ML20056D968
| ML20056D968 | |
| Person / Time | |
|---|---|
| Issue date: | 05/19/1993 |
| From: | Taylor J NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| To: | Shewmon P Advisory Committee on Reactor Safeguards |
| References | |
| PROJECT-669A ACRS-GENERAL, NUDOCS 9308190096 | |
| Download: ML20056D968 (25) | |
Text
{{#Wiki_filter:r 0, i f 4+i L - o Dr.PaulShewmon,Ch/'T!)eactorSafeguards May19,199?] ( an Advisory Committee on U.S. Nuclear Regulatory Commission Washington, D.C. 20555
Dear Chairman Shewmon:
I am responding to the letter you sent to the Chairman on April 26, 1993, in which you commented on staff positions contained in SECY-93-087, " Policy, Technical, and Licensing Issues Pertaining to Evolutionary and Advanced Light- } Water Reactor (ALWR) Designs." I am pleased that the discussions we have had on these complex and important issues over the past several months have resulted in the Committee's general agreement with the staff's positions in - SECY-93-087. In your letter, you provided additional comments on 7 of the 42 issues discussed in SECY-93-087. Those issues are fire protection, hydrogen control, core debris coolability, containment performance, equipment survivability, control room annunciator reliability, and defense against common-mode failure in digital instrumentation and control systems. I have enclosed a staff response to your comments and recommendations on these issues. [ As noted in SECY-93-08:', the staff will soon be finalizing its position on i many of the remaining policy issues that affect the passive plant designs. We will continue to work with you to solicit your views on these additional topics. i Sincerely, Originalsigned by James H. Sn!ezek James M. Taylor M' Executive Directorf for Operations
Enclosures:
1. Response to ACRS Lir of 4/26/93 2. Memo for Chairman & Comm. fm Taylor dtd 8/31/90 3. Hydrogen Concentration Limits for Proposed New Designs of Nuclear Power Plants 9308190096 930519 .i cc w/ enclosures: PDR PROJ The Chairman 669A PDR Commissioner Rogers -Commissioner Curtiss gl "lT p T {b 5 L8A U Q 8p T A p n g1/ kh E I Commissioner Remick b Commissioner de Planque SECY DlSTRIBylL0Th See next page
- See previous concurrence OFC-TA:AQtk TECH ED (A)D/fDST:ADAR (A)ADAR:NRR6' ADT:NSR v i
s-5, p.s w g NAME MJCase:bt MMejac* WRP# chardt JACalvo //10 WTRussell DATE 05//o/93 05/7/93 05/fl/93 05/ ///93 05/0493 7 t s' g j f 4 DA RS E1R / o 0FC V 030042 ^ c\\\\ I ["Murl ey i 4Ja? lor NAME I f., q?d[ I F [h/93 DATE 05]h/93 0 [ f)FFICIAL jECORD COPY !DOCUMEl1 NAME: GT8528.MJC g -t y f} /,h /? f &La - G (W~2) A' y
/] g
- e i
U w Dr. Paul Shewmon Distribution: Central File (w/ incoming) NRC & Local PDRs (w/ incoming) EDO 8828 EDO R/F JMTaylor, EDO JHSniezek, EDO JLBlaha, EDO HLThompson, ED0 MTaylor, ED0 EBeckjord, RES RBernero, NMSS EJordan, AE00 JScinto, OGC TEMurley/FJMiraglia, NRR JGPartlow, NRR WTRussell, NRR ATGody, NRR DMCrutchfield, NRR JACalvo, NRR AThadani, NRR RBarrett, NRR JKudrick, NRR JMonninger, NRR CMcCracken, NRR BBoger, NRR JWermiel, NRR MChiramal, NRR HPastis, NRR MJCase, NRR RWBorchardt, NRR OGC OPA OCA NRR Mailroom - GT8828 (w/ incoming) BJToms, NRR - GT8828 i PMHagnanelli, NRR - GT8828 ADAR R/F (w/ incoming) i a
a w w RESPONSE TO ADVISORY COMMITTEE ON REACTOR SAFEGUARDS LETTER OF APRIL 26. 1993 In its letter of April 26, 1993, the Advisory Committee on Reactor Safeguards (ACRS) provided comments and recommendations on the staff's positions concern-ing fire protection, hydrogen control, core debris coolability, containment performance, ec;uipment survivability, control room annunciator reliability, and defense against common-mode failure in digital instrumentation and control systems. The following is the staff's response to these comments and recom-mendations. Fire Protection ACRS Concern: The ACRS remains concerned that a common normal ventilation system (such as that proposed for the Advanced Boiling Water Reactor) will be difficult to design to prevent the effluent from a postulated accident in one train of engineered safety features from reaching essential mitigating equipment in the other trains and creating conditions that exceed their environmental qualifications. Of particular concern is the capability of ventilation dampers to isolate the effects of high energy pipe ruptures in confined compartments served by the common heating, ventilation, and air conditioning system. Staff Response: The staff is cognizant of and agrees with the ACRS concerns concerning potential common-mode failures through common ventilation systems. This issue is being evaluated in the advanced reactor reviews and will be addressed in both the applicant's safety analysis report and staff's safety evaluation report. Final closure of this issue has not been reached on either the ABWR or the System 80+ designs. Hydroaen Control ACRS Concern: The ACRS is concerned about the staff's claims that it has sufficient basis for understanding hydrogen behavior to go forward with licensing criteria. The ACRS notes that the staff has not demonstrated that this basis is as extensive, or applicable, as the staff believes.
- Further, the AP600 and ABB-CE System 80+ designs have containments that are more susceptible to significant damage from hydrogen detonation than most existing and evolutionary plants.
This requires that the licensing criteria for this ) issue be reconsidered. Staff Response: The staff has continued to assess the basis of its position on hydrogen control for advanced light-water reactors (ALWR) since it was initially taken in SECY-90-016 " Evolutionary Light-Water Reactors (LWR) Certification Issues and Their Relationship to Current Regulatory Require-ments." In support of its decisions on hydrogen control issues, the staff has relied on a team of technical experts from Los Alamos National Laboratory and Sandia National Laboratory. Supplementing these efforts, was work performed by Dr. Joseph Shepherd of Rensselaer Polytechnic Institute and a committee of the National Academy of Science. The staff had previously sent the Commission a report (Enclosure 2) summarizing the conclusions drawn by these experts from detailed reviews of the detonation experimental data base and modeling ENCLOSURE ]
(l v w
- efforts. The staff also indicated that unless significant new information becomes available to warrant reconsideration of the hydrogen control provision of SECY-90-016, then these provisions should be used for all advanced reactor designs.
In 1990, Dr. Shepherd prepared an independent study (Enclosure 3) of hydrogen concentration limits for ALWRs and substantially documents much of the experimental data base in support of this issue. The staff has undertaken a project to update this independent study with any new information relative to hydrogen control that has taken place since 1990. The staff will provide the results of this review to the ACRS. At this time, the staff does not antici-pate any changes from the current position based on research since 1990. Core Debris Coolability ACRS Concern: The ACRS notes that the staff has weakened the position taken in SECY-90-016 by not requiring that the core debris be adequately quenched. The Committee believes that the present criterion for coolability, namely a 2 cavity floor area greater than 0.02m /MWt, is not soundly based. It recom-mends that the staff validate containment response to core-on-the-floor accident sequences by independent analyses using, for example, MELCOR or CORC0f4 and CONTAlfl. Staff Response: The staff agrees with the ACRS recommendation on this issue. It will validate the containment response to accident sequences involving core-concrete interaction using available state-of-the-art codes. For the evolutionary light water reactors, the staff has contracted with the national laboratories to perform such analyses. The results of these analyses will be documented in the final safety evaluation reports for the particular designs. With respect to the Electric Power Research Institute (EPRI) cavity floor 2 sizing criterion of 0.02 m /MWt, the staff generally agrees with the ACRS. The staff position neither supports nor disputes this criterion. Rather, the staf f believes that vendors should provide a large cavity floor space to enhance the potential for spreading. To determine if vendors have met this criterion, the staff will evaluate, among other aspects, (1) the amount of subcompartmentalization within the reactor cavity affecting the capability of core debris to spread, (2) the potential for core debris accumulation in sumps or low areas, (3) the effect of core debris impacting structural side walls within the cavity and the bottom and sides of the containment liner, (4) analytical results of core-concrete interaction, and (5) the design impact of increasing the floor area. The staff believes that the updated position in SECY-93-087 on core debris coolability is a reasonable extension of the original position proposed in SECY-90-016. At that time, both the staff and the ACRS recognized the need to further evaluate this issue in light of the best available technical insights. In SECY-90-016, the staff indicated that the issue of core debris coolability was an area in which experimental research was ongoing and that the staff would continue to evaluate core debris coolability and any specific cavity sizing criterion as more data and information became available.
O O In its letter of April 26, 1990, the ACRS indicated that quantification of l what constitutes sufficient. reactor cavity floor soace was still an open question, as was the means by which one quenches.the core debris. In addi-tion, the ACRS indicated that resolution of this issue would require engineer-t ing judgment because many of the physical processes were not fully understood. After SECY-90-016 was issued, MACE test results were inconclusive regarding core debris'quenchability. Future MACE tests are planned; however, the likelihood of them proving quenchability is uncertain. In light of this uncertainty and differing views of experts regarding core debris coolability, the staff concluded that better definition of acceptance criteria was needed for evolutionary and passive light water reactors to deal with the potential i for extended core-concrete interaction. The criteria in SECY-93-087 were l intended to meet this objective. Particularly, the staff concluded that if the containment could withstand core-concrete interactions for approximately 24 hours, an acceptable level of mitigation would be provided. .C_gntainment Performance ACRS Concern: The ACRS agrees with the requirement that containment stresses not exceed ASME Code Service Level C for metal containments, but it is not clear how electrical penetrations through the containment should be consid-ered. i Staff Response: The s;aff has considered the integrity of electrical penetra-tions under severe-accident conditions. Testing of electrical penetrations under these conditions has been performed at Sandia National Laboratories as documented in NUREG/CR-5334, " Severe Accident Testing of Electrical Penetra-tion Assemblies," and NUREG/CR-3234, "The Potential for Containment Leak Paths Through Electrical Penetration Assemblies Under Severe Accident Conditions." t The results of the testing program indicated that the leak integrity of-1 electrical penetration assemblies was not compromised by severe-accident
- loads, Eauipment Survivability I
ACRS Concern: The ACRS agrees that passive plant design features provided only for severe-accident mitigation need not be subject to the environmental qualification requirements of 10 CFR 50.49. It believes, however, that such . mitigation features must be designed to provide reasonable assurance that they will operate in the severe-accident environment for which they are intended l and over the'timespan for which they are needed. j Staff Response: The position taken by the ACRS on this issue is in exact f agreement with the staff's position as stated in SECY;90-016 and SECY-93-087. During the review of the evolutionary and passive light water reactors, the staff will evaluate the. vendors' determination of equipment. relied on for j -severe-accident mitigation and ensure that it is designed to operate in the severe-accident environment for the timespan during which it will be relied on. In addition, consideration will be given to the applicable initiating. event. i u 1 i b ?
g . Control Room Annunciator ( Alarm) Reliability ACRS Concern: The ACRS believes that the staff needs to clarify and provide additional justification for the position that alarms that are provioed for manually controlled actions for which no automatic control is provided and that are required for the safety systems to accomplish their safety functions shall meet the applicable requirements for Class IE equipment and circuits. Staff Response: The staff's recommended position is that the alarm system for advanced light-water reactors (ALWRs) should meet the applicable requirements of the EPRI Utility Requirements Document (URD). In addition, the staff position as stated in SECY-93-087 is that " alarms that are provided for manually controlled actions for which no automatic control is provided and that are required for safety systems to accomplish their safety functions, shall meet the applicable requirements for Class IE equipment and circuits." Section 2.2.10 of the EPRI URD states that (1) "the MCR [ main control room) shall contain compact, redundant, operator workstations with multiple display and control devices that provide organized, hierarchical access to alarms, displays, and controls" and (2) "the display and control features shall be designed to satisfy existing regulations, for example: separation and independence for Class lE circuits (IEEE Standard 384); criteria for protec-tion systems (IEEE Standard 279); and requirements for manual initiation of protective actions at the system level (Regulatory Guide 1.62)." The additional requirement that the staff proposed is also consistent with the criteria of IEEE Standard 603, "IEEE Standard Criteria for Safety Systems for Nuclear Power Generating Stations." The 1980 version of this standard is endorsed by NRC Regulatory Guide 1.153, " Criteria for Power, Instrumentation and Control Portions of Safety Systems." Compliance with the provisions of IEEE Standard 603-1980 is considered to satisfy the provisions of IEEE Standard 279-1971, which is referenced in 10 CFR 50.55a(h). All ALWR I&C system designers have committed to meet the guidelines of IEEE Standard 603. IEEE Standard 603-1980, Section 5.8.1, " Displays for Manually Controlled Actions," states: "The display instrumentation provided for manually con-trolled actions for which no automatic control is provided and which are required for the safety systems to accomplish their safety functions shall be part of the safety systems and shall meet the requirements of IEEE Std 497-1977." In light water reactor designs there may be a few instances when the necessary display instrumentation incorporates a particular Class IE alarm for alerting the operator to take the required manual control action. Such situations include low-temperature overpressure protection (LTOP) arming alarm, contain-ment hydrogen concentration - high alarm, boron dilution alarm, and residual head removal suction valve / reactor pressure high interlock alarm. In these instances, if the safety system is not automatically actuated and the alarm is the only indication provided to the operator to initiate manual action, the alarm is also to be part of the safety system and shall be designed to Class IE criteria.
g 3 w w Defense Acainst Common-Mode Failure in Dioital Instrumentation and Control Systems ACRS Concern: The ACRS is concerned about the staff's second recommendation on this issue that the vendor or applicant analyze each postulated common-mode failure for each event that is evaluated in the accident analysis section of the safety analysis report (SAR). The ACRS recommends that the scope of this assessment include consideration of common-mode failures during all events postulated in the SAR (e.g., fire, flood, pipe rupture, and extensive loss of essential power sources) and not be restricted to those events discussed in Chapter 15, " Accident Analysis." Staff Response: The thrust of Point 2 of the staff's recommendation is that the applicant assess the defense of the plant's instrumentation and control (l&C) system against cnmmon-mode and common-cause failures - particularly software errors. In performing the assessment, the applicant should analyze each event that is evaluated in the accident analysis section of the SAR. The staff considers the scope of the required assessment adequate in that it envelopes all events for which the software is required to respond. Specifically, the staff believes that safety system actuations for Chapter 15 events envelope I&C requirements posed by other events postulated in the SAR and that the defense against common-mode failures in the I&C system due to software errors would be identified appropriately by an analysis against Chapter 15 events. Common-cause or common-mode failures due to physical phenomena such as fire, flood, and environmental effects of pipe rupture are addressed in the SAR in requirements pertaining to physical separation and electrical isolation and equipment (computer) qualification. The events analyzed in Chapter 15 of the ABWR standard SAR involve (1) a decrease in core coolant temperature, (2) an increase in reactor prcssure, (3) a decrease in reactor coolant system flow rate, (4) reactivity and power distribution anomalies. (5) an increase in reactor coolant inventory, (6) a decrease in reactor coolant inventory, and (7) radioactive release from subsystems and components. As indicated above, in assessing the defense against common-mode failures by analyzing I&C system response for each of these events, the staff believes that the effects of all events postulated in the SAR are adequately enveloped. For example, the evaluation of failure of small lines carrying primary coolant outside the containment (one of the events that lead to a decrease in reactor i coolant inventory) would envelope all small-break events outside the contain-ment, including a break in the reactor water cleanup system. ] i
.Y. T : O o [f,....,[", + UNITED STATES Be, g NUCLEAR REGULATORY COMMISSION
- U
' E WASHINGTON. D. C. 20555 AUG $ ! 1923 HEMORANDLRt FOR: Chaiman Carr Comissioner Rogers Comissioner Curtiss Comissioner Remick FROM: James M. Taylor Executive Director for Operations
SUBJECT:
MAY 22, 1990, STAFF REQUIREMENTS MEMORANDUM ON BRIEFING OF EV01.UTIONARY LIGHT WATER REACTOR CERTIFICATION ISSUES AND RELATED REGULATORY REQUIREMENTS In the May 22, 1990, staff requirements memorandum (SRM), the Comission requested the staff to report to the Comission how the views of the staff's ~ consultant (s) on hydrogen detonation compare with those of the staff. Since the accident at Three Mile Island, Unit 2 (THI-2), the NRC has regularly-used a team of hyorogen experts from the Los Alamos National Laboratory (LANL) and the Sandia National Laboratory (SNL) to assist the staff with its review of hydrogen detonation. To supplement those efforts, the staff has used input from a consultant, Dr. Joseph Shepherd of Rensselaer Polytechnic Institute (RPI) and a comittee of the Natiunal Academy of Science (NAS). Herbers of the NAS comittee are listed in the enclosure. The contractors from the national laboratories and the RPI consultant are both recognized internationally as authorities in the field of hydrogen combustion and have published extensively in journals and other scientific literature. This paper includes a sumary of the current-conclusions drawn by these experts from detailed reviews of the detonation experimental data base and modeling efforts. In 1987, the NAS comittee published the report, " Technical Aspects of Hydrogen Control and Combustion in Severe Light Water Reactor Accidents," which provided a detailed review of the NRC's hydrogen programs. With regard to detonations, the NAS comittee used a method to calculate the width of a detonation cell that extrapolated from the existing data base to provide predictions for full-scale hydrogen detonation in a light-water reactor containment. To avoid the potential for hydrogen detonations, the NAS comittee recomended that hydrogen concentrations should not exceed 9 to 11 percent by volume. Dr. Joseph E. Shepherd of RPI developed the Zeldovich-von Neumann-Doering chemical-kinetics model to predict detonation cell width. This model has been used extensively by Dr. Shepherd and others including the NAS comittee and the SNL experts. The methodology Dr. Shepherd used to determine the detonation CONTACT: T. Kenyon, NRR 2-1120 ENCLOSURE 2
(7 v v \\ The Comission limit assumed that the deflagration-to-detonation transition (DDT) will not occur if the size of the detonation cell is greater than two channel widths. DDT has been observed experimentally when the detonation cell size is smaller l than one channel width. To avoid the potential for detonations, Dr. Shepherd recomended that the maximum concentration of hydrogen should not exceed 10 percent by volume. The sethod used by Dr. Shepherd was similar to the NAS method. The hydrogen authorities from SNL have conducted small, intermediate, and large-scale hydrogen combustion experiments to investigate diffusion flames, ) j deflagration, accelerated flames, DDT, and detonation phenomena. The detonable i range of hydrogen in a hydrogen-air mixture for their experimental facility (Heated Detonation Tube, HDT) is 11.6 percent to 74.9 percent at 20*C and 1 9.4 percent to 76.9 percent at 200*C. The detonation Ifmit is between 38.8 percent and 40.5 percent for steam at stoichiometric hydrogen-air-steam mixtures at 100'C and I atmosphere. SNL recorsnends that the hydrogen in the i containment should be maintained at lower concentrations than those for which a i detonation has been experimentally observed because of the uncertainties associated with scale and temperature on detonation limits and the' potential for accelerated flames to generate impulsive loads. The data base used by SNL for detonation limits includes data from all U.S. programs and from international programs (i.e., Italy, Federal Republic of Germany, and Canada). In addition, Professor John Lee of McGill University, recognized internationally as a leading authority in the area of combustion, was under subcontract to the SNL project. In its review of hydrogen detonation, the staff has used LANL only in the development of a finite-difference code to perform detailed hydrogen transport and mixing calculations. The conclusions of these authorities were considered in the staff's develop-1 sent of its position on hydrogen detonation as discussed in SECY-90-016, ' Evolutionary Light Water Reactor Certification Issues and Their Relation-i ship to Current Regulatory Requirements," and in the draft safety evaluation report on Chapter 5 of the Electric Power Research Institute (EPRI) Advanced j Light Water Reactor (ALWR) Requirements Document. In those papers, the staff stated its position that, because of uncertainties in the phenomenological Enowledge of hydrogen generation and combustion, as a minimum, evolutionary ALWRs should be designed to limit containment hydrogen concentration to no greater than 10 percent. Therefore, the staff recommended that the require-ments of 10 CFR 50.34(f)(2)(ix) remain unchanged for evolutionary ALWRs. Part (A) of this regulation requires that: Uniformly distributed hydrogen concentrations in the contain-tent do not exceed 10% during and following an accident that releases an equivalent amount of hydrogen as would be generated from a 100% fuel clad metal-water reaction, or that the post-acci-dent atmosphere will not support hydrogen combustion.
a. m. i The Ccanission. [ Unless significant new information becomes available to warrant reconsidera-tion of this position, the staff expects to employ the same criteria on all advanced reactor designs. i Origin:1 Signed Byi James M.Taylot' James M. Taylor Executive Director for Operations
Enclosure:
As stated cc w/ enclosure: SECY OGC ] ACRS i .i I 'I l i l 1 \\ a
q 4 ,l*['[.- I a. ' fi Et: CLOSURE -NAS C0tt11TTEE uT.BERS Morran C. Aasmussen, Chairman Herbert Kouts -Eli K.-Dabora Gerard M. feeth Antoni K. Oppenhein Daniel J. Seery Robert G. Zalosh James J. Zucchetto $NL EXPERTS Marshall Berir.an [ , 'Lloyd Nelson-Lichung Pong Martin Sherman =! Scott-Slezak l Douglas Stamps Sheldon Tieszen C. Channy Wong LANL Ex>ert Join Travis RPI_ Expert t Joseph ShepherC 'i ^ I ~ i b I I i t )
t 6
- l
. a, a, Hydrogen Concentration Limits for u Proposed New Designs . 1 of l t Nuclear Power Plants I l i 1 Joseph E. Shepherd l Jepartment of Mechanical Ensincering, .i Aeronautical Engineering, and Mechames Rensw!aer Polytechnic Institute Troy, NY j January 6,1990 i l i -i Prepared for OfEce of Nucl-ar Reactor Regulation, US Nuclear Reph / Commission. 3 1 Work performert vt 'e 4 ontract for l Sandia National Laborate Lituquercue, NM. i ) - r; -t
- ENCLOSURE 3_
-' Y. . - + d r
e, a, Hydrogen Concentration Limits for Proposed New Designs of Nuclear Power Plants J. E. Shepherd Introduction Water-cooled nuclear power plants can have a hydrogen combustion hazard if a lossmf-coolant accident occurs and oxidation of the core metal results in a release of hydrogen into the containment atmosphere. Extensive research into this hazard has been con-ducted over the last decade. The results and unresohed issues as of 19S6 are discussed in the report (NAS 1987) of the Netional Academy of Sciences review committee. A nore recent assessment of the issues and the priorities for U.S. Nuclear Regulatory ' Commission sponsored research in the 1990s are given in the Severe Accident Research Program Plan (NRR 1989). The specific issue considered in the present report is: What should be the maximum allowable hydrogen concentration (under severe accident con-ditions) within the containment of the next generation of nuclear power plants? Hydrogen Combustion HazaiJs The specification of concentration limits for explosive gases in air has traditionally focussed on eliminating hazards by keeping the concentration below the fiammability limit through dilution or inerting. While some traditional hazard-mitigating strate-gies have been censidered for nuclear power plants, the focus of most previous efforts i Berman et al 1982) has been in managing the hazard via deliberate combustion rather than eliminating it entirely. This is apparently based on cost considerations and the low perceived risk of this type of accident. The current hazard management strategy is based on burning the hydrogen at lean concentrations (4-13% by volume) to limit the overpressure on the containee -+ shell anu the neat loads to the struc-and equipment within it. This strategy relies on a benign burning process, either a standing or propagating diffusion fiame, that produces low containment atmosphere pressures (1-3 bars overpressure) and low heat loads on The structural components and safety-related equipment. The disculty with this appmach is in assuring that a benign combustion process always occurs. For many types of combustible gas hazards, there has been a growing recognition over the last two decades that under certain conditions h,w-speed fiames can spontaneously become unstable and accelerate to a very high velocity (1000-2000 m/s), simultaneously generating dan 5erously high pressures (10-20 bars). The most severe possibility is that a transition to detonation or DDT event (Lee and Moen 1980) occurs. While some scale-model tests (Tamanini et al 19S9, Thompson et al 1938, Berme.n et al 19S2) of the deliberate burning technique have been successful. other scale-model 1
e. w w tests (Tieszen et cl 1989, Sherman et al 1989, Guirao et al 19S9, Cummings et al 1987) have shown that flame acceleration and high overpressures do occur. This issue has never been fully resolved due to seveml complicating factors: a) the appropriate combustion tests have never been performed on realistic geometric configurations at true scale; b) the effects ofinitial temperature, pressure, hydrogen and steam concen-trations are very significant and the values expected in the course of an accident are known with great imprecisioo: c) the models (physical, analytical and numedeal) avail-able for ant.lyzing transient combustion cannot predict the spontaneous transition from deflagration to detonation. While these issues remain outstanding, the phenomenon of flame acceleration and the relevance to severe accidents in nuclear p1wer plants has been clearly demonstrated. Qualitatively, four types of combustion behavior have been observed for lean mix-Between concentrations of 4-13% H, low-speed propagating difusion flames tures. 2 have been observed in a number of spherical or cylindrical vessels (with no internal compartments) between 1 and 2000 m in volume. These flames were initiated by low-energy spark or thermal igniters. Standing diffusion flames have also been observed in both small an'd large-scale tests when a steady flow of hydmgen into the vessel exists during the experiments. Hydrogen concentrations above 4-5% are required to obtain ignition of the flame. Detonations have been deliberately produced in tubes up to 0.43 m in diameter, the largest being the HDT facility at Sandia Labs (Tieszen et al 1987). In tests at room temperature, detonations have been initiated in hydrogeo-air mixtures at concentra-tions as low as 11.5% H ; at higher temperatures (370 K), down to a concentration of 2 9.5%. Detonations have also been produced in this tube for various compositions of steam-air-hydrogen mixtures. Spontaneous transition from low-speed flames to detonation has been obsened in laboratory (Guirao et cl 1959. Lee 19S6) and large-scale experiments (Sherman et cl 1959) at hydrogen concentrations dc,wn to ;5% H. Most transition experiments are 2 carried out in channel or tubular geometry with periodic obstacles (partial obstruction of the flow) to generate turbulence and promote transition to detonation. Transition in such a partially obstructed chnnel occurs very rapidly, within 2 - iuneters from the initiation point. Transition in smooth tubes also occurs but requires a much longer distance, often over 100 diameters from the initiation point. Even without transition to detonation. very high-speed flames can be generated from weak ignition sources within partially obstructed tubes. The gas pressures as-sociated with such high speed motion can be quite substantial (Shennan et sl 1989), up to 5 times the initial pressure or about 50% of the maximum detonation pressure. Structural loads due to these events may be as severe as those due to detonation itself. Flame acceleration tests in partially obstmeted tubes (0.15-0.3 m diameter) at McGill Unive:sity (Guirao et cl 1956) demonstrate a rapid increase in maximum turbulent flame speed with increasins; hydrogen concentration, frorn 100 m/s at 10% H2 to over 1000 m/s at IS% H. Similar tests (partial blockage. no venting)in the FLAME facility 2 o i I
A w w ~ ~ ~. _ at Sandia have resulted in tu6ulent Bame speeds of 500-700 m/s in mixtures containing about 157c Hr. Our current knowledge of hydrogen combustion behavior for lean mixtures at am-bient conditions can be summarized by the following table: Table 1. Lean Hydrogen-Air Combustion Behavior. Hz(7c) Geometry Size (m) Ignition Observations 4-13 sphere or cylinder .2-17 glowplug low-speed Hames (1-10 m/s) >4 various .5-17
- glowplur, standing diffusion Hames 10-15 obstructed tube
.3-1 weak spark high-speed Bames (100-1000 m/s) > 15 obstructed channel 2.4 weak spark transition to detonation >11.7 tube .43 high explosive detonation Note that the size given is a characteristic minimum length such as a tube, cylinder, or sphere diameter or a channel width. These minimum dimensions appear to be one of the controlling factors in determining the potential for transition from low speed Hames to detonation. Clearly, a wide range of behavior is possible for lean mixtures and a potentially hazardous situation exists if the hydrogen concentration is above 107c. The concentra-tion range between 10-157o is of particular interest since many loss.of-coolant accidents will result in hydrogen concentrations in this range. Unfortunately, this is the concen-tration rance in which the interpretation of the existing data is most controversial. A concentration of 157o or greater is clearly hazardous and concentrations below 9-107o are apparently safe. The worst hazard appears to be accelerated Bames and possibly, transition to detonation. in partially-obstructed or compartmentalized conngurations similar to the lower portions of many reactor containment buildings. Concentration Limits and Detonation Cell Size The key issue would then appear to be: At what concentration of hydrogen does the transition from low speed Bames to detonation cease to be a hazard? The following variables or parameters have been demonstrated to hase an important inSuence on this phenomenon: absolute scale (size); degree of condnement; composition of the con-tainment atmosphew: stratification or other inhomogeneities; thermodynamic state of 3
-[ i i l i atmosphere; mean and fiuctuating (turbulent) velocity flow field; type ofignition source; geometric configuration of compartments, connecting passageways, and obstructions. 1 To make any progress on analyzing this hazard, the problem must be grossly simpli-fied to a single configuration and set of initial conditions. A commonly--used worse case is a tube or channel, open only at the ends, and partially obstructed by obstacles protruding out into the channel at intervals along the length. For initial conditions, a dry (no steam) hydrogen-air mixture at an elevated temperature (370-450 K) probably represents the most hazardous case. Having made these choices of a configuration and initial conditions, a criterion for transition from low speed flune to detonation is now required. Analyzing expenmental results in both rough and smooth tubes, Lee (1986) has suggested that a characteristic detonation prol,erty of the gaseous mixture, the detonation cell size S, must be less than some multiple of the channel width TV before transition can occur. Observations of accelerating fiames and transition to detonation within tubes (Lee 1986. Sherman et ~ l.1989) suggest a tentative transition criteria: If S 5 IV, then a fiame may accelerate c and become a detonation. ( Using thir transition criteria, we now chose a reference channel width IV and by extrapolating the existing cell size data, determine the mixture composition which corresponds to the limiting case for transition. A standard engineering practice is to incorporate factors of safety into evaluations of the present type, particularly when the computation involves so many assumptions and uncertainties. This implies that a " safe' mixture is one which has a cell size exceeding the limiting value by some safety factor, i.e., S 2 21V. What channel widths IV are possible in a reactor containment? A wide range of values (0.5-50 meters) are present in existing containments,5 meters is chosen in the present analysis as a representative width for proposed designs. Cell 2ize data from the HDT experiments at Sandia were obtained from published reports (Tieszen et al 1987) and provided directly by the researchers (Stamps et al 1990) at SNLA. The data for lean mixtures are shown in Figure 1 for both cold (300 K) and hot (373 K) tests of hydrogen air mixtures. Note the large amount of uncertainty and the lack of data below 13.67o H for cold mixtures and 12.57o H in hot mixtures. 2 2 Cell size mereases with decreasing hydrogen concentration for lean nurturer En the cell size approaches the value of rD (D is the tube diameter), only a single instability mode is present and the detonat;on is said to be " spinning". Spinning persists for lower concentrations but the detonation is highly unstable and becomes harder to initiate as the H concentration is decreased. The leanest mixtures that could be initiated were 2 11.77o H for the cold cases and 9.57o H for the hot cases. 2 2 In order to infer the mixture composition at cell size of 5 m, the present data must be extrapolated. This extrapolation was carried out using the relationship S = AA between cell size and idealized reaction zone length A. An idealized reaction zone length A is determined by a chemical reaction kinetics computation and the ZND model of one-dimensional, steady reaction zone structure of a detonation traveling at the Chapman-Jouguet velocity. Details of this model and results for hydrogen-air-4
A w w diluent mixtures are described in Shepherd (1980). This model and the correlation S = AA are reasonably successful at correlating data over a wide range of compositions and thermodynamic states. It has been extensively applied to other fuel-oxidizer systems (Westbrook and Urtiew 1982) with equal success. In order to use the correlation S = AA, the factor A must be empirically determin,A This factor depends on the mixture composition and the method for determining A. Previous work on cold hydrogen air mixtures (Shepherd 1986) yielded values between 5 and 50 for A, with a systematic depeadence on equivalence ratio, shown in Figure 2. For the leanest mixtures, the value of A appeared to sharply decrease with decreasing hydrogen concentration. The trend shown in Figure 2 for the leanest mixtures is incorrect since it was based on measurements taken in a smaller tube (0.3 m diameter tube at McGill, see Guirao ef al 1989) under conditions for which the detonation was spinning; these cell size measurements are now known to be in error (too small) for concentrations less than 15%. Analysis of the more recent HDT data indicate that for H concentrations between 11.5 and 14%. the factor A is between 5 and 15, with an 2 average value of about 10. The factor A is shown for the lean cases in Figure 3; note the large amount of scatter in the data and the higher values for more recent results. Using a value A = 10, the S data for cold mixtures between 15 and 13.5% H2 and the S data for hot mixtures betwe<:n 15 and 12.5% H can be reproduced by the 2 correlation S = AA. The comparison between data and the correlation is shown in Figure 1. Extrapolation to leaner mixtures than measured (larger cell sizes) is shown for the cold mixtures in Figure 4, for the hot mixtures in Figure 5. A cell size of 5 meters (reaction zone length of 0.5 m) corresponds to a cold mixture of 11.9% H, or a 2 hot mixture of about 10.4% H. Applying a safety factor of two yields a minimum cell 2 size of 10 m this corresponds to cold mixture of 10.6% H2, or a hot mixture of 9.8% H. 2 Recommendation I recommend that the maximum hydrogen concentration should be limited to less than 10% H2 by volume. This value is based on constraining the composition to mixtures with cell sizes greater than 10 meters. This dimension is set by choosing a cell size twice the minimum required for transition to detonation in a partially obstructed channel with a characteristic width of 5 m. The transition criteria is based on our best knowl-edge at the present time: the cell size must be less than the tube or channel width for transition to be possible. Note that this recommendation is completely consistent with the detonation limit estimate of 9-11% H given in the NAS report (1987) several years age. That result was 2 based on similar reasoning but used a characteristic dimension of 50 m. The present results are somewhat sharper due to the improved cell size data base available today. 5
g w w i l i Summary I We conclude that hydrogen concentration limits should be set on the basis of the po. tential for fiame acceleration and deflagration-to-detonation transition. The marimum safe concentration is determined by equating the detonation cell size to a minimum characteristic dimension of the compartments within the containment. Concentrations are calculated by extrapolating presently available cell size data using establiabi cor-relations between computed reaction zone lengths and cell size. Choosing a minimum cell size of 10 m yields a maximum safe H concentration of 10% for a dry mixture at an elevated initial temperature (370 K). The analysis is limited to dry compositions at moderate initial temperatures. Cell sizes have been obtained by extrapolation of existing data using a methodology that is reasoncbly reliable. There are uncertainties in the data and also introduced by our assumptions. However, this approach represents the state of the art in hazard analysis at the present time. The approach that has been used has not been specially created for nuclear power plants but is a generic approach developed and used for many gaseous I combustion hazards. There are a number of other complicating factors that could be considered: effect of steam in the containment atmosphere; effect of elevated temperature in the containme nt atmosphere; plant-specific geometry; accident sequence particulars; etc. We have not considered the actual stress loading produced by a detonation or accelerated flame. The stress history and the structural response should clearly be a factor in any final hazard analysis. At the present time, structural response and failure under conditions of impulsive loading are not well understood for containments. Consideration of any of these issues requires a much more elaborate analysis than the present study. However, the simple but fundamental considerations of the present work would still be valid in such future studies. Acknowledgement I would like to thank Dourlas Stamps and Sheldon Tieszen for providing their most re-cent cata on detonation tell size and discussing the interpretation of their cxperunental results. l l I 1 6 i )
j. g g w w References M. Berman, J. Larkins and L. Thompson 1982 Hydrogen and Water Reactar Safety proceedings of the Second International Conference on the Impact of Hydro-gen on Water Reactor Safety, Albuquerque, NM, October 3-7, 1982, Sendia NationsJ-Laboratories Report SAND 82-2456. J. C. Cummings, J. R. Torczynski, W. B. Benedick 1987 Flame Acceleration in Mixtures of Hydrogen and Air, Sandia National Laboratories Report SAND 86-00173. C. M. Guirao, R. Knystautas, J. E. Lee 1989 A Snmmary of Hydrogen-Air Detonation Experiments, NUREG/CR-4961, Sandia National Laboratories Report SAND 87-7128. J. H. S. Lee and I. Moen 1980 "The Mechanism of Transition from Deflagration to Detonation in Vapor Cloud Explosions.' Prog. Energy Combust. Sci. 6, 359-389. J. H. S. Lee 1986 "The Propagation of Turbulent Flames and Detonations in Tubes," in Advances in Chemical Reaction Dynamics, Reidel,345-378. NAS 1987 Technical Aspects of Hydrogen Control and Combustion in Severe Light-Water Reactor Accidents, available from Energy Engineering Board, Commhsion on Engneering and Technical Systems, National Research Council,2101 Constitution Av-enue, NW. Washington, DC 20418. NRR 1989 Revised Severe Accident Research Program Plan, Division of Systems Re-search. Office of Nuclear Regulatory Research, U.S. Nuclear Regulatory Commasion, NUREG-1365. M. P. Sherman, S. R. Tieszen and W. B. Benedick 1989 Flame Facility, NUREG/CR-5275, Sandia National Laboratories Report SANDS 5-1264. J. E. Shepherd 19S6 " Chemical Kinetics of Hydrogen-Air-Diluent Detonations," AIAA Progress in Astronautics and Aeronautics, 106,263-293. D. W. Stamps, W. B. Benedick, S. R. Tieszen 1990 " Hydrogen-Air-Diluent Detona-tion Study for Nuclear Reactor Safety Analysis," Sandia National Laboratories Report SANDS 9-23S9, to be published. F. Tamanini, E. A. Ural. J. L. CharTee, J. F. Hosler 1989 " Hydrogen Combustien Experiments in a 1/4-scale model of a Nuclear Power Plant Containment," in Twenty- - Second Symp. (Intl.) Combustion. The Combustion Institute, 1715-1722.
z. w w R. T. Thompson 1988 Large-Scale Hydrogen Combustion Expedments, EPRI Report, NP-3878, Vol.1. S. R. Tieszen, M. P. Sherman, W. B. Benedick, and M. Berman 1987 Detonability of H -Air-Diluent Mixtures, NUREG/CR-4905, Sandia National Laboratories Report 2 SAND 85-1263. S. R. Tieszen, M. P. Shennan, W. B. Benedick 1989 Flame Acceleration Studies in the MINIFLAME Facility, Sandia National Laboratories Report SAND 89-0859. C. K. Westbrook and P. A. Urtiew 19S3 " Chemical Kinetic Prediction of Critical Pa-rameters in Gaseous Detonation," in Niniteenth Symp. (Int 1.) Combustion, The Com-bustion Institute, 615-623. l 8
[ ] y o 9 -~. 1 i l 10 A 10 A - j 1.60 - e - e- -- c - e +e-S = -p: ? 1.20 ~ <3 e =l i .80 trj o o -.s us .I p c o .40 j O T, = 370 K O T. = 300'K ,y U E spinning detonation ,, 3 + spinning detonation -{ I l i l i i .00 .08 .10 .12 .14 .16 ~ t 0 He mole fraction t ? 1 Figure 1. Detonation cell sizes measured in lean hydrogen-air mixtures at two-initial temperatures. Data from Tieszen et al. (1987) and Stamps ct. al. (1990). Curves shown~ are predicted cell sizes based on S = 10 A, where A is the calculated reaction zone length. Half filled data points- -[ - refer to spinning detonation cases. .l t ~9 ? c
w-l I C i S 'i C o T f i 6 t 4o I = yo O" a O CO ..O C! i =c w m-f ? cr.o ? c. o 0 10 Equivalence ratio l 4 e Figure 2. Ratio A of detonation cell size S to calculited reaction zone length ' A for hydrogen-air mixtures at I atm and 300 K as a function of the [ equivalence ratio. Reproduced from Shepherd (1986). 10 i ~
,o 1, _ .j 1 .,.y....J O. a, 3i i T t 6 i .20.00 O McGill.3 m tube P .a SNLA HDT (300 K) O O O SNLA HDT (370 K) a o 15.00 l O O a i O. 10.00 G ^ l 3 a 3 A Ca O a a a, a m I I -5.00 .30 .40 .50 equivalence ratio p d d 1 ro 3. Ratio A of detonation cell size S to calculated reaction zene length i a for lean hydrogen-air Inixtures (hot and cold) as a function of the. equivalence ratio. The filled data points are McGill data taken under l conditions of spinning detonation and were not used in the evaluation of A.- 11 N
o o I 2 10 a stable detonation - 10 A E spinning detonation O 1 5 a failure to initiate 1 <1 10 = ~ u o m c .g s _ _ s _ s _ - S = rD = 10 c u c 10-1 .08 . 10 .12 .14 .16 H mole fraction 2 Figure 5. Detonation cell size S and scaled reaction zone length 10A as a func-tion of hydrogen mole fraction: hot lean mixtures. 370 K. Curve shown is identical to upper curve shown in Figtue 1. The onset of spinning detonation is indicated by the dashed line labeled S = ; D, = 1.43 m for the HDT. No cell size data are available for lengths larger t.han this. 13
- y. '., ;*8
'g UNITED STATES pA LEAR REEULATCRY COMMISSIO J EDOPrinci$aNobeNoEenceControl { ~ ~w L' - i FROM: DUE: 05/20/93 EDO CONTROL: 0008828 DOC DT: 04/26/93 FINAL REPLY: Pcul Shewmon ACRS i TO: q p g 93 Ch31rman Selin ~ FOR SIGNATURE OF: -** GRN CRC NO: Ex cutive Director y DESC: ROUTING: ' SECY-93-087, " POLICY, TECHNICAL, AND LICENSING Taylor ISSUES PERTAINING TO EVOLUTIONARY AND ADVANCED Sniezek LIGHT-WATER REACTOR (ALWR) DESIGNS" Thompson Blaha DATE: 04/27/93 Nat Taylor Beckjord, RES ASSIGNED TO: CONTACT: Bernero, NHSS NRR Murley Jordan, AEOD Scinto, OGC SPECIAL INSTRUCTIONS OR REMARKS: PREPARE RESPONSE TO ACRS FOR EDO S:GNATURE. PUT COMMISSIONERS AND SECY ON CC (SHOWN ON 3 ORIGINAL) FOR REPLY.- P NRR RECEIVED: APRIL 28,1993 ACTION: DAR: CRUTCHFIELD ACTION NRR ROUTING: TEM ras DUE TO NRR DIRECTOR'S OFFICE JP' YG BY ba<x /7, 'L3 -NRR FAIL ROOM j /> P P W 4 n l l y '.
4.. r '4 g j ft W -\\ / ~a trz4 %'n-UNITED STATES - NUCLEAR REGULATORY COMMISSION l g ,I ADVISORY COMMITTEE ON REACTOR SAFEGUARDS f o,, WASHINGTON, D. C. 20Elb5 j April 26, 1993 I i The Honorable Ivan Selin Chairman e U.S. Nuclear Regulatory Commission Washington, D.C. 20555
Dear Chairman Selin:
i
SUBJECT:
SECY-93-087, " POLICY, TECHNICAL, AND LICENSING ISSUES PERTAINING TO EVOLUTIONARY AND ADVANCED LIGHT-WATER REACTOR (ALWR) DESIGNS" During the 396th meeting of the Advisory Committee on Reactor Safeguards, April 15-17, 1993, we discussed the NRC staff posi-
- tions, delineated in SECY-93-087, on policy, technical, and licensing issues pertaining to evolutionary and advanced light-water reactor designs.
During this meeting, we had the benefit of discussions with representatives of l the NRC staff and of the documents referenced. We have discussed' these issues during several of our previous meetings and provided comments and-recommendations in the reports referenced. We are in general agreement with the staff's positions in SECY 087; however, we have concerns regarding some issues and offer our 'omments and recommendations as follows. (The section titles and c letter designations correspond to those in SECY-93-087.) I. SECY-90-016 ISSUES l E. Fire Protection In our April 26, 1990 report, we pointed out that redundant train separation is likely to be the -most' significant i feature leading to reduced fire risk. We recommended that the proposed fire protection enhancements include separa-1 tion of environmental control systems (i.e., separate j heating, ventilating, and air conditioning-(HVAC) systems j for each train). The staff responded by conceding:that i separate HVAC arrangements may be needed, although other options may be available to the. designer. The commission endorsed the staff's response. We remain concerned that a common normal ventilation system (such as that proposed for the ABWR) will be difficult to design to prevent the affluent from'a postulated accident in one train of engineered safety features from reaching. essential mitigating ' equipment in the other - trains and 2 0 ' 008828 N - U 3 s y. p.,.c
~,q o o w w The Honorable Ivan Selin 2 April 26, 1993 creating conditions that exceed their environmental qualifications. Of particular concern is the capability of ventilation dampers to isolate the effects of high energy pipe ruptures in confined compartments served by the common IWAC system. G. Hydrocen Control The staff claims that it has sufficient basis for under-standing hydrogen behavior to go forward with licensing criteria. It has not been demonstrated to us that this basis is as extensive, or applicable, as the staff be-lieves. Further, the AP600 and ABB-CE System 80+ designs have containments that are more susceptible to significant damage from hydrogen detonation than most existing and evolutionary plants. This requires that the licensing criteria for this issue be reconsidered. H. Core Debris Coolability The staff has weakened the position taken in SECY-90-016 by not requiring that the core debris be adequately quenched. We believe that the present criterion for coolability, 2 namely a cavity floor area greater than 0.02m /MWt, is not soundly based. We recommend that the staff validate containment response to core-on-the-floor accident sequenc-es by independent analyses using, for example, MELCOR, or CORCON and CONTAIN. J. Containment Performance We agree with the requirement that containment stresses not exceed ASME Code Service Level C for metal containments, but it is not clear how electrical penetrations through the containment should be considered. Such penetrations utilize nonmetallic electrical insulation as a portion of the containment boundary and need further consideration. L. Eauiement Survivability We agree that passive plant design features provided only for severe accident mitigation need not be subject to the environmental qualification requirements of 10 CFR 50.45. We believe, however, that such mitigation features must be designed to provide reasonable assurance that they will operate in the severe accident environment for which they are intended and over the timespan for which they are . needed.
6 O C' The Honorable Ivan Selin 3 April 23, 1993 II. OTHER EVOLUTIONARY AND PASSIVE DESIGN ISSUES Q. Defense Acainst Common-Mode Failure in Dicital Instrumenta-tion and Control Systema The staff's second recommendation is that the vendor or applicant analyze each postulated common-mode failure for each event that is evaluated in the accident analysis section of the safety analysis report (SAR). We recommend that the scope of this assessment include consideration of common-mode failures during all events postulated in the SAR (e.g., fire, flood, pi essential power sources) pe rupture, and extensive loss of and r.ot be restricted ' to those events discussed in Chapter 15, " Accident Analysis." T. Control Room Annunciator (Alarm) Reliability The staff's basic recommendation is that the Commission approve the position that the alarm system for ALWRs meet the applicable EPRI requirements for redundancy, indepen-dence, and separation. These requirements do not include the use of Class 1E equipment and circuits. The staff also seeks approval of an additional position that goes beyond the EPRI requirements. This position is that " alarms that i are provided for manually controlled actions for which no automatic control is provided and that are required for the safety systems to accomplish their safety functions, shall meet the applicable requirements for Class 1E equipment and circuits." We believe that the staff needs to provide clarification and additional justification for this position. Collectively, our identified issues represent a significant array of incompletely addressed concerns. We urge that they be addressed on a timely basis to ensure their early consideration by the design teams. Sincerely, Paul Shewmon Chairman Referenqgg: 1. SECY-93-087, dated April 2, 1993, for the Commissioners, from James M.
- Taylpr, Executive Director for Operations,
- NRC,
Subject:
Policy, Technical, and Licensing Issues Pertaining to Evolutionary and Advanced Light-Water Reactors (ALWR) Designs
g _ s r g a. The Honorable Ivan Selin 4 April 23, 1993 2. Report from Paul Shevmon, ACRS Chairman, to Ivan Selin, NRC Chairman,
Subject:
Computers in Nuclear Power Plant Opera-tions, March 18, 1993 3. Report from David A. Ward,-ACRS Chairman, to James M. Taylor, Executive Director for Operations,
- NRC,
Subject:
Draft Commission Paper, " Design Certification and Licensing Policy Issues Pertaining to Passive and Evolutionary Advanced Light Water Reactor Designs," September 16, 1992 4. Report from David A. Ward, ACRS Chairman, to Ivan Selin, NRC Chairman,
Subject:
Digital Instrumentation and control System Reliability, September 16, 1.!92 J 5. Report from David A. Ward, Aq:RS Chairman, to James M. Taylor, Executive Director for Opepations,
- NRC,
Subject:
Issues Pertaining to Evolutionary ani Passive Light Water Reactors and Their Relationship to Current Regulatory Requirements, August 17, 1992 6. Report from David A. Ward, ACRS Chairman, to James M. Taylor, Executive Director for Operations,
- NRC,
Subject:
Issues Pertaining to Evolutionary and Passive Light Water Reactors and Their Relationship to Current Regulatory Requirements, May 13, 1992 7. Report from Carlyle Michelson, ACRS Chairman, to Kenneth M. Carr, NRC Chairman,
Subject:
Evolutionary Light Water Reactors Certification Issues and Their Relationship to Current Regula-tory Requirements, April 26, 1990 e e em}}