ML19351F130
| ML19351F130 | |
| Person / Time | |
|---|---|
| Site: | Crane  |
| Issue date: | 12/09/1980 |
| From: | Capra R, Ross D Office of Nuclear Reactor Regulation |
| To: | |
| Shared Package | |
| ML19351F129 | List: |
| References | |
| NUDOCS 8012290562 | |
| Download: ML19351F130 (63) | |
Text
.
. ~.
f s
O-s OUTLINE 4
S l
l This testimony of Dr. Denwood F. Ross, Jr. and Robert A. Capra supplemerts j
the NRC staff testimony of Walton L. Jensen, Jr. relative to the Board i
i Question Regarding UCS Contention 8.
a j
The purpose of this testimony is to specifically. respond to one aspect of the Board Question Regarding UCS Contention 8, namely addressing each of 1
the recommendations in NUREG-0565 and NUREG-0623.
f 9
t t
a
'9 s
I a
f 8012290 5p
w INDEX Item Page Board Question Regarding UCS Contention 8..........................
1 1
i Introduction..............................
NUREG-0565 Background....................................................
3 Interrelation of Associated Staff Documents...................
5 Recommendations:
11 2.1.2.a...............
2.1.2.b.......
14 2.1.2.c..................................................
16 2.1.2.d..................................................
17 2.1.2.e..................................................
16 2.2.2.a..................................................
18 4
2.2.2.b..................................................
18
-4 2.2.2.c..................................................
22 2.3.2.a..................................................
.v4 1
i 2.3.2.b..................................................
26 2.3.2.c..................................................
28
}
2.4.2.a..................................................
30-32' 2.5.2.a..................................................
2.6.2.a..................................................
33 36 2.6.2.b..................................................
38 2.6.2.c..................................................
i l
40 2.6.2.d..................................................
42 2.6.2.e..................................................
44 2.6.2.f..................................................
46 2.6.2.g..................................................
47 2.6.2.h..................................................
49 2.6.2.i..................................................
NUREG-0623-51 Background....................................................
53 Recommendations...............................................
~
54 I
Current Status................................................
l
,, -, ~
.,e
,g,
-,,,m, a
.--n.-
4,,,,,-
.--gnw w
..p,n,...m.,
yy y,.
e---g
O UNITED STATES OF AMERICA.
NUCLEAR REGULAT03Y COMMISSION BEFORE THE ATOMIC SAFETY AND LICENSING BOARD In the Matter of
)
METROPOLITAN EDISON COMPANY, ET AL,
)
)
Docket No. 50-289 (Three Mile Island Nuclear
)
Station, Unit No. 1)
)
NRC STAFF TESTIMONY OF DR. DENWOOD F. ROSS, JR. AND ROBERT A. CAPRA IN RESPONSE TO THE BOARD QUESTION ON UCS CONTENTION 8 Board Question Regarding UCS Contention 8:
"The Board directs the staff and the licensee to present experts and the fundamental documents involved in the small break LOCA analysis, and to have very complete testimony on this subject.
The recommen-dations of NUREG-0565 and NUREG-0623 should be addressed.
It appears from the small break LOCA analysis that there is a large amount of reliance upon operator action and on non-safety grade equipment.
The Board wants that issue explored by testimony including why such reliance is proper.
Tr. 2374-85."
Response
INTRODUCTION l
Testimony on the subject of small break LOCA analyses and the supporting fundamental documents have been filed by the NRC staff in response to UCS Contention 8.
The recommendations of NUREG-0565 and NUREG-0623 as well as the reliance upon operator action and nonsafety grade equipment during a small break were addressed in NRC staff testimony on the Board Question Regarding UCS Contention 8 previously filed by Mr. Walton L. Jensen, Jr.
I L
o
. The purpose of this testimony is to supplement the latter testimony, by specifically addressing the NRC's position vp the licensee's position regarding the recommendations contained in NUREG-0565 and NUREG-0623.
In addition, the implementation of these recommendations as well as the inter-relation of NUREG-0565 and NUREG-0623 with other staff documents will be discussed.
In particular, the following documents are discussed:
(1) NUREG-0660, Volumes 1 and 2, "NRC Action Plan Developed as a Result of the TMI-2 Accident," May 1980; Revision 1, August 1980; (2) NUREG-0694, "TMI-Related Requirements for New Operating Licenses,"
June 1980; and (3) NUREG-0737, " Clarification of TMI Action Plan Requirements,"
November 1980.
This testimony is divided into two major parts:
Part 1 addresses the recommendations of NUREG-0565 and Part 2 addresses the recommendations of NUREG-0623.
l l
o
. PART 1 NUREG-0565, " Generic Evaluation of Small Break Loss-of-Coolant Accident Behavior in Babcock & Wilcox-Designed 177-FA Operating Plants," January 1980 BACKGROUND Shortly after the accident at TMI-2, several meetings were held between the NRC staff, the B&W licensees, and B&W.
It was during these meetings, held in April 1979, that most of the information regarding small break LOCAs was requested by the staff.
The majority of the analyses were performed by B&W in the April /May 1979 time frame.
The areas to be addressed by B&W were:
(1) Extend the small break spectrum to include analyses of very s-all breaks, giving special attention to failure of pressurizer valves to close; (2) Systematically evaluate plant response to small break LOCAs to provide a more realistic plant response compared with the response obtained using the conservative assumptions used in licensing calculations; (3) Analyze degraded conditions where auxiliary feedwater (AFW) was not available; (4) Validate computer codes, used for small break LOCA analyses, by benchmarking them against the response of TMI-2 during the accident;
. (5) Address the concerns expressed by ACRS consultant Mr. C. Michelson of TVA regarding small break LOCAs*;
(6) Prepare design changes aimed at reducing the probability of a small break LOCA produced by the failure of a PORV to close; and (7) Develop generic guidelines from which detailed emergency procedures I
for coping with small break LOCAs could be developed.
In early May 1979, the Bulletins & Orders Task Force (B&OTF) was formed within the Office of Nuclear Reactor Regulation under the direction of Dr. D. F. Ross.
The B&OTF was responsible for reviewing and directing the TMI-2 related staff activities on loss of feedwater transients and small break LOCAs for all opera-ting reactors to assure their continued safe operation.
The B&OTF was divided into three groups:
Projects, Systems, and Analysis.
R. Capra was the Project Manager assigned to the B&W-designed operating plants.
W. Jensen was a member of the Analysis Group.
It was, therefore, the responsibility of W. Jensen and f
R. Capra, under the supervision of D. Ross, to evaluate the information requested of B&W related to small break LOCAs and to publish the results in NUREG-0565.
l During the months of May through July 1979, the bulk of the small break LOCA l
analyses performed by B&W were reviewed'by the B&OTF.
In August 1979,
References:
Letter from D. R. Patterson (TVA) to J. McFarland (B&W),
" Transmitting Report by C. Michelson entitled, ' Decay Heat Removal During a Very Small LOCA for a B&W 205-Fuel Assembly PWR,'" April 1978 and Draft report by C. Michelson (TVA) entitled, " Decay Heat Removal Problem Associated with Recovery From a Very Small Break LOCA for a CE System 80 PWR," dated May 15, 1977.
i
. attention was shif ted to evaluating the same type of small break LOCA analyses i
for other vendor-designed operating reactors.
This work was completed in December 1979.
Therefore, it was not until all vendor small break information was reviewed that the vendor-specific generic reports were completed.
In January 1980, NUREG-0565 was published along with similar reports on Westinghouse, Combustion Engineering, and General Electric-designed plants.
The recommenda-tions contained in NUREG-0565 and the other vendor-specific reports represented the views of the Task Force.
Additional work, especially that of coordinating and exploring interactive effects, had to be done (as noted below) before the B&OTF work product was adopted as NRC policy.
INTERRELATION OF ASSOCIATED STAFF DOCUMENTS Concurrently with work of the B&OTF, many investigations into the accident at TMI-2 were conducted.
Each investigation and special task force developed its own independent set of recommendations.
As one would expect, many of the recommendations overlapped and intertwined with the recommendations of other l
groups.
Therefore, in order to develop a complete and orderly set of TMI-2 l
related recommendations, the TMI Action Plan (NUREG-0660) was developed.
(See NRC staff testimony relative to Board Question 2 for a more detailed explana-l tion of the development of the Action Plan.) The recommendations spawned by l
the investigative bodies and special task force reports, including NUREG-0565, were thus incorporated into NUREG-0660.
Where individual recommendations over-lapped with others, the intent of the individual recommendations were incorporated into items of the Action Plan that had a larger and broader scope.
1 l
. All 22 of the recommendations contained in NUREG-0565 were incorporated into s
the Action Plan as subitems of II.K.2 or II.K.3.
Table 1, Page 7 of this testimony, cross-references the recommendations listed in NUREG-0565 with those contained in NUREG-0660.
As discussed in NRC staff testimony in response to Board Question 2, the Action Plan contain's the total set of TMI-related actions, but not all have been approved for implementation. The follcwing documents identify TMI-related requirements from the Action Plan which will be implemented on TMI-1:
(1) Commission Order of August 9,1979 (the short-and long-term portion of the Order encompass many of the items contained in NUREG-0660);
(2) NUREG-0694, fuel load and full power requirements (these items will be required on TMI-1 prior to restart); and (3) NUREG-0737, requirements for operating reactors (these items are listed in Enclosure 1 to the document).
l Referring again to Table 1, it can be seen that of the 22 recommendations originally contained in NUREG-0565, 20 apply to TM1-1. Two of the recommenda-tions do not apply.to the licensee (2.4.2.a is a staff action item and 2.5.2.a applies only to Davis-Besse 1).
Of the 20 remaining recommendations:
Table 1 IMPLEMENTATION OF RECOMMENDATIONS CONTAINED IN NUREG-0565 ON TMI-1 NUREG-0565 DESCRIPTION ITEM IN NUREG-0660 IMPLEMENTATION:
METHOD & SCHEDULE FOR THI-1 2.1.2.a Automatic Block Valve Closure System II.K.3.1 NUREG-0737 See II.K.3.1 & II.K.3.2 2.1.2.b Evaluation of PORV Opening II.K.3.7 & II.K.2.14 NUREG-0694 See II.K.2.14 2.1.2.c PORV Reporting Requirements II.K.3.3 NUREG-0694 See II.K.3.3 2.1.2.d Evaluation of Safety Valve Reliability II.K.3.2 NUREG-0737 See II.K.3.2 2.1.2.e Safety Valve Reporting Requirements II.K.3.3 NUREG-0694 See II.K.3.3 2.2.2.a Analysis Methods for SBLOCA-Appendix K II.K.3.30 NUREG-0737 See II.K.3.30 4
2.2.2.b Plant Specific Analysis-10 CFR 50.46 II.K.3.31 NUREG-0737 See II.K.3.31 2.2.2.c Effects of CFT Injection II.K.3.35 NUREG-0737 See I.C.1 (Incorporated) 2.3.2.a Automatic Trip of RCPs During SBLOCA II.K.3.5 NUREG-0737 See II.K.3.5 2.3.2.b Reliability of Nonsafety-Grade Equipment II.K.3.4 NUREG-0660 See II.C.1/II.C.2/II.C.3 7
2.3.2.c
. Simulator Improvements II.K.3.54 NUREG-0660 See I.A.4.1 2.4.2.a Staff Audit Calculations II.K.3.36 NA Licensee 2.5.2.a Diverse DHR System for Davis-Besse 1 II.K.3.8 NA Licensee 1
2.6.2.a Verify Two-Phase Natural Circulation II.K.3.32 NUREG-0737 See II.K.3.30 (Incorporated) 2.6.2.b Instrumentation for Natural Circulation II.K.3.6 NUREG-0737/ Order
- See I.C.'1/II.F.2 (Incorporated) 2.6.2.c Analysis-Isolated SBLOCA II.K.3.37 NUREG-0737/ Order
- See I.C.1 (Incorporated) 2.6.2.d Analysis-SBLOCA in Spray Line II.K.3.38 NUREG-0737/ Order
- See I.C.1 (Incorporated) 2.6.2.f Evaluation of RCP Seal Damage II.K.3.40 & II.K.2.16 NUREG-0694 See II.K.2.16 2.6.2.g Predicti'as of LOFT Test L3-6 II.K.3.41 NUREG-0737 See II.K.2.5 (Incorporated) 2.6.T Information on Noncondensible Gases II.K.3.42 NUREG-0737/ Order
- See I.C.1 (Incorporated) 2.6.2.1 Effects of Slug Flow on OTSG Tubes II.K.3.43&II.K.2.15 NUREG-0694 See II.K.2.15
- Tha Commission Order of August 9, 1980 directs the licensee to perform all recommendations contained in NUREG-0578.
R: commendation 2.1.9 of NUREG-0578 was expanded in scope when incorporated into the Action Plan and appears as Itea I.C.1 in NUREG-0660.
Therefore, the various recommendations of NUREG-0565 which are now included in the scope of Item I.C.1 are also loosely tied to the order.
8-(1) Five of the original NUREG-0565 recommendations have been extracted directly from the Action Plan and will be implemented on TMI-1 in accord-ance with the schedule shown in NUREG-0737:
2.1.2.a Auto. Block Valve Closure System 2.1.2.d Evaluation of Safety Valve Reliability 2.2.2.a Analysis Methods for SBLOCA-Appendix K 2.2.2.b Plant Specific Analysis-10 CFR 50.46 2.3.2.a Auto. Trip of RCPs during SBLOCA (2) Eight of the original NUREG-0565 recommendations have been incorporated into other items of the Action Plan.
As such, they do not have an indi-i vidual implementation.
They will be implemented on TMI-1 in accordance with the scope and schedule of the recommendation into which they have been incorporated.
These items also appear in NUREG-0737:
2.2.2.c Effects of CFT Injection 2.6.2.a Verify Two-Phase Natural Circulation Models 2.6.2.b Instrumentation for Natural Circulation 2.6.2.c Analysis-Isolated SBLOCA 2.6.2.d Analysis-SBLOCA in Pressurizer Spray Line 2.6.2.e Effects of CFT & HPI Slugging 2.6.2.g Predictions of LOFT Test L3-6 2.6.2.h Information on Noncondensible Gases
.r
..,, - -, -. ~..
,.c y--
,---,y-
..,. ~.,.,
, + -,
.g-(3) Five of the original NUREG-0565 recommendations have been extracted as part of the fuel load and full power requirements stated in NUREG-0694 and as such, they will be required to be implemented on TMI-1 prior to restart:
2.1.2.b Evaluation of PORV Opening 2.1.2.c PORV Reporting Requirements 2.1.2.e Safety Valve Reporting Requirements 2.6.2.f Evaluation of RCP Seal Damage 2.6.2.i Effects of Slug Flow on OTSG Tubes (4) The two remaining original NUREG-0565 recommendations have been incorporated into other items of the Action Plan for which plant specific requirements have not yet been developed.
Therefore, until further guidance is issued by the staff, no licensee action is required:
2.3.2.b Reliability of Nonsafety-Grade Equipment 2.3.2,c Simulator Impretements.
RECOMMENDATIONS l
l The following section of this testimony discusses each of the 22 recommendations listed in NUREG-0565.
For each smmendation the following information is presented:
Recommendation:
States the recommendation as originally presented in NUREG-0565.
-y r._.
_v,.. _ _
4 m
4 10 Licensee's Position:
i States the licensee's position on the recommendation, as presented in Licensee's Testimony of Robert C. Jones, Jr and T. Gary Broughton in Response to Board Question on UCS Contention 8.
t NRC's Position:
Explains where the NUREG-0565 recommendation has been incorporated into the Action Plan (NUREG-0660).
4 Clarifies the scope of the recommendation.
?
Describes what additional information, if any, will be required of the ifcensee in order to comply with the recommendation.
i Implementation:
i Identifies the method / mechanism through which this recommendation will be implemented on TMI-1.
Presents the currently approved schedule for completing the work associated 4
with the recommendation.
4
. =.
i j
i I #
NUREG-0565, Recommendation 2.1.2.a Recommendation:
Provide a system which will assure that the block valve protects against a stuck-open PORV.
This system will cause the block valve to close when RCS pressure has decreased to some value below the pressure at which the PORV should have reseated.
This system should incorporate an override feature.
Each licensee should perform a confirmatory test of the automatic block valve closure system.
Licensee's Position:
" Design and installation of an automatic PORV block valve closure system is not being pursued at this time.
The need for such a system has no's oeen determined by appropriate analysis, which is called for by item II.K.3.7 of NUREG-0660.
Furthermore, it is not obvious that the addition of a closure system would be a modification which would provide greater safety, since the system may result in an increased probability of challenge to the pressurizer safety valves.
Until the evaluations in response to Item II.K.3.7 are com-pleted, the need to design and install an automatic block valve closure system has not been established."
NRC's Position:
This recommendation has been incorporated into the Action Plan as item II.K.3.1.
Implementation of this item is dependent upon the results of the analysis called for under item II.K.3.2 of the Action Plan.
Item II.K.3.2 requires that each licensee perform an analysis of the probability of a small break LOCA caused by a stuck-open PORV or safety valve.
The analysis
1 -
should consider modifications which have been made since the TMI-2 accident to decrease this probability. Modifications to reduce the likelihood of a stuck-open PORV will be considered sufficient improvements in reactor safety if they reduce the probability of a small break LOCA caused by a stuck-open PORV such that it is not a significant contribution to the probability of a small break LOCA due to all causes.
(According to WASH-1400, the median probability of a i
small break LOCA is 10 3 per reactor year.)
The resuits of the analysis should then be used to determine whether the modi-fications already implemented have reduced the probability of a small break LOCA sufficiently or whether the automatic PORV isolation system specified in item II.K.3.1 is necessary.
The analysis must consider the effect of an automatic PORV closure system on the potential of causing a subsequent stuck-open safety valve and the overall effect on safety (e.g., effect on other accidents).
Implementation:
l This recommendation will be implemented on TMI-1 if the results of the analyses performed under II.K.3.2 deems it appropriate.
Both item II.K.3.1 and II.K.3.2 are requirements of NUREG-0737 for all operating reactors and apply to TMI-1.
The scheduled specified in NUREG-0737 calls for:
I l
(1) Submit analysis required under item II.K.3.2 by January 1,1981; i-l l
l l
l -~. -
\\
] !
r j
(2) Submit proposed design modifications for staff approval, if required by results of analyses by July 1, 1981.
i l
(3) Install modifications and perform confirmatory tests of the system at the
}
next refueling outage following staff approval of the design, unless this outage is scheduled within 6 months of the approval date.
In this event, 2
modifications will be completed during the following refueling outage.
i l
1 l
)
1 i
f I
i I
i J
l l
l l
l l
l l
l
. NUREG-0565, Recommendation 2.1.2.b Recommendation:
Most overpressure transients should not result in the PORV opening.
Therefore, licensees should document that the PORV will open in less than five percent of all anticipated overpressure transients using the revised setpoints and antici-patory trips for the range of plant conditions which might occur during a fuel cycle.
Licensee's Position:
(Summarized)
Licensee provides a qualitative assessment of challenges to the PORV.
The discussion covers the following transients:
loss of feedwater, loss of external load, turbine trip, uncontrolled control rod withdrawal from startup conditions, inadvertent closure of main steam isolation valves, and inadvertent moderator baron dilution.
Licensee states that anticipated transients which have occurred on operating plants will not result in lifting of the PORV due to the revised setpoints of the PORV and high pressure trip and the addition of the'anticipa-tory reactor trip.
Other transients which can lead to PORV opening have not occurred on operating plants.
Therefore, it is apparent that this fraction is less than five percent.
NRC's Position:
This recommendation has been incorporated into the Action Plan as items II.K.3.7 and II.K.2.14.*
- Note:
Information/ analyses requirements regarding PORV lift frequency is described under both items of the Action Plan.
A consolidation of the information required is provided under item II.K.3.7 of NUREG-0737.
1 I 4 t
The staff also believes that the probability of opening the PORV has been reduced substantially by the addition of the anticipatory reactor trip for loss of feedwater and turbine trip combined with the reduced setpoint of the I
high pressure reactor trip and the increased setpoint of the PORV.
- However, i
the staff is unable to make a quantitative judgment of the expected lift
- [
frequency.
Therefore, we have requested the B&W licensees to perform additional l
analyses of anticipated transients which indicate the sensitivity of PORV chal-lenges to (1) the variation in core physics parameters which may occur in the plant cycle; (2) single failures in mitigating systems; and (3) transients which do not actuate the anticipatory reactor trip.
The analytical assumptions should include those specified in the plant FSAR.
The results of these more-detailed and extensive analyses should be used to determine the expected frequency of PORV openings for overpressure transients.
The frequency should be less than I
five percent of the total number of overpressure transients.
I Implementation:
i This recommendation (Item II.K.2.14) is a full power requirement for near term operating license applicants, as described in NUREG-0694.
As such, the analyses described above will be required to be completed by the licensee prior to restart.
It may be possible for the licensee to perform the analyses as part of the anal-yses required under Action Plan item II.K.3.2, which has been previously discussed in this testimony on page 11.
. NUREG-0565, Recommendations 2.1.2.c and 2.1.2.e Recommendation 2.1.2.c:
All failures of PORVs to reclose should be reported promptly to the NRC.
All challenges should be reported in annual reports.
Recommendation 2.1.2.e:
All failures of safety valves to reclose should be reported promptly to the NRC.
All challenges should be reported in annual reports.
Licensee's Position:
" Licensee will propose changes to the TMI-1 Technical Specifications that sill require reporting of failures or challenges to the PORV and safety valves."
NRC's Position:
These recommendations have been incorporated into the Action Plan as item II.K.3.3.
The licer see's commitment to mc;: *y its Technical Specifications to incorporate this requirement is acceptable t; the staff.
Implementation:
Recommendation II.K.3.3 is a full power requirement for near term operating.
license applicants, as described in NUREG-0694. As such, the reporting require-ments described above will be required to be in place prior to _ restart.
. NUREG-0565, Recommendation 2.1.2.d Recommendation:
Licensees should submit a report to the NRC which discusses the safety valve failure rate experienced in B&W operating plants.
Licensee's Position:
" Licensee is unaware of any instances of failures of Reactor Coolant System j
safety valves at any B&W plant.
See Licensee's testimony in response to the Board Question on UCS Contention 6."
NRC's Position:
This recommendation has been incorporated into the Action Plan as part of item II.K.3.2.
(Item II.K.3.2 was discussed previously in this testimony on page 11.
i As part of the licensee's response to Action Plan item II.K.3.2, the safety valve _ failure rates based upon past history of the operating plants designed i
by the specific nuclear steam supply system (NSSS) vendor should be included.
Presumably, the data presented in that section of the report, submitted under l
II.K.3.2, will be based upon a more detailed record search than the licensee j
has indicated above.
Implementation:
This recommendation will be implemented on TMI-1 as required by NUREG-0737.
The analyses required under item II.K.3.2 is to be submitted for NRC staff review by January 1,1981.
i i
. NUREG-0565, Recommendations 2.2.2a and 2.2.2.b Recommendation 2.2.2.a:
The analysis methods used for small break LOCA analysis by B&W should be revised, documented, and submitted for NRC approval.
Recommendation 2.2.2.b:
Plant-specific calculations using the NRC approved model for small breaks should be submitted by all licensees to show compliance with 10 CFR 50.46.
Licensee's Position:
"The small break LOCA analyses which were performed after the TMI-2 accident were done to provide an improved analytical basis for emergency procedures for small break LOCA's. These analyses were not done to demonstrate compliance with 10 CFR 50.46.
NUREG-0565 states that the post-TMI-2 analyses are beyond those normally considered in small break analyses and that the NRC staff has some concerns relative to the use of the currently approved small break model for these purposes.
However, NUREG-0565 (Section 2.2.1) also contains the following conclusion:
"The small break analysis methods used by B&W are l
satisfactory for the purpose of predicting trends in plant behavior following small break LOCAs and for training reactor operators." NUREG-0565 does nnt l
l state that the approved B&W small break evaluation is deficient for demon-l strating compliance for TMI-1 with respect to 10 CFR 50.46 and Appendix K.
While further code development may be performed and model modifications may be made, the changes are not expected to result in a substantial change to the Appendix K evaluations performed for THI-1."
. NRC's Position:
These two recommendations have been incorporated into the Action Plan as items II.K.3.30 and II.K.3.31, respectively.
As a result of the Bulletins & Orders Task Force review of the post-TMI-2 small break LOCA analyses performed by B&W, a number of concerns were developed regarding the adequacy of the current models to correctly predict certain phenomena.
These concerns are listed on pages 4-1 through 4-3 of NUREG-0565.
4 We were concerned mostly with a model that would correctly predict what the operator might see following a small-break LOCA, or loss of feedwater (or both).
In some instances the qualifier " correctly" should not be associated with Appendix K models, as Appendix K is supposed to be biased towards a conservative end product.
Some notable conservatisms are decay heat production, initial lienar heat rate, initial stored energy, and metal-water reaction rate.
For the small-break LOCA, and below the metal-water reaction rate threshold, the analysis may not differ too much from reality (i.e., it may be " correct").
In working with B&W (and the other vendors) to develop a small-break methodology that was more of a best-estimate calculation, we saw several features that, in our opinion, should be inserted (at a later date), into the Appendix K model.
These features should then be tested against experimental data.
In addition to the modeling concerns, additional systems verification of the small break model as required by Section II.4 of Appendix K is required as well as providing i
experimental verification of the various modes of single phase and two phase natural circulation predicted to occur during small break LOCAs.
l l
7 4
y c.
. Based on the cumulative staff requirements for additional small-break LOCA model..
verification, including both integral system and separate effects verification, the staff considers model revision as the appropriate method for reflecting any potential upgrading of the analysis methods.
The purpose of experimental verification is to provide the necessary assurance that the small-break LOCA models are acceptable (or at least conservative) to calculate the behavior and consequences of small primary system breaks.
The staff believes that this assurance can alternatively be provided, as appropriate, by additional justi-fication, using recent test data,0f the acroptability of present small break LOCA models with regard to specified staff concerns.
Such justification could supplement or supersede the need for model revision.
Recent Integral Systems tests include the entire Semiscale small break test series and LOFT Tests L3-1, L3-2, L3-7, and L3-5.
The staff believes that the present small break LOCA models can be both qualitatively and quantitatively assessed against these tests.
Other separate ef fects tests (e.g., ORNL core uncovery tests) and future tests, as appropriate, should be factored into this l
assessment.
l Based upon the information discussed above, a detailed outline of the proposed program to address this issue should be submitted.
In particular, this sub-mittal should address (1) which areas of the model, if any, the licensee intends to upgrade, (2) which areas the licensee intends to address by further justifi-cation of acceptability, (3) test data to be used as part of the overall veri-fication/ upgrade effort, and (4) the estimated schedule for performing the necessary work and submitting this information for staff review and approval.
This submittal will form the basis for a meeting with the B&W licensees and y
w-
- -m
-4
. B&W, scheduled for December 16 and 17, 1980, aimed at staff review and approval
- of the overall plan.
Following staff approval of the revised models, plant-specific analyses using the revised model will be required.
Implementation:
These recommendations will be implemented on TMI-1 as required by NUREG-0737 on the following schedule:
(1) Detailed outline of scope and schedule for meeting this requirement should be submitted by November 15, 1980.
(No schedule has been filed by the licensee as of the date of filing of this testimony.)
(2) Model revisions and/or additional information required to address these concerns should be submitted by January 1, 1982.
(3) The plant-specific analyses using the revised models should be submitted by January 1,1983, or one year after any model revisions are approved.
t
- :2 -
NUREG-0565, Recommendation 2.2.2.c Recommendation:
The effects of core flood tank injection on small break LOCAs should be fur-ther investigated to determine the amount of condensation realistically expected and to determine its effect on heatup and core uncovering.
The condensation model and modeling procedures (i.e., injection lo:ation used in the computer analyses) require further investigation to assure that the effects of CFT injection are biased in a conservative manner.
Semiscale and LOFT test data should be used to verify the models.
Licensee's Position:
"This staff concern relates to the potential for a large underprediction of system pressure due to the analytical assumption of instantaneous steam con-densation on the cold CFT water delivered to the RCS during a small break.
Contrary to this concern, the small break analyses performed for TMI-1 do not predict large pressure oscillations caused by CFT injection.
Thus, while further examination of this phenomena may be performed, the small break pre-dictions are not expected to be substantially altered."
NRC's Position:
This recommendation is listed in the Action Plan as item II.K.3.35.
This recom-mendation has further been incorporated into item I.C.1 of the Action Plan, entitled, " Guidance for the Evaluation and Development of Procedures for Transients and Accidents." As stated in NUREG-0737 under item I.C.1, "In the course of review of these matters on Babcock & Wilcox (B&W)-designed plants, the staff will follow up on the bulletins and orders matters related to analysis methods and results as listed in NUREG-0660, Appendix... Table C.3, item 35..."
In addition, this matter will be evaluated by the staff as part c
-r
t 1
l l 1 1
i of its review of item II.K.3.30 (NUREG-0565, Recommendation 2.2.2.a) " Analysis Methods Used for Small Break LOCA Analysis." Item II.K.3.30 has been previously 1
discussed in this testimony on page 18.
l J
Implementation:
I t
l This item will be implemented on THI-1 as required in NUREG-0737.
Reanalysis of transients and accidents and inadequate core cooling and preparation of l
guidelines for development of emergency procedures should be completed and 1
f submitted to the NRC for review by January 1, 1981.
The NRC staff will review the analyses (including the effect of the phenomena I
of CFT injunction) and guidelines to determine their acceptability by July 1, j
1981, and will issue guidance to licensees on preparing emergency procedures from the guidelines.
Following NRC approval of the guidelines, licensees should' revise and implement their emergency procedures by the first refueling outage af ter January 1,1982.
I r
r i
t a
f f NUREG-0565, Recommendation 2.3.2.a Recommendation:
Tripping of the RCPs in the event of a LOCA is not an ideal solution. The licensees should consider other solutions to the small break problem, for example, an increase in the HPI flow rate.
In the interim, until a better solution is found, the RCPs should be tripped automatically in the case of a small break LOCA.
The signals designated to initiate the RCP trip should be carefully selected in order to differentiate between a small break LOCA and other events which do not require the RCPs to be tripped.
Licensee's Position:
J "The THI-1 Restart Report, Supplement 1, Part 3, response to question 11, presents the design characteristics of our proposed reactor coolant pump trip i
system.
This system is based on a coincident loss of sub-cooling margin and high pressure injection actuation.
The NRC staff has accepted this approach as described in NUREG-0680 (SER at p. C2-18)."
NRC's Position:
This recommendation has been incorporated into the Action Plan as item II.K.3.5.
l Implementation of this item was revised in the May 1980 version of NUREG-0660 l
l to provide for continued study of criteria for early reactor coolant pump trip.
A complete discussion of this subject is provided later in this testimony under NUREG-0623, beginning on page 51.
With respect to the licensee's position, as stated in the TMI-1 restart SER, the staff has reviewed the licensee's commitment for automatic pump trip.
i l
l l-
...~ -.
However, the staff would not determine the acceptability of the proposed method for pump trip until the detailed system design and design evaluation were submitted.
Implementation:
See page 55 of this testimony.
. ~
NUREG-0565, Recommendation 2.3.2.b Recommendation:
The B&W small break LOCA analyses rely on equipment which has not previously been characterized as part of the reactor protection system or part of the engineered safety features.
The equipment used to provide the necessary RCP trip, the pressurizer PORV and PORV block valve, and equipment used to actuate the PORV and PORV block valve fall into this category. The reliability and redundancy of these systems should be reviewed and upgraded, if needed, to comply with the requirement of Section 9 of NUREG-0585, "TMI-2 Lessons Learned Task Force Final Report," regarding the interaction of non safety and safety-grade equipment.
Licensee's Position:
"The equipment used in the post-TMI-2 accident small break LOCA analyses which is not part of the Reactor Protection System or part of the engineered safety features is identified in licensee's testimony in response to UCS Contention 8 and ECNP Contention 1 (additional LOCA analysis) (pages 3, 4, 8 and 9).
The specific items utilized in the analyses are the emergency feedwater system and the equipment used to provide reactor coolant pump trip.
The pressurizer power operated relief valve (PORV) and PORV block valve have not been relied upon in the LOCA analyses."
NRC's Position:
(
I This recommendation is listed in the Action Plan as item II.K.3.4.
Item II.K.3.4 has been further subdivided and incorporated into three other items of the Action Plan for continued study and evaluation:
II.C.1, " Interim Reliability Evaluation i
Program (IREP);"
II.C.2., " Continuation of IREP;" and II.C.3, " Systems Interaction."
All three of these programs are long-term programs which may eventually lead i
l i
. to additional requirements for modifications at TMI-1; however, at the present time there is no specific applicability or schedule yet developed for TMI-1.
In the interim, other programs have been implemented with respect to TMI-1 which address specific staff concerns.
Action Plan items II.E.1.1 and II.E.1.2, " Emergency Feedwater Evaluation" and " Emergency Feedwater System 1
Automatic Initiatic7 and Flow Indication," respectively, are being implemented 1
on TMI-1 as discussed in NRC staff testimony regarding Board Question 6.
In addition, items II.K.3.1 and II.K.3.2, " Installation and Testing of Automatic Power-Operated Relief Valve Isolation System," and " Report on Overall Safety Effect of Power-Operated Relief Valve Isolation System," respectively, are also being implemented at TMI-1 as discussed previously in this testimony beginning on page 11.
Implementation:
No specific requirements regarding item II.K.3.4 have yet been developed by the staff.
The staff is actively pursuing IREP and Systems Interaction studies on a plant specific basis as well as developing plans to implement IREP studies on all operating reactors.
The interim program described above are applicable to all operating reactors, including TMI-1, as discussed in NUREG-0737.
__.._.,_. _ _ _. _ _.. ~._ _, _ _ _._, _. _.._,_...- -. ___
. NUREG-0565, Recommendation 2.3.2.c Recommendation:
Plant simulators used for operator training should offer, as a minimum, the following small break LOCA events:
(1) continuous depressurization; (2) pressure stabilized at a valve close to secondary system pressure; (3) repressurization; (4) stuck-open PORV; and (5) stuck-open letdewn valve.
Each of these cases should be simulated wit h RCPs running as well as tripped.
The first three events should be simulated for both cold and hot leg breaks.
In addition to the usual assumed single failures in the ECCS and feedwater systems, complete loss of feedwater should be simulated in conjunction with the above events.
It is important that training programs also expose the operators to various kinds of system transients on inadequate core cooling as discussed in Section 2.1.9 of NUREG-0578.
Licensee's Position:
" Operator training, including the use of simulators, will be addressed in licensee's testimony on management competence."
NRC's Position:
i This recommendation is listed in the Action Plan as item II.K.3.54.
Item II.K.3.54 has been further incorporated into item I.A.4.1, " Initial Simulator Improvement." Specific requirements for simulator improvement have not yet been developed.
As part of its program to develop an initial set of simulator requirements, the NRC contracted with Oak Ridge National Laboratory to perform an evaluation of the capabilities and use of nuclear power plant simulators,
-~
i,
either built or being built by the U.S. nuclear power industry, to determine the adequacy of existing standards for simulator design and for the training I
of power plant operators on simulators as well as to assess the issues about simulator training programs raised by the accident at TMI-2.
That report has been completed and was published as NUREG-CR-1482, " Nuclear Power Plant Simulators:
Their Use in Operator Training and Requalification," (July 1980).
It is the staf f's intention to forward this document to all vendors, licensees, and applicants for a 90-day comment period.
Following the comment period, the staff will develop its initial set of requirements for improving simulator capabilities and training.
Implementation:
i No specific requirements related to this item have yet been developed for the j
l licensee.
I
. NUREG-0565, Recommendation 2.4.2.a Recommendation:
While certain modeling differences and assumptions exist as well as differences in the results of the analyses, it does not alter the staff's primary conclusion of the suitability of the CRAFT-2 program to generate the required information upon which operator guidelines are developed.
Therefore, it is not recommended, at this time, that the staff perform additional audit analyses to correct these differences.
Licensee's Position:
Since this item is only applicable to the NRC staff, no licensee position is stated.
NRC's Position:
The staff utilized a modified version of RELAP4/ MOD 7 computer code to audit selected analyses performed by B&W using the CRAFT-2 computer code.
While some differences existed in the modeling assumptions and the results of the analyses, the staff concluded that the CRAFT-2 code could predict the expected plant response to depressurization, pressure stabilization, and repressurization transients.
In addition, the CRAFT-2 code could predict loss of natural circulation phenomenon.
Therefore, reasonable assurance was provided that the calculated system response using CRAFT-2 could be used as a basis for developing emergency procedures and operator training aimed at detecting and mitigating the consequences of a small break LOCA.
For the range of break sizes evaluated by the staff with RELAP4 (less than 2
0.07 ft, including PORV failures), no uncovering of the core was calculated
. 2 to occur.
The staff analysis of the 0.07 ft break did show some core uncovery,'
but no significant clad heatup.
In this case, CFT injection was calculated to occur and a rapid refill of the reactor vessel prevented any additional heatup.
Therefore, at the time NUREG-0565 was published, the Task Force did not believe that additional audit calculations were necessary to validate the CRAFT-2 code's usefulness in generating the analyses necessary to support operator guideline development.
However, since NUREG-0565 was published, two additional programs have emerged from the development of the Action Plan which may lead to addi-tional staff audit calculations.
The staff may perform additional audit calculations as part of its review of the licensee's submittals associated with item I.C.1, " Guidance for the Evaluation and Development of Procedures for Transients and Accidents." This program is described in detail on pages I.C.1-1 through I.C.1-5 of NUREG-0737.
In addition, should any model changes be made, as part of compliance with Action Plan item II.K.3.30, " Revised Small Break Loss-of-Coolant Accident Methods to Show Compliance with 10 CFR, Appendix K," the staff will perform any audit calculations it believes necessary to verify the correctness of the model changes.
Licensee action required under item II.K.3.30 was previously discussed in this testimony beginning on page 18.
Implementation:
Not applicable to the licensee.
. NUREG-0565, Recommendation 2.5.2.a Recommendation:
The NRC TMI-2 Action Plan should consider the need for a diverse decay heat removal path independent of the steam generators for Davis-Besse 1.
Con-sideration of diverse systems should include, for example:
(1) increased PORV relieving capacity; (2) higher shutoff head HPI pumps; or (3) installation of a high pressure residual heat removal system.
If a system which manually depressurizes the reactor coolant system below the HPI actuation setpoint is selected, the time available to the operator to decide if system depressuri-zation is necessary (i.e., feedwater cannot be restored) should be greater than 20 mirptes.
The staff believes that times less than 20 minutes do not provide the operator sufficient time in which to fully antlyze the situation and could result in incorrect action being taken.
Licensee's Position:
Not applicable to TMI-1.
NRC's Position:
Not applicable to TMI-1 l
l l
NUREG-0565, Recommendation 2.6.2.a Recommendation:
The various modes of two phase natural circulation, which are expected to play a significant role in plant response following a small break LOCA should be demonstrated experimentally.
In addition, the staff requires that the licen-see provide verification of their analysis models to predict two phase natural circulation by comparison of the analytical model results to appropriate integral systems tests.
Licensee's Position:
"The B&W small break LOCA evaluation model includes appropriate consideration for the mechanisms responsible for natural circulation.
The computer code utilized, models both density changes and flow losses under single-and two-phase fluid conditions.
Thus, the evaluation model should reasonably predict the various modes of two phase natural circulation.
Additionally, for small break LOCAs, the steam generators do not have an important influence on the transient except for those cases where the break size is insufficient to discharge energy at least equal to that added by the core decay heat.
As noted in licensee's testimony in response to UCS Contention 1 and 2 (Natural and Forced Circulation) (pages 6 and 7), this break size would be less than approximately 0.02 ft.
Breaks smaller than 0.02 ft2 will retain substantially 2
more system inventory than the design basis small break, which is approximately 0.07 ft, and have large margins relative to the potential for core uncovery.
2 Therefore, while further examination of two phase natural circulation phenomena may be performed, TMI-I is still expected to conform to 10 CFR 50.46."
NRC's Position:
This item is listed in the Action Plan as item II.K.3.32.
Action required by the licensee to provide experimental verification of its analysis models to predict two phase natural circulation has been incorporated into Action Plan item II.K.3.30, " Revised Small-Break Loss-of-Coolant Accident Methods to Show Compliance with 10 CFR 50, Appendix K."
A more detailed description of II.K.3.30 requirements is addressed beginning on page 18 of this testimony.
In addition to action required by the licensee, the NRC has incorporated into the Action Plan item II.E.2.2, "Research on Small Break LOCAs and Anomalous Transients."
This research program focuses on transient and small break experimental research with the purpose being to gain a better understanding of the thermal-hydraulic phenomena experienced in light water reactors during these conditions.
One aspect of the research will be to provide the experimental verification of the various modes of two-phase natural circulation using the semiscale and LOFT test facilities.
In addition to these programs, the Office of Nuclear Regula-tory Rsearch is coordinating with Japan and the Federal Republic of Germany for tests on small breaks, flow blockages, and natural circulation.
To date, experimental evidence has been obtained that two phase and reflux-boiling modes of natural circulation can adem:ately remove decay heat from the core.
This data has been obtained at both the Semiscale and LOFT facilities in the U.S. and in the PKL facility in Germany.
For example, test results from LOFT test L3-2 (conducted in February 1980) indicated that the steam generator transitioned from liquid natural circulation to two phase natural circulation, and possibly to reflux-boiling and then back again to liquid
I 1 i i
,i natural circulation with no evidence of instability.
Curing another LOFT test conducted in September 1980 (L3-5) the pressure vessel liquid level was lowered 1
far below the hot leg and flow conditions in the hot leg were monitored to fur-tner study the question of reflux-boiling.
All flow measurements showed that even for this case, flow continued in the positive direction and adequately provided core cooling.
For these tests, staff codes predicted the major i
phenomena in the proper sequence.
Each of the facilities mentioned above have recirculation ("U"-tube) type steam generators.
An integral systems test facil-ity utilizing a OTSG does not presently exist.
)
Implementation:
4 Licensee implementation of this recommendation will be required as part of Action Plan item II.K.3.30 as described in NUREG-0737.
The implementation schedule for II.K.3.30 is provided on page 21 of this testimony.
NUREG-0565, Recommendation 2.6.2.b Recommendation:
Appropriate means, including additional instrumentation, if necesst.ry, should be p.ovided in the control room to facilitate checking whether natt,ral circu-lation has bean established.
Licensee's Position:
4
" Checks that natural circulation has been established are included in appropriate plant procedures and require observing primary system hot and cold leg tempera-tures for a constant differential and observing that cold leg temperature approaches secondary system ssturation temperature.
The instrumentation used in this determinatio7 ar.s located in the control room."
NRC's Position:
lhis recommendation is listed in the Action Plan as item II.K.3.6.
The Action Plan further incorporates this item into two additional items which are presently being implemented on TMI-1.
Item I.C.1, " Guidance for the Evaluation and Develop-I ment of Procedures for Transients and Accidents," (previously discussed in this testimony beginning on page 22) is being pursued by the staff and licensee.
Item II.F.2, " Instrumentation for Detection of Inadequate Core Cooling," is also being pursued jointly by the staff an'1 licensee.
Both of these items require detailed analyses, guidelines, and procedures, as well as the identi-fication of aeditional instrumentation that may be needed by the operator to reconize and evaluate off-normal conditions in the' plant during accident and transient situations.
To date, no additional instrumentation has been identi-fied, by the staff,as being required to directly monitor natural circulation conditions in the reactor coolant system.
L
. The staff has approved the current guidance given to TMI-1 operators with respect to verification of natural circulation (TMI-1 Operating Procedure 1102-16, Rev. 3, "RCS Natural Circulation Cooling).
That guidance directs the operator to verify natural circulation by checking:
1.
RCS AT increases to
- 30*F and stabilizes.
T stays essentially constant c
with T increasing and then leveling out.
h l
2.
Verify heat removal from OTSG's:
a.
Turbine bypass valve positions, b.
Atmospheric dump valve positions, c.
Feedwater valve positions, and d.
Feedwater flow.
3.
Incore thermocouple temperatures stabilize.
While the staff believes that the instrumentation needed to perform this veri-fication is adequate, until the analyses, evaluations, and reviews of items I.C.1 and II.F.2 are completed, the staff has not foreclosed on the possibility that some additional instrumentation may be needed in the future.
l l
Implementation:
i Both items I.C.1 and II.F.2 are being implemented on TMI-1 as directed by Com-mission Order (both items were originally part of the Lessons Learned NUREG-0578 i
recommendations) and as identified in NUREG-0737.
I
. NUREG-0565, Recommendation 2.6.2.c Recommendation:
Licensees should provide an analysis which shows the plant response to a small break which is isolated and the PORV fails-open upon repressurization of the reactor coolant system to the PORV setpoint.
Licensee's Position:
"A specific analysis providing the plant response to a small break which is isolated and the PORV fails-open upon repressurization of the RCS to the PORV setpoint has not been performed.
However, based on the analyses discussed in licensee's testimony in response to UCS Contention 8 and ECNP Contention 1(e)
(Additional LOCA Analysis), the response to this event can be described.
Initially, as a result of the small break, the system will depressurize.
Actuation of the High Pressure. Injection system (HPI) will automatically occur, assuming feedwater availability, prior to the loss of natural circulation.
Should break isolation occur after natural circulation is lost and prior to the establishment of the boiler-condenser mode of steam generator heat rer. oval,
[
system repressurization would occur.
Assuming that the repressurization reaches l
l the PORV setpoint and that the PORV sticks open, a transient very similar to that calculated for a PORV initially stuck open would then occur.
Adequate core cooling would be continuously maintained for this transient by the fluid provided by HPI."
i f
(
I
. U NRC's Position:
This recommendation is listed in the Action Plan as item II.K.3.37.
As stated in the licensee's position, a specific analysis of the exact scenario described in the recommendation has not been performed.
However, by letter from J. G. Herbein to H. R. Denton, dated June 30, 1980, the licensee 2
submitted the results of a detailed, quantitative analysis of a 0.01 ft break in the cold leg pump discharge piping, with no auxiliary feedwater to the steam generator resulting in repressurization of the RCS to the PORV setpoint.
In the analysis the PORV is assumed to stick-open.
This analysis is presently undergoing staff review and appears to provide a more limiting break scenario than that described in recommendation 2.6.2.c.
Staff review of this analysis is expected to be completed in late December of this year.
Item II.K.3.37 has been incorporated into the scope of Action Plan item I.C.1,
" Guidance for the Evaluation and Development of Procedures for Transients and Accidents." A discussion of item I.C.1 has been provided earlier in this testimony beginning on page 22.
Implementation:
At the present time, no additional information is required by the licensee regarding this recommendation.
However, should additional follow up analyses be required, it will be folded into the staff's review and evaluation of item I.C.1.
Item I.C.1 is being implemented on THI-1 as directed by Commission Order and as describes in NUREG-0737.
. NUREG-0565, Recommend;.j?on 2.6.2.d Recommendation:
Licensees should provide an analysis which shows the plant response to a break in the pressurizer spray line with a failure of the spray isolation valve to close.
Licensee's Position "A break in the pressurizer spray line along with a failure of the spray isolation valve to close results in inventory loss from both the RCS cold leg and the top of the pressurizer. The leak rates from the cold leg would be limited by the area of the spray line, 0.025 ft, and from the pressurizer the 2
leak would be limited by the flow area of the spray nozzle in the pressurizer, 0.072 ft.
The small break LOCA analyses performed for TMI-1 to demonstrate 2
conformance to 10 CFR 50.46 envelop the total leak flow area for this case.
Thus system inventory losses similar to that which would occur for this scenario have already been considered in the LOCA analyses.
However, for this accident, liquid inventory would remain in the pressurizer while the TMI-1 small break analyses empty the pressurizer.
The effect of the stored inventory in the pressurizer for this event is expected to be offset by the increased inventory availability of HPI for core cooling.
In the analyses performed for TMI-1, less than 70% of the HPI was calculated to enter the core due to the direct bypass of the injected fluid out the break, which was assumed to be located in the bottom of the cold leg pump discharge piping between the HPI nozzle and the reactor vessel.
For the spray line break, no i
HPI fluid would bypass out the break without first entering the vessel. The
increased HPI flow for the spray line break would establish long term cooling earlier, relative to an equivalently sized pump discharge break, and is expected to offset the effect of the stored inventory in the pressurizer.
Therefore, an analysis of this accident is not expected to provide results which are in excess of 10 CFR 50.46 limits."
NRC's Position:
This recommendation is listed in the Action Plan as item II.K.3.38.
Item II.K.3.38 has been incorporated into the scope of Action Plan as item I.C.1,
" Guidance for the Evaluation and Devclopment of Procedures for Transients and Accidents." A discussion of item I.C.1 has been provided earlier in this testimony beginning on page 22.
The staff has required that as part of the analyses done to support guideline and procedure development for transient and accidents, the licensee consider this scenario.
The analysis should be carried out far enough into the event to assure that all relevant thermal / hydraulic /neutronic phenomena are identified.
Implementation:
As stated above, item I.C.1 incorporates this recommendation.
Item I.C.1 is being implemented on TMI-1 as directed by Commission Order and as identified in NUREG-0737.
i l
NUREG-0565, Recommendation 2.6.2.e Recommendation:
Licensees should provide confirmatory information to show that HPI and CFT flows during small breaks are insufficient to form water slugs, of if they do, to show that the structural design bases of the primary system includes loads due to:
(1) water slug inertial motion; (2) water slug impact; and (3) pressure oscillation due to steam condensation.
Licensee's Position:
"During small breaks, water slugs are not expected to be formed as a result of HPI and CFT flows.
The HPI flow would be less than 140 ft / min during a small break transient.
Since the piping volume from the HPI nozzle to the reactor 3
vessel is 280 ft, it would take two minutes to fill the pipe.
Also the reactor vessel internal vent valves will continuously equalize pressures throughout the primary system.
Therefore, the HPI water will drain into the vessel and there is no mechanism available to hold the HPI water in the cold leg pipe.
Thus slug flow as a result of the HPI will not occur.
The water injected from the CFT's also is not expected to produce slug flow since the fluid is directly injected into the reactor vessel downcomer.
- Also, the internal vent valves minimize pressure gradients within the vessel such that no holdup of injected CFT water will occur.
Thus, no water slugs will occur as a result of CFT injection."
i
i 43 -
NRC's Position:
This recommendation is listed in the Action Plan as item II.K.3.39.
Item II.K.3.39 has been incorporated into the scope of item I.C.I.
The response provided by the licensee (as stated above) is being reviewed by the staff If a more quantitative analysis, supported by test data, is required it will be pursued through the implementation of item I.C.1.
Implementation:
If additional documentation and supporting that data is required by the staff, regarding this concern, it will be done through the staff's review and evalua-tion of item I.C.1.
Item I.C.1 is being implemented on TMI-1 as directed by Commission Order and as identified in NUREG-0737.
t
=
l 4
1 i
.-,,,-.-.,,,c
,y-
,r,,
.,,,,,,.~
..p.__....v..
. ~,,,,. - - -,
,y.
.g..r...
NUREG-0565, Recommendation 2.6.2.f Recommendation:
Licensees should provide an analysis of the possibility and impact of RCP seal damage and leakage due to loss of seal cooling on loss of offsite power.
If damage cannot be precluded, licensees should provide an analysis of the limiting small break LOCA with subsequent RCP seal failure.
Licensee's Position:
"This recommendation was addressed in Licensee's response to R. W. Reid's letter of November 21, 1979. which was provided by letter No. TLL-285, dated June 30, 1980.
In this response, a description of the RCP seal system and its cooling was provided along with a discussion of the probable degradation mechanism, the time and methods available to restore seal cooling, and the result of loss of cooling for up to 60 minutes.
The results of that analysis did not indicate that excessive seal leakage would occur within 60 minutes."
NRC's Position:
This recommendation is listed in the Action Plan as items II.K.3.40 and II.K.2.16.*
The analysis stated above by the licensee is presently undergoing review by l
the staff.
The licensee's analysis shows that following a complete loss of seal cooling (i.e., no seal injection and no component cooling water) the total leakage per pump would be approximately 5 GPM after 30 minutes and approximately l
l
- Note:
Information/ analyses requirements regarding RCP seal degradation is described under both items of the Action Plan.
A consolidation of the information required is provided under item II.K.2.16 of NUREG-0737.
f I
J j 10 GPM after 60 minutes. The analysis assumed the RCPs were in a static condi '
tion (i.e., not running).
+
i Implementation:
4
}
This item (II.K.2.16) is a full power requirement for near term operating license applicants, as described in NUREG-0694.
As such.this item must be completed prior to restart. The staff expects to complete its review of the licensee's submittal by late December 1980.
1 h
a 4
l
.f O
i t
t
46 -
NUREG-0565, Recommendation 2.6.2.g s
Recommendation:
Licensees shall provide pretest predictions of LOFT test L3-6 (Reactor Coolant Pumps Running).
l Licensee's Position:
"GPU is a participant in the B&W Owners' group program to predict LOFT L3-6.
This analysis will be performed by B&W and provided to'the NRC."
NRC's Position:
r This recommendation is listed in the Action Plan as item II.K.3.41.
The purpose of this recommendation and the action required as well as the schedule for sub-mission of test predictions is discussed later in this testimony under NUREG-0623 beginning on page 54.
Implementation:
See page 55 of this testimony.
1 e
w-w-
--wg-,,,-w-,...~,,__y,_.,
NUREG-0565, Recommendation 2.6.2.h Recommendation:
With regard to the effects of noncondensible gases during a small break LOCA, the licensees should provide the following information:
(1) The technical justification for omitting the radiolytic decomposition of injected ECC water as a source of noncondensible gas; and (2) Confirmatory information to verify the predicted condensation heat transfer degradation in the presence of noncondensible gases.
Licensee's Position:
" Analyses of the effect of noncondensibles on the condensation heat transfer process in the steam generator during a small break LOCA have been performed.
These analyses, which included the effects of radiolytic decomposition, deter-mined that sufficient condensation surface would remain within the steam generator and that the broiler-condenser mode would not be prohibited.
Additionally, even under a postulated condition that noncondensible gases prohibited conden-sation, HPI can be operated in the feed and bleed mode to supply adequate core cooling - see licensee's testimony in response to UCS Contentions 1 and 2 (Natural and Forced Circulation).
Thus while further examination of the effect of noncondensibles on the condensing heat transfer process within the steam generator may be performed, provisions are available at TMI-1 to assure adequate core cooling."
. NRC's Position:
e This recommendation is listed in the Action Plan as item II.K.3.42.
This item as been further incorporated into item I.C.1, which has previously been discussed in this testimony beginning on page 22.
The only analysis submitted by the licensee (that the staff is aware of) dealing with the sources and effects of noncondensible gases, was included in a letter from J. G. Herbein to H. R. Denton, dated June 30, 1980.
This analysis is the same generic analysis performed by.B&W that is discussed in Section 4.2.11 of NUREG-0565.
While the staff does not believe that the additional volume of noncondensible gases which would be added by including this term would change the overall conclusion of the analysis with respect to the effect of noncon-densible gases on inhibiting natural circulation, the staff still requires that the licensee provide justification why this source can be omitted from the analysis.
In addition, the licensee has not provided the confirmatory information to verify the predicted condensation heat transfer degradation in the presence of noncondensible gases.
Implementation:
I The licensee will be required to address these items in its supporting analyses for item I.C.1, " Guidance for the Evaluation and Development of Procedures for Transients and Accidents." Item I.C.1 is being implemented on TMI-1 as directed by Commission Order and as identified in NUREG-0737.
NUREG-0565, Recommendation 2.6.2.i Recommendation:
By use of analysis and/or experiment, address the mechanical effects of induced slug flow on steam generator tubes.
Licensee's Position:
" Analyses of the effect of induced slug flow on the steam generator has been performed.
The analyses assumed that a sudden front of water impacted the tube sheet with a flow equivalent to that of normal operation.
It was assumed that this load was suddenly applied and that the entire load was absorbed by the tubes directly under the inlet nozzle of the steam generator.
The loading on the steam generator tube was calculated to be 21.5 lbf, in comparison to the theoretical buckling load of approximately 700 lbf.
Thus, induced slug flow will not affect the integrity of the steam generator tubes."
NRC's Position:
This recommendation is listed in the Action Plan as items II.K.3.43 and II.K.2.15.*
The analysis described by the licensee was submitted to the staff in a letter from J. G. Herbein to H. R. Denton, dated June 30, 1980.
The analysis is presently undergoing staff review.
Certain details of the calculations were not shown in the analysis; therefore the staff will ask that the licensee
- Note:
Information/ analysis asked for in this recommendation is listed under both items of the Action Plan.
A consolidation of the information required is provided under item II.K.2.15 in NUREG-0737.
l '
submit the complete set of. calculations which support their findings.
Implementation:
This item (II.K.2.15) is a full power requirement for near term operating license applicants, as described in NUREG-0694. As such, this item must be completed prior to restart. The staff expects to complete its review of 1
this matter by late December 1980.
i e
i I
l t
i i
PART 2 NUREG-0623, " Generic Assessment of Delayed Reactor Coolant Pump Trip during Small Break Loss-of-Coolant Accidents in Pressurized Water Reactors BACKGROUND Shortly after the accident at TMI-2, the NRC issued IE Bulletins to all pres-surized water reactor licensees that instmJcted them in the event of HPI initiation to maintain forced flow in the reactor coolant system.
At that time, forced circulation with the RCPs was thought not only to be acceptable but pre-ferred.
More extensive analyses were performed by all PWR vendors and the staff in the months that followed the accident.
In some cases, vendor analyses con-cluded that either delayed trip or continuous operation of the RCPs during a small break LOCA could possibly lead to predicted fuel cladding temperatures in excess of the current licensing limits (2200 F).
The B&W analyses showed that if the RCPs remained running during the accident, the core would remain acceptably cool.
However, the continuous operation of RCPs resulted in the primary coolant system evolving to a high system void fraction early in the accident and remaining relatively high until the system depressurized enough to actuate the low pressure injection system and recover the system liquid inventory.
Because the system void fraction evolved to such a high value, B&W examined what would happen if the pumps were tripped at some time into the accident when the system void fraction was high.
At the time of the pump trip, the liquid that was previously dispersed around the primary system, through pumping action, now collapsed down to the low points of the primary system, such as the bottom of the vessel and steam generators.
For small breaks' 2
between 0.2 and 0.025 ft, this resulted in a significant uncovery
of the reactor core.
Since the HPI system could not refill the reactor vessel in time, an insufficient amount of liquid was available to provide acceptable core cooling.
Based upon our review of these analyses and similar analyses performed by the other vendors, the staff issued IE Bulletin 79-05C/06C on July 26, 1979.
In addition to follow up analyses required by the Bulletin, each licensee
- was instructed to immediately trip all operating RCPs upon reactor trip and initia-tion of HPI caused by low reactor coolant system pressure.
All licensees were further required to provide two licensed operators in the control room at all times during operation to accomplish this action and other follow up actions required during such an occurrence.
As part of the long-term action required by the Bulletin, each licensee was required to propose and submit a design which would assure automatic tripping of the operating RCPs under all circum-stances in which this action may be needed.
The purpose of publishing NUREG-0623 (November 1979) was to present the results of the staff review of the vendor analyses submitted in resoonse to IE Bulletin 79-05C/06C and to present the staff's conclusions based upon that review.
Based upon the conclusions, presented in Section 6.0 of the report, it was quite clear that additional research and analyses would have to be done by both i
l industry and the staff in order to completely answer all questions associated l
with the pump trip issue.
l
- Westinghouse has always instructed its customers to trip RCPs on SBLOCA, l
even before the TMI-2 accident.
l
D e Recommendations The Board has asked the staff to address the recommendations of NUREG-0623.
As stated above, the document does not present a specific set of recommen-dations; however, from the conclusions presented in Section 6.0, the recommendations can be summarized as follows:
(1) Tripping of the RCPs during small break LOCAs is required at this time:
- (a) The pump trip should be automatically initiated from equipment that is safety grade (to the extent possible) and (b) The criteria and requirements for RCP trip should minimize (to the extent practicable) the probability of initiating a RCP trip for a non-LOCA transient.
(2) Verification of small break models against appropriate integral system tests (small break with RCPs running) should be performed.
(3) Continued exploration by industry to find other solutions to the small break problem, other than RCP trip, should be pursued.
- Section 7.3.1 of NUREG-0623 presented a proposed schedule for design submittal,' evaluation, and installation of automatic RCP trip circuitry.
1
o CURRENT STATUS In its review of NUREG-0623, the Advisory Committee on Reactor Safeguards (ACRS) did not totally agree with the conclusions presented by the staff.
The ACRS agreed that tripping of the RCPs upon primary system depressurization was an acceptable procedure; however, it did not agree with the urgency of the proposed schedule for installing an automatic pump trip.
The ACRS position was based primarily on its concerns regarding the limited experimental data available to verify the analytical models.
In addition, the ACRS expressed concern that by tripping the RCPs during non-LOCA transients some degree of operator control was lost with respect to primary plant pressure control (most PWRs rely on the differential pressure across the core produced by RCP operation to provide spray flow to the pressurizer for RCS pressure reduction).
Because of these concerns, the staff has delayed the requirement for licensees to install an automatic RCP trip.
The decision of how and when to trip the RCPs in the long-term will be made following the completion of additional experimental data as described below.
Integral cystem tests to verify that a one dimensional equilibrium models could reasonably predict the qualitative behavior of small break LOCAs in a PWR with RCPs operating was needed.
This assurance was obtained in tests conducted earlier this year on the Semiscale facility where small break LOCA experiments were done with RCPs operating and tripped (S-SB-P1 through 5-SB-P6 Test Series).
In September of this year a small break LOCA with RCPs tripped was run on the LOFT facility (Test L3-5).
The results of data from L3-5 will be compared with data obtained from Test L3-6 (presently scheduled'for December 10, 1980).
Test L3-6 will be a small break LOCA with RCPs running through the test.
, All holders of approved ECCS models (including B&W) have been requested to perform a " blind" post-test analysis of L3-6 using the actual test conditions.
It was also requested that the models which will be used to evaluate the test be documented with the staff by December 3, 1980.
Test results and the capabilities of the vendors' ECCS models to correctly pre-i dict the plant behavior experienced during L3-6 will provide a strong input into the staff's determination of:
(1) Whether RCP trip is required, (2) If RCP trip is required, the criteria to be used as a basis for the trip, and (3) If RCP trip is required, is their sufficient time available to the operator to warrant consideration of manual tripping or is an automatic trip required.
As previously discussed in this testimony, under NUREG-0565, Recommendation i
2.3.2.a (page 24),the RCP trip issue has been incorporated into the Action Plan as item II.K.3.5.
Clarification and implementation of this item is pro-i vided by NUREG-0737.
The present schedule for resolution of this issue is as follows:
(1) Vendors are to document the models to be used for analysis of LOFT Test l
L3-6 prior to December 3, 1980.
t
. (2) DOE /NRC conduct LOFT Test L3-6 (scheduled for December 10, 1980).
(3) NRC will distribute initial conditions of the test to vendors approximately 4 weeks after the test.
l (4) Prediction results will be submitted approximately 4 weeks after receipt of the initial conditions.
1 (5) NRC determination of model acceptability is due by April 1, 1981.
(6) If required, licensees propose necessary design modifications by July 1, 1981.
(7) If required, licensees shall implement necessary design modifications by March 1, 1982.
l o
PROFESSIONAL QUALIFICATIONS DENWOOD F. ROSS, JR.
OCTOBER 1980 I am presently employed with the U. S. Nuclear Regulatory Commission, within the Office of Nuclear Reactor Regulation, as the Director of the Division of Systems Integration.
My work address is 7900 Norfolk Avenue, Bethesda, Md.
The functional assignments of the Division of Systems Integration include six systems areas (Reactor Systems, Instrument and Control Systems, Auxiliary Systems, Effluent Treatment Systems, Power Systems, and Containment Systems) as well as Accident Evaluation, Radiological Assessment, Core Performance, and Systems Interaction.
Work assignments prior to this present prsition include:
Acting Director, Division of Project Management Director, Bulletins and Orders Task Force (a post-THI-2 group)
Deputy Director, Division of Project Management Previously I served as Assistant Director for Reactor Safety (from 1/76 to 10/78).
This included supervising the activities of the Analysis Branch, the Core Performance Branch, and the Reactor Systems Branch which, together, formed-the Reactor Safety group in DSS.
The work assignments performed by Reactor Safety included evaluation of emergency core cooling system response, as well as reactor core and primary coolant system response to transient and other accident conditions.
l
. Prior to that assignment I served as Branch Chief of the Core Performance Branch for about 2-1/2 years.
Other job assignments,since ca.ing to USNRC (then AEC) in August 1967 include project manager assignments (or major con-tributor to) for several projects, including Three Mile Island (Units 1 & 2);
Crystal River Unit 3; Oconee 1, 2, and 3; and Quad Cities 1 & 2.
In addition, I served on a special task force reviewing ECCS performance, including extended service at the ECCS rule-making hearing.
Prior to joining NRC I worked at the General Dynamics nuclear research facility at Ft. Worth, Texas for 10 years, including 4 years as operations supervisor for three research and test reactors.
I also worked for 1-1/2 years at the MTR-ETR operations at the NRTS, Idaho.
I have degrees in Civil Engineering (BS, 1953); Mathematics (MS, 1963); and Nuclear Engineering (MS 1960, and D. Engr., 1974),
PROFESSIONAL QUALIFICATIONS ROBERT A. CAPRA NOVEMBER 1980 Since June 1980, I have served as Technical Assistant to the Director, Division of Systems Integration, Office of Nuclear Reactor Regulation, U.S. Nuclear Regulatory Commission (NRC).
In this capacity, I work for the Division Director on various technical and administrative matters associated with review and regulation of operating nuclear power plants and plants under review for operation licenses.
I enlisted in the United States Navy in July 1964 and served in that capacity for three years.
During that time my duties included attending the Enlisted Naval Nuclear Power School, Mare Island, California foliowed by subsequent study and qualification as a reactor operator and staff instructor on the Navy's " DIG" reactor located in West Milton, New York.
1 i
Following my enlistment, I attended the United States Naval Academy where I graduated in June 1971 with a Bachelor of Science degree in Marine Engineering and was commissioned as a line officer in the United States Navy.
Additional graduate level studies in nuclear reactor theory, thermodynamics, electrical engineering, health physics and other related engineering fields were completed 1
f in 1972 at the Officer Naval Nuclear Power School, Bainbr#dge, Maryland.
I i
subsequently returned to West Milton, New York where I studied and qualified l
as Engineering Officer of the Watch on the Navy's " DIG" reactor.
l l
s From 1973 to 1976, I served aboard an operating nuclear submarine, during which time my duties included standing watch as Engi.neering Officer of the Watch as well as directing and training and personnel under my supervision in the operation, maintenance and repair of various equipment and systems, primarily associated with the ship's nuclear reactor.
During this period my assignments included supervision of the Operation Department, Electrical Division, Reactor Controls Division, Main Propulsion Division and the Chemistry and Radiological Control personnel.
In addition, I was certified as Chief Engineer for the supervision of operation and maintenance of Naval Nuclear Propulsion Plants.
From 1976 to 1978, I was assigned as a Company Officer at the United States Naval Academy where my duties included supervising, directing and evaluating the training and activities of 130 cfficer candidates (midshipmen).
I joined the NRC staff in July 1978, where I served as a Licensing Project Manager in the Division of Project Management.
In this capacity, I coordinated 1
the safety review of one operating license application (South Texas 1 and 2),
two construction permit applications (New Haven 1 and 2 and Haven, Unit 1) and served as the Project Manager for two plants under construction (Nort Anna 3 and 4).
In addition, I served as the Licensing Topical Report manager for General Electric Topical Reports.
l In June 1979, following the accident at Three Mile Island, Unit 2, I joined the Bulletins and Orders Task Force (B&OTF) and served as the Project Manager for B&W-desi aed operating reactors.
In this capacity, I coordinated and established t
1
_ ~
3
.s for the s.ork being dcne by the lask Force v.hich was assoc ated with i
- s i i.. I:J,',l-de signed plants.
I coordinated the scope and schedule of the work
,hmd by the B&W licensees and served as the pridcipal liaison between the L force, the licensees and the EEW 0..aers' Gcoup.
During this period, I i
, sed as editor of NUREG-0565 and fURI.G-0667 and authored the B&W section of II.3if Fir.31 Report, l'UREG-CMS.
I served with the Task Force until I was n:ig v d te ry present position in June 1990.
i l
l t
l l
l
[
l l
l t