ML20203B086
| ML20203B086 | |
| Person / Time | |
|---|---|
| Site: | 05200003 |
| Issue date: | 11/04/1997 |
| From: | Joseph Sebrosky NRC (Affiliation Not Assigned) |
| To: | Liparulo N WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP. |
| References | |
| NUDOCS 9712120329 | |
| Download: ML20203B086 (9) | |
Text
. _ _ _ _ _ _ _ _
l November 4, 1997 Mr. Nicholas J. Liparulo, Manager l
Nuclear Safety and Regulatory Analysis Nuclear and Advanced Technology Division Westinghouse Electric Corporation P.O. Box 355 Pittsburgh, PA 15230
SUBJECT:
OPEN ITEMS ASSOCIATED WITH CHAPTER 19 0F THE AP600 SAFETY EVALVATION REPORT (SER)
Dear Mr. Liparulo The Ccntainment Systems and Severe Accident Branch has provided an SER for a portion of Chapter 19. However, the input to these sections contained some open items. These open items have been extracted from the SER and can been found in the enclosure to this letter.
Specifically, the enclosed open items concern in-vessel steam explosions.
You have requested that portions of the informi. tion submitted in the June 1992, application for design certification be exempt from mandatory public disclosure. While the staff has not completed its review of your request in accordance with the requirements of 10 CFR 2.790, that portion of the submitted information is being withheld from public disclosure pending the staff's final determination. The staff concludes that these follow on ques-tions do not contair those pcrtions of the information for which exemption is sought. However, the staff will withhold this letter from public disclosure for 33 calendar days from the date of this letter to allow Westinghouse the opportunity to verify the staff's conclusions.
If, after that time, you do not request that all or portions of the information in the enclosures be withheld from public disclosure in accordance with 10 CFR 2.790, this letter will be placed in the Nuclear Regulatory Commission Public Document Room, if you have any questions regarding this matter, you may contact me at (301) 415-1132.
Sincerely, original signed by:
Joseph M. Sebrosky, Project Manager Standardization Project Directorate Division of Reactor Program Management Office of Nuclear Reactor Regulation Docket No.52-003 g[9
Enclosure:
As stated cc w/ encl:
See next page
"-i DJSTRIBUTION: See next page DOCUMENT NAME: A:lVSE.RAI
- 1. r...w..
. w.
.m.
. w... w som c. cm wa u.u.ch nu.nces. r - Cm we.nachmenU.nckswo "N' = No copy l
OFFIC[
PM:PDST:DRPM lJCSB:DSSA
.C D:PDST:DRPM l l
l NAME JMSebrosky:sg & M MSnodderly pk'$ TQuay T W lDATE 11/'6 /97 V
11/3 /97 11/ 4-/97 acclCIAL RECORD COPY y yq yy y yy y
p l
9712120329 97;104 gDR ADOCK 0520 0 3
.,1 I,141,lill, il,
,1,11,
i Mr. Nicholas J. Liparulo Locket No.52-003 l
Westinghouse Electric Corporation AP600 cc: Mr. B. A. McIntyre Ms. Cindy L. Haag Advanced Plant Safety & Licensing Advanced Plant Safety & Licensing Westinghouse Electric Corporation Westinghouse Electric Corporation Energy Systems Business Unit Enorgy Systems Business Unit P.O. Box 355 Box 355 Pittsbu gh, PA 15230 Pittsburgh, PA 15230 Enclosure to be distributed to the following addressees after the result of the proprietary evaluation is received from Westinghouse:
Mr. Russ Bell Ms. Lynn Connor Senior Project Manager, Programs DOC-Search Associates Nuclear Energy Institute Post Office Box 34 1776 I Street, NW Cabin John, MD 20 ;8 Suite 300 Washington, DC 20006-3706 Mr. Robert H. Buchholz GE Nuclear Energy Dr. Craig D. Sawyer, Manager 175 Curtner Avenue, MC-781 Advanced Reactor Programs San Jose, CA 95125 GE Nucleat Energy 175 Curtner Avenue, MC-754 Mr. Sterling Franks San Jose, CA 95125 U.S. Department of C'iergy NE-50 Barton Z. Cowan, Esq.
19901 Germantown Road Eckert Seamans Cherin & Mellott Germantown, MD 20874 600 Grant Street 42nd Floor Pittsburgh, PA 15219 Mr. Charles Thompson, Nuclear Engineer AP600 Certification Mr. Frank A. Ross NE-50 U.S. Department of Energy, NE-42 19901 Germantown Road Office of LWR Safety and Technology Germantown, MD 20874 19901 Germantown Road Germantown, MD 20874 Mr. Ed Rodwell, Manager PWR Design Certification Electric Power Research Institute 3412 Hillview Avenue Palo Alto, CA 94303
O f
(
RISTRIBUTION:
Letter to Mr. Nicholas J. Lioarulo. Dated! Novomber 4. 1997
- Docket File
- Enclosure to be held fcr 30 days
- PUBLIC PDST R/F TQuay JSebrosky TKenyon BHuffman JNWilson DScaletti JMoore, 0-15 B1B WDean, 0-5 E23 ACRS (11)
JKudrick, 0-8 H7 RPalla, 0-8 H7 MSnodderly, 0-8 H7 SBasu, T-10 K8
I AP600 Open Items Associated With In-vessel Steam Explosions BACKGROUND:
In its review of the IVSE report, the staff his identified two general areas of the overall formulation which were not sufficiently robust and require additional analysis and justification. The two areas are: (1) quantification of melt relocation, and (2) bounding approach to premixing and explosion calculations. The first area of quantification of melt relocation includes partitioning of heat flux in the melt pool and investigation of the bounding melt release rates. Additionally, this area includes an open item oncerning the possibility of downward relocation of melt as identifiec by the v.)E peer reviewers, in addressing the first area of quantification of melt relocation, the authors claimed to have taken a " bounding" approach with regard to the location and the size of melt release, but the robustness of the supporting analysis was questioned by the staff as well as the DOE independent peer reviewers for several reasons.
The authors did not provide information on the up, down, and sideward partitioning of heat flux in the melt pool, and a demonstration that the relative partitioning remained constant as the melt pool evolved.
Inis information is necessary in determining the stability of crust surrounding the melt pool, in particular, the bottom crust or blochge above the lower core support plate. The information is also necessary in determining the timings of downward vs. sideways meltthrough.
Pending the receipt of this information from the authors and its satisfactory review, the item is considered to remain open. Concerning the bounding approach to melt release rates, the authors conceded that the release rate was an intangible parameter, but did not provide any analysis or argument to demonstrate why release rates higher than the ones considered would not be possible. A higher release rate may be produced by a larger failure area or from failure at inultiple locations. The latter example was pointed out by the DOE peer reviewers.
The staff recom-mends that additional analysis using higher release rates is needed to demonstrate the robustness of the overall conclusion.
Specifically, the staff recommends that the authors consider a high enough release r;te that would breach the lower head and demonstrate, based on arguments pertaining to AP600 geometry and other design features, that such a rate is " physically unreason-abl e. "
DOE sponsored a peer review of the IVSE report. Based on its % view of the DOE neer review coments, the staff supports the position of t*e reviewers regarding the possibility of downward relocation. The authors d'd not provide any analysis to demonstrate that a downward relocation of salt due to the failure of lower blockage from a primary steam explosion (i.e., an explosion caused by the interaction of a small melt mass with water producing energetics sufficiently '.arge to cause blockage failure, promote downward relocation, further premixing, and subsequent secondary explosions, hence the distinction) was not possible. Absent such demonstration, the authors should add a new splinter (downward relocation) in the R0AAM framework, and invcstigate the effect of melt release location and rate from this splinter on the explosion
-load. This constitutes an open item.
Enclosun
. The second area of premixing and explos'on calculations encompesses two other open items from the staff's review, i.e., sensitivity analysis involving melt length scale, and investigation of one case of explos'an calculation that produced a very high peak pressure (~ 1000 MPa).
In addresting this second area, the authors claimed that they took a bounding approach to premixing and explosion calculations. Justification fcr this claim was based largely on the
" fitness-for-purpose" verification of the PM-A' PHA and the ESPROSE.m codes and the analytical models therein, as well as the use of the verified codes to assess in-vessel steam explosions in AP600 in a " conservative" manner.
However, sensitivity studies involving melt length scale and investigation of one case of explosion calculation that prnduced a very high peak pressure were not addressed by the authors. Therefore, the staff recommends sensitivity studies involving me'It length scale in fragmentation models be performed so that the conclusions in the IVSE report can be generalized for a reasonable range of values of these parameters. The recommended studies would 1so provide additional verification of fitness for purpose of the codes.
Doen Items:
In sumary, the staff's review identified five open items and recommended additional analysis and discussion to address these items.
The items and the coiresponding recommendations are:
720.424F Partitioning of heat flux in the melt pool; information on parti-tioning as well as demor tration that the partitioning remain constant 720.425F Bounding approach to melt release rates; consideration of higher release rates and sensitivity studies 720.426F Splinter scenario involving downward melt relocation; demonstra-tion as to why the scenarie is " physically unreasonable" or consideration of the scenario within the R0AAM framework 720.427F Choice of melt length scale in premixing analysis; sensitivity analysis involving melt length scale 720.428F High peak pressure in one explosion calculation; investigation of the case The open items and the recommendations to address these items are documented below.
Clarification of Open Iten,s 720.424F. 720.425F. and 720.426F:
The authors claimed to have taken a " bounding" approach with regard to the location of melt release. Specifically, the authors concluded that in an AP600 geometry, the melt release would be through a sideways breach of the crust surrounding the melt pool followed by failure of the reflector and the core barrel, and the melt would then flow out of the pool into the lower plenum water. The location was predicated upon a melt relocation scenario
- that would lead to a stable blockage (crust) formation at the lowest region of the active fuel (i.e., on top of the lower core support plate) thus making the downward relocation path unavailable. Calculations were provided for the timing of core barrel and reflector meltthrough as well as the timing of core plate dryout, and it was shown that the sideways failure would occur prior to the core plate dryout. Note that these calculations are dependent on the physical properties of crust (e.g., thermal conductivity, porosity), its growth rate, and heat flux distribution in the melt pool (i.e., up, down, and side).
The AP600 design has a relatively flat radial power profile and a high aspect ratio. Also, the core plate is much thicker in the AP600 design (about twice that of operating reactors) so that it acts as a substantial heat sink. These design features make the sideways meltthrough more likely to precede the core plate meltthrough. However, given the uncertainties in the current under-standing of late phase melt progression, it is difficult to completely rule out, as the authors did, the downward relocation of melt. The staff has commented on the uncertainties in crust properties, heat flux distribution, etc., with regard to their implication on the likelihood of downward reloca-tion, and recommendtJ that sensitivity studies involving these parameters be performed.
Several DOE peer reviewers also commented on this issue.
The authors, in response to these coments, performed additional sensitivity studies involving crust porosity and thermal conductivity and concluded that the crust was indeed robust in all cas;s and that the downward heat flux (relevant to core plate meltthrough timing) was still within the range estimated previously. However, the authors did not provide any response to an earlier staff comment on the relative partitioning of up, down, and sideways heat flux.
Instead the authors performed calculations to show that the core plate provided a substantial heat sink and a further margin to the " race" between the sideways failure and the possible bottom failure.
This confirms their earlier assertion that the sideways failure is more likely, but provides no additional argument to rule out the bottom failure altogether.
Addition-ally, the staff notes that the authors summarily dismissed the notion of crust (blockage) failure from a primary explosion, but did not provide any analysis of crust structural response under an explosion load to support their posi-tion.
Pending this, the staff believes that the authors should consider a splinter scenario involving downward relocation.
Some DOE peer reviewers commented on the possibility of failure at multiple locations and on the possibility of bottom crust (blockage) failure from a primary explosion.
In the former case, the release rate is likely to be higher than those considered in the report.
In the latter case, even though the primary release rate may be small, the secondary release rate (i.e., melt release upon core plate failure) is likely to be much higher, possibly leading to stronger secondary explosions that would challenge the 'ower head integrity.
The authors rejected the notion of failure at multiple locations based on an argument, i.e., "that once a relocation begins, by local meltthrough, the melt height drops and there is less opportunity of other melt breakthroughs azimuthally.
Rather, we think the path, once opened will continue to enlarge
4 and melt will be released basically from the same location." The authors did not explain why azimuthal breaks in other locations are not possible as the melt height drops.
Concerning the possibility of blockage failure from a primary explosion resulting in stronger secondary explosions from a much higher secondary release rate, the authors summarily dismissed the possibility of blockage failure as noted previously, and claimed that the subsequent process of secondary explosions would be physically unret.sonable.
Calcula-tions were not provided to support the claim. Therefore, the staff considers this to be an open item related to tht possibility of a new splinter scenario, as discussed previously, involving downward relocation.
For the sideways meltthrough, melt release rates considered (100, 200, and 400 kg/s) are comparable to the TMI-2 scenario (- 160 kg/s).
These rates were calculated based on an exit velocity of 1 m/s undtr gravity draining and exit hole sizes of 10 cm x 10 cm, 10 cm x 20 cm, and in cm x 40 cm, respectively.
The authors claimed these numbers formed a reasonable range to bound the release rates, but provided no eviduce to support the claim.
For reference, the hole size in THI-2 was much larger (60 cm x 150 cm).
The staff noted that the peak impulse loads calculated in the report were close, at least in one case, to the fragility limit and thus recommended that parametric calculations with higher release rates (corresponding to larger hole sizes) be made to determine the release rate that would breach the lowar head, given all other input conditions the same as in the base case.
The authors conceded that the release rate was an iatangible parameter that was difficult to bound, but did not perfom parametric studies with higher release rates.
Instead, the authors claimed that the lack of r.onservatism, if any, in bounding the release rates is compensated by the conservative treat-ment of other intangible parameters, namely, trigger timinc and melt breakup parameter.
Further, the authors claimed that there was no basis to assume that the impulse would be proportional to the release rate citing that they found no discernible dependence of impulse loads to the release rates in the range of 200 kg/s to 400 kg/s, and thereby implying that the higher release rates would not c.ecessarily mean higher impulse loads and consequent lower head failure.
Besides, the authors pointed out that sufficient safety margin was already provided by the non-intersecting load and fragility curves. The authors did not demonstrate that at release rates higher than 400 kg/s, a discernible dependence would not be found.
the staff believes that at a sufficiently high release rate, the safety margin the authors alluded to earlier may be substantially reduced. Of course, a higher release rate means a larger failure hole size assuming that the flow velocity remains constant.
As noted previously, the TMI-2 had a much larger hole through which the melt was released. The authors agreed to address why the hole size seen at THI-2 (60 cm x 150 cm) was not considered for the AP600 analysis.
Stating that "the geometry in AP600 is so different from that in THI-2 as to forbid any attempt at relating respective failure areas," the authors basically eliminated further consideration of a larger hole size. The staff believes that ore cannot summarily preclude, without a firm basis, consideration of larger failure areas and correspondingly, release rates higher than 400 kg/s.
Therefore, the staff recommends that Westinghouse consider a high enough release rate that would breach the lower head and demonstrate, based on arguments pertaining to AP600 geometry and other design features, that such a l
e
) -
rate is " physically unreasonable." For example, the authors could investigate if a release rate of 1600 kg/s (based on a 40 cm x 40 cm openk.g and repre-senting a rate that is an order of magnitude higher than that of THI-2) would breach the vessel and then argue its unreasonableness.
Clarification of Open Item 720.427F,.
The approach to quantification of premixtures, taken by the authors, involved specifications of a single value (20 mm) of the initial melt length scale, a range of values (from 10 to very large or no breakup, denoted as nb) of the breakup parameter, B, and formulation of a causal relation (based on PM-ALPHA) for the quantity of fuel mass in a prerrixture. The basis for the choice of a 20 mm melt length scale for AP600 was not stated. 00E/ID-10504 Report,
" Premixing of Steam Explosions: PM-ALPHA Verification Studies," chose differ-ent length scales for MIXA06 (6 mm) and FARO L-14 (40 mm) assessment.
It appears (see DOE /ID-10502) that one can virtually construct any number of combinations of the melt length scale and the breakup parameter, B, which will produce code results that are comparable to experimental data.
The staff has recommended additional PM-ALPHA calculations with a range of values of the initial melt length scale.
Some DOE peer reviewers made comments as well on the melt length scale and more generally, on fragmentation modeling in PM-ALPHA. Moreover, they raised the question of maturity of the code, i.e., that the code was not considered ft ily developed and verified with regard to all its modelling capabilities.
With regard to the issue of melt length scale, the authors explained that "using a larger initial size would be even more limiting," presumably for the strength of an explosion. Besides, the authors claimed that a larger initial size would break up to smaller sizes during the FCI and thus, "we would end up with the same kind of premixtures as in the 20 mr case considered."
Intu-itively, this is a valid argument provided one can ignore the possible effect of initial breakup from the larger size to say, the 20 mm size, on the subsequent melt-coolant interactions. The authors should demonstrate that the effect is indeed insignificant by performing additional sensitivity studies with a larger initial melt length scale. The staff considers this to be an open item.
Clarification of Open Item 720.428F:
The approach to the quantification of exf;osion loads, taken by the authors, was based on the a::sumotion that a given premixture was always triggerable (i.e., the probability of triggering is unity) and, as such, involved specifi-cation of a trigger of sufficient strength (~ 100 bar) to initiate explosions.
Further, the approach involved consideration of a range of values of trigger timing (0.05 s to 0.19 s for B = 10, 0.04 s to 0.155 s for # = 20, 0.05 s to 1.0 s for B = nb i.e., no breakup) and formulation of a causal relation (based on ESPROSE m) for the impulse loading from steam explosions. A total of 24 loading calculations were performed initially. Of these, 7 cases producet a peak pressure in the range between 200 MPa and 1000 MPa (an indication of the degree of severity). However, the calculated impulse. loads in these seven cases were between 90 kPa-s and 190 kPa-s, i.e., below the fragility limit of the lower head material.
' The staff questioned the conclusion that " peak imnulses do not depend strongly on the size of the mixing zone." The authors agreed that one particular calculation (release ratt of 400 kg/s, # - 20, trigger timing of 0.12 s) seemed to be the outlier to the above conclusion. The authors further acknowledged their concern about this case since the peak pressure was about 1000 MPa, and stated that they would reinvestigate this calculation. Addi-tional calculations supporting the original conclusion were provided by the authors in an addendum to Chapter 6 of DOE /ID-10541. The staff notes that the additional data enhances the basis of the original conclusion.
However, the staff could not find the disposition of the reinvestigation of the 1000 MPa case in this addendum as claimed by the authors. Therefore, this is still considered an open item.
720.429F Although the report " Lower Head Integrity Under In-Vessel Steam Explosion Loads," DOE /ID-10541 is referenced in PRA Chapter 39 the following companion reports are not referenced in the PRA: " Pre-mixing of Steam Explosions:
PM-ALPHA Verification Studies,"
DOE /ID-10504, and " Propagation of Steam Explosions:
ESPROSE.m Verification Studies," DOE /ID-10503. The staff believes these reports should also be listed as a reference to PRA Chapter 39.
_..____.-___.______.__ _____ _ _ __ - _ m