ML20149H279
| ML20149H279 | |
| Person / Time | |
|---|---|
| Issue date: | 03/29/1985 |
| From: | Ross D NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Myers H HOUSE OF REP., INTERIOR & INSULAR AFFAIRS |
| Shared Package | |
| ML20149B626 | List: |
| References | |
| FOIA-87-728, FOIA-87-853 NUDOCS 8802190198 | |
| Download: ML20149H279 (31) | |
Text
{{#Wiki_filter:, B RBMinogue, RES SBassett, RES
'~
- RBernero, RES OCA -
ISR 2 5 W TRehm, EDO DISTRIBUTION: Subj chron c hossrdgfile
)Dr.HenryMyersScienceAdvisorSubcomittee on Energy and the Environment
/ Comittee on Interior and Insular Affairs United States House of Representatives Washington, D.C. 20515
Dear Dr. Myers:
SUBJECT:
FOLLOWUP TO TELECON OF 3-25-85 I agreed during the subject telecon to provide references concerning various statements by NRC bodies on the safety aspects and research aspects of the B&W boiler-condenser mode of heat transfer. Attachment I has the Appeal Board statement. The yellow hi-lights are the relevant parts. (Recall that feed & bleed is gerinane only to a total loss of main & emergency feedwater systems). Attachment 2 contains connents made by the Office of Research, NRC, on the purpose of the integral systems testing proposed by us. Attachment 3 is a statement on the need for integral systems testing taken from a report by the test advisory group, in June 1983. Attachment 4 is an extract from the ASLB Partial Initial Decision on THI-1 restart. It contains their evaluation of my testimony during the restart proceedings, in early 1980. I highlighted some of the points relevant to our conversation. The total research on B&W intnral system testing is estimated to be in excess of $30 million. The total share of Reserch, NRC, is about 42%, or about $14 million. The others in this cooperative project are B&W, the B&W owner's group, and EPRI. Sincerely. Odginal Signed i% Denwood F. Ross, Deputy Director Office of Nuclear Regulatory Research
Enclosures:
As stated 8802190190 880216 720 PDR
$0$NOO DD/RES OCA /
DFRoss: bas EDO[ I I 3/26/85 3/A?' /85 3/ /85 , g 1
0 l ATT/lCHMENT 1 . , i l 1 The follcuing are relevent quotes taken from the Atomic Safety and
!.icensing Appeal Board proceedings on restart of THI, Unit 1: l The Appeal Board Onds, with some express qualiGcations, that all the systems. structures and components it etamined are suMciently rel6abic to permit restart of the Umt. Because, howeser, there are other safety issues that are before the Commission for its separate and exclusive determination, the Appeal Board does not decide the overall question of
~ h 8%i{
the opersbility of Unit I, leaving it to the Commission to decide after it has examined all systems and considered information within and outside the record in thi> proceeding. Generally speaking, the Licentina Board found various deGciencies in design and procedures that must be correcied before the plant is permitted to restart. The Board concluded, however, that, if the deficiencies are
~
corrected, Unit I can be operated in the short term without endangering { } E 2. \ the public health and safety. The Board also found that the licensee has made reasonable progress with respect to various necessary and sufficient long term actions which provide reasonable assurance that TMI I can be operated in the long term without endangering the heahh and safety of the public.
- c. Bodcr ConJenserProcess in summary, the analyses by the hcensee (which hase now been ap-proved by the staff in Board NotiGeation BN 83 21 A, dated March ll, 1983) indicate that the boiler condenser process is capable of removing core decay heat such thal HPI now will exceed break now before core un-covery occurs. We believe that the heat removal calculationsinclude sum- ~
gg cient conservatisms to make a full scale test of the boiler condenser pro-cess at TM11 unnecessary before restart. However, we recommend that this cooleng process be studeed further as part of continuing research in order to increase the current knowledge of thermal hydraulic behavior
, durin6 small break loss of coolant accidents.8"
, ed. Ferdcod hred Based on the testimony of several staff and licensee witnesses,'* the Licensing Board found that, in the event of a failure i f the emcrgersy feedwater system, the core could be adequately cooled using feed and hiccd while repairs to the emergency feedwater system were being niedr,."'
Furthermore,in its investigation of the rchability orthe emergency feedwa. ter system, the Licensms Board assumed "an additional safety factor of 100" because of the feed and bleed option."8 We conclude that there is in-
' D pg sufficient evidence of record to support the Licensing Board's reliance on feed and bleed to provide core cookng at TMI 1. In reopening the record on decay heat removal issues, we anticipated that supplemental evidence would enable us to make a Gnal decision on the viabihty of feed and bleed cochng at T Mi 1. Howes er, as we will disciiss,it is not possible on the bases of Ihe supplemenied record Io reach a final conclusion on this cooling process.
In the absence of a rehable emergency feedwater system, it has been suggested that dway heat could still be removed by the "feed and bleed" method. This is discussed in Section IlliA), utfra. The licensee has not relied on the adequacy of this backup method of removing decay heat. It argues that modifications to the emergency feedwater system now make it safety grade for the transients that are the subject of the restart proceeding, and that feed and bleed is needed only for accidents beyond the design basis of the plant and thus need not be examined for the pur. pose of determining whether TMI l may be safely restarted. It chose, therefore, to demonstrate that, mthm the scope of the issues ni this case, natural circulation is reliable. As a result, it was left to the staff to demons-trate the reliabihty of feed and bleed. We find that the staff was unable to make such demonstration. Plainly the feed and bleed process is conceptu. - 9 ally valid. The staff, however, was unable to resolve numerous analytical uncertainties with regard to the process. As a consequence, we are unable on this record to endorse feed and bleed as a reliable backup system of decay heat removal. We believe that further analysis or testing is required to prove the viabilety of feed and bleed as a decay heat removal method. In sur,: mary, we find that all the systems, structures and components we have examined are sufnciently rehable to permit ressert, with some qualiGcations as expressly noted. in light of the bifurcation of issues be. tween the Commission, on the one hand, and the adjudicatory boards, on the other, the Commission must determine - after examining allsystems and considering the information that is both within and outside the record before us - whether there is reasonable assurance thai Three Mile Island Unit No. Pcan be operated without endangering the health and safety of the public. 4
ATTACHMENT 2 I t O IST PROGRAM BASIS / PURPOSE General Aim Obtain Integral System Test data to study themal-hydraulic phenomena unique to the B&W NSSS. Particular Objectives The IST program is formulated on the basis that valid dJth *111 be generated to (a) satisfy requirements of TMI-2 Task Action Plan Stem !!.K.3.30 which requires that the SBLOCA model be compared to applicable data. (b) gain technical insight and allow assessment of B&W abnonnal transient operator guidelines, SBLOCA scenarios and steam generator tube rupture (c) benchmark and assess advanced safety computer code. (2 -(m e % .;-< h AJ ~ Istfe ~ d h. d p *~7 r, t f i- 5 "C m m a n r o e Li
~
en osw a t .a s P,4. J ' Question 1
!s the purpose of this research to demonstrate that B&W plants meet NRC requirements?
Response
Yes, one of the purposes of the B&W Integral System Experiment Program is to satisfy the requirements of TM1-2 Task Action Plan (TAP) Item II.K.3.30, which requires that the small-break model be compared to applicable data. While the primary emphasis of the program is on satisfying II.K.3.30, it also addresses other NRC issues, such as technical insight into, and assessment of, the B&W Abnormal Transient Operator Guidelines (ATOG), steam generator tube rupture in a once-through steam generator (OTSG), and secondary side blowdown in generators of this type. It is important to reafize that this program will produce experimental data that will be used to perform an assessment of how well computer codes predict the phenomena in these transients. B&W has the sole responsibility to perform the assessment of their computer codes? NRC will use the data to assess its computer codes to allow independent audits to be performed.
. This dual use of the data from the B&W Integral System Experimental Program forms the base for the cooperative program.
k
Qwestion 3 ! l How necessary is the research to $4W owners ability to demonstrate compliance with safety requirements? i i
Response
The integral systems test data is consid,ered long-term confirmatory research and does not directly relate to present licensing issues.* However, we believe the closing of present outstanding licensing issues i must be predicated on the expectation that the technical judgments we make today will be tested by a longer term confirmatory research program. This approach is not unlike that previously used by NRC to justify the acceptability of Appendix K 10 CFR Part 50, tending completion of longer i term confirmatory research (e.g., LOFT and FLECHT). Therefore, the experimental data to be obtained from the B&W research program is considered necessary to ultimately demonstrate compliance with safety requirements, i l k i I l I l 1 1 i
, -- , - -- - - -- ,-- - ---n- ,--n-.. . ,- - -,--,, .. . - -- - --,- -,. . ~ - -
ATTACHMENT 3
@ e [h b - N 6- /2c p , 7% Iit3 f
i 3. THE NEED FOR INTEGRAL SYSTEM TESTING In late 1982, the TAG met several times to develop a list of testing needs and to review existing test data. These data would be used to gain addition-al confidence in understanding transient thermal-hydraulic phenomena of B&W plants. Some of these concerns focused on understanding slow cyclic pres- - sure changes postulated by B&W to occur during certain small break loss-of-coolant accidents (SBLOCAs), assessing the revised SBLOCA models, character-izing natural circulation, and in particular, conditions occurring in the hot leg U-bend region. From these early discussions, the TAG developed a list of issues that it felt should be addressed through an IST facility. The B&W Owners were of the opinion that the GERDA 1 x 1 f acility was ade-quate, while the NRC felt that a 2 x 4 test facility was needed to adequate- ' ly address the issues. ' The resultant composite list of issues was categorized into four basic phenomena as shown in Table 3-1. These phenomena are: natural circulation (NC ) , SB LOC A, fee 0 and bleed cooling (FB), and steam generator tube rupture (SGTR). 3.1. Natural Circulation Due to the unique configuration of the B&W NSS, previous large integral test facilities did not model the unique B&W hot leg configuration or the OTSG, and, as a result, did not simulate the appropriate natural circulation condi-tions. In particular, there was uncertainty about the ef fects of two-phase flow, non-condensable gases, and the validity of the boiler-condenser mode ! i of heat remoital. In addition, the hydraulic stability, effects of high l point vents, and internal reactor vessel vent valves (RVVV) were items of interest. 3.2. Small Break Loss-of-Coolant Accident Since a major phenomenon in an SBLOCA in a B&W plant is the natural circula- l tion mode of heat removal, this issue should also be addressed in an IST. j , 3-1 Babcock & Wilcox
.w.,
1
, S,iQilar to otner hoLOCA testing procedures, parameters such as break size j and location, isolation of the break, and sensitivity to e<nergency core cool-ing system (ECCS) operation should be addressed. Also, the ef fect of reac- l tor coolant pump (RCP) operation was to be evaluated. As in the !C phence.
enon, the effects of the RYVVs, which are unique to the B&W design, were to be verified on an integral basis. 3.3. Feed and Bleed Mode of Heat Removal While not identified as a high priority issue, it was determined that cer-tain tests should be perfomed to assess the ccnputer code capability of pre-dicting this particular mode of heat renoval. 3.4 Once-Through Steam Generator Tube Rupture Understanding of the phencnena occurring during SGTR has been receiving in-creased attention. Since the B&W once-through steam generator (OTSG) pre- , sents a significantly different configuration from a U-tube SG, it was con-cluded that integral testing of tube ruptures would be beneficial. The effects of single and multiple ruptures, RCP operation, and asymetric loop ef fects were to be considered. The sequence of a tube rupture with a stean line break was also to be consioered in the proposed testing. The B&W Owners and the NRC each evaulated the list of issues above and rated them according to their importance in understanding one of the four basic areas of interest and the current knowledge of that issue. Subsequently, the TAG developed a caprcnise prioritization of the issues, and the results are shown in Table 3-2. This evaluation used a rating sys-ten from A to D to represent a measure of consensus of prioritization. An "A" rating was defined as top priority. A "D" rating was used to indicate a lower priority, although the issues rated as "0" were still considered to be of sufficient importance to warrant investigation. The issue ratings were used to set the essential features of the rectnmended test loop. A compromise was reached for all issues except the use (understanding) of the high point vents as they affect nautral circulation. In this case, the Owners rated the vents "D" and the NRC rated this issue as "A". The discrep-ancy is based on a difference of opinion about the state of knowledge of 3-2 Babcock a Wilcox _ _ __ _ ,_. _ . - - = w m mm m- - - - sv w'*- 'n'-%
such a device coupled with the intended use of high point vents in an operat-ing B&W reactor. The Owners felt that no new phenomenon needs to be tested regarding high point vents in order to challenge the adequacy of a computer code. In the case of an RVVV, the Owners felt that scaling in the proposed facility would preclude meaningful test data and could produce confusing or misleading behavior. The NRC, on the other hand, believed that the testing of high point vents was necessary since no data are presently available on the integral response of the B&W systems to such venting. They also believe that a computer code capable of reproducing effects seen in testing could be considered reliable enough to predict plant perfo rmance. However, the TAG agreed to include head and hot leg vents in the 2 x 4 facility and test ma-trix. The only other area of disagreement between the Owners and the NRC arose over the question of whether or not the categories of "Natural Circulation Cooldown" and "Cyclic Issues" should be listed on Tables 3-1 and 3-2. The Dwners considered that these issues are really particular tests under the heading of natural circulation and need not be further identified. The NRC considered these issues important enough to be identified as an issue and rated as "B" and "D", respectively. - 1 i j l l I
-~
3-3 Babcock & Wilcox
. me a -
4
I 1
' ~
ATTACHMEfiT 4 Cite as 14 NRC 1211 (1981) LBP 8159 UNITED STATES OF AMEAiCA l NUCLE AR AEGULATOAY COMMISSION l ATOMIC SAFETY AND LICENSING BOARD Before Administrative Judges l Ivan W. Smith, Cha rman ., 1 Dr. Walter H. Jordan ,, Dr. Linda W. Little l ,, I
'! . Matter of Docket No. 50 289 SP '
l (Restart) ' 4 l tOPOLITAN EDISON .#cey
, 1PANV -
e Milo Island Nuclear -M; tion, Unit No.1) December 14,1981 ;. APPEARANCES i o censee, General Public l'ailities Nuclear Corporation: l George F. Trowbndge. Thomas A. Batter. Ernest L. Blake. Jr.. Robert E. Zahler. Deborah B. Bauser, and Delissa A. Ridgway. Esquires. Shaw. Pittman. Potts & Trowbridge l Jclear Regulatory Commission Staff: l James R. Tourtellotte James \1 Cutchin. IV. Joseph R. Gray, j Daniel T. Swanson. and Lucinda Low Swartz Esquires
! immonwealth of Pennsylvania:
l Robert W Aster and \fichele Straube. Esquires \1r. William ! P. Dornsife 1 ! 'rs. \lerjorie St. Asmodt and Ntr. Norman O. Aamodt, pro se I l iti Nuclest Group Representing York: l
\fs Gail B Phelps l 1211 1
i 1 I l 1 l 4 l l
~
M a y' w%% - p,mmemme --
- 2
. . f.
m a " 7_ I , I, I . I I ~. Jones, (f. Tr. 4588, at 12,13: Tr. 7770 73 (Keaten). Licensee propmed j finding i 22. It L; i is the Board's view that the actions described here
- ad
- quately respond to the concerns listed in UCS Contention 2(c).
629 In summary the Board finds that operation of the reactor coolant pumps following a small break 1.OCA is not required to assure adequate ' i cooling of the core. In the event that natural circulation is interrupted, the decay heat can be dissipated by other means until natural circulation is i restored. Provisions are being made to operate in the feed and bleed mode l even if the lesel of radioactisit) in the coolant exceeds that experienced in the TNil 2 accident. 24*
- 8. Detection of inadequate Core Cooling (ICC) .*
- 1. Background and Findings on the Merits
! 630. The Board admitted three contentions which were concerned with j the detection of inadequate core cooling. UCS Contention 7 argued that j the public health and safety would not be adequately protected unless a d direct measure of water level was provided. Sholly Contention 6(b) states ] that the August 9 order was inadequate in that it did not require .
.yggi,
,v.
' "completion of the installation of instrumentation for the detection of inadequate core cooling' ANGRY Contention V(B) claims that the NRC **
j order does not protect the health and safety of the public in that it fails to , i require "Installation of instrumentation providing reactor operators direct % . information as to the lesel of primary coolant in the reactor core" 631. Although LCS and Ntr. Sholly ultimately withdrew their conten-tions on this inue, both the Staff and the Licensee responded to all three contentions. The Licensee's testimony was prepared by Robert W. Keaten,
'['
'V
}
\fichael J. Ross, and Robert C. Jones, (f. Tr.10.619. The Staff presented its testimony through Lawrence E. Phillips, (f. Tr. 10,807 (two sets j
hereinafter Phillips 1 and Phillips 2) and Deeweed F. Rosa, Jr., ff. Tr. 15,915. LCS, ANGRY, and Nir. Sholly did not submit any direct l testimony and did not participate in cross esamination. The Common- 4 i
*calth of Pennsylvania did cross-ciamine both the Staff and the Licensee witnesses and submitted proposed findings (tt 79101). See generally, Tr.
10,728 56,10.866 907,15,98716,000, and it. 036 39. 632. No intenenor participated in the evidentiary sessions at w hich (
- Licensee and Staff testimony was heard, nor did any intenenor submit i
proposed findings on inadequate core cooling issues. Consequently, this pertion of the decision is not directed to the intenenor contentions but rather to the issue of compliance with the Commission's August 9,1979 order, a matter of dispute among Staff, Licensee and the Commonwealth. i 1233 l i
,1 b
O
- . .. . . - . - - . . _ _ _ _ - - . _ _ - - - - _ _ - _ _ _ . - - - _ - _ _ _ _ _ - - . . . . ~l
i e d
,a + g w
- <- , ?,, - '
q m o a
=
633. Instrumentation for detection of inadequate core cochns (ICC) is " a mandatory issue in this proceeding in that it is included as Section
'# 2.1.3.6 of NUREG 0578. The TMl 2 Lessons Learned Task Force (Task Force) adopted the following positions: .
i e a
!. Licensees shall deselop procedures to be used by the operator to "
recognize inadequate core cooling with currently available in- 3 g ,, strumentation. The licensee shall provide a description of the g existing instrumentation for the operators to use to recognize these g conditions A detailed description of the analyses needed to form
-a
', the basis for operator training and procedure development shall be c provided pursuant to another short term requirement, "Analysis of ],I Off Normal Conditions, including Natural Circulation" (see Sec.
j~ g, tion 2.1.9 of this appendit). 3 in addition, each PWR shall install a primary coolant saturation " 7~ meter to provide on line indication of ecolant saturation condition. Operator instruction as to use of this meter shall include con- M N ', , 1 sideration that is not to be used esculsise of other related plant parameters. j
-- =,-
- 2. Licensees shall pro ide a description of any additional instrumen.
~
tation or controls (primary or backup) proposed for the plant to ; L supplement those desices cited in the preceding section gning an unambiguous, easy to interpret indication of inadequate core cool-ing. A description of the functional design requirements for the Q system shall also be included A description of the procedures to n 4 be used with the proposed equipment, the analysis used in de-seloping these procedures, and a schedule for instalhng the equip- % - ment shall be provided. {.- NLREG 0578 at A II, A 12.
' ' 634 On p 8 of NUREG 0575 the Task Force summarized item 213 b _
a fo N u Perform analyses and implement r rocedures and training for prompt recognition of low reactor coolant lesel and inadequate core -
;g cochng using esisting reactor instrumentation (now, temperature. 3 power, etc.) or short term modifications of esisting instruments De-scribe further measures and preside supporting analyses that will yield ]
j more direct indication of low reactor coolant lesel and inadequate core ; cooling such as reactor sessel mater lesel instrumentanon lemphasa s added). M 1234
=
0 5
,qr w' = ,
. =
~,g( '
, - e em.
m . _
j }
-.4x
. a
- 635. On pp A II 12 the Task Force discussion mentions a number of
, suggestions for directly measuring coolant lesel or soid fraction but con-cludes "that detailed engineering evaluation is required before design l , . j requirements for a direct lesel measurement system can be specified." l 636 Table B 1 of NUREG 0378 is the Task Force recommended implementation schedule for each of the recommendations. Position 1 of , 2.1.3 b, the use of esisting instrumentation plus the subcooling meter to j detect ICC, is considered a Category A item scheduled for early com- - pletion. Position 2, the design and implementation schedule for new in-j strumentation that will yield a more direct indication of low reactor sy
. cNlant lesel, is also a Category A item; only the installation of the new ..
) instrumentation is placed in Category B. We note, however, that under item 11 F 2 of NLREG 0737, it appears that the Staff no longer considers j that the requirements of position 2 must be completed prior to the j prepesed restart dates. However, the Staff does maintain that reasonable ] progress must be demonstrated prior to restart and that the Licensee has failed to demonstrate such progress Staff Es.11, at 28 30. 637 Positions 1 and 2 of 2.1.3 b will be discussed indiddually in the . following paragraphs. The measures taken by Licensee to meet the re-
! quirements of position I will be briefly described, and Staff concurrence in the adequacy will be documented Following that discussion we will esplore the dnagreement between Licensee and Staff concerning compliance with position 2 on the need for additional instrumentation. In the discussion that .I
' 4 l follows we will use the term "coolant level instrumentation" in a broad sense to denote a system that measures coolant level, coolant inventory, ,'
) I cxlant density, or some parameter closely related to the foregoing. 638- The instrumentation asailable at TMl1 which indicates in-4 adequate core cooling consists of core esit thermocouples which indicate
~
coola nt su perheat associated with escessive fuel cladding temperature, reactor coolant pressure sensors, cold leg and hot leg resistance tem- - r4rature detectors (RTDs) which proude inputs to compute the margin to ~ coola nt saturation conditions, subcooling meters which will display the margie to saturation, and reactor coolant pump current which provides indication of increasing coolant quality while the pumps are running. t, Further, prior to the restart of TMI 1, the Licensee will upgrade the 2 44 cuting instrumentation systems, information displays, and operating pro-cedures which relate to the detection of and response to inadequate core exling conditions These modifications in conjunction with improsed
- crerator training will substantially enhance the capability of the operator ~
J to recognite and respond to conditions of inadequate core cooling. Phillips-
- 1. If Tr 10.507. at 5, 6 7, Keaten (Detection of ICC), ff. Tr.10.619, at 79 l
1 l ms l i N +' S*W D en
- e may,e g
] *P *
,3\
1 4
- __m m_ m- -
-w i*
639. In addition to the upgrading of existing systems, the Licensee has ,, , i committed to the installation of a primary coolant saturation meter, has ;nge. described t*o short term modifications to existing instruments, and has 6; i proposed new emergency procedures to aid in the detection of inadequate dess core cooling. Phillips 1, ff. Tr. 10,807, at 6 7, The two short term gee. M modifications insolve routing the in core thermocouple signals outside of pyy containment and connecting the 52 in-core thermocouples to the plant the computer (for display purposes), and posiding an estended range for l, ade J. reactor outlet (Tu) temperature measurement (from $20' 620*F to 120' per 920* F). This modification will be made to four Tu channels, two in each l co, reactor coolant loop in addition, it is intended to isolate the new wide i , range Tu signal from the existing control signals. These signals will then be (o, seismic Category I and separated for use as redundant signals All gi modifications required for esisting instrumentation will be implemented 33 prior to TMl.1 restart. Id., at 7; Keaten, er al., ff. Tr.10,619, at 6 9. ge
'I 640. The proposed emergency procedures for inadequate core cooling p, (EP 120139) and 1202 68) rely on the information asailable from the g core exit thermocouples, reactor coolant system pressure, reactor sessel u outlet temperature, and the new saturation (subcooling) meters to identif) 7 y the approach and esistence of inadequate core ccoling and to specify the i e
g operator actions required to present or recoser from inadequate core cooling. Keaten, er ol.. ff. Tr.10,619, at 9. These procedures are under i 1 resiew by the Staff and resised submittals hase been required from the 1 Licensee. Although the Staff has not completed its review at this time,it is , confident that procedures acceptable for TMI l restart without reliance on water lesel measurement can be deseloped The Staff has found in-adequate core 'mling emergency procedures based on instrumentation t~ similar to that which will be provided prior to TMI l restart to be acceptable for other PWRs while a lesel measurement system to further enhance the operational safety is being deseloped Phillips 1, ff. Tr.10,507, at 6. 641. As we base discussed in our PID on management issu'es t" 196 204), the Licensee has also included specific training in heat transfer and fluid dynamics, plant operating characteristics, plant res;ense to tran-sients, and guidance for operator response to LOCAs in its Operator Accelerated Retraining Program (O ARP). All of the licensed TMI l operators will be required to complete the O ARP, This training, along with
.- the ongoing requalification training program (ser PID 't 180-195), is to I assure that the operators will recognize and respond to reactor coolant conditions approaching and following saturation. The training prosided to TMI I operators is intended to assure that the operators are aware of
~
.. a,
i i
- e i
assilable information on the status of core cooling and know how to I interpret it correctly." Keaten, ef al., ff. Tr.10,619, at 7.1415.
! 642. This Board agrees with Staff and Lietnsee that the measures l described above meet the requirements of position I of "Lessons Learned"
) Section 2.1.3 b and will be adequate to protect the health and safety of the , j public in the short term. In our opinion, those changes in equipment plus the changes in procedures and operator training, if effective, will provide adequate protection for the health and safety of the public for a limited
! period of operation. We now turn to a discussion of the basis for our '
l conclusion that a water level meter is required in the long term. ' 643. It is Licensee's position that the abose modifications are adequate 4 for the long term - that no additional instrumentation is needed. The
- Staff insists that further instrumentation to measure coolant level is neces.
! sary and that Licensee's plans for such instrumentation must be adequate j to demonstrate reasonable progress prior to restart. The Commonwealth of 1 Penns>hania recognizes that a cmlant level meter "would be desirable for , l
- the long term" but urges that further generic studies and testing be
, undertaken by the Staff prior to a commitment by Licensee. Licensee ;
J proposed finding 1 53; Staff proposed finding i 102; Commonwealth l proposed findings 11 98 100. l 644. In support of their position that instruments to measure water lesel are not needed to detect ICC, the Licensee relied heasily on the l testimony of a panel of espert witnesses, Messrs. Kesten, Ross, and Jones, }3 { _, ff. Tr 10.619f' Mr. Keaten defined ICC as: " . inadequate core cmling -
; is considered to esist when the fuel is uncoscred to an estent and/or for a E time such that the limits of 10 CFR .50 46 would be exceeded." /I, at 7.
hj'. -' 645. We base some problems with Licensee's definition of ICC. We j prefer Staffs definition: "When the two-phase froth level begins to drop 4 i below the top of the core, the esposed fuel begins to heat up and will pp l ultimately reach temperatures at which fuel damage occurs. This is in. j adequate core cmling." Phillips, ff. Tr.10,807, at 3. The differences in definition of the beginning of ICC do have some bearing on the differences (.
! of opinion on the need for cmlant lesel instruuentation. %
I 646 The Licensee urges us to reject the Staffs definition. Licensee PF { l T 34 This we decline to do. In our opinion ICC occurs prior to fuel damage - continued ICC leads to fuel damage and should be detected l when the fuel heat up process begins. j .. l l j
" Tbc depcadeve upm correct omf ator resNases in compnction o rth ibe ne* instrumentation wa dmro t hc orportawe of the rehabihty of NRC and Latases admnistered crersior p* i ed iSc a=NetaNe of the reomeed reca:eedieg on che Dag on trest tests Tke oarioiv s of.ibe e =>tnesses are attacked to ibe r nest moey Their etMrtise is j rc@ red by this Beard I
4 1237 l , i k i 1
l p , m L_._._.... 1 i l l l
/ 647 Staff witness Phillips has pointed out that neither the instrumen-tation proposed by the Licensee nor the coolant lesel instrumentation urged
- g. a b) the Staff are direct measurements of ICC, Phillips 1, if, Tr.10,807, at
- 4. A direct measurement of fuel ciadding temperature would be preferable i but is impractical. Lacking any direct indication of ICC the) urge that 4 Licensee's instrumentation be supplemented by cmlant lesel instrumen-tation - a second string to the bow.
A 648. TMI will have devices to measure the temperature of the coolant l l j
*- at the core exit and in the bot legs abose the core. If the measured exit l l
temperature is below the saturation temperature, the core is coscred with water. Jensen, er ol., ff. Tr. 7548, at 10 Since saturated conditions must 48.\ occur in the reactor coolant system hot legs before there is danger of b inadequate core cooling, the instrumentation available to the operators to ! detect a loss in the subcooling margin, including the new saturation meter which was not asailable at the time of the TMI 2 accident, prosides 1C ,k information anticipatory to an inadequate core ccoling condition. Thus, the j instrumentation provides the operator with knowledge that action should bc , i y( M taken to maintain or reestablish the subcooling margin and that an l i inadequate core cooling condition is being approacArt See Keaten, er ol..
.4 ff. Tr. 10,619, at 8 (Keaten); Tr. 10,7 9 30 (Kcaten), Tr. 10,8 8 30
- (Phillips). t 649. If an accident occurs which nesertheless results in the uncosering ;
of the core, superheated reactor cwlant conditions would be indicated by core esit thermocouples and the espanded reactor coolant hot leg tem-M, perature instrumentation Keaten, er al., If. Tr.10,619, at 5 (Jones); ; Phillips 1, (f. Tr.10,807, at 4 The Staffs witness Jensen testified that the ! ranges of this instrumentation used to monitor core cooling are adequate .
, for the eperator to determine if the coolant in and abose the core is subemled. saturated or superheated. Jensen, er at. ff. Tr. 7548, at 9. The Staff has suggested, nesertheless, that while core esit thermxouples can l . In preside an indication of the esistence of inadequate core cooling, the measurement of superheated steam temperatures by the core esit' ther-l mocouples indicates inadequate core cooling imminent or already present i
i . Staff Es. I, at C8 21. Thus it becomes apparent to the Board that the
.* differing definitions of ICC lead to differing conclusions concerning the I l l ,
need for additional instrumentation. ' 650. Both Licensee and Staff agree that the saturation meter would be i ' the first to indicate the need for operator action in the esent of a I ! small break LOC A Staff agrees with Licensee that the core esit tem-l perature indicators (RTDs) would preside evidence of core uncoscring and oserheating Howeser it is Staffs position that a lesel meter is need:d to I
< coser the period between the initiation of the break and the uncosering of the core Tr.15.992 (Ross). This period may be as long as 30 rmnutes to b
e4 f l
*V g
. gy g g-{ -
. . a. .
- .. s . "
A* y.Att. 4: WW " ~ .
.enie we-MM%. - ~2.rsw4ds.64h&
l l oser three hours. Phillips 2 (f. Tr.10.807, at 3. The tesel d meter obdates the need for the operator to fly blind for an estended period. It prou e4 an additional and divene method.o( determuting ICC. Tr.15,995 (Ross). 651. In order to asoid the onset ofto inadequate ensure that thecore cooling conditions, operators Licenice has taken specific steps at TMI.) , I understand the requiremer"s for adequate core cooling and are provided , l the cedures at necessary information to esaluate core coolant conditions. Plant pro l maintaining an adequate saturation margin in the reactor coolant sptem d and to proude guidance for steps to be taken if theat saturation 10.619 7. 8 margin is x f less (Keaten) than the required salue. Keaten, er al. ff. Tr.The reused procedures defin ble from the core esit thermocouples, reactor coolant system temperatures and the new saturation meter in identifying when inadequate core cooling, , by Licenscei definition, is approaching and to specify the operator action ' required to promptly enhance core cooling. // at 9. 652. For esample, in the immediate and follow up action requirements of TNil li procedure for loss of reactor coolant causing high pressure 48), strong emphasis is placed on maintaining injection (Licensee Esreactor coolant system pressure temperature relationships .
- to assur subcochng condition of at least 50' F esists. Specifically, the procedure .
requires that upon automatic initiation of HPl(1) allthe reactor coolant pumps low pressure I are tripped and HPI shall not be terminated unless injection sptem is in operation, flow is at a rate in escess of 1000 gpm in 4 each line and the situation has been stable for 20 minutes; or (2) the . degree of subcooling is at least 50' F (as determined by the saturation meter or the fne highest in core thermocouple readings) and the action is . *' necenary to present pressuriter lesel from going off scale high. If 50* f subecchng cannot be maintained, the procedure requires that full HPl shall be reinitiated. Licensee Es 48, at 2. 8. 65). The T\111 prc<edures, using the instrumentation described abose, suure that the operaters take the following key actions during any ap. proach to an inadequate core cooling condition: e a Initiate n:gh preuure injection. b \isintain steam generator lesel:
. i ' ",
- c. Trip the reactor coolant pumps if the engineered safety features actuation signal is initiated by low reactor coolant >> stem preuure; and, d
\ionitor core esit thermocouple temperatures to assure that ad- .
f equate core coating esists. 1239 e
- l l
l
b.s a m m , m & &. M No further 10,619, at 9. action IO. is required for design basis egents. Keaten, er al. ff. Tr. 654 Licensee has objected to the water lesel meter, partly occause it - would not aid the operator, but also in that the operstor might be misled and improperly turn off the HPI system. Licensee propmed finding t 65. Tr.16,649 50 (Jonest later questioning by the Board led to a restatement that the improper operator action would not occur if he were properly tr,iined in using water lesel instrutnentation " We taliese a better state. ment of Licensee's position (also adduced by Licensee) is that without an identifiable use for the instrumentation, its installation would detract,
. rather than add to safety from a human factors point of view. Licensee Es.
23, Appendis A at 2; Tr.10,644 45 and Tr.10,703 (Keaten): Tr.10,706 ( \1. R oss t 655 The Staff has resiewed the Licensee's justification for no ad-ditional instrun entation and found it unacceptable and therefore not in cumphance with position 2 of 2.1.3 b of NUREG 0578. Phillips 1, IL Tr. 10,bO7, at 9. Mr. Phillips gase a number of reasons why he belieses that a water lesel meter (or equivalent) is necessary at TNf t 1 for the protection of the health and safety of the public. We will summarire his testimony in the 10bl4J following paragraphs taken larfel) from Sta; proposed findings " " 656 The Licensee's procedures for detection of inadequate core ccchng rely primarily on the saturation meter and core exit thermocouples The saturation meter, while presiding a basis for initial actions, dxs not
. s;.. dminguish between anomalous transients which can drain the pressurizer and c.tuse primary twp saturation due to cochng and shrinkage of primary
,, cwlJnt sersus losb of Ccolant insentor) which could lead to inadequate core cooling if it continues. Phillips 2, ff. Tr.10,807, at 2.
657 ) g Licensee witness, in rebuttal, pointed out that in either esent, the
^'
proper esent action is HPl actuation, and that the operator can diagnose the from a knowledge of the secondary side parameters Tr. 10,711 16 (Jones). 655 The Staff has pointed out that the T\111 Emergency Procedure l tp 1202 6B describe > the different operator responses to small break LOC A
*ctsus l oscrcochng esents which cause automatic high pressure injection.
l These ritueedures now require the operator to distinguish between the
, transients based on indirect indicators from existing instrumentation Ves.
sel lesel instrumentation, if asailable, would permit a much quicker and i - 4 "in ru. .e Rm statet it sen ,Pd s.sc% .bca be iemr,c: for the Suff "I d mot hiswg l P h e r 6 P 11) ihat d 'i opctd! Jt c.P in t c rf ec t te t'P M 41 w f g 0;* a'd f c svec hl figt fi 4,C a
, T A.
vs .mJ, pd.d i 's t by
- 6ikc b oibet r c'assa ihtccN~ e Trrmcic's le One i d e da b riem i a' hc r c e '. (s . > 3 Pc .
- M N ce V
s 1240 a 5. 1 v-- L_w w w ~..w.y- n. . . ~ 1 ; ~ c w.WA J g: ( f.Nk." \")9. . 9' *. , j h n *M W'.9%QI
,,?-
> i$. ',
.. \
..e .
i
I a 4 1 l i I more reliable diagnosis of the conditions. For small break LOCA, an i orderly cooldown is required, but not necessarily for an overcooling tran-
- sient. In both cases, a sessel lesel meter, if asailable, would preside i coordinating information to assist the operator in restoring the water solid l primary system (possibly u"ng the upper head sent) and the normal water
! lesel in the pressurizer. Ph. ys.2, ff. Tr.10,507, at 2 3. 659 for a small break LOCA, the primary system will continue to lose coola nt insentory, at a rate and duration dependent on the size and I location of the break. until the safet) injection make up flow etcceds the } rate of coolant loss. For some conditions, the time intenal from the instant i of primary system saturation conditions until the occurrence of superheat ! indication on the core esit thermocouples or hot leg RTDs is in excess of 30 minutes, and possib!) up to three hours or more. The superheat , condition does not occur until the core is partially uncosered and fuel ' ! heatup has begun. /d. at 3. - 1 660 ff level instrumentation were asailable, the effectiseness of HPI in I recaerirg the system and the trend of leici indication (continuing to lose ) coolant or refilling the system) would proude saluable diagnostic infor- , 1 matien on the nature of the transient before the lesel drops into the core. The lesel indication would also proside c>idence that the core is cosered during recosery from a TMI 2 type flow blockage cono. tion, esen though C superheat may persist at the core esit thermocouples. None of the process a I parameters monitored by esisting instrumentatior, provide equnalent infor. O mation on a continuous basis. /d. at 3. W 661 In the siew of the Staff. sessel lesel inferraation is important and f. F I pombly essential to proper emergency procedures relating to use of the t { reactor scuel head sent required by the TMl Action Plan. Vessel lesel ,g , l information would indicate the esistence of a soid in the upper head so
- that the need for senet senting could be evaluated Id. at 4 5. .b l 662. The Staff belieses that while esisting equipment may be sufficient ...r -
j to respond to TMl type accidents, it may not be sufficient to respond to ' other u0dentifiable accidents. Tr.10.892 (Phillips). One of the lessons - j learned from the TMI 2 accident was that there need to be more diverse , instruments that would let operators cope with anomalous transients for which procedures and training do not esist. Tr. 15.994 ( Ross). Thus, Y instrumentation other than that currently being used is needed. Reactor sessel lesel is an additional and diverse method of determining inadequate core cooling. Tr.15,995 (Ross). Just what the Staff had in mind with . reference to anomalous transients was not made clear etcept that they - t were outude the scope of proposed procedures and training Licensee's ) poution a that the present procedures are adequate for any small break LOC A Keaten, er al., (f. Tr.10.619. at 5.12.14 and 19.
.j ., f
^? I l . p. '
l 1241 *i i, l i
^
i 9
- j, p
- 2 4 .- - -
-s.g o.- ,. T' l
l
' .. 663. In support of their position the Staff cites two recent incidents, l .
one at St. Lucie, the other at Sequoyah I. Both were small break LOCAs. g in the St Lucie incident of June 11, 1980 a bubble was formed in the reactor sessel head at a pressure beliesed to have been above saturation pressure There was an estended period of operator confusion concerning i h3 h.
. A.q.
gI_ the status of the system in the Sequoyah I loss-of coolant incident of February ll,1951, it required 35 mint tes for the operator to diagnose the esent D. Ross ff. Tr.15,915, at 3-4. 664 In a sense the St. Lucie and Sequoyah esents can be used in S . . support of both Licensee and Staff By following procedures, the operators
'~2 took action that terminated the events successfull). But procedures can
/ neser be an adequate substitute for operator understanding. The failure to diagnose cach esent quickly can only be laid to lack of information on the j part of the operators as to what was happening.
+
.; 665 We are consinced that a meter capable of measuring reactor
~
coof. int insentory from 100 percent to zero would be a useful and saluable l O j operating adjunct and is needed in the long term. To that extent we agree with the Staff Howeser, we do not agree with the Staff that the state of l
,. the art is sufficiently advanced to require a demonstration of reasonab!c i progres; by the Ucensee to the estent of requiring the completion of sis
[ ) *. items (hsted in Staff Es 14. at 29 30) prior to restart. ! h, 666. Staff witness Ross has described the state of development of i - ', coolant lesel meters by the nuclear industry. D. Ross, (f. Tr.15,915, at 11 12. It appears that Westinghouse may be able to adapt a pressure l differential system to their reactors but there was no eudence that such a i l system could be fitted to a B&W plant Combustion Engineering is looking ! !- - at heated junction thermocouples but has not completed design and testing Tr 15.977 (Ross). B&W has not committed to any des!gn Although many other plants appear to be further along than TMI I in their research and
.s des clerment e fforts, none hase met the requirements of position 2 of
** 213b 667. Licensee did not ignore the long term recommendations cf Section f
- ..l.3 b of NL:R EG 0578 Licensee's Restart Repart includes B i W's
! Evaluation of Instrumentation To Detect inadequate Core Cochng, Prepar-l ed for i?? O*ners Group. August 15. 1950. The following methods of
~ ,q detecting inadequate core cochng were esamined in this esaluation (1) l esisting core thermocouplet (2) additional asial core thermocouples, (3)
) - - ultrasonic reactor sessel lesel indication, (4) neutron or gamma beam reactor senel lesel indication, and (5) differential pressure transmitters for
~
j , i reactor sessel lesel indication The B&W evaluatien concluded that none j of the proposed methc6 of detection would meet all of the Staffs criteria. l . The report also concluded that each proposed reactor sessel lesel measure-j - - ment system concept fails to proside any additional aid to the operator for i
- 1242 i
i 4
! 3
- .. . . . l
x t . v
.'~,
~
~
i l
- 1. Supp. l. Part 2, h record includes the .
) detection of inadequate h core cooling. Licensee Es. Ans i n by Westinghouse and testimony of Licensee's witnesses don t e s T r. 10,724 25 l systems evaluated by B&W and under consi erat o10.70910 (Jones); Combustion Engineering. See Tr. i 10.759 67 (Keaten, Jones) f h elements of the - (Jones); Tr. 668. Licensee has been following the efforts o ot erh Institute, to investigate industry. including the Electric Power tResearc ms. Tr. 10.707-09 l ]c tinue to pursue potential reactor water level instrumentation lif they prose to be sys e ! IKeaten). Licensee has also espressed its intent to con l i, possible methods of measuring levelintethe with reactor and vesseking with the* other . l reasonable. Tr.10.919 (Keaten). In addition to worB&W ow l assist a professor at Pernsylvania State University inncept for measuring water le to pursue, first on a research reactor, Licensee a co has ndent also sought a evalua. d on the basis of using esisting neutron detectors. proposal f i tion of the ongoing work to develop reactor w rogress' ! bl 16.521 23 (Keaten). i mentation. As because it elected to inige4e the need 6es waaer le I , ! i litigative position
- can be seen from the Chairman's separate statement, tsiled on the issue, althoug '
was not frivolous; Licensee could have preva s find that before grounds narrower than it argued.inter The alla. Staff present;would have u reasonable erogress can be found, the Licensee must, 13) Evidence of a tangible commitment ecute the defined to performance or participa. y .j tion in the appropriate test rrograms to es development program. .s
' > c, .
Staffs witness Staff PF ? 104, citing StaffhiEs.14,i at i g it 29 30.670. We perceive 1 ] l commease6 4 hat "plant O 07owners 37." D. Ross, (f. who have devoted t l nel 1 espect to meet the ockeduled requirements c(PetJRE s in . ';j Tr.15.915,resistance at 7. While we understand why the Staffs technica person to their objectives as reasonable progrested above we refused g would not view meeting them, we view it somewhat l differently. As nol indication would mislead to accept Licensee's testimony that water eve i d But we beliese cperators. whom we espect and require tok be well tra ne .and B&W could not LOCA. l their witnesses when they testined h control room that Licenseeidentify a . and that unneeded instrumentatior could detract from t e l 1243 1 l l ,-
B T , f_ ,
,n g good place the emphasis uren Licensee s g, ,
human factor deegn Here ea presented as compared to the actual hv F
=
f ait h in behesing the testimony it W e would not espect Licensee te has a deur abihty of water lesel indication i D $ resistance when it 7 comm > ie ihn prop 3 sed deugn change withouta would be uselen and coun I j
! % w behef thatfind Licensee concern about the practicably of such f,instrumen.
! su g W e d. , net B&W analysa Licensee Es
" be unreasonable. gn en its ow n nad iaten te 10.' 24 2 5 t Jonesi e 5
Pa r t 2. T r 10 e45 a supr _- o' Nio coser esen the Staffi ow *W n witnew conclude thatteshfred no ssstem that there is st 10J 3 3 i Phi ~ ps . i ine pe b m t y that the Staffles ulumatch el is acceptable < T r prepese d to mea >ure w ater ble it wth res e* be'ere the Staff determines whether any system it agairstis arsacceptaprasided and we t he potentiat use of the informanon to be found acceptaSe a de:nmerts iTr* W!Ohas%: e to6:bei foundPh dtohpresi rs idehanen overallorder enhancement y _ preNsed sy stem make sucr a determinaton untd tre W - to d ets and the Staff wm not the operating methWs h a <. e be e n idenufied these the ssscs ac ,rcahed f
- N: - and test data are asariable. and the Sta f isTecertain that v'm gp__ s and *W ny lead te unsafe ach a s N ; ' te" s a e indeed a pha te safets*
e4 to 4% < Ph a ps i E = I
=
the Ln.crace
- m r.ut meet the requiremers frum o
- c. o': It u a p p.u ent tha tm Recommendanon 213 b before! restart r ma i
] Ns.t.on 2 d NUREG-0. toward tr a: gN n m a-
of s ,e a :n progress f rt Pe "
-- te eng ree r.g p riprogressed as rapidh as we thewould hasehas art Licensee hked L n.cn cc has not of the state pont ' s>e. and m siew meet >rg one : of Recomme-
- p -
regu a; n n ,' a t ed reawnable progren m C g cew instal cc - : Ib hus h p egreu a u,f%,ent for restartregard ng the need to CC EgL j ir s p - a's the Baard s nmt'on f#ews g = r' s >v-e a
- i de:c, .m adeva:e s . 'e s = a ng is J+
-f eca ' ae ad r
EIFF ' ala
- p ace at the m e a -- d p s e d . r e s me - tre-, e I
h ts s~t : .: - mea % e
- a m e
h l l eu . t
' the sC < term A meters -g term reo ed ;r the t_
}t *W am the ope a:er m : hat m' m
I e,aer s R ,'s a-d recm e -g t re u ra nt s ga;ed trar s.cr:s to reg mng of the f 6 age s j t r e ter te core uncesery C h_ y the s , ;-: m e-te', f*e* meet the Staff cntena at sene A though the licensee need not the deselopment ad n
~
,Y rea, 4 high priority shoutd be gner pa" ce a' ee frame s
_j the u mc vi restart. les ci mete N. requ re *r m 3 B_ 2g . inanuun of a reactor want bC6ft) 6et UC eas e 1. *re 5ta-a-j 'he(.~~ o t<
,e-
- .f -e, .
- c e e. -
- ,.s k A 4. s.
*y e,e [>*
N -e _ = 7 e P m p . r
~
g g 4 = e se .ie- _ ud W ' w s sl"ee M ' _ M M - 4 r
t l
)
u. 1 I
\
installation at TMI l consistent with the treatment tors ac- of other similar re i I l
- 2. l Separate Discussion on Standard for Determining Need
! 67J. ' focus for the first time in this decision the difncultyL the Board ' ! had with the meaning of the ultimate issues to be er decided the und ' Commusion's August 9.1979 hearing order- ! (b W hether the long term actions
- recommended e Director of by th '
Nuclear Reactor Regulation (set forth in Section 11 of this Order) \ i are necessary and suf6eient to proside reasonable assurance that ; 4 the facility can be operated for the long term without :ndang ering s { i the health and safety of the public, and should Nensee as soon as practicabic. of the be required , 1 1 { 10 NRC, at 148 A similar definition of necessaryains and sufficient' obt v with respect to the short term recommendations .
; .y graphs we h.ise esplained the factual basis for our conclusion on water 41 instruments levelIn the fore ditionally here We at docuss and explain the background of the dispute ad - , ,
) j gnen to the 'necessary" standard in other plant modification explains how ,
$ j.<
issues a issues we arrned at the criteria for "necessary' in deciding ese th 5 675 , i with the absolute conceptThe term necessary' in normal English would Sjh, be synonymous
'necenary" measure under the Commission's order cof "indispensable' and "essentia A gnen a gf 4 1 cd6*6a> a une q.a non to reasonable assurances of public safet)ould fairly be regard-
{ the long term the Staff firstIn support of its position that water lesel indication is needed ' Thermal Hsdrauhes Section whose testimony ey we hase discussed estensis -@f lpresent abose. Lnder persistent questtoning by the parties . Mr. and the Board }' Phdhrs described enhancement water lesel indication in the of the safety margin' (Tr. a long e term as "a desir bl 3 l '*- j 10.360); useful to enhance additional margin of safetysafety' (Tr.10.861; Tr.10.864; Tr.10,8891. -[nlecess which we feel is needed* (Tr. 10.68 $n
'necenary on the long range to proside this modification n ance the to e h the 677 operation of the reactor.". See . . .
, ,at5.
aho to Philhps 2 ff Tr 10 807 safe n t ducern from \fr. Phillips thatongthe o Staff regarded its lFrom these and oth term demand for w. iter lese! indication as indispensable to the Commissio ' n s requirement 1245 p t i m
t ' of "reasonable assurance that the facility can be operated in the long term without endangering the health and safety of the public.
- 10 N R C, at .
148. We could infer from Mr. Phillips' testimony that the public health and safety would not in the long term be endangered without water lesel indication already safe in that the Staff seeks only an incremental improsement in an condition i Mr. Phillips' testimony also raised the concern to .. the Board that the Staff, not understanding the limitations of our junsdic. tion, was using inappropriately the TMt 1 restart hearing to force a ' rouune backfitting measure upon Licensee. Tr. 10.886 63. Staff counsel correctly perceised that the Board wanted assurance that the Staff was not holding the restart of the plant as hostage to enforce its will upon Licensee in a dispute beyond the scope of the hearing Tr.10.682 (Cutchin). 64 Later. Dr. Denwood Ross, Director of NRR'S Division of Sptems g- integrahon. testified to justify the Staffs position on water level in-strumentation D Ross, ff Tr 15.915 at 2. He noted that, until the 4 Board's concerns carne to his attention, the Staff had not focused on the distinction
.- @ ^.. between "necessary" and "desirable" and that it was g incumbent -.
upon the staff to determine in some unainbiguous fashion
' whether sirabic.' Tr.(water 15,929 30.lesel iridication] was necessar) or whether it was de-D* Ross' conclusion is that the Staff belieses that water lesel instrumentatior. is "necessary to provide reasonable assurance of no undue risk to the health and safety of the public." D. Ross, ff. Tr.
e 15.915. at 2. This statement we take to be a restatement of the
~
sion's hearing order. its chief salue is to assure us that the Staff belieses
.. that6'9water lesel instrumemation (or its equisalent) must be installed .
Members of the Board, and apparently some of the parties, had difficult) with the concept of compararne necessity and with the idea that a design modification would be 'necessary' to reasonably the protect
. ~ .
health fea sible and safety of the public only if the modification proses to be 3 " E g. Tr.10.854 (Phillips. Smith) There was. therefore, a ten. dency by the Board and parties to test the limits of the Staff witnesses' s ie n pla nh.of "necessary" parnally in terms of whether this plant, or other should esen be permitted to operate in the long term without water lesel instrumentation Tr E r . Tr.10,552.10,885 56 (Phillips. Smith). 10.556 (D Ross, Jordan) (Phi!!qs, Jordan). Tr.15.956 (D Ross, Baster); Tr.16,030-)!
- 3 650 It is fair, we beliese, to summarize Mr. Phdlips' siew of the o
benefits of water lesel indication as a reasonabic enhancement to safety, the need for which would have to be reconsidered if it turns out t water 651. lesel indication is not feasible Tr.10.555 s6 (Phdlips) The testimony of Dr. Ross, as well as his purpose, was tv express
, in much stronger terms the view that water levelinstrumentation should bc required of Licensee and the industry in the long term:
1246 7,# UN, * * *"
].
. - .h*- w4 "
- Y* . M;#
. n. . ..s y .!;. - 1 %
3 ;- . y ' &'& . N -
___-- - - ^ - ~ ~ ~- -_ _ ______-__-_-
.~_ ~ - ., ,,e . -- 8.n e m W 4 aw 5 % ,-.q r. w . n=o-'f,.
- _. -
\ l l 1 l l ! I think it is necessary that additionalinstrumentation other than ) the ones present!) esisting in operating reactors is needed Reactor . j sessel level seems to me to be an appropriate instrument to add to the armaments that the operating staff hat. if that turns out in a ) car to be technically infeasible - and I , do not regard that as hkely at all - then I would - my position would be that the regulators and the regulated are going to have to do T ,, some more searching and find some other way to provide the diversity 4 and the conGrmation and the diagnostic capability for the operating staff, if not lesel something else. ' l' ! Tr.15.995 (D Ross. Dornsife). ! es2. The clusne nature of the "necessary" requirement in the Commis-i sion's hearing order is esidenced by the frustration apparent in the cross. . . . - j esamination of Dr. Ross by the Commenwealth's nuclear engineer Mr. D. . I ! Dornufe Although Dr. Ross was willing to describe water level instrumen-
- tation as "necessary", he talked at redundant necessity.
j Q .. the staffs concernWouldaboutit be correct to characterire the Commisuons' - l sessel water lesel in that, although not , I abolutely necessary to ensure pubhc health and safety, it increates - i the margin useful ? or decreases the risk from accidents and therefore it M- ! i ! 'W. (Pause > $eet .i { A ,.. 1' I have not used those words, ' absolutely necessary " Q @** l k n.ow That is why I used them
,~
Tr 15.993 94 (Ross. Dornsifet . H3 We do not make hght of the Staffs effort to apply reasonable CN
'4 qu.i9hfication to the uwally abolute meaning of the word "necessary" in the Commisuon's order; our inquiries attempted the .
i ersa thenc But ne th Commisuon's order, nor the technical witnesses presided useful general staedards tFe Board should employ in determining what " is
. necessary to proude reasonable assurance that the facihty can be operated e for th
- long term without erdangering the hestth and safety u of c" For the p bh I ga
- dance we have koked to the Commisuon's backfitting regulation which at 10 CFR 5010ta t proudes in pertinent part: ,
The Commission may, in accordance with the procedures speciGed in this chapter, require the backGtting of a facihty if it Gnds that such action will proside substantial, additio.1al protection w hich is i i j i 1247 im _w*"ul""** *=W
*F ge* e
_ _ - _ _ _ _ _ \--
a 4 s
/
t required for the public health Iid safety or the common defense and security. 654 Here again the Commission uses the absolute term "required"
.' which we equate with "necessary" as used in the hearing order. We do not know if the use of the word "necessary" in the hearing order is a studied distinction from the word "required" in 10 CFR 50.109. Probably no "kg
' difference was intended We beliese the Commission intended the same consideration in its hearing order as is intended in the backfitting reg-ulation, and we hase construed "necessary" accordingly.
' 685. In practice the Commission requires substantial improsements in *
,; the safety of nuclear facilities esen where, under precesting technology, 8' the facility design had been considered adequate to protect the public l' 0
health and safet). That is, the Commission need not find first that a nuclear facility is unsafe before it requires substantial irngrovemeritt in safety whe'e such improsements are practical. 6% Approaching the consideration from the other direction, Secdon i l 18: of the Atomic Energy Act empowers and requiris the Co nmission to
- proude " adequate protection to the health and saf
- t> of the pubhc "
l In Citi:en for Safe Pomer v Nuclear Regulatory Cowminin.1:4 F.2d
~
3 1291, 1297 (D C. Cir. 1975), the Court recognizeJ. that "absolute or l' pe rfect assurances are not required [b> the Act), and neither present
- i _
technology nor public pohey admit of such a standard. c' 667 As to esisting licenses, the mandate of Section 132 cf the Act is k
' ^ embodied in the backfitting regulation, Section f1109. Despite its seem- t(
ing!> absolute language, Section 50.109 does not preside for "substantial, o' additonal protection" without which "perfect" asturances of safety will A want Similarly "necessary" actions referred to in the Commission's hear. ing order are not actions indispensable to perfect and abso!wte assurances that the pubhc health and safet) will not be endangered. 655 The Board has taken additional direction from the Commission's December 26, 1980 Revised Statemert of Policy on further Commission l Guid,ince for Power Reactor Operating Licenses (46 Fed. Reg ?!40. January 23, 1951), which approves NL: REG 0737 as a baus for respond- _ inF to the TMl 2 accident with respect to NTOLs Obsening the need ' for a balance between safety significance and practicality, the Commission " stated
# 3
*- \s discussed abose, many actions were taken to improse safety c immediately or scen after the accident These actions were generally a considered to be interim improsements in scheduling the remaining C
.,L a
[ N 1 1248 i k
.fh
a
,,,,.a - --
l I e i i ! improvements, the availability of both NRC and industry resources ' was considered, as well as the safety significance of the actions. Thus, the Action Plan approsed by the Commission presents a sequence of actions that will result in a gradually increasing improvement ir, safety as inoividual actions are completed and the initial immediate actions are replaced or supplemented by longer term improvements. '
/d. Policy Statement PS 54.
689. Therefore, we have adopted a standard that "necessary modifica- > tiens as stated in the Commission's hearing order ne modifications which would produce a substantial and additional protection to the public health and safety and which, based upon the record, are reasonable in view of the technology. resources and risk involved. In other words, we have done .' exactly what Staff witnesses base done, i.e.. measured necessity partially in terms of feasibility."
- 3. Sep***te Statement on Water Level Indication by Mr. Smith ,
690. Although I join in the decision on water level ind: cation, par. . ticularly the result, there was a weakness in the evidence which ! believe warrants discussion. The decision on this issue depends partly upon the espert and salue judgments of the technical members of the Board and not 4 solely upon the opinions of the expert witnesses. I do not suggest that the technical members base reached beyond the evidentiary record to arrive at , our decision; they base not.' But, if this had been a private litigation *
,r -
between the Licensee and the Staff as adversaries without a strong public 1 s 3 i
, in Iwnew I .
m *"
...a mm ta o,xration lo test the Staffs posioon.1
.nwhc to the si,indards me have develogd. a r . in has also directed its htigation ,'
+
cng.necrmg pdgment reasonableness, practkahty, and .N
*['
Cr' l iren ihn pem. in light of Sra.voit An Polh,noa Leapr
- Castle. $12 F 2d $72 tlst 1
the %ved Ivr cigrt aduce, as et Stocoasi v. Costle. a i ' rig here and 1he consideration preva i
=here v ihc decneon m kers apply their oen esgruse to the evidentiary record The latter -', .
.irre a n authonicd by the NRC's unNue preregao+e under Secoon 19f a of the Atomic Ewrp Wi he gemtong Adminatranse Procedure Act* beanngs by f unum a three. member board *ith ';
ar;ropriate to the nsues to be decided . l 1249 1 i ) i 1 a _ ge w =go.. a ,
%- i
, I
^
.1 -
.d$bM 6 1 interest in the result, and without built in adjudicators' expertise, the Staff ;7;]94 7, might have lost on the issue of the long term need for water level in- ble versu-dication in this proceeding. . g g It requires no scientific analysis to understand that operators of 691. " #
PWRs waght to know whether their nuclear fuel is covered with coolant ;n, and, that direct instrumentation for this indication should be provided if it f is safe, unambiguous, and practical. Howest.r, the Commission did not vest * 'Y' ! us witn general backfitting jurisdiction. The Commission agreed with the I Board's concept of its jurisdiction as having a "reasonable nesus between nesus to in its unpublished order of March 14, ; , the issue and the TMI 2 accident 696 1980. This is the jurisdictional standerd we have applied throughout the nesus te proceeding. PID ' 24. If water level indication does not have a reasonable would in nesus to the TMI 2 accident we a e without authority to require it at f ,'t 6 Tb re is no doubt that the factual matter of water lesel in the anywas. reactor ses>el has a indicationclose nexus to :he accident. core coolingThe is anneed for unam-of inadequate ei@er t biguous, easy to interpret g g7 p, undisputed learned lesson from the accident as stated in 2.1.3.6 of 697 NU REG 0578. The question before us now, however, is whether, in vie" jg9;f of the post accident plant design and operating procedures modifications, we base the authority to impose remedies which may not have a reasona- 1980 a$ discusse ble netus to the accident. 693. The key word is "remedy" If the remedy cannot enhance the threw ~s safety of TMI l with respect to a circumstance having a close nexus to the TMI 2 accident, we are without authority to impose it. The Staff may 3d base that right outside this proceeding, but we do not. d 694. The debate begins to unfold where the Licensee takes the position posed f l that water lesel indication would not be helpful in an inadequate core 698 i cooling condition; that no additional or earlier action beyond procedures the Sc, presently in place, including the procedures which justify resiart in the term. can be identified. Keaten, ei ol., ff. Tr.10,619. Licensee's
>hort Proposed Finding i 63. The Board was neser persuaded by Licensee's
*"I
,c ;
g mation position that, because it could mislead the operators, water lesel indication is worse thin useless; if anything this testimony was disconcerting. How-point.' the sig ever, the question of whether water level indication is helpful or not is ser} 699. much relesant to the question of the nesus to the TMI 2 accident. and "r I reacto-condit the Li e i , 1250 i o%--
695. The NRC Staff started out by recognizing that it has not iden-tified differences in operator actions if water lesel information were availa-4 ble sersus those actions now required by esisting guidelines for inadequate core cooling. Phillips 2, ff. Tr.10.807, at 2. Staffs witness Phillips called , for water lesel indication to distinguish between anomalous transients due to cooling and shrinkage of coolant as compared to loss of coolant insen-1 tory. /d. This purpose has a connection to the TMI 2 accident. The Board has accepted any small break loss of coolant accident as having sufficient nesus to the accident, in the contest of this issue, and I do not question this standard. 696. The Staff goes on to identify other possibilities with a stronger nesus to the accident. The Staff stated that water level instrumentation would indicate the effectiveness of HPl in recovering the system. Id.. at 3. Licensee rebuts this testimony with its B&W espert, Mr. Jones, w ho testified that the operators would not do anything with this information any w ay. Tr. 10.687 88. The Staff has not addressed Mr. Jones' testimony, ' either by rebuttal testimony or by reference to it in proposed findings. Staff Proposed Findings T 111. Mr. Jones' testimony remains unrefuted. ' 697 Staff witnesses Phillips (2nd set, ff.10.807, at 4) and Ross (ff. Tr. 15.915, at 3) refer to a small break LOCA event at St. Lucie in June 1930 as an example of the usefulness of water lesel indication. Mr. Jones 4 discuoed the esent at length in rebuttal (Tr. 10.688 91). and at least threw some doubt into the utility of water level indication during that i transient. The Staff, however, does not discuss this testimony in its pro-s. posed findings on the subject. It does not esen discuss the St. Lucie esent escept to identify it and to assert its summary belief that water level indication would have contributed to safe operation. Compare Staff pro-posed finding ' 107 with Licensee proposed findings 't 68 69. , 69t
,1 the Sequoyah incident in February 1981 - another small break }LOCAA similar situat esent Licensee points out that Dr. Ross did not identify any additional actions the operators might have taken on the basis of water lesel infor-mation (Licensee proposed finding 5 70). But the Staff is silent on this point. Staff proposed finding ! 107. It was left to the Board to interpret y the significance of the St. Lucie and Sequoyah events.
g% 699. The Staff also states that water lesel information might be useful, and "possibly essential" in emergency procedures relating to the use of the
~
,7 reactor sessel head sent, but concedes that it has not evaluated the , . .
conditions for which the head vent should be opened Staff depends upon the Licensee for this information. Staff proposed finding ! 112. Staff again j
*}
s 1251 '- -
. o , ' 4a .
~- s< w~ . ,, . ; n ;,. .
+. e 3. n..
4 1
,; g y ,
ignores the testimony of Mr. Jones who testified that B&W guidelines ing f under development for vent use do not rely upon water level indication. Tr. TM M,4 10,692. Licensee proposed finding 171. y 700. I make these observations to point out, as I stated at the outset, mas R$ f, - that the Board depended on its internal expertise to arrive at its conclusion Y' on water lesel indication, not to praise the Licensee's presentation which, nee ( a no-in fact, we have criticized above. Having decided initially that there was
~
not I no use for water lesel indication in a B&W reactor for an inadequate core g l cooling situation, it seemed that nothing could stimulate the witnesses' for 2 imagination as to the possible uses. 7 701. Licensee makes the observation that the Staff in its proposed ig. findings has accurately cited the record, but that the findings are incom- this ! plete in that they do not discuss much of Licensee's testimony and the cha- l Lice. ace's cross examination c.f Staff witnesses. Licensee's reply findings l (1st Set) i 43. This comports with my observations. Moreover, even though the Licensee submitted extensive proposed findings on the ICC issue (t' 24 91) raising many contested sub-issues, including the deficiencies in Staff > case. the Staff filed no reply findings whatever on this issue. Even
,. though this is a contested matter between the Licensee and the Staff, the y were not joined in the proposed and reply findings. The Staff is issues largely in default 7 702. When I state that the evidence of nexusI intend is weak, between no the need for water level indication and the TMI 2 accident critienm of the efforts of Staffs witnesses. To the contrary, Mr. Phillips ,
supplemented his original testimony in an effort to support the Staffs ' position and Dr. Ross, after examining the hearing transcripts, came to the 'CW hearing to emphasize the Staffs position that water level instruments are fou needed. The fact that neither would exaggerate the particular uses to be tha made of wa'er leselenhancedindicationtheir in acredibility TMl 2 accident situation is commen- W dable and. radeed, it as to the generic de- 5 h sirability and feasibility of water lesel indication in PWRs. The problem is
- that the facts do not seem to strongly support the Staff as to the particular *I issue in this proceeding. This was rescaled in the cross-examination of the
- l Staff, witnesses. In cross examining Mr. Phillips on this issue,isthe Com-accep-M monwealth's nuclear engineer probed the criteria of what " ' " I' acceptable for the long term." Tr. 10.875 I'
- table for restart. but not .
(Phillips, Dornsife). The tenor of Mr. Phillips' testimony is that water lesel indication, or its equisalent, is needed in the long term to detect anomalous co,
- sic "On ike other ban d. Ucensec faded to addren the Staffs tesumory and proposed find,rgs f w%rmng tbc unhi> of lesel mdicanon that the core is coscred durmg blMage condioon Phdh&2. (f.10.70s ai 4. Suff Proposed Fmdmg ' 111 00*
a becmec Ser TNil.2 tyrc th. Profwd Fmdags " M 72 1252
situations. Tr. 10,877; 10.891 92. Mr. Phillips believed that "
- current l instrumentation that is available at TMI-I is sufficient to respond to TMI 2 type accidents and is adequate for those types of accidents, but may not be adequate for other unidentifiable accidents." Tr. 10,891 92. Dr.
Roo belined that water lesel indication is necessary in that " there needed to be more diverse instruments that would let operators cope with i anomalous transients" and that level indication is required because " not all transients could be postulated in advance " Tr.15.994; See aho Tr.16.01718. for w hich the Staff sees a need for water level indicatio.1.Thus it is the anomalous 703. I do not believe that I hase over emphasized the implications of the Staff testimon) nor taken it from context. In its proposed finding on tha issue i 114) which we adopted serbatim in our decision the Staff itself characterizes the need as: The Staff belieses that while existing equipment may be f sufficient to respond to TMl type accidents, it may not be sufficient to respond to other unidentifiable accidents. Tr.10,892 (Phi!!ipn One . of the lesx ns learned from th; TVil accident was that there need to ~~, be more diverse instr sments that would let operators cope with an-omalous transients for which procedures and training do not exist. Tr. ! 15.994 (Ross) Thus, instrumentation other than that currently being .) j u>cd. is needed. Reactor sessel level is an additional and diverse i method vf determiaing in:dequate c :e cooling. Tr.15.995 (Ross). ! 704 ... i
% hy then do I join the technical members in the decision to y %.
)
require implementation of position 2 of Sec. ion 2.1.3.b? First the Board , I found. based upon the convincing testimony of Mr. Phillips and Dr. Ross , l that. in a small break LOCA transient, water lesel indication would be , useful for as long as 30 minutes to 3 hours between the time that the l saturation meter at first, and the core exit temperature indicators later, 4,_ . wouldit proside While may be that accurate core information to the operators. T 659 supra
- this interim information is needed only for an-. ,
g omalous, unidentified episodes, these episodes would be within a TMI 2 9. 4 type transient. The determination of the Board that operators should bc
- informed under these circumstances falls within the appropriate expertise
- of the technical members and is within our jurisdiction.
705. Secono, the Board found from Mr. Phillips' testimony that water-lesci indication is useful to distinguish between anomalous accidents which
- cools and shrink the primary coolant as compared to loss-of coolant tran-sient
- It is true that the state of the evidence, as adduced by Licensee, is i that operator action between the two type transients would remain the same. Nesertheless, the expert judgment of the Board's technical members 2
1253 p g.* [ i 4 l
e U e
*s 4
+ . ,%.
- . . . , . ,. . . . . . m,. m ., ..
ws. 709. 3 that operators should have the instrumentation needed to diagnose the TMl 2 accident, even if for unidentified purposes, is sufficient, and this that th. 4 ' purpose falls within the scope of our jurisdiction. des elops Ow ners Abnormal Transient Operating Guidelines of NUI C. loss of-t i ' guidelir. 706. Board Question 11 states: breaks,
# The board is not satisfied with the staff findings in the SER dition, with respect to Recommendation 2.1.9.c (transients and accidents) of than de NUREG 0578. The Staff concludes that satisfactory progress has they ar been made and the item is complete. SER, pp. B 10, C8 49. Accord- directio-ing to Table B 2, the analyses and procedures were scheduled for 2 3.
completion by early 1980. We obserse that in May of this year 710. [1980), it was reported that the "Staff is perfore ing a generic review diagnos of transients and other accidents in accordance t th Recommendation the pre 2.1.9 of NUREG 0578' (NUREG 0667, p. 5-26). conditit great a We expect the licensee and the staff to present evidence that pregrar the requirements on p. A 45 of NUREG 0578 will be met and to hase as explain the schedule for meeting those requirements. The board, as 7 g i, well as the staff, must have sufficient information to decide whether procedt, satisfactory progress is being made. a naly se-design t 707. Recommendation 2.1.9.c of NUREG 0578 asks that the Licensee actise f
"[p]roside the analysis, emergency procedures, and training to substantially also be improse operator performance during trarsients and accidents, including functior esents that are caused or worsened by inappropriate operator actions.-
^~d-Page A 45 of NUREG 0578, which discusses Recommendation 2.1.9.c, 7 I 2' l sets out the requirements of the analyses of transients and accidents. The with th I anahses are to include the design basis events specified in Section 15 of NURE' the FS AR, single actise failures and consequential failures for each system IO'; ^I insohed in a particular esent, operator failures to perform required control tube tu manipulations and operator actions leading to the loss of function of a
' "0 " '
j safety system. Further, the analyses are to incorporate esent trees, com-7 I 3-I puter calculations, and reactor simulators. was to l 708. Both the Licensee and the Staff presented direct evidence in
] response to this question. The Licensee's testimony was sponsored by T.
accider l ,, Gary Broughton (ff. Tr.10,941), and the Staff's by Walton L. Jensen. Jr. (ff. Tr.11.005). No other direct evidence was presented, but the Common-peratg to mity g accider wealth of Penns>hania participated in cross examination of both witnesses. OPIC There was also extensise examination bs the Board. es ents, were st 1254 4 [
* =}}