ML20040E206
| ML20040E206 | |
| Person / Time | |
|---|---|
| Issue date: | 12/08/1981 |
| From: | Ebersole J, Mark C, Plesset M Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-1928, NUDOCS 8202030289 | |
| Download: ML20040E206 (11) | |
Text
.
00/h-/9.AV DATE ISSUED:
12/8/81 qw?
,dj))? l-I'l-b o'
[
j ACRS ECCS SUBC04:11TTEE MEETING MINUTES DECEMBER 2-3, 1981 IO LOS ALAMOS, N.M.
The ACRS ECCS Subcommittee met at the National Security and Resources Study Center in Los Alamos, New Mexico. The purpose of the meeting was to review selected portions of the NRC Safety Research Program in the areas of LOCA-ECCS for the Committee's annual Report to Congress. Notice of the meeting was published in the Federal Register on Monday, November 16, 1981.
Principal meeting attendees are noted below:
ACRS NRC M. Plesset, Chairman H. Sullivan J. Ebersole L. Shotkin C. Mark (2nd. only)
F. Odar D. Ward W Beckner I. Catton (2nd only)
A. Serkiz A. Dukler K. Gar 11d INEL V. Schrock V. Ransom T. Wu W. Weaver Y. Chen - ACRS Fellow P. Boehnert - DFE*
LASL
- Designated Federal Employee D. Liles K. Williams SANDIA W. Defiuth N. Evans A complete list of meeting attendees is attached to the office copy of these minutes.
Meeting Highlights, Agreements, and Requests 1.
Mr. L. Shotkin (NRC-RES) provided an overview of the RES efforts for the code application, development, and assessment program. There are 5 main program elements associated with the code application effort. These efforts and 8202030289 811208 PDR ACRS 1928 PDR
ECCS Meeting December 2-3, 1981 assocaited significant results given in parentheses are:
(1) 20/3D program support with the TRAC-P code, (TRAC predicted flow reversal in German PWR hot-leg injection plant before phenomenon verified in later tests); (2) use of advanced codes for evaluation of thermal hydraulics for pressurized thermal shock problem (an on-going major effort; (3) support calculations for NRR (BE codes show for PWRs PCT occurs during blowdown not reflood, - tests are confirming this fact); (4) TRAC-B used to predict multideminstonal effects seen in SSTF; and (5) use of RELAP-5 for predicitons of LOFT and Semiscale tests.
The code imrpovements (needs) were also reviewed for the advanced syster codes based on current and planned applications (Figure 1). Mr. Ward questioned the usefullness of applying TRAC or RELAP to plant simulator development.
He felt that it was preferable to rely on the industry to develop suitable codes in this area. Mr. Shotkin noted that RES will try to make effective use of Industry products in this area, if at all possible.
2.
Mr. V. Ransom (INEL) discussed the status and applications of the RELAP-5 one-dimensional best estimate advanced code.
The code is being applied to descriptions of operational and LOCA transients and may also be applied to descriptions of severe core damage transients. The current version released for public use is M00-1.
A MOD-2 version will be completed in FY 82 and released to the public in early FY 83. MOD-2 will be the last major development effort with the code going into a maintenance /
enhancement mode for FY 83 thorugh 85.
During this time, a severe core damage fuel model may be added, and the code will be put in an inter-active real time mode for possible simulator application.
Details of the REALP code features were provided, highlighting the code models/ capabilities, user conveniences, and run-time improvements.
3.
Mr. D. Lilies discussed the status of the TRAC-P development and assessment effort.
He focused on the key problem areas addressed by TRAC development and the development plans for FY 82-83.
Key problem arers (and their solu-tions) noted were:
(1.) reflood modeling (addition of i.verted annular
.s ECCS Meeting December 2-3, 1981
.I flow regime and 2-D conduction solution), (2) lack of mass conservation (tighter level of convergency in solution procedure); (3) excessive run time (use of new two-step numberics for one-dimensional component model'ing),
(4) modeling of small break LOCA and slow transients (use of full two-phase-fluid model in both 1-and 3-dimensional models plus addition of a non-condensible gas field), (5) code useability (addition of user convenience features), and (6) improvement of constitutive models (continual process).
Referring to the mass conservation problem noted above, Dr. Catton suggested LASL check the TRAC code, on a global level, to assure there are no energy conservation problems with the code.
Work underway in FY 82 includes improving the code's ability to analyze operational transients.
For FY 83, LASL plans to release - PF1/ MOD-1 for operational transient use and begin work on - PD3 (last major TRAC-P version pl anned).
LASL detailed the applications of TRAC PD2 as noted by L. Shotkin above.
Comparisons of TRAC PD2 post-test calculations of the first round of JAERI Slab Core Test Facility tests snoe reasonable agreement with the data.
4.
Mr. W. Weaver (INEL) discussed the program objectives and status, model development and results, and future plans for the TRAC BWR code.
Presently, development of the BD1 version is complete and it has been released for public use.
Development of the BD1/ MOD-1 version is underway.
BD1 has a basic LOCA capability. BD1/M00-1 will include modeling capability for LOCA, operational transients, and ATWS.
BD1/ MOD 1 development should be complete by October 1982, with developmental assessment and public release scheduled for early FY 1983. There are tentative plans for a BD2 version which would include spatial kinetics modeling ability. However, NRC has not given a fian1 approval for devleopment of this code version.
Dr. Plesset asked RES if there are any plans to develop a BWR version of RELAP-5.
Dr. Sullivan replied that this topic is under active discussion at RES.
ECCS Meeting December 2-3, 1981 0.
5.
Dr. F. Odar (RES) detailed the status of the RES code assessment program o
including results, to date, of the TRAC PD2 large break LOCA assessment.
Dr. Odar noted that independent assessment is essential to quantify code accuracy and provide understanding of code capabilities. Feedback from such assessment also aids code development.
In FY 82 independent g
assessment of TRAC PD? will conclude, while RELAP-5/ MOD-1 and TRAC BD1 assessment will ' continue.
Assessment of TRAC-PF1 will also begin in FY 82.
However, it w:.s also noted that the assessment schedule may be impacted by the effort underway on the pressurized thermal shock calculations. Most of the assessment effort is scheduled to conclude by late 1983.
Results of the TRAC-PD2 large break assessment conducted to date were described. Key results are noted below:
0
' PCT prediction is reasonably accurate (2 6 uncertainty
- 80 K).
PCT is accurately predicted during blowdown but is not accurately predicted during reflood.
' Accumulator injection and CHF are predicted accurately.
'The entrainment model predicts excessive entrainment and needs improve-ment. The critical flow model also requires improvement.
i
' Limited qualitative assessment of small break LOCA phenomena in LOFT L3-5 and L3-6 tests are reasonably well predicted.
'PD2 assessment will continue in FY 82.
Dr. Schrock suggested that various code Users run comparisons of a given l
small break LOCA test as part of the assessment effort.
RES agreed this
{
would be a worthwhile effort.
l l
l l
ECCS Meeting December 2-3, 1981 6.
Mr. L. Shotkin provided a summary of key comments received by members of the NRC Advanced Code Review Group on the code assessment program. These comments include:
1.
The LB LOCA problem is solved. Completion of LB LOCA code assessment is a financial and political issue.
2.
The concept of truly independent code assessment shold be implemented.
3.
Safety margins should be defined, and these should determine the allow-able error in key parameters.
4.
It is important for (NRC) codes to predict system phenomena in actual LWR's.
Therefore, they may have to be " tuned" to large-scale data and integral system tests.
As a consequence, they may fail miserably on small-scale test data.
NRR comment was also solicited. Mr. B. Sheron commented that RES should define acceptance criteria to allow an end to code development and assess-ment.
RES is in basic agreement with this sentiment but they are seeking a consensus from the " reactor safety community" that all important phenomena are simulated, and that this simulation has been adequately assessed.
NRR also provided comments on the RELAP-5 code development effort. NRR places high priority on User support and want to use RELAP-5 for BWR ATWS and pressurized thermal shock calculations. They also place a high priority on continued development of the code's reflood model.
7.
Dr. Plesset noted that a recent presentation on a COBRA-TRAC UHI calculation 0
gave a PCT of u l600 F.
He asked RES if this PCT value is " correct", or is an artifact of the code. He felt this value was high for a best-estimate calcul tion.
Dr. Sullivan said he would 'look into this item and report to the Subcommittee at a future meeting.
8.
Dr. Sullivan provided an overview of the Semiscale program. Currently a series of natural circulation tests are underway. Future tests scheduled include:
3 intermediate-break tests (Jan.-Feb. 82), 5 secondary-side transient tests (June-August 82), 7 loss of electrical power tests (Sept.-Dec. 82) and 10 steam generator tube rupture tests (Feb.-June 83).
ECCS Meeting December 2-3, 1981 Key results from the natural circulation tests, to date, include:
'For a falling primary inventory level, reflux cooling initiates at 70% inventory, and core cooling is adequate to at least 50% inventory.
'For a decreasing secondary inventory:
there is little effect on core cooling for a collapsed level above 50%, and core cooling is adequate.
'For a case of inert gas in the primary system:
nitrogen injection to 10% of system volume decreased two-phase flow by 50% and similiar injection during reflux boiling increased the reflux rate by 100%.
In response to a question fron Dr. Dukler, RES said that while the above results are characteristic of Semiscale only, it is hoped to use this data to verify code predicitons for full-size PWRs.
Responding to questions from Mr. Ebersole, Dr. Sullivan said the transition from single to 2-phase flow is always smooth. Mr. Ebersole urged RES to investigate the effect of nitrogen injection into a competely solid primary system (inadvertent gas injection).
9.
The status and plans of the joint W/NRC/EPRI FLECHT-SEASET program were discussed.
Figure 2 shows a schematic of the facility. The facility is limited to low pressure (ev150 psig). The 21-rod blockage tests, which had as their objective determination of the most severe blockage configuration, were recently completed This wost-case blockage shape (long non-concentric, non-coplanar) will be used in upcoming 163-rod blockage tests.
A series of natural circulation (NC) tests are planned. These tests will examine core cooling transitions, system response, and steam generator behavior during single, 2-phase, and reflux condensation NC modes.
RES plans to conclude all the above testing during FY 82.
In response to questions from Mr. Ebersole, Dr. Sullivan said RES has plans to evaluate cooldown transients in conjunction with W and EPRI.
ECCS Meeting December 2-3, 1981 10.
The status and results of the 2D/3D program were detailed. The first round of tests in the JAERI Cylindrical Core Test Facility (CCTF) are complete, and the first round of tests in JAERI's Slab Core Test Facility (SCTF) are underway. The German UPTF (Upper Plenum Test Facility) design is complete and construction has begun. Dr. Sullivan noted that problems with UDTF have severely impacted the 20/3D program schedul e.
Present estimates project a 3-year slip due to UPTF delays, forcing program completion out to 1988.
Significant results of the CCTF and SCTF reflood tests run to-date include:
CCTF
'Significant upper plenum deentrainment seen (10-15%) leading to core top quench.
The core is effectively cooled by 2-phase flow and completely quenches in 3-10 minutes. These results are similar to what is seen in FLECHT.
SCTF
'There is no significant effect on core cooling with 50% core flow blockage simulated.
PCTs seen in tests were " low" (
- 1475 F).
The core is completely quenched in approximately 5 minutes after the
~
lower plenum is filled.
Significant deentrainment is seen in the upper plenum (similar to CCTF).
Dr. Sullivan also noted that TRAC PD2 has predicted the above data well, and the NRC advanced 2-phase flow instruments have been functioning properly in both facilities. RES noted that there have been problems with collection and analysis of the JAERI data and this is receiving attention.
- 11. Dr. W. Beckner (RES) discussed the status of the joint NRC/GE/EPRI BWR Refill /Reflood Program and the Full Integral Test Facility (FIST). For 0
the Refill / Reflood program, core spray tests at the 30 SSTF (Steam Sector Test Facility) have concluded and data anlaysis is in progress.
Final reports on the test program are scheduled for the end of FY 82.
Both separate effects and system effects tests simulating the BWR 4/5 and BWR 6 ECCS systems were conducted.
ECCS Meeting December 2-3, 1981 Results from the SSTF test program have shown that significant multi-dimensional and multi-channel effects are seen and that these effects appear to be beneficial to core cooling. The ROSA III facility tests in Japan support the SSTF results. This was cited as significant because ROSA-III uses electrically heated rods whereas SSTF simulates reflood using steam only.
Among the specific test results noted was that LPCI injection into' the jet pump greatly aids core reflood (this design is used on the BWR/4 plants.
The FIST facility is an upgrade of the TLTA (Two Loop Test Apparatus) and is designed to eliminate many of the scaling compromizes suffered by TLTA.
The FIST Program is jointly funded by NRC/GE/EPRI. A limited Phase I test program has been negotiated which includes large-and small-break L.0CA tests plus some ATWS-type tests. Phase I tests will start in November 1982.
In response to questions from Dr. Chen, RES said there will be no simulation of radial power distribution or the power feedback phenomenon (FIST is a full height, single bundle facility). A Phase II test series is under negotiation and would include transient-type tests (Figure 2).
12.
Dr. Beckner provided an update on the topic of BWR core spray distribution based on questions raised by Dr. Okrent at a recent ACRS Meeting. Tests run in Japaneese test facilities have shown that central bundles receive low spray rates.
However, RES has very little specific information on these Japaneese tests.
US tests run in a 360 air-water facility simulating the BWR/6 spray system has shown that the spray water " piles-up" in the center of the core. This was also assumed to occur in the BWR 4/5 cores.
However, RES has no data to confirm this assumption. Based on their understanding of the Japaneese. tests, RES said the spray overlap probably doesn't occur in the BWR 4/5 plants.
Dr. Beckner discussed the significance of these findings. He said that based on Appendix K assumptions, the " required" minimum spray flow rate to cener bundles (N1.0 -?.5 gpm) probably isn't seen in EWR 4/5 plants.
l He also said NRR is evaluating the significance of this finding. He went 1
i l
ECCS Meeting December 2-3, 1981 on, to note however, that tests in TLTA, ROSA III in Japan, and SSTF demonstrate that several different modes of core cooling exist, and that the BWR/4 LPCI system rapidly refloods the core. GE has performed a sensitivity study that shows that an earlier reflood results from the core spray water " thrown away" in Appendix K calculations. his earlier reflood offsets a lack of spray cooling to the center bundles.
Dr. Beckner said NRR tentatively plans to discuss this item with the ACRS at the December meeting.
13.
Dr. Sullivan discussed RES's integral test plans after the LOFT facility is shutdown.
He said RES will rely on the Semiscale (PWR) and FIST (BWR) facilities and also plans closer cooperation with Germany and Japan in order to obtain test results from their integral facilities (PKL/ LOBI, and the planned Large Scale Test Facility). RES will attenpt to secure the overseas information by way of formal agreements. Dr. Dukler noted that the LSTF will be providing data that should address inportant questions on 2-phase flow phenomena and RES should try to get access to this information.
14.
Dr. W. Evans (Sandia) discussed the status of steam explosion research at Sandia. The objectives of the Sandia program is to estimate the probability and consequences of in-and ex-vessel (pressure vessel) explosions and describe the phenomena associated with non-explosive nolten fuel-coolant interactions in-and ex-vessel. We Sandia program consists of 3 main elements:
(1) small scale experiments; (2) intermediate-scale experiments (N 20 kg moltent mass); and (3) modeling and analysis work. Signifcant experimental results include:
(1) evidence of a low pressure explosion trigger " window". (Figure 3); (2) evidence of a mass ti.reshold (> 2 kg) needed for steam explosion (for iron-alumina tests);
(3) evidence of multiple explosions (also iron-alumina tests); and, (4) corium behavior is described as "quite explosive". We tests with nultiple explosions produced energy conversion ratios of > 3%. Sandia estimates the probacility of an in-vessel steam explosion that results in contain-t
~4 ment failure to be N 10 / reactor year for a PWR (Zion) and *10-3 for a BWR (Browns Ferry).
ECCS Meeting December 2-3, 1981 Dr. Schrock questioned the validity of the above probability estimates without first investigating the probability that the necessary initial conditions exist for the event.
Dr. Plesset said he believed that an in-vessel steam explosion is unlikely; he believes the vessel would most likey be dry at this point in the accident sequence.
15.
Dr. Sullivan discussed, in closed session, the status of the US/FRG negotiations on the UPTF (a part of the 2D/3D program), and the status of the funding for LOFT.
The Upper Plenum Test Facility project has been plagued with delay.
As a result, NRC's estimated costs have escalated.
RES has scheduled meetings with FRG representatives for late January to resolve differences in the test program scope and set a fixed program cost.
The status of LOFT funding for FY 82 is still uncertain. The "last" version of the FY 82 funding bill required funding of the full test matrix specified by the LOFT Special Review Group.
If RES must fund LOFT at this level in FY 82 (re$44 M) then there will be a severe impact on the FY 83 budget. Dr. Sullivan said he would provide ACRS the latest information on the LOFT program upon his return to Washington.
16.
Dr. A. Serkiz (NRC-NRR) discussed the status of the NRC effort on Unresolved Safety Issues A-1 (Water Hammer), and A-43 (Containment Emergency Sump Performance).
Concerning USI A-43, Dr. Serkiz said the original concern was that sump vortices resulted in unacceptable results (loss of pump function).
Later, concern arose about the effects of debris (insulation) blocking the pump intakes.
In response to a question from Mr. Ebersole, Dr. Serkiz agreed that NRC has to evaluate the effects on pumps from ingestion of debris.
The sump test program (including full scale tests) has shown that vortices are not a significant problem.
Debris assessment is also underway; preliminary results indicate that most operating plants do not have a problem in this regard.
Some of the older plants however
ECCS Meeting December 2-3, 1981 should receive further review.
A draft NUREG, detailing A-43 resolution.
l is due to be issued for public comment, in July 1982.
Addressing water hammer (USI A-1), Dr. Serkiz said there have been 150 events reported since 1969. These break down to 67 PWR events (includes 1
27 steam generator events) and 82 BWR events. The causes of the events are about evenly divided between plant operation effects and systems / design probl ems.
Significant conclusions noted by Dr. Serkiz include:
' Water hammer cannot be eliminated; NRC's goal is to minimize its occurrence and mitigate its effects.
' Steam generator water hammer (SGWH) appears to be under control via use of J-tubes on feedrings and limits on auxilliary feedwater rates.
Post-TMI auxiliary feedwater procedures are also expected to further reduce SGWH.
A NUREG addressing SGWH resolution is under preparation.
'BWR water hammers have been centered in a small number of plants (8 plants have had 65% of the events).
A large percentage of events (52%) have occurred in safety systems, usually the RHR system. NRR is focusing its efforts on the RHR system to try to reduce such events.
'HRR is planning to issue a NUREG on A-1 resolution in late November 1982.
17.
Prior to adjournment, Dr. Plesset solicited comments from Subcommittee Members and Consultants, in particular, on the RES code development and assessment programs.
He requested that the comment be provided prior to I
December 25, 1981.
- 18. The meeting was adjourned at 3:05 p.m., December 2,1981.
NOTE: Additional meeting details can be obtained from a transcript of this i
meeting available in the NRC Public Document Room,1717 H Street, N.W.,
Washington, D.C., or can be purchased from Alderson Reporting Company, Irc., 400 Virginia Avenue, S.W., Washington, D.C. 20024, (202) 554-2345.
- _ _ _ - _ _ _ _ _ _ - _ _ _ _ _