ML19306H146
| ML19306H146 | |
| Person / Time | |
|---|---|
| Issue date: | 03/09/1988 |
| From: | Morris B NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Nulton D ENERGY, DEPT. OF |
| References | |
| PROJECT-672A NUDOCS 8803140427 | |
| Download: ML19306H146 (17) | |
Text
... _.. _.
\\;
i
.]
l t
\\
ci
>l
(
l!
MAR 0 91988 sj e
/.
f
]
Mr. David Nulton, Deputy Director l'
[
j ijj Office of Advanced Reactor Programs c
1 Office of Nuclear Energy
('
[
U.S. Department of Energy
~
Washington, D.C.
20545 j
a
Dear Mr. Nulton:
j i
i A mesting to discuss fuel'and other key' issues remaining in our MHTGR+ review' H
1 (Project 672) is scheduled to be held beginning at 9. am in.NL/S, Room 014 on j~
fire, augmentation of Amendment 8 responses, reactor vessel perfomance, March 17. 1988 for. fuel issues and on' March 18, 1988 for the issues of graphite mitigative. features of the reactor building, and comparative safety enhancements.
This is anticipated to be'the final.NRC/ DOE meeting on MHTGR issues prior to
.i
)
issuance ~of our Safety Evaluation Report. We. understand that;the agenda for s
each day will be proposed by DOE on the basis of the enclosed list of draft.
comments and requests for additional information.- The final agenda will.be
'j l
issued as a supplement to this notice as soon as it is available.
If you have any questions, please contact Pete Williams, the Project Manager on-492-3736.
,.y Sincerely, U
j l
4 I
Bill M. Morris, Director i
Division of Regulatory Applications j
Office of Nuclear Regulatory Research
Enclosure:
4 j
Comments and Requests for 4 -
Additional Information.
).
See attached list for distribution i
- See previous concurrences
/hl 0FFC:ARGIB/DRA*
- ARGIB/DRA*:ARG 3/ /A bu /b1
,h NAME: WILLIAMS:sy : WILSON
- TP g.mcatir DATE:3/4/88
- 3/4/88 3/1/88 3/ N8 S.
L OFFICIAL RECORD COPY L
R gy==;
(k
)' '
j t
J
. w
LikR c g 1ges David Nulton DIL;RIBUTI9N RES Ciec
~
Chron ARGIB Rfi E. Beck.iord T. Speis i
<3. Morris Z. Rosztoczy T. King J. N. Wilson B. Hardin 1
R. Landry P. Williams M. Dey J. Flack
- 0. Gormley
~
R. Baer N. Anderson F. Cherny.
S. Shaukat
- k. Johnson D. Thatcher J. Fulman J. Glynn J. O'Briet L. Soffer
~
J. Read D. Cleary A. Murphy G. Arndt R. Kirkwcod E. Podolat, i -
R. Erickson B. Mendlesohn H. VanderMolen E. Chellich r
L. Beltracchi F. Congel C. Hinson O. Lynch D. Matthews R. Senseney M. Spangler F. Coffman S. Ball, ORNL P. Krceger, BNL G. VanTuyle3 BNL R. Ireland, Reg. D:
M. El-Zeftawy, ACR$/H-1026 PDR - Project 672 grddect\\fifej672$ { ChdrS('} Files)
+
-y
~
- v
7
.;1-j o -
- 1
._m_,
ycpg I s'
,.(
s' a(Y j i. A).
gye l
p Enclosure j
,s u
\\
e 1-5 g
l y
- a'
\\ ; V Draft Comments and Requests for Additianal Information
]
d 4-45 We have included in our review of-feel design 3nd phrformance a j
i document referenced in the PSID, "US/FRG Accident Condition Fuel Performance Models," HTGR-85-107, December 1985. As a result of this
)
4 review, we have the following coments ana requests' for additional information pertaining to this report. WenotetJStDOEconsiders
)
this' report the most.recent reference. supporting-the MHTGR fuel.but i
it does not provide data on the UC0 refccene i fuel.
I
(
/
The report does deal with fuel that is signifNantly more advanced,,
4 afA
).,
and that is considered an improvement over ol'dertfuel; but notes that 7
~
s.
q j
"The (fuel failure) model needs. further evaluatiop with fuels of current design, particularly in' the. temperature. range from 1200*C.to 1
l 1800 C where the amount of relevant testing datai s. limited.". As i
i m
t described in more detail below and in later comments. it is, the -
i i
staff's general concern that this need for additional testing is not -
adequately reflected in the RTDP. Therefore, it 'is' t!he purpoaa of '
f j
the following questions to identify those areas that may nee 6g,/
f.
1 additional R &.D.
It is not expected that technical solutions to l
each of these items be available at the conceptual design stage but !
rather it is expected that a plan or commitment to obtain a technical
]
solution can be agreed upon.
g i
e A.
While the range of 1200*C to 1800"C is the most important
[
temperature range for the acceptance of-the MHTGR, as this f
covers the temperature range which'the fuel wpuld exper'ance during anticipated cmnt~ and acciderit scenarios, cdeqapie data r
up to the thermal decomposion range of about 2000 C is Elsr,-
o_
4 t
needed to gain knowledge of failure mechanisms and margins,
]
including wet and air-oxidizing conditions.
1 4
4 f
..i y s
J,b m #1 d
(
{
r o
4
.j
i Enclosure B.
There are two basi: assumptions underlying the data of Benz and others in the report:
(1) Cs release is a direct indicator of j
sic failure; and (2) the delay in the release of Kr after sic failure is due to a diffusive transport mechanism in the remaining intact PyC layer.
Both assumptidns'should be explained and their significance made clear.
C.
All experiments cited for high-temperature fcilure appear to be out-of-core simulated heating tests.
It should be explained and justi.fiedwhytheseexperimentshavevalidity'Nrthein-pile I
situation. For example, external heating has the reverse temperature gradient from in-pile heating, anc any synergistic effects due to radiation and overtemperature would not be
{
q'
)
present. Also, short-lived gaseous products have long since decayed, and the internal pressure may be significantly j
dif ferent from the pressure during operation. The latter effect may be of particular significance for the diffusion of gases.
1 D.
The crityria used for experimental data acceptance, correction o~. rejectic(should be described with respect to their ability to detect both systematic faGEfes and obscure failure mechanisns of safety signift:arte. This observation is important.sincefuelintegrityisclainhdtolevelsof10-6 with admitted need ftr further validatlou.
It is exactly the rare occurrences th d are being sought.
Furthermore, " careful consideration" is insufficient justificatlon for omittim the appalently well-established pressure vessei failure model trom the overall failure algorithm. Please discuss.
_A -
E.
Significant ternel migration, or a lack thereof, must be quantif.sd and backed by evidence.
F.
Choosing a Veibull statistical distribution for " reasons of simplicity" is not acci,.coble-it must be shown that it is applicable and valid. Discuss possible means for confirming t
4 this distribution experimentally.
f 2
s
=
y
)
1 Enclosure I
G.
The footnote:on' page 17 notes. the exclusion of initial defects.
l
~
Have the initial defects been lost in the shuffle? What is the
-l definition of a= defective particle? Is 'a' " weak" particle a l
defective particle?
3 H.
To the list on page 18, it should be added that Cs has been --
assumed to be a representative fission product..Along with this' j
should be an explanation and justification as to why this is;so..
l I.-
Figure 3 (a) "..100% weight loss" is nonconservative. Do the.
J
. points-confirm the curves? =Are they 30 h points? nDoes "no
-i
[
change" mean baseline?. With respect to what? Extrapolating to
'l a delta log k of-.3 and a ramp of 8000_ h (one year at normal ~
f
^
core operation with a very slight temperature drift) will'
{
readily take us to full failure at temperatures as low as 1000 C
]
s to 1200*C. Please explain.-
l i
?
a J.
Figure' 5 - It is unclear what the experimental points confirm.
t Please explain.
l 3
K.
The process described in 3.2.4 may " average out" a's a-systematic'
)
1 failure. What assures that this does not happen?
)
)
L.
Table 2.
Choosing average values appears contradictory to
.l 4
safety interest. A single poor batch may dominate safety.
Safety (and licensing) may require choosing the highest value of m.
Please discuss.
l M.
Figure 12 shows insufficient range of the MHTGR. Fuel integrity to 10 is claimed. The second 12b figure (paga 54) and Figure 13 show that the fuel fails to comply with the model in the range of interest and importance to the MHTGR and is not adequate for the MHTGR. Please explain.
4-46 While DOE acknowledges that more fuel perfermance data'needs to be' I
obtained in the 1200'C to 1800"C range, DOE has also stated that a' 4
a.
I 3
.j
i' Enclosure very larae data base suppor'ts the expectation'that.the sic layer reference fuel can perform the primary containment function for the MHTGR. Please provide a summary of selected existing data ~in i
sufficient detail that the staff can assess the degree of. support this data base provides. The summary should include:
(1) q composition of the particle (kernel, coatings, matrix conditions),
(2) significant particle dimensions, (3) exposure conditions-(burnup, l
fast fluence, irradiation and temperature _ history), (4) chemical-l attack conditions, (5)' irradiation method (in-pile, in-reactor),-(6) method and cnnditions for. testing exposed individual particles and i
particles in compacts (out-of-pile ramo heatup, in-pile heatup, l
l other), (7) means for detecting failure or survival, (8) other l
factors contributing to knowledge of the failure mechanism, and (9) the source and a reference (GA, ORNL, FRG, other).
i 4-47 Provide a discussion that explicity relates the data base presented in Comment 46 to the expectation that the planned fuel particle design and corresponding research program will successfully result in the reference fuel meeting its design goals, which the staff judges will be more rigorous than for older fuel.
Include in this discussion consideration that the only data apparent to the. staff-supporting the f6 m n of the sic barrier by a mechanism of fission-
)
product corrosion '(t.% pr.bciple failure mode thought to be j
appropriate for MHTGk ruel) cppears to have been obtained with UC 2
fuel in thermal gradients of d e type not 1tkely to be encountered.in i
the MHTGR.
4-48 Provide a summary type identifi<ation of the main issues in the fuel particle failure and pe; forman' e model to be resolved by future testing and describe the relative importance of these issues to the i
MHlGR design requirements.
Identify the specific TDNs in the RTDP j
that will address these issues.
For each TDN identified, discuss how.
j questions of statistical adequacy, laboratory quality assurance,-
)
I changes of fuel composition with burnup, external chemical' attack, and manufacturing consistency will be approached. We anticipate that the RTDP will be revised before a later review stage to provide this <
4
....-.....~..
......J
Enclosure i
information in detail. This response should be coordinated with.
Comments 4-49,4-50, 4-51, 4-52, and 4-53.
4-49 With respect to fuel statistical performance on.the basis of
)
laboratory testing, the achievement of the stati'sti'cally low failure probabilities at a satisfactory confidence level and the multitude of affecting parameters require a rigorous research and development program that complies with a systematic statistical approrch commen:, urate with the number of parameters and the required accuracy.
This is necessary for the staff to accept the fuel quality level proposed. This concern was noted'previously.in NUREG-0111. More specifically, please discuss-the following two areas in which statistical information on the production fuel performance,to be supported by laboratory testing, will be needed.
1.
The statistical validity of experimental evidence is a well established branch of statistics. The RTDP does-not currently l
explain how quality assurance in accordance with 10 CFR 50' Appendix B 's to be achieved. The~ quality assurance plan shoulc' be made a part of a revised RTDP plan.
2.
The means for achieving 95 percent and 50 percent confidence levels needs-to be confirmed. The associated probability distribution should be established and verified. An analysis needs to be presented that will show that the 5 percent and 50 t
percent levels of "non-confidence" do not result in exceeding the stated performance requirements. This needs to be done in such a manner that it will satisfy the intent of the SRP, 4.2,1(c).
4 4-50 With respect to the effects of fuel composition on performance, the significant failure mode of the silicon carbide layer at around 1600 C is stated to be caused by internal chemical corrosion.
Consequently, the fuel failure mocel and experimental data should include fuel composition as an explicit parameter and consider the effect of these. changes over the irradiation 11fetime. Most;of the 5
- -.... ~,.. ~
I Enclosure t
- data available relate _to highly' enriched fuel while the MHTGR utilizes a ful of 19.9 percent enrichment. which results in the fuel j
eventually containing significant quantities of. plutonium. The RTDP
. states that the various fuel perameters are to be covered, but j
implies that " representative sample" testing will suffice for the overall poof. We do not believe such an approach can be justified as there is a need to cover, in a statistically valid way, the entire i
range of parameters and their combination. We would expect that a.
matrix of these validations would be planned and executed. Please discuss.
4 1
4-51 The lists of " service conditions" in the RTDP with respect-to-the effects, or lack themof, of chemicals to which the fuel may be externally exposed need to be fully identified and consistent with safety analysis parameters. The chemicels to be considered must i
include, at least, water vapor, oxynen, and nitrogen, and the potential for synergistic effects from trace chemicals and radiation.
In consideration of long term normal exposure, the ranges of j
impurities which are either necessary or which must be avoided should be defined. The planned experiments should consider the full range I
of exposures to chemical attack as can be derived from long term normal operation, ir.cluding A00s, followed by the appropriate bounding events described in our letter of November 10, 1987, i
4-52 With respect to achieving extremely low defects (6 x 10-5) in production fuel, statistical quality control and quality assurance plans for manufacturing, including acceptance criteria, need to be developed to ensure that manufactured fuel is of the specified quality and will perform as predicted.
In particular, the quality control program must contain a fuel particle and fuel compact sampling scheme and inspection technique that reflects the allowable defect rate, wi.ich is about 20 times less than for Fort St. Vrain.
The verification of such a plan is expected to require empirical confirmation. We believe the development of an acceptable plan is of such significance to MHTGR safety that the development and 4
6
s Enclosure description of the plan should be included in the RTDP. Please discuss.
j i
)
4-53' We consider " weak" fuel-particles as those particles that do not fail i
like particles with cracked sic layers under normar operat4'1 (which gives evidence of the amount of defective fuel), but would fail-unexpectedly under postulated transient conditions.
For this reason, the particle failure model and the manufacturing specifications need to be developed to account fnr particle weaknesses and that the ac -
]
companying quality assurance program should be capable of their -
detection with the same reliability as for fully defective particles.
q
- -~
l Please discuss.
i
.4-54 Much of the present data is separate effects ano in vitro. 'The correlation to in situ fuel must be firmly established, again considering the low probabilities and high accuracies required. The RTDP recognizes the need for integrated proof testing.to indicate any
]
weakening of fuel integrity by long-term exposures at normal operating conditions but would not perform this with full scale fuel l
elements. Discuss how the planned testing program will provide an l
equivalent level of overall confidence that wof.d be derived from full scale testing.
i j
4-55 Cracks in Fort St. Vrain fuel blocks due to unanticipated thermal stress were judged by the staff to be acceptable for th'at reactor because the cracks were seen as not affecting the safety performance.
j l
Although calculations of stress-to-strength limits have been made for j
the MHTGR fuel, which would indicate a margin against cracking, the PSID states the fuel design is not intended to preclude limited j
i cracking. The staff has deferred judgement on the acceptability of j
fuel block cracking in the MHTGR pending further experience with Fort
]
St. Vrain. Please discuss your views on this matter.
4 4-56 The PRA assumes a release fraction of 0.02 percent of the halogens for categories DC-1 and 2 without supportir.g information in any documentation yet available. to us, including.HTGR-85-107. Describe q
7 v
e.ww.-
Enc 1osure the basis for this assumption in terms of fuel failure data now available and data expected to be obtained by the RTDP.
5-43 In'your response to Comment 5-38 you reported the sensitivity of peak vessel, fuel and the Upper Plenum Therral Protecti6n Structure:
(UPTPS) temperatures to variations in' the UPTPS insulation properties.
Please be prepored to discuss this study at the March 18, 1986 meeting including how these results can be experimentally confirmed.
5-44 In your response to Comment 5-39 you cited design and analytical reasons why means were not provided for depressurizing the reactor -
vessel prior to the elevated temperature operations that would occur in a conduction cooldown event. We further understand that if such a depressurization did occur the resulting higher temperatures would cause on investment loss.
It is the staff's opinion that some means should be provided to ensure reactor vessel depressurization at temperatures significantly above current code values to diminish the vessel's vulnerability to a catastrophic pneumatic mode of failure for the case of when vessel temperaures exceed 800 F.
In addition, the analysis of event G.2 and the depressurized cases'in certain of the BES require a deliberate means to cause depressurization when the vessel temperture _ exceeds 800 F, the value above which your-proposed code extension would not apply to pressurized conditions. Please discuss our concern that this provision is needed to be consistent with the requirements for protection against vessel overpressure as required by the ASME. code.
5-45 We require that the primary system steel vessels meet at least the same level of integrity as LWR steel vessels. The following comments and requests for additional information pertain to this issue.
A.
On the basis of probabilities and expert engineering judgement, the staff in conjunction with the ACRS established for LWR plants the probability of a steel reactor vessel failure of sufficient. size and location to defeat the performance of the 8
. ~
Enclosure Emergency Core Cooling Systems at l' x 10'I per year. On the basis of an equivalent amount of electric power generation per plant, the MHTGR requires in comparison to a modern BWR plant ~
(which has a reactor' vessel of dimensions comparable to the MHTGR)'about 10 reactor vessels per plant.' HEnce, it could be argued that tM likelihood of a large vessel failure for an MHTGR plant should be.10-6 per plant year. Please discuss.
B.
A catastrophic vessel failure (see Comment L) would put the reactor core in a physical state judged by the staff as undefined within the reactor cavity.-Can D0E define this state with respect to the potential for a graphite-fire and the continued removal of decay heat without significant additional fission product release?
C.
We agree with the response to Comment 5-16 that the MHTGR vesse' will not be subjected to the LWR conditions of pressuized thermal shock, intergranular stress corrosion cracking, and water hammer and also agree that in estimating the level of vessel integrity credit can be given for the absence of these conditions.
Indicate degree of credit'in terms of failure probability that you believe is appropriate and' discuss this in
~
connection-with your response to A above. However, we do not agree that credit can be given for lower neutron embrittlement effects until more experimental evidence is available. We note that in the response to Comment 5-15, DOE stated that " final confirmation of an effect of neutron flux must likely ewait the results from an MHTGR surveillance program."
D.
We believe that the significance of challenges from extreme repetitive loads, thermal fatigue, and thermal _ stress will be dependent on the duty cycle for conduction cooldown cvents. We notice from Table 3.9-1 of the PSID and Table 8-2 of the RDTP that only a single occurrence is stated for each of the conduction cooldown events listed. Are we to infer that reactor re-start will not occur following any of these events except for 9
Encic:ure Event 317 If-this is not the case, please gi' e (1) the pemitted duty cycles planned for both pressurized and I
depressurized cases, (2) the bases for the permitted duty cycles, and (3) an estimate of the actual nunher of conduction cooldown' events expected to occur over the' lifetime of the j
reactor.
i
~
E.
.In general, we see no potential for. pressurized thermal-shock y
events similar to'those that have occurred in LWRs. However, a localized thermal shock by water ingress.from a failed Standby _
j Cooling Heat Exchanger (SCHE) might' occur during a dep essurized.
conduction cooldown event. While this is not a pressurized I
thermal shock event, the vessel is at an elevated temperature where it may be vulnerable to failure by thermal shock r
nevertheless. Please discuss and also include in this
'I discussion the potential for this failure mode to provide a chimney effect for a graphite fire potential.
F.
In response to Comment 5-17, it is stated that reactor vessel l
inspections may be deferred to the end of the Code-specified inspection intervals. Justify tMs procedure with respect to maintaining the same level of vessel integrity as for LWRs.
G.
In Appendix A to the PRA, it is stated that a 7-inch long flaw in the reactor vessel can be detected and the reactor scrammed l
(on low-pressure) because the helium make-up rate would be exceeded. Discuss the safety significance of this capability from the standpoint of (1) this flaw size as a precursor to vessel failure, (2) the leak-before-break philosophy, and (3) meeting the intent of SRP Section 5.2.5, " Reactor Coolant Pressure Boundary Leakage Detection."
i J
H.
!!as ASME approval been cbtained regarding the adequacy of Technical Development Need 8-2, " Properties of SA533B at Elevated Temperatures." If not, is it being sought? Are creep 10
,...... ~. - -.
,e Me4:*4i J.
-A a
-Enclosure rupture tests on a single heat of metal' sufficient? Are high temperature fatigue and creep-fatigue interaction data needed?
I.
Will the MHTGR vessels be designed to the fionfiandatory-Appendix G,Section III, Division 1 of the ASME Code to provide a degree of flaw tolerance? Discuss the pros and cons of designing to this Code Section.
l J.
Experience has shown that tMw W stresses and strains in many I
cases have turned out to be muut higher than the results obtained from generally accepted design engineering methods; thermal problems are not always solved with sufficient accuracy by code calculations. Please comment on this statement with respect to (1) how unknown or underestimated thermal stresses could contribute to the probability of vessel failure and_(2) the methods, including possible development items, for ensuring that the MHTGR vessels will be designed without hidden thermal i
stresses, particularly under Event Category III conditions.
K.
Experience has shown that creep-fatigue interactions have r
resulted in failures which fell short of the lives predicted by use of the-linear cumulative damage rules given in Code Case i
N-47.
Please comment on this statement with respect to (1) the probability of vessel failure and (2) design' methods for the MHTGR vessel that will take this concern into account.
L.
Differences between hydrostatic and pneumatic vessel failures are well known.
Please refer to papers by Karl Kussmaul, et al., Hutin and Churier, and other authors in Volume G of the
" Transactions of the 7th International Conference on Structural Mechanics in Reactor Technology," Chicago, 1983 for-illustrations of these phenomena and a discussion of conditions causing stable tearing or catastrophic failure, and to NUREG/CR-2570, April 1982 for a description of how system compliance must be accounted for.when. determining the expected i
11
j
~
Enclosure.
)
i degree of stability in a metal fracture process._ Based on j
information in these and possibly other references,' comment with ~
-]
regard to what degree you consider catastrophic, ~ pneumatic-type -
vessel failures a concern for the MHTGR.
In the event of the' i
-7 occurrence of a low probability (10 per year) vessel failure would you predict stable tearing or catastrophicL failure and why?
)
M.
Evaluate the potential for ASME Code and Code' Cases to include j
the MHTGR cross duct design,'considering that the duct significantly differs,from usual reactor vessel-design geometry.
In addition, discuss further the considerations and identify the major issues facing a plan of action for admittance of SA 533 and SA508 to Code Case N-47 coverage.
6-6 In response to Comment 15-5, DOE stated that the reactor and steam-
)
g<enerator supports and the reactor building would be' designated
' safety-related. " Discuss your position whether or.not this safety
]
related designation is to apply to the case of total RCCS failure.
It is the staff's concern that the design bases for this case may. net 3
adequately reflect the environmental conditions,. including ~
j consideration of the necessary performance requirements for the cont. rete forming-the cavity. These design bases should be reviewrfd by that staff and be supported by the sensitivity study requesten in Comment 15-4. At present, the staff considers the' consequences of conduction cooldown without RCCS operation as undefined and not j
necessarily bounded by the DC-1 and DC-2 events described in the'PRA.
4 i
6-7 Our independent studies estimate that concrete temperature in the I
reactor silo could reach peak temperatures as high as 700 C in 3
several regions for the case of RCCS failure.
In anticipation that your.'esponse to Comment 6-6 will show that performance r' ;uirements for concrete carnot be met by ordinary concretes, discuss the options for resolving this concern. Because there is likely to be i
uncertainty in the efficacy of concrete at these temperatures, l
12
I Enclosbre i
provide the effects of concrete failures on fuel tempe*/atures and i
i pressure vessel integrity.
j j
1 j
6-8 In the safety analyses of depressurization events preset.ted in.the.
~
PSID credit was taken for plate-out and settling of fission products within the reactor building. The staff has not reviewed the modeling j
for this credit, and har'not discussed with DOE (1) the nature of the
" torturous path" that the reactor building prov Ees and (2) the~
research program related to plate-out and settling to be performed as I
described in the RTDP. DOE.is asked to provide a summary of the reactor building's features that contribute. to its retention--.
capability, indicate whct retention capability would exist for I
bounding Event Category III sequences, and indicate whether the research program will include reactor building. environments consistent with the EC-IIIs considered.
i i
6-9 In the DC-2 event described in the PRA the reactor building is 1
assumed to be " sealed" after 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br />. Such a capability is seen as providinp sn important degree of added protection and accident i
mitigation for the NHTGR concept.- Describe how this could be accomplished and discuss whether this feature should be considered by the stsff in its review of MHTGR safety.
15-8 On the bases o a position paper being developed to obtain Ccamission guidance on key safety issues, the staff has reformulated its bounding event criteria. This formulation has been generalized for advanced reactor designs but remains basically the same for HTGRs, as expressed in Comment 15-8, with the exception that the 36 hour4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> time durations have been replaced by "x hours."
In performing the requested examination of the bounding events DOE has the option of assum'ng failure of safety grade 9quipment for a period of time consi; tent with previous experience (unless a lesser time can be jntified) or it can use longer times to demonstrate MHTGR safety margins. We need to finalize the bounding eveats for the MHTGR, i
i 13
\\
Enclosure based on the above, and request DOE be prepared to discuss. this at i
the meeting.
In addition, we would like to call attention to our letter of January 4,1988 to Francis X. Gavigan from Bill N.Torris where it was stated: "While further consideration is and will continue to be given to the methodology, our ability to make a licensability determination l
on the MHTGR is impacted by the lack of information on the response of the plant to what we consider are bounding events which should be evaluated at this stage of the design." We have yet to receive the requested infonnation and request that you be prepared to discuss these events at the meeting.
1 15-9 Carbon dioxide and to a lesser extent carbon monoxide would diminish radiation heat transport, depending on their concentrations,.if present in the regions between the core barrel and the inner surface of the reactor vessel or between the outer vessel surface aivd the
]
RCCS cooling panels. Provide as a function of time for moderate cases of moisture ingress (SRDC '., for example) and for BES-4 and BES-5 the amount of-CO and C0 that could be present in these regions 2
and the consequences with respect to maximum ft 1 and vessel temperatures. These results should include pressurized and
{
depressurized conditions. '
1
]
15-10 The graphite fires and significant releases of radiation that occurred in the Windscale reactor in England in 1957 and during the course of
{
the Chernobyl accident in 1986 led to substantial investigations into I
the nature of graphite fires in nuclear reactorr and the conditions necessary to sustain combustion. DOE should assess the potentials for a graphite fire in the MHTGR making use of the information ceveloped in these investigations. Two documents that summarize this information and provide important references are: NUREG(Draft) 1251,
" Implications of the Accident at Chernobyl for Safety Regulation of Commercial Nuclear Power Plants in the United States," and NUREG/CR-4981, "A Safety Assessment of the use of Graphite in Nuclear Reactors licensed by the U.S. NRC " (Schweitzer et al., BNL, September 1987).
In using NUREG/CR-4981 consideration should be 14
. _ _ ~. -.
~
1 Enclosure I
given to the fact that data for sustained combustion do not appear to j
be available for graphite at temperatures above 800 C.
Also include j
in your graphite fire study considerations of the reactor building including its " sealed" state as disci ssed in Comment 6-8.
15-11 DOE should describe the MHTGR regarding the following criteria'with respect to enhanced safety:
i A.
Enhanced safety characteristics and margins from: (1)long response time, (2) reduce potential for operator error, (3) capability to retain fission products', (4) highly reliable safety systems (passive / inherent characteristics),-(5) l simplification (system / analysis).
B.
Potential improvements in safety are to be considered when the margins are small or when large improvemente in safety can be realized with reasonable cost.
In particular, documentation regarding the addition of a containment building is requested, l
C.
Demonstrate
.hanced safety / margins via testino on a
~!
first-of-a-kind plant.
3 This description.should be'done in comparison to current generation LWRs, such as the ABWR.
1 I
i 15
- - - - - - -,