ML20236P138
| ML20236P138 | |
| Person / Time | |
|---|---|
| Issue date: | 11/10/1987 |
| From: | Morris B NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Gavigan F ENERGY, DEPT. OF |
| References | |
| PROJECT-672A NUDOCS 8711170163 | |
| Download: ML20236P138 (11) | |
Text
l/
i
~
I p uo
~
. UNITED STATES 8.
"o NUCLEAR REGULATORY COMMISSION y
- a WASHINGTON, D. C. 20666 o
NOV 'l 01987
'Mr. Francis X. Gavigan, Director-Office'of' Advanced Reactor Programs Office,of Nuclear Energy U.S. Department'of Energy
(
Washington, D.C. 20545
.c
Dear Mr. Gavigan:
On'0ctober 16, 1987, members of the NRC staff and its contractors from ORNL and BNL met.with representatives of DOE and its contractors to discuss selected areas of our review of the Modular HTGR, Project 672. These areas' included reactor physics, the reactor cavity cooling system (RCCS) and the means for 1
L developing site suitability source terms'(SSSTs).- In addition, an overview was-presented of the' status and approach being considered to address'the-key issues of standardization and design certification, accident = selection, containment, and emergency planning. lThis presentation included the. identification of seven boundingAventsJntended-to mechanistically _def_ine_acc.idents_and_ bound _uncer-
' tatiifies for selecting'SSSTs.
Information regarding the plant response to wthesCennts-irrequested.. Our contractors from ORNL-and BNL presented the results of independent calculations pertaining to core cooling with the RCCS operational and with the RCCS having failed. These calculations explored the sensitivity to various parameters including effects of insulation in the upper plenum, asymmetrical fuel loadings, decay heat rate and graphite thermal conductivity.
The following items are enclosed with this letter: Enclosure 1, List of Attendees; Enclosure 2, Agenda;' Enclosure 3, Action Items and Clarifications (including requests for additional'information).
DOE should respond to Enclosure 3 by November 27, 1987. An additional meeting will be.needed with DOE prior to the staff's finalizing its SER, primarily on the topics of fuel, PRA findings, site suitability source terms, and
'certain miscellaneous topics including fire protection and primary system pressure relief. We will contact you separately regarding scheduling this meeting.
j If you have any questions, please do not hesitate to contact Dr. Peter l
Williams, Project Manager for Project 672.
Sincerely,
'm 8711170163 871110 I
l'&. lNlYWO PDR PROJ 672 A PDR Bill M. Morris, Director Division of Regulatory Applications
/,
'h
Enclosures:
- 1. List of Meeting Attendees
- 2. Agenda 3.' Action Items and Clarifications J
NOV 101987 DISTRIBUTION RES Circ Chron
'ARGIB '4/F.
E. Beckjord
.T.:Speis-B. Morris Z. Rosztoczy.
T. King J. N. Wilson C. Allen.
R. Landry P. Williams.
J. Flack
.i B. Hardin
- 0. Gormley R. Baer N. Anderson F. Cherny S. Shaukat R. Johnson Di Thatcher J. Hulman J. Glynn
~J. O'Brien L. Soffer J. Read D..Cleary A. Murphy G. Arndt
-R. Kirkwood
'E. Podolak R. Erickson B. Mendleschn H. VanderMolen E. Chelliah 2~.'Beltracchi F. Congel
- 0. Lynch D. Matthews R. Senseney D McPhearson F. Coffman S. Ball, ORNL P. Kroeger, BNL G. VanTuyle,.BNL R. Ireland, Reg. IV M. El-Zeftawy,.ACRS/H-1026 PDR - Project 672 frojecti Fils !672 f(Central; Files)
.____---______-_________--_--_-____--_____-___-____-____-_x
- \\
' ATTENDEES NRC/ DOE. Meeting on MHTGR October _16, 1987-1 L,
LP.M. Williams RES/ARGIB.
(301).492-9613 1
Jerry N.. Wilson NRC/RES/ARGIB-(301) 492-4727-i
' John.H. Flack-NRC/RES/ARGIB-(301) 443-7767'
_Syd Ball.
ORNL.
(615):574-0415 q
- Greg VanTuyle.
BNL (516) 282-7960 i
(FTS) 666-7960 David Moses-ORNL/NRC Program (615) 574-6103 Uri Gat
_0RNL-(615) 574-0560
~ Peter G. Kroeger-BNL (516) 282-2610-Ti (FTS) 666-2610 Moni'Dey_
NRC/RES/ARGIB (301) 492-8100 Jim >Glynn_
NRC/RES (301) 443-7360.
Gunter Arndt NRC/RES (301) 443-7893 Brad Hardin:
'NRC/RES/ARGIB (301) 492-8986 Paul Kasten ORNL (615) 574-6093 Arkal Shenoy GA' (619) 455-2552 j
. John Cunliffe BECHTEL (415) 768-2227 Tony Neylan-GA (619) 455-2580 Fred Silady
'GA (619) 455-4320' Donald V. Graf
~MHTGR-PDC 0 (619)-455-4294 Andrew C. Millunzi D0D-HQ (301) 353-3405 J. M. Kendall GCRA (619) 455-9500 Lloyd P.. Walker SWEC (619) 455-9500 John O'Brien RES (301) 443-7854
\\
i
g.
g f.
- ?
y? q
. c.
o AGENDA NRC/ DOE Meeting'on MHTGR PSID Review-(Project 672) s
' October.16,11987 Room P-ll8
Purpose:
'To discuss review status and' key issues /open items resulting from MHTGR-PSID review.
Agenda:
a.
1 i.v 9:00.
Introduction and status of review T. King 9:30~
Reviewfof specific. items for PSID Chapters:
f
. Chapter 4-(comments'4-41 thru 4-43 attached)
-P. Williams /
D. Moses.
1
- Chapter 5 (comments 4-44 thru 5-41 attached).
.P.' Williams /'
J. O'Brien/
p
-S. Ball 12:00.L1:00 Lunch
'1:00 Chapter 15:(comments 15-4 thru 15-7 attached)
P. Williams /
P. Kroeger 3:00 Other areas to be addressed at a future meeting P. Williams /-
T. King 3:30 Overview of NRC criteria being developed for T. King addressing key issues:
- adequacy of containment
- treatment of severe accidents
- emergency planning
- source term 4:00 Ajourn l
1
4 '
ACTION ITEMS AND CLARIFICATIONS NRC/ DOE MEETING ON MHTGR OCTOBER 16, 1987 4-41 We have reviewed the responses to our comments on reactor physics concerns presented in PSID Vol. 5 (as categorized under "neutronics" on page R 4-iii) and in DOE-HTGR-87-085, "MHTGR Core Nuclear Uncertainties". We find the information presented conditionally acceptable in support of the MHTGR reactor conceptual design for use in the transient and accident analyses at the conceptual design stage and for illustrating the feasibility of the MHTGR passive shutdown characteristics. However, it is our position that this acceptability can not be extended to more advanced stages of design without substantial improvement of the data base. Data base improvements are needed in recognition of modern standards of accuracy in experimental techniques, the uniqueness of the inner reflector geometry, scarcity of experimental work with LEU fuel, and the growth of the plutonium fraction with burnup.
In DOE's Response 4-15, DOE "comitted to validate MHTGR nuclear physics codes consistent with NRC regulations and industrial standards relevant to the MHTGR as the design development proceeds. This long-term commitment will take considerable time to complete and will be developed during preliminary design."
In view of our finding that the reactor physics data base will need improvement, DOE agreed to the inclusion of reactor physics in the Regulatory Technology Development Plan (RTDP).
DOE will describe l
what reactor physics data could become available from cooperative programs with West Germany and Japan and whether or not DOE believes that such programs will be sufficient in themselves to provide the necessary improvements in the reactor physics data base. The reactor physics plan should be der.cribed with respect to background, objectives, approaches and acceptance criteria as currently available and should be expanded later as new information becomes available.
7
)
4-42 In studies of conduction cooldown with and without RCCS availability, performance results in terms of peak fuel and vessel temperatures were seen to be highly sensitive to long term values of decay heat.
In Response 4-18 it was stated that more recent and better qualified i
data were being evaluated in comparison to the " original" PSID data.
]
DOE will describe progress being made in this area and indicate the
{
degree and effects of uncertainties in more recent data.
In this description, DOE should indicate the approaches being considered, including collection and analysis of existir.g data and the plans for experimental validation for the decay heat values.
a 4-43 In Response 4-25 it was stated that if flux mapping detectors should fail, plant operation could continue and that no ISI was planned for these detectors.
It is our understanding that these detectors monitor the core for long term burnup effects and assure that undesirable fuel temperatures do not occur in the lower core regions.
Therefore, it would appear that these detectors should be considered "important to safety" (with the intent that they be built to quality standards, receive periodic testing and calibration and have appropriate Technical Specification requirements to monitor flux distributions and to assure their availability and performance) unless DOE can show that this function is not required or is performed by other neans.
The details of the safety standards that need to be met could be developed during preliminary design.
In a like manner it j
i would appear the same treatment should be given the startup monitors, discussed in Response 4-26, since they perform an operational safety function during refueling.
4-44 DOE will discuss the data base referred to in General Atomic's " Graphite Handbook" supporting the values used for thermal conductivities in the core and reflector graphites.
This discussion should include the thermal annealing effects on the conductivities of irradiated graphite. The data base should be presented in sufficient detail that NRC can make a preliminary assessment of the quality of the data base, including whether or not additional measurements may be necessary.._____ ____-______ _ _ -
4 <
t 5-38 In 'our. independent calculations of RCCS performance under pressurized conduction cooldown we encountered two major concerns. The first is that DOE does not model downward by-pass flows in the region external to the core, but rather considers that all convection cells are
- i confined to the core region itself. The second is that the upper plenum thermal protection structure may contain sufficient insulation to cause more heat than intended to be transferred through the vessel
. side wall from the vessel head, resulting in a potentially unac-ceptable temperature peaking at wall. locations. Discuss these concerns and their affects on your conclusions regarding vessel integrity and RCCS performance.
5-39 Discuss'if vessel integrity could be better guaranteed and' uncertain-ties reduced if the system is depressurized before the vessel reaches elevated temperatures.
Identify the means available to depressurize the primarily system, if this becomes desirable.
5-40 Provide the results of calculations or detailed discussions to illustrate the sensitivity of peak and average fuel temperature and peak and average vessel temperature (under RCCS only heat' removal conditions) to the following uncertainties:
(1) graphite thermal conductivity, including radiation annealing effects, (2) surface contact and gap resistances between adjacent graphite islocks and between graphite blocks and the inner surface of the core barrel, (3) effects of convection flows in the core and reflector, (4) emissivities on the core barrel inner and outer surface, (5) effects of helium
.l convection, seismic keys and helium ducts on the heat transfer across i
the core barrel, (6) the emissivity on the inner and outer surfaces of the reactor vessel, (7) effects of convention flows exterior to the j
reactor vessel, (8) emissivity of the RCCS panel surface, (9) geometrical and asymmetrical effects of fuc1, reactor internals, and RCCS local structure, (10) influence of the upper plenum thermal protection structure and other reactor internals that could cause shifting of thermal gradients or otherwise cause temperature or stress peaks, and (11) any other modeling practices or assumptions that could effect RCCS performance.
In evaluating the above, parameters that can be shown by sinple analysis or a confirmed data base to have a trivial
' _________________-______m
n.n L
o v
n k
m effect can be omitted. The overall goal of this study is to demonstrate
.that the.RCCS has the potential of meeting its. performance requirements with acceptable margins and to identify any; potential area that should 1
be included in the Regulatory Technology Development Plan to reduce
,l uncertainties.
5.. A very ' low value f or,the-seismic failure probability of the RCCS.is giveniin the PRA (Vol. 2).
From our previous experience with:
I structural _ support systems-and the subsurface location of most of the-RCCS components,'we believe that such a low failure probability has the potential of being achieved.
Conservative' design rules (using NRC-
' approved structural codes with input determined from Stan'ard Review d
Plan Sections 2.5, 3.7.1, 3.7.2, 3.8.1, 3.8.2, and 3.8.3) would have-to be used. Also, inservice inspection would have to be very thorough and'perhaps some disassembly of components would be required..The current status of;the RCCS design activities regarding seismic.
l integrity, particularly in terms of fragility assessments, should be; discussed in light of the above.
5-42 During an event.'in which only the RCCS is used for decay heat removal, the reactor vessel could be exposed to temperatures that possibly approach or even' exceed extended code allowables. Therefore, for reasons of accident progression monitoring, decisions regarding depressurization and assessing the reuse capability of the vessel, it appears desirable to have instrumentation capable of monitoring vessel temperature history during a conduction cooldown event. This would avoid putting the utility and the NRC into a position of having to
. infer whether continued operation is safe.
It is strongly suggested that such instrumentation be provided.
15-4 We have reviewed material provided in Appendix G of the PRA study that l
addresses complete failure of the RCCS. For this severe event, DOE should provide a sensitivity analysis similar to that requested in Comment 5-40 for the RCCS. This study should be augmented to include -_-_
such modeling factors as the thermal resistance of the RCCS panel itself, convection currents in front and behind the panel, emissivities of both sides of the panel and the cavity surface, methods for modeling cavity and vessel geometry in two and three dimensions, time dependent consequences of failure to depressurize the vessel, and soil conditions external to the reactor cavity.
{
l 1
15-5 The Appendix G calculations for RCCS failure are made on the basis that there are no structural failures other than the RCCS or the cross duct. Other failures that coincide or follow these failures that might result in further elevation of fuel or vessel temperatures should be discussed qualitatively and quantitatively. These might include local or gross vessel failures from over temperature and changes in geometry which could affect heat transfer within and exterior to the vessel, as might be caused by earthquake, after shock, and additional over heating. Furthermore, concrete failure, particularly above the vessel, either from earthquake or over heating, thould be considered from the standpoint of causing additional structural failures, combustible gas generation, or significant changes in the heat transport mechanisms in the reactor cavity. Also, DOE should indicate whether it plans to commit to design the reactor cavity, vessel support, and any other critical structural items to the same integrity standards as the RCCS itself.
15-6 In Appendix G, consequences of the various severe events explored are presented in reference to the silicon carbide coating degradation temperature of 2000 C.
Explain why this parameter was chosen, in lieu of the 1600 C value used elsewhere in the PSID, and describe the safety consequences, including time dependent off-site doses, if this i
}
limit is exceeded.
15-7 The staff has concluded that, for plant designs with long response times and the capability to withstand many low probability events, it might be possible to develop mechanistic bases for site suitability source terms (SSSTs) rather than following the customary approach of _-
o h.
hE,
. postulating non-mechanistic. source terms.- We anticipate that such SSSTs would be developed by best estimate calculations of releases in l
terms of a time-history that reflects the unique features of the fuel and the MHTGR design. This time-history approach should also define the extent of accidents which need to-be considered and/or certain-end-of-sequence reactor states at times when it is clearly evident' that additional fuel failure and fission product release is not-credible. Therefore, development of mechanistic SSSTs requires that all credible event. initiators and sequences are identified or bounded-and that there is. margin to other less likely scenarios that could lead' R
to releases larger than those from the accident envelop considered.
The staff believes that to achieve this end the approach initiated in
. Appendix G to the PRA document should'be: continued and expanded in accordance with the guidelines given below.
The staff suggests that-DOE perform a best estimate analysis of the releases from the following set of Bounding Event Sequences (BESS).
These BESS are intended to be a bounding spectrum of initiators and sequences to account for uncertainties in design, failure 4
probabilities, R & D program results, reduced operating experience and to provide conservatism to account for the shift in emphasis
.j from accident mitigation to accident protection / prevention and to limit the reliance on non-safety grade equipment. They address the following key event categories: reactivity insertion, heat removal, loss of coolant, chemical attack, low probability seismic and other external events (flood, fire, wind, aircraft ). The time-history of off-site doses I
l should be evaluated over the full course of the event presented for the j
2 hour2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> and maximum dose (including the time at which it occurs) at the site boundary and the 30 day dose at the LPZ.
- ________________________________n
p
~
y^
1
' BES~ :1 Inadvertent withdrawal of al1~ control' rods-w/o scram for 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />
-(one-module)
W/ forced. cooling-
.W/RCCS cooling'only (pressurized +'depressurized)
- BES 36 hour4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> station blackout (all modules) w/ failure to scram in y
one module.
pressuH zed + depressurized BES --3 Loss of forced' cooling plus RCCS for 36. hours (one module):
with scram (pressurized ~+ depressurized) without. scram (pressurized + depressurized)
BES -'4.~2 S.G. tube rupture (all tubes) with failure to isolate or dump S.G. + failure'to scram (one' module):
with forced circulation cooling'(depressurized) w/o forced' circulation cooling (depressurized)
BES - 5 Large He leak-(one module):
Double ended guillotine break of cross duct with' failure
'to scram (assume RCCS failed)
BES - 6 External events consistent with those imposed on LWRs.
' Refinement or adjustment of these events may be' desirable after further discussions with DOE and should reflect DOE's progress in assessing low probability accident potentials.
The' staff. developed the above list from four separate considerations, which are:,(1) events that have occurred or nearly occurred over the~ entire history of nuclear reactor technology, (2) events that take into account the MHTGR's
- design emphasis on accident prevention as opposed to mitigation, thus
' significantly reducing the effects of mitigation uncertainties in the analysis, (3) events that test the MHTGR's passive and inherent safety features, as opposed to complications developing from numerous active safety l
systems and erroneous human actions and (4) events that assume worst case
.l
' failure of non-safety grade systems. -.
- Y