changes (SRV discharges) are present. A time delay of at least 10 minutes is also associated with core exit thermocouples whose calibration would require prior knowledge of vessel level drop rate, therefore core exit thermocouples are a poor indicator of fuel temperature.
Additionally, in-core thermocouples are a poor back-up level indicator for the same response-time and insensitivity reasons.
The sensitivity of thermocouples becomes worse as the break size gets larger.
L.31-20
As a core temperature monitoring device, thermocouples are not useful nor accurate. As a back-up level indicator, in-core thermocouples are useless. Their use for any operational decisions has the potential for confusion and erroneous or deleterious operator actions.
2.0 Heat Transfer Analysis of In Core Thermcouples for Small Break s The usefulness of the thermocouples mounted inside the in-core neutron flux monitors (thimbles) was examined during a small break LOCA. In this event the core is initially covered with water and the reactor has scrammed. Decay heat in the fuel rods continues to boil the water in the core until, eventually, the water level drops to the top of the core because it is assumed that mak e-up w ater is not available. As the w ater level drops further, to the plane where the in-core thermocouples are located and beyond, the rods become uncovere,d and begin to heat up. Heat flows outw ard to the channel wall, to the thimble, and finally to the thermocouplc inside. The heat balance on the steam as it rises through the uncovered portion of the core also provides the self consistent equations for determining the time-temperature output of the thermocouple.
This analysis is reported in SLI-8117 (October 1981) by S. Levy, Inc. It includes radiative heat transfer to the thermocouple fom the thimble wall; it includes radiative heat transfer from the fuel rods to the channel wall as well as convective heat inputs to the channel. Heat transfer among the fuel rods is by convection to the steam and by radiation from rod group to rod group (interior to exterior groups). The analysis used the practical location'for the thimble at I foot blow the top of the core.
The results indicate the extreme conservatism in the NRC's analysis which assumed adiabat ic temperature rise and no time delay between thermocouple response and input decay heat (NUREG 0519 Supplement 1). The Levy analysis indicates that the plane of the thermocouples is uncovered about 150 seconds after the top of the core uncovers. The rod heats up adiabatically but the rate of temperature rise drops off due to convective and radiative heat transfer to the channel. As the foam level in the bundle drops, more and more of the core below the thermocouple plane becomes uncovered and the temperature of the steam passing the thermocouple location rises. The channel w all, thimble, and thermocouple all rise in temperature and the thermocouple is 450F above saturation temperature some 780 seconds (13 minutes) after the start of core uncovery. The time lag between the thimble and the thermocouple is extremely small; direct contract between the two will not significantly change the time delay for the thermocouple to respond.
Reasons f or differences between the NRC calculation and the Levy calculation are that NRC used a significantly faster fuel heat up rate 2.7 to 3.80 F/sec compared to 10F/sec) and the NRC neglected convective and radiative heat transfer as previously noted. A further simplification was made by the NRC calculation -
L.31-21
that of constant vessel pressure - which is not realistic for a BWR in this transient situation. During a small break where loss of coolant occurs, the pressure will most likely also change. As relief valves operate the pressure will vary considerably. Voids will f orm in the saturated liquid when the pressure drops, such voids raise the w ater level by a magnitude of 8-12 f eet depending upon the reactor system parameters: core height, vessel volume, vessel and core cross-sectional areas, amount of water below the core, etc. The effect of the vessel level change due to pressure variation is to periodically cover and uncover the plane of the
' thermocouple until the core is almost completely uncovered.
Clearly, the location of thermocouples inside the ARM thimbles does not provide meaningful data on core temperature conditions during transients which have accompanying pressure variations.
At core exit, the steam temperature follows the temperature of the top of the f uel bundle f airly closely. Since poder at the top of i the bundle is lower than that further dodn during the transient under discussion, the use of exit thermocouples can not indicate start of core uncovery nor extent of uncovery. Approximately seven
' minutes af ter start of uncovery the exit plane thermocouple would indicate a temperature 450F above saturation temperature.
At a position in the steam dome, a thermocouple would read 450F above saturation at about 9.2 minutes after uncovery provided that no accumulation of saturation-temperature steam is present to dilute the superheated steam entering the steam dome. Such a provision is not realistic, hence the response time of a steam-dome thermocouple would be even slower, markedly slower. .
3.0 Conclusions Detection of the propagation of core damage was shown to not be within the capability of thermocouples inside a BWR. The time responsiveness of thermocouples is not adeauate for effective response to thermal transients for operational control nor for t analytical responses to postulated long term cooling conditions i because of pressure and level variations in the reactor vessel.
Variations in vessel pressure and w ater level would cause erratic j
thermocouple readings dhich would not track the fuel temperatures.
t The sensitivity of thermocouples in a break situation depends upon the size of the pipe break and the level drop rate. Essentially, thermocouples in a BWR cannot provide reliable nor meaningful -
indications of core cooling status.
L.31-22
- . , y PART D FOURTH RESPONSE: CONTINUED EVALUATION AND COMMITTAL The BWR-Owners Group met with the NRC Staf f on January 26, 1982 to discuss the continuation of activities under NUREG 0737 Item II.F.2. Inadequate C re Cooling. The definition of NRC interest in the detection of the approach to ICC, confirmation of the existence of ICC, and definition of the recovery from ICC conditions w as explicitly given. The specific job which the ICC instrument (s) is to perform was outlined and several suggestions were made concerning assumptions and coverage of the ICC evaluation. .
As a member of this BWR Owners Group, CECO not only participat'es in this evaluation of ICC but also commits to endorse any practical and cost-effect ve resolution to the ICC question as approved by I the NRC Staff. Such resolution is, of necessity., scheduled subsequent to the fuel loading of LaSalle reactors. -
In the meantime, for LaSalle, the unique problem of inaccurate I w ater-l e ve l indication during situations where boil-of f can occur '
in the cold reference leg due to high containment temperatures is' , ,
handled by emergency procedures (LGA's) on which the operators are' '
specifically trained. Recognition of high drywell temperature (and pressure) which gives rise to reference leg boil of f/ flashing ~ is +
a provided via a Class lE temperature monitoring system with alarms c' in the control room. The operator's procedures cover a wide 'range '
of applicable temperature and pressure conditions such that .
application of emergency procedures precedes the actual boi1 off/ flashing conditions in the drywell. The LGA's cover the situation for inventory make-up when there is no reliable level indication available.
Interim engineering resolutions for this boil off/ flashing condition leading to degraded water level measurement are being pursued with the NSSS vendor in accordance with vendor recommendations in NEDE-24801 (April, 1980). At LaSalle Yand ay instruments are not used for safety functions, only cold leg instrumentation is used and operator procedures are relied upon for the boil off/ flashing conditions as indicated by the containment monitoring instruments. Recovery is normally made by s depressurization and refill with low pressure ECCS's.
,(;
(. i,
\
s, ,
N L.31-23 GRC/ji ,
2-4-82 0839L* ,
e.
s t
60
, INDICATED RPV WATER LEVEL
............. AVERAGE BUNDLE TEMPERATURE AT MIDPLANE WATER LEVEL
- --- BYPASS THERMOCOUPLE AT CORE MIDPLANE INSTRUMENT RANCE FUEL NARROW WIDE ZONE
< / ~ !.EVEL 3 (SCRAM) .. ..
j V LEVEL 2 (INITIATION OF HIGH-PRESSURE ECCS. DETECTABLE H
/. '
l RECIRCULATION PUMP TRIP) '
-PRODUCTION ,
OPERATOR WOULD DEPRESSURIZE VESSEL . ,
. 40 - '
. /.
,. , SIGNIFICANT H 2 '
g ,
, LEVEL 1 , PR03UCTION -
W (START OF LOW-PRESSURE Er.C DETECTION OF; FISSION rb ,
SYSTEM PUMPS, ISOLATION)
PRODUCTS IN kE
- ELL ,
a AIRSPACE W
r- :* ..
j TOP OF
' d FUEL
" ' -DELAYE0 f,CCS g -
DiJECTION b g
. z .e -< -.
O . .
WATER LEVEL ***..
o a Y v
'omOg 20 -
Oax r _, ,,,...** , -
2000 ,C E ~
9 > BOTTOM OF ASSUMED BLOCKED CHANNEL ,.;..****h q
~~
$mn
-a xn 2m FUEL
- h #
w mo >>p ' >*, \ $ b Q moi:.-
F *.***...****/' TEMPEPJTURE g *
-< o > m m f
\ $
m%* *
A
- e s
I u
n.
w~ m m -
1000 W x mlmm
>. OW C
m
< O . ' */ * * . * */,/.4:1-WITHOUT C68E SPRAY \
H
._. - m O ..
'/
- ~]H=
o=>m o
o .
$C
, z * *' * * *' * * ** * ** w w aa w n... 4. d .a*_,--------_ #-
\
\
-a
^
c ) m U CH t'ITH CORE SPRAY OR BYPASS FLOODED _________t*""""
" -==
>$$ i 94 0 t I '
- O t $;c 'v"o r > m 1000 2000 3000 4000
- ld hN
- g w
= 0 a p H
, TIME AFTER SCRAM (SECONDS) .< m g
g r- m om I -
- mg O mc mm m$p 3 Z ~2 m
CD a
i