ML20033F437
| ML20033F437 | |
| Person / Time | |
|---|---|
| Site: | Zion File:ZionSolutions icon.png |
| Issue date: | 03/06/1990 |
| From: | Licciardo R NRC |
| To: | |
| Shared Package | |
| ML20033F435 | List: |
| References | |
| NUDOCS 9003210034 | |
| Download: ML20033F437 (32) | |
Text
,
)
Enclosure l
SllMMARY EVALUATION BY ROBERT B. A. LICCIARDO i
0F RESPONSES TO HIS DIFFERING PROFESSIONAL VIEW (DPV)
AND DIFFERING PROFESSIONAL OPINION (DPO) CONCERNING (A) ZION 1/2 CONTAINMENT PURGE ISOLATION VALVES, AND (B) METHODOLOGY USED FOR CALCULATING OFFSITE DOSES Robert B. A. Licciardo Registered Professional Engineer, California Nuclear Engineering License No NU001056 Pechanical Engineering License No. M015380 DATE: March 6, 1990 Af5 t
.p A bock cSo
l' 11 Y
EXECUTIVE SLNMARY
1.0 BACKGROUND
By memorandum to Dr. T. E. Murley, Director, NRR, dated May 12, 1989, the writer submitted a Safety Evaluation Report (SER) and Differing Professional View (DPV) concerning a proposal by management to issue a TS amendment allowing the containment purge isolation valves at Zion 152 to remain open during power operation.
This DPV was accepted by Dr. T. E. Murley, Director, Office of Nuclear Reactor Regulation for review by a Standing) Review Panel from which Dr. Murley finally issued action requirements to a effectively review all new and advanced fuels for the possibility of a rupture during the blowdown phase of a LOCA, and b) to revise the Standard Review Plan to clarify the relationship between DNBR and fuel failure. This review however continued to recommend the issuance of the management SER permitting the large containment isolation valves to remain open at power. As a consequence the writer submitted a Differing Professional Opinion (DP0) to James M.
Taylor, the Executive Director for Operations (EDO), dated October 19, 1989, to require that these valves remain closed under these conditions. The EDO j
appointed an Independent Review Comittee the input from which the EDO finally issued recommendations on January 2, 1990.
This evaluation represents the first opportunity the writer has had to be able to contribute further to this problem based upon his review of the substantial references and coments made available to the writer by the review comittees. This sumary addresses the outstanding issues only, and in tems of the five recomendations made by the Independent Rev_iew Comittee for the DPO.
2.0 FUEL CLAD RUPTURE DURING A LOCA BLOWDOWN In their sumary report, the DP0 review committed concludes that " contrary to the DPO, fuel failure is extremely unlikely during the first seven seconds of a design bases loss of coolant accident":
Summarily: The Zion reactor with a (licensed) maximum power rating for LOCA calculations of 17.9 kw/ft and fuel designed to be at a clad pressure of 1400 to 1700 psi at normal full power conditions gives calculated temperatures over the first seven seconds of a LOCA blowdown of 1500-1700 F and " evaluated" pressure differences across the clad which could result in fuel ruptures; and because of the lack of capability of existing Reactor Coolant System (RCS) and fuel rod models to calculate the maximum cladding temperates and related small pressure differences across the clad sufficiently accurately to ensure that rupture will indeed not occur under these conditions, the writer proposes that there is a sufficient probability that it will occur to the extent necessary to justify his primary position that for Zion the containment purge valves should remain closed in operational Modes 1 (power), 2 (startup),
3 (hot standby), and 4 (hot shutdown).
3.0 DNBR AND FISSION PRODUCT RELEASE: CURRENT REGULATORY POSITIONS In their sumary report the Panel concluded that "The concept of specified acceptable fuel design limits does not apply to such an accident (that term is reserved for anticipates transients)
This is a residual issue of major significance, because it determines the importance of any containment isolation valves open to both the containment and the environment during a LOCA, and as a consequence of the writer's DPV, Dr. T. E. Murley has issued a directive that this be resolved for clarification within the Standard Review Plan by October 1990.
iii Sumarily:
The concept of fuel rod failure as a loss of herr'eticity of the fuel rod cladding giving a release of gap activity with a value of 10% core activity, and caused by an infringement of Safety Analysis Fuel Design Limits, a measure of which is infringement of DNBR criteria for the fu51 rods, was initially regulated by 10 CFR Appendix A Criterion 10 to only Normal Operating Conditions and Anticipated Operating Transients; but has been extended by Regulatory Guidance and Precedence to all " Postulated Accidents" as described and fully referenced in his DP0 material.
4.0 0FFSITE DOSES AND THE IMPORTANCE OF THE TIMING OF FISSION PRODUCT RELEASE In their sumary, the DP0 review committee proposed that "The dose associated with a conservative iodine " spike" release during the time that the valves are closing is well within Part 100 limits."
Sumarily: This value of offsite dose provides only for fission product product activity in the coolant of the Reactor Coolant System, and depends for it's use on the proposition that fission product during a LOCA is released only at the time of fuel clad rupture as calculated to meet Appendix K requirements and that this fuel rupture is evaluated as generally occurring after 5-15 secs. into the event by which time the purge valves are closed.
Based on existing Regulatory positions and the review of the substantive test references reported elsewhere in his DP0/DPV, the writer proposes their can be no fundamental Regulatory and related Guidance support for this proposition.
Also, that existing Regulations would require the offsite dose over the first seven seconds to be calculated at 158,000 rem.
In the event non regulatory provision for no fuel melt was allowed this could be reduced to 64,000 rem; a normally operating design basis fuel failure at IT of the core alone would 634 rem which is over twice the Allowable value. The overwhelming importance of fuel failure and it's related fission product release to offsite dose during the blowdown phase of a LOCA provides no margin for any uncertainty about surety of containment to prevent its release to the environment.
5.0 RISK PERSPECTIVES:
In their summary, the' Panel proposes "There is little incremental risk associated with operation of a reactor such as Zion with the purge valves open, as their is a likelihood that the valves (or at least one of the redundant valves) will close on demand."
Summarily, it should be recognized that the prime concern of the writer is not the failure to close of these valves but the failure to keep them closed during Modes 1, 2, 3, and 4 so that they remain fully open to discharge the L
contents of containment to the environment for the first 7 seconds of blowdown.
' The writer's CP0/0PV proposes, that on a Regulatory basis any permission to grant l
an Amendment allowing the valves tc remain open in these Modes for subseouent successful automatic isolation within 7 seconds will give offsite doses during that 7 seconds, far in excess of allowable values and is thereby unacceptable.
l l
Furthermore, any other decision based on probability estimates of contribution I
l
iv to totality of risk from 611 potential contributors at the facility would result in offsite doses exceeding these allowable values and thereby in violation of Regulatory requirements and thereby the current " Safety Goals for the Operations of Nuclear Power Plants; Policy Statement" in the Federal Register to Reference 11.
6.0 CURRENT NRC GUIDANCE AND POLICY ON THE OPERATION OF CONTAINMENT p0RGE 150L/T10R VALVES In their sumary report the DP0 review comittee also concludes that "There seems to be some confusion as to the regulatory times and conditions for operation of these large valves.
(We noted that the issue may be moot; Zion does not actually envision routine operation with valves open, we were informed).
Perhaps the need to routinely purge reflects some basic design deficiency which should be addressed.
In any case, it is clear that the containment leak tight criterion is better served if one does not ha've to open large butterfly valves from time to time, or leave them open indefinitely.
In our opinion, NRR-should reexamine policy in this area."
Sumarily, this proposition of confusion is not an uncomon regulatory experience of multiple criteria setting different Safety limits from which a conservative approach to containment purce valves based on regulatory reouirements for fission product release does reveal the most limiting conditions which must be applied and which have been evaluated and presented by the writer. Unfortunately, much of the SPP on containment purge valve 1 solation times and thereby extended use.into Modes 1, 2, 3, and 4 has been predicated on fission product and related release criteria which are much less conservative than the Reg. Guide 1.4 criteria, and also SRP 4.2 criteria l
for fuel clad hermiticity, without any appropriate safety evaluation for J
l departure therefrom.
Furthermore, I also confirm by detailed reference l
that Zion was designed and ifcensed so that the containment purge valves are to remain closed in Modes 1, 2, 3, and 4 and that provision has already been made for. access to containment, and those other features of plant operations, normally proposed as a reason for there use; and it is the writer's experience L
that this may also apply to other Westinghouse facilities. Thereby, the writer proposes that the current related uncertainties in fission product i
release and the potentially excessive offsite doses arising therefrom, cauld never justify en alternate proposal to allow them to remain open at this time.
7.0 CONCLUSION
The Zion reactor with a licensed maximum power rating for LOCA calculations of 17.9 kw/ft and fuel designed to be at a clad pressure of 1400 to 1700 psig at normal full power conditions, gives calculated temperatures over the first seven seconds of a LOCA blowdown, of 1500 - 1700"F, and evaluated pressure l
differences across the clad which could result in fuel ruptures. Furthermore.
l the overwhelming importance of fuel clad failure, both by DNBR infrinoement and related loss of hermeticity and or clad failure and it's related fission product release to offsite dose during the blowdown phase of a LC0A provides no margin for any uncertainty about surety of containment to prevent it's release to the environment. The writer proposes that for Zion there is a sufficient probability that fuel failure will occur to the extent necessary should thereby remain closed in operational Modes 1 (purge valvespower),2(start to justify his primary position that the containment (hot standby), and 4 (hot shutdown).
Furthermore, it has been confirmed by detailed reference that the Zion facility was designed and licensed so that
n,,
j 1
-1 D,.
the containment purge valves are to remain-closed in' Modes 1, 2, 3, and 4 and 1
l i
. that. provision has already been made for access to containment, and those other-features of plant operations, nonnally proposed as a reason to amend this requirement at an unacceptable risk to Public Health and Safety.
1 l
1 1
o l
l l
l l
l-I
\\
\\
s.
yj TABLE OF CONTENTS 1
EXECUTIVE
SUMMARY
11 j
l
1.0 INTRODUCTION
................................................... 1 2.0 FUEL CLAD RUPTURE DURING A LOCA BLOWDOWN....................... 1 l
3.0 DNBR AND FISSION PRODUCT RELEASE: CURRENT REGULATORY POSITIONS...................................................... 6 4.0 0FFSITE DOSES AND THE IMPORTANCE OF THE TIMING OF FISSION PRODUCT RELEASE................................................ 10 5.0 R IS K PERS PE CTI VES.............................................. 13 6.0 CURRENT NRC GUIDANCE AND POLICY ON THE OPERATION-OF CONTAINPENT PURGE ISOLATION VALVES............................. 17
7.0 CONCLUSION
..................................................... 22
1.0 1NTRODUCT10N The writer wishes to take this opportunity to respond to the substantial response by the NRC in it's review of the his Differing Professional View (DPV) submitted to Dr T. E. Murley, Director, Office of Nuclear Reactor l
Regulation on May 11, 1989, Ref. ), and his following Differing Professional Opinion (DPO) submitted to James E Taylor, Executive Director For Operations on October 19, 1989, Ref. 2, concerning Zion 1/2 Containment Isolation Valves, i
and Methodology Used for Calculating Offsite Doses, and to do this by reference to his submittal entitled: An Evaluation of the Criteria and the Calculation of Offsite Doses Deriving from Open Containment Purge Valves During a LOCA at Zion Units 1 & 2 Dated July 20,1989, Ref. 3 and related Responses to refs. 4 j
& 5.
This sumary addresses outstanding issues only, and in terms of the five recomendations made by the Independent Review Comittee for the DPO.
2.0 CLAD RUPTURE DURING A LOCA BLOWDOWN In their sumary report, the DP0 review comittee concludes that " contrary to the DPO, fuel. failure is extremely unlikely during the first seven seconds of a design bases loss of coolant accident":
I note that this conclusion rests largely upon the LOC test series, Refs. 11 &
12, through the LOFT program Ref.16, and ultimately the Code Scaling, Applicability and Uncertainty Evaluation Methodology (CSAU) program, Ref. 7.
The writer's coments on this proposal derive from his own extensive research development and applications experience in the High Temperature Gas Cooled Reactor field, the Advanced Gas Turbine Aircraft Engine Field, and.especially
i
... J in-the detailed design of turbines, combustion chambers, and compressors and a'dvanced duty heat exchangers and steam generators. And also in the design and optimisation of similar equipment and systems for fossil fueled hioh temperature power production systems of generally advanced design, The LOC 3-6 test series refs. 11 & IP, were designed to produce Temperature / Time profiles for fuel cladding giving failure in the Reflood phase of a LOCA, and the resulting related power profiles, time of break, break flow characteristics and thennal hydraulic details of the fuel, do not represent in a valid manner the behavior of fuel in a Zion reactor during a LOCA blowdown.
Furthermore, peak power ratings for the fuel varied from 6-13 kw/ft compared with the Zion FSAR which categorically shows that for LOCA calculations - and the current TS confirm this - whilst the average power rating is 6.9 kw/ft the 1
maximum rating is 17.9 kw/ft. And this is for first cycle fuel, not third cycle; and with design starting pressures inside the clad gap under power conditions of approx.1400 psig (obtained through the Zion FSAR references) to 1700 psig (Lauben Ref. 8) and calculated maximum temperatures of 1500 to 1750'F within 2 to 7 secs. of the initiation of a LOCA. The specific reference to Zion fuel calculations giving a peak clad temperature (PCT)'of 1543*F in ref. 6 was calculated for 15x15 Zion fuel with a peak power rating cf only 13.26 kw/ft compared with 17.9 kw/ft for Zion.
Further, a more recent review by the writer of this specific reference shows that it also calculates the gap pressures during the same blowdown, and at a starting pressure of 1260 psig in the gap.
If these pressures are scaled up by the ratio of the absolute starting pressures for 1700 psig (Ref 8 Enclosure by Lauben for a 15 by 15 fuel), or 1400 psig for a Zion UFSAR fuel (Ref. 3 Licciardo), the resulting
3 pressure drops across the clad during the LOCA blowdown are such as to cause rupture of the Zion fuel at temperatures of 1550-1750'F for 17.9 kw/ft max
~
power in the fuel rods. As Lauben reports, the 15 by 15 fuels at 13.26 kw/ft and a gap starting pressure of 1260 psig with a maximut; temperature of 1543*F do not rupture the clad, but evaluation of the same data by the writer shows that this could occur for the higher power first fuel cycle rods of Zion 1&2.
In response to my earlier clarification of many of these positions, and a recommendation of the Standard Review Panel for the DPV, the Office of NRR established a policy to require that safety evaluations for all new fuels perform a detailed check of the potential for rupture in the blowdown phase of the LOCA, Refs. 4 & 15.
In respe:t of this action it must be said that the ~ ore recent m
evaluation by the writer of the Ref. 6 material by Siefken now shows a greater risk for current first cycle fuels of 15 by 15 designs at 17.9 kw/ft, than for later new fuels of potentially 17 by 17 design at approximately 13.3 kw/ft.
-The additional reference material provided by the DP0 on the LOFT tests, and references to the CSAU program, confirm the above conclusions.
The unfortunate feature of LOFT, ref.16, is that whereas it may well represent 1
in an effective manner, the thermal hydraulics of the Reactor Coolant System itself, it does not however, represent the detailed thermal hydraulic conditions surrounding the fuel in a Zion fuel bundle in an effectively valid l
manner; a test core which uses a coolant velocity 1/2 that of the operating reactor (and a rod length of also 1/2) cannot be a sufficiently valid thereby accurate representation of the critical features ultimately required to
calculate the time of fuel element rupture as the time of fission product release from the clad gap as well the probability of fuel reuwetting during the' blowdown phase.
Furthermore, the only test representing the Licensing Basis LOCA/ LOOP accident is test number L2-5 which apparently is done at peak rod power of approx.11.9 kw/ft (compared with the Zion value of 17.9 kw/ft) and even then, considering all these limitations, shows measured valves of clad temperature during blowdown of 1430'F i.e., in the region of significant influence of the alpha / alpha plus beta transition region for Zircalloy. And that from the CSAU program, for the higher power peek power loading of 17.9 kw/ft for Zion, higher temperatures in the 1600-1800*F are expected with a reasonable probability; Ref. 7, Fig. 33. And also, considering measuring (and testing) errors this covers the range of transition temperatures over the alpha / alpha plus beta / beta phase transitions with all its substantive variance in material properties for the clad.
Further, the experimental evidence strongly suggests that the accurate calculation of clad gap pressures (as
-well as Reactor Coolant System pressures) which is essential to a reliable determination of the relatively small clad gap pressure differences which can cause rupture under the related high temperature conditions, 350-500 psi, l
gives a low probability of a reliable estimate that rupture will not occur under these conditions. Whilst the FRAP T-6 and the BALON 2 code (Ref.13 & 14),
by the reviewers represents to the writer a substantial advance over the Zion Licensing basis references with which the writer initiated this DPV/DP0 process, and now includes corrections for many of the perceived deficiencies of the older fuel rod models which caused him to so comment on page 4-4 of his DPV evaluation, it still has not generally been assessed appropriately, against all the other realities of a nuclear fuel installation in a reactor over the full
^
n.,
,c.
5 range of expected operating conditions up to some 55,000 to 60,000 mwd /Te fuel j
exposure, and including especially the residual substar.tive elements of SRP Section 4.2.!!.A.I fuel system damage and A2 fuel rod failure - necessarily considered to ensure fuel clad integrity for ell oper6tional fuels.
The writer's judgement at this time is that the existing set of parameters used in the models, have not been proven to the necessary degree of accuracy cgainst a starting basis of a known ccid pressure in the fuel clad through to power operation and then into a LOCA and at varying levels of burn up. After reading the material referenced by the reviewers, the writer concludes that these models depend upon a sebstantial diverse set of experimental data, and because of this requires substantial justification before it's application to a particular facility such as Zion. And'even then as shown by the calculations versus test results for'this limited Set, cannot provide results with the necessary accuracy to substantiate their use for calculating the timing of fission product release due to calculated rupture in the blowdown phase of a LOCA during which their would be no additional level of protection against it's catastrophic release to the environment by reason of open containment purge valves.
Summarily: The Zion reactor with a (licensed) maximum power rating for LOCA 1
calculations of 17.9 kw/ft and fuel designed to be at a clad pressure of 1400 to 1700 psig at nonnal full power conditions gives calculated temperatures over the first seven seconds of a LOCA blowdown and " evaluated" pressure differences across the clad which could result in fuel ruptures; and further because of L
the lack of capability of existing Reactor Coolant System (RCS) and fuel rod L
models to calculate the maximum cladding temperates and related small pressure differences across the clad sufficiently accurately to ensure that rupture will L
i i
....,.. j l
1 indeed not occur under these conditions, the writer proposes that their is a sufficient probability that it will occur to the extent necessary to justify his primary position that for Zion the containment puree valves should remain closed in operational Modes 1 (power), ? (startup), 3 (hot standby), and 4
]
(hotshutdown).
3.0 DNBR AND FISSION PP0 DUCT RELEASE: CURRENT REGULATORY POSITIONS In their sumary report the DP0 review comittee concluded that "The concept f
of specified acceptable fuel design Itmits does not apply to such an accident (that term is reserved for anticipated transients)."
This is a residual issue of major significance because it detemines the
)
importance of any containment isolation valves open to both the containment and the environment during s LOCA and as a consequence of the writer's DPV it is now 'the subject of review by the NRC for resolution by October 1990 as required by Dr. T. E. Murley Ref. 4.
l The current licensing basis for the timing of fission product release has been fully described and interpreted by the writer in his DPV submittal.
The outstanding issue is whether the writer's interpretation that the first significant release of the fission product source tem of a LOCA does occur on thr infringement of DNBR for the fuel, required to be considered by Appendix K as occurring at the first one tenth second into a LOCA, is correct, or are the regulations to be interpreted to mean that the DNBR fuel failure criteria do not apply under this one postulated hypothetical accident condition and that timing of the first fuel rupture as calculated by the Appendix K requirements
7 was additionally meant to represent this timing of not only the first fission product release but also of the TID Source Tem.
Their has been an apparent substantial misunderstanding by the review comittees of the current complete Regulatory Practice in the fom of Regulations, Guidance and Precedence and this has resulted in an incomplete representation of the writer's current propositions in this regard and also resulted in the use of contradictory practice by the NRC in evaluating offsite 1.0CA offsite doses for all facilities other than for Zion. The writer has provided a full representation of the current Regulatory and related Guidance and Precedence positions on this subject as it relates to the concomitant release of fission products, in his DPV Ref. 3. Section 1.3 The concept of fuel rod failure as a loss of hemeticity of the fuel rod cladding giving rise to a potential release of gap activity, with a necessarily conservative value of 10Y core activity, caused by an infringement of safety analysis fuel design limits, the measure of which is infringement of DNBR limits for the fuel rod, was initially regulated by 10 CFR 50, Appendix A Criterion 10 to only nomal operating conditions and Anticipated Operating Transients; but has been extended by Regulatory Guidance to all Postulated Accidents as described and fully referenced in his Dp0 material.
Furthermore it has been used as such in evaluating the radioactive source terms for every comercially licensed facility in their FSAR, and also the related generic and plant specific evaluations for fuel supply both in the licensee safety analysis reports (SAR's) and the NRC safety evaluation reports l
j
.c.
(SEP's.); the writer has been fully aware of these criteria from his earlier three years as a technical reviewer in the Reactor Systems Branch and from his review of the licensing basis references for Zion for the preparation of his 3
DPV. As described in his DPV, this criterion has been universally used for 1
meeting regulatory requirements for all Accidents except for the LOCA when it has been stated that the legal source term has been otterwise set by 10 CFR 100 Requirements.. However in responding to NRC requests in Reg. Guide 70 for more realistic but non regulatory realistic evaluations of offsite LOCA doses to be also provided in the FSAR, the Ifcensees, as a precedent, have resorted to this same same basis for the r 9 ted source term when their is no calculation of %el melt in the rods from LOCA calculations.
The early background tr.aterial in which the proposition that the RCS inventory with an iodine spite be used as the source tenn for evaluating offsite doses from open containment isolation valves, is absent any evaluation of the alternate criteria which were available and this must be be considered an inconsistent and thereby a significant omission in the rationale for its choice.
Its use represents a direct contradiction of every past practice for accidents and could represent a precedent for substantive relaxation of current licensing limits in a very substantive way and especially if it is used by the NRC to justify dose reductions on a LOCA from etherwise catastrophic values, to relatively benign values.
It could also be used as a basis for relaxation from licensing basis requirements for the more severe transients, and postulated accidents, which might otherwise be unacceptable, and their are on file, precedents for this type of request.
As the writer's DP0/DPV has reported on page 1-9, the standard review plan (SPP) states "for postulated accidents, the total number of fuel rods that exceed the criteria (DNBR) has been assumed to fail for radiological dose calculation purposes. Although a thermal margin criterion is sufficient to demonstrate the avoidance of overheating from a deficient cooling mechanism,
it is rot a necessary condition (i.e., Di$ is not a failure mechanism) and I
other mechanistic methods may be acceptable. There is at present little experience with other approaches, but new positions recommending different l
criteria should address cladding temperature, time duration, oxidation and i
embrittlement".
Further, as the writer has proposed in his OP0/DPV, the purpose of Appendiy K as generally described in the related regulations, is to
+
identify that particular rupture which would have the most conservative effect with respect to meeting the reovirements of 10 CFR 50.46 and for this end it models, and uses factors, to calculate conservative values for the related i
purposes.
The regulatory purpose of Appendix K calculations is not to determine and identify when failure by clad rupture (bursting) first occurs as an otherwise conservative evaluation of when fission product is first released from the fuel during a LOCA for the purpose of calculating related offsite doses.
The writers comprehensive review of the Blowdown Test references provided by the reviewers and the parallel development or calculational models using these
- results, further confirms that view. The writer proposes that, at this time they are not adequately and conservatively designed to perfom this function to the level of importance required to ensure absence of the catastrophic offsite dose consequences which could otherwise occur, for Zion 1/2.
~
..~
30 Sumarily:
The concept of Fuel Rod failure as a loss of hermeticity of the fuel rod cladding giving a release of gap ar.tivity with a value, of 101 core activity, and caused by an infringement of Lafety Analysis Fuel Design limits, a measure of which is infringement of DNBR criteria for the fuei rods, was i.nitially regulated by 10 CFR 50 Appendix A Criterion 10 to only Nonnal Operating Conditions and Anticipated Operating Transients; but has been extended by Regulatory Guidance and Precedence to all
- Postulated Accidents" as described and fully referenced in his DP0 material.
4.0 0FFSITE DOSES AND THE IMPORTANCE OF THE T!NING OF FISSION PRODUCT RELEASE In their sunrnary. the DP0 review consnittee proposed that "The dose associated with a conservative iodine " spike" release during the time that the valves are closing is well within Part 100 limits."
Whereas the writer's DPV/DP0 Exhibit 2 Rev. I does indeed show that the offsite dose contribution fttm the conservative Iodine 131 EOU. spike over the first 7 seconds is well within Part 100 limits at 40 rem (this must be added to a 2 hr offsite dose from other sources of 123 rem), the important issue is, what can be added to offsite exposure as the potential for fuel failure
' increases.
It should be noted, that the Iodine release in this so called
" conservative" spike represents only.00% of the (10%) gap activity of the total Zion core; i.e., the staff effectively proposes that no more than.08" of the fuel fails during the blowdown phase of a LOCA.
As the writer's DP0/DFV report has stated, fuel failures on LOCA would have to be 1imited to 0.2" of the core gap activity to limit the offsite dose to PART 100 limits.
.. f -
.. When these values of futi failure are matched against the design basis failure of 1% for fuel in the core during nomal operations, and when under LOCA blowdown conditions we should reasonably expect rupture of this same fuel (Ref. 11, ' No.11 of the LOC-5 tests), the addition to offsite dose alone would be 634 Rem which when added to the otherwise calculated offsite dose of 123 rem gives a total of 757 rem which is over twice the allowable value of 300 Rem. And in cJdition to that, we have short half lived fission product sources which now become important as potentially significant additional doses for our nece:sary consideration. If one considers the Guideline limiting fuel failure criterion of the the control rod ejection accident, of 10% fuel failure because of related infringement of DNBR, the addition to offsite dose of such a level of failure, under LOCA conditions, would be 6340 Rem i.e., twenty one (21) times the Part 100 limits. What is the rationale behind the assumption that what is valid for a Condition 3 Accident is not velid for a licensing basis LOCA, and especially when the calculeted marginal risks of such a rationale are so catastrophic? The NRC could well be placed in the position of re-defining the most limiting accident in terms of offsite doses. As my DP0/0PV report has shown, the reouired use of Regulatory Guide 1.4 criteria which requires an immediate release of fission product to containment, would result in an additional contribution to thyroid dose over 7 seconds of 158,000 rem; using DNBR failure criteria only, with 10% fission product gap release, would reduce this to 64,000 rem: It should be noted, that after dilution inside containment, these doses are j ccused by the related fission product. I 131 E0, release of only 0.3% and O.157% core inventory over the 7 secs closure time of the partially opened 42 inch Containment Purge valves; Ref. 3 Exhibits P-5. x l
l 1 . Sumarily: This value of offsite dose given by the review comittee provides only for fission product activity in the coolant of the Reactor Coolant System, and depends for it's use on the proposition that fission product j during a LOCA is released only at the time of fuel clad rupture as calculated to meet Appendix K reoutrements and that this fuel rupture is evaluated as generally occurring after 5-15 secs. into the event by which time the purge valves are closed. Based on existing Pegulatory positions and the review of the substantive test references reported elsewhere in his DP0/DPV, the writer proposes their can be no fundamental Regulatory and related Guidance support for this proposition. Also, that existing Regulations would require the offsite dose over the first seven seconds to be calculated at 158,000 rem. In the event non regulatory provision for no fuel melt was allowed this could be reduced to 64,000 rem; a nomally operating design basis fuel failure at 1% of the core alone would 634 rem which is over twice the Allowable value. The overwhelming importance of fuel failure and it's related fission product release' to offsite dose during the blowdown phase of a LOCA provides no margin for any uncertainty about surety of containment to prevent its release to the environment. 5.0 RISK PERSPECTIVES: In their sumary, the DP0 review comittee proposes "There is little incrernental risk associated with operation of a reactor such as Zion with the
l s purge valves open, as their is a likelihood that the valves (or at least one ) of the redundant valves) will close on demand." 10 Code of Federal Pegulations 50 are the minimum set of legal requirements of the United States of America (USA) to be met before a licensee can be issued for ] l the construction and operation of a nuclear power plant in the USA. As of this time their is no Regulatory provision in that Code which would allow the j writer as a Federal Employee to infringe those regulations, except by the issuance of an o emption from the regulations. In performing a Safety Evaluation for any facility for the express purpose of amending the releted operating licensee, the writer is obliged by law to see that these regulations are met. 11 is the writer's proposition that Probability Risk Assessments (PRA's) can be used to show relative improvements which are possible from amendments to plant engineering and operations but that any such changes must i result in a facility which still meets as a minimum the requirements of the current Regulations. This is a primary subst A.ive element of the only ) Regulatory Policy Position the writer is aware of in this matter; namely the " Safety Goals for the Operations of Nuclear Power Plants; Policy Statement" in the Federal Register to Reference 9. Further, it may not be generally recognized that the current licensing basis for all commercial facilities already has a clearly identifiable and stated risk package incorporated into it's license and which is based on both regulatory, and related regulatory guidance. For all pressurized water
k 34 reactors (PPP's) the detail is providtd in the beginning sections of Section 15 (or 14) of the related Final Safety Analyses Report. The structure of the risk criteria derives from the fact that all facilities are required to be designed to be able to withstand 4 different sets of occurrences (events) of increasing operational severity, varying from Condition 1 occurrences called nontal operating transients, through to Condition 4 occurrences which are the postulated major accidents. Each cf these 4 sets has a specified frequency of L occurrence which in general decreases with increasing severance of the event and together with related increases in allowable offsite dose. And the plant is required by law to meet the related multi-faceted safety criteria and safety limits. For a LOCA, their are specific regulatory requirements under 10 CFR Part 100 which calls for protection against a hypothetical accident designated as the design bases accident and which is specified to occur at least once in the 1f fetime of the licensed facility, and to ensure that related offsite doses calculated on the basis of the conservative criteria called for in the regulations and guidelines limit the offsite exposures to criteria and related specified values of 10 CFR 100. For the case of Iodine, which has been used as an example onli in this DP0/DPV, the offsite dose limit at the site boundary is limited to 300 rem at the boundary of the exclusion zone over the first 2 hours. These risk elements define the bcsis for all the hard real engineering design end construction without which the Zion 1/2 facility would never have be built and operated. It has often been proposed for consideration, that our nuclear power facilities are conservatively designed and that because of this we,can make licensing decisions outside the framework of the existing regulations and
l ++ A I j l 1 including those derived from probability studies. This position is not generally valid and especially in the calculation of expected approaches to licensing basis safety limits for the design basis transient and accidents of Section 15 or (14). The so called conservatism is not really conservative in the sense that we can accurately calculate expected values of our safety criteria (during all design basis occurrences), and we then multiply that value by a factor of safety to establish a design point for which we will ultimately design the engineering features of the facility so providing a large real margin. In fact, what we have is a very complexed geometry nuclear steam supply system with the further complex set of interacting themal hydraulic and nuclear core and plant process and protective elements, And we have to physically, and then mathematically, model the equipment and the imposed four (4) sets of transients and accidents, in an attempt to calculate the dynamic responses for this system so that we could know how to ultimately protect the facility to achieve acceptable varying levels of fission product release. The substantive limitetions in our modelling capabilities has caused us to use conservative models for which we can calculate plant responses which will be within the safety analyses limits for the safety criteria -- but for which we do not know the available margins to these safety limits. So the proposition of large available safety margins is empty of substance because we cannot calculate what they really are. As improved modelling capability is developed, the more uncertain models are replaced and so the real margins will be reduced, but to l what actual extent remains unknown.
, a The Independent Review Committees Risk Perspective, provides evaluations and proposals, with varying degrees of admissibility into the licensing process. The writer proposes, that fundamentally it appears that the risk study cannot deal with what is effectively a 6.93 square ft, opening at the initiation of the LOCA event, and which could blow down the whole of containment to the environment in matter of minutes, if the valves failed to close. Whilst I respect the information provided by the Committee, the real significance of large open containment isolation valves in reaching hard decisions on the safe engineering, construction and operation of this facility can only be reached by the type of licensing basis evaluation presented on page 1-6 cf his DPV/0P0 in which it is revealed that for a technical specifications leakage limit of 0.3% per day from containment, the use of large open containment isolation valves closing inside seven (7) seconds would increase the offsite dose by a factor of 541,000 over the first seven seconds and by 32000 over the first P hours. For a fission product release from the clad gap of 10f core inventory, this would cause offsite exposure due to I 131 to increase from virtually 0 to 63, 400 rem; and to be in conformance with Regulatory / Guidance requirements, this would increase further to 158,000 rem. Summarily, it should be recognized that the prime concern of the writer is not the failure to close of these valves but the failure to keep them closed during Modes 1, 2, 3, and 4 so that they remain fully open to discharge the l contents of containment to the environment for the first 7 seconds of blowdown. The writer's DP0/0PV proposes, that on a Regulator:. basis any permission to l l grant an Amendment allowing the valves to remain olen in these Modes for subsequent successful automatic isolation within 7 seconds will give calculated offsite doses during that 7 seconds, far in excess of allowable values and is thereby unacceptable. Furthermore, any other decision based on probability
17 4 4 estir'ates of contribution to totality of risk from all potential contributors at the facility would result in offsite doses exceeding these allowable values and thereby in violation of Regulatory recuirements and thereby the current " Safety Goals for the Operations of Nuclear Power plants; Policy Statement" in the Federal Register to Reference 11. 6.0 CtirRENT NRC Gl'IDANCE AND p0LICY ON THE OPERATION OF CONTAINMENT PURGE ISOLATION VALVES In their sumary report. the DP0 review comittee also concludes that "There setns to be some confusion as to the regulatory tirnes and conditions for operation at these large valves." (We noted that the issue may be moot; Zion does not actually envision routine operation with valves open, we were infonned). Perhaps the need to routinely purge reflects some basis design deficiency which should be addressed. In any case, it is clear that the containment leak tight criterion is better served if one does not have j to open large butterfly valves from time to time, or leave them open indefinitely. In our opinion, NRR should reexamine policy in this area." The writer's DPV provides the current SRP position that very large lines penetrating containment (about 42 inches in dia), should be restricted to Cold Shut Down conditions and refueling operation and they n r.t be seal closed in all other operational modes. Also that purge system designs thit ere acceptable for use on a non routine basis during normal plant operations can be achieved by providing additional purge lines of a size that should limit LOCA radiological consequenced calculated in accordance with Reg. Guides 1.3 & l.4 not to exceed the 10 CTR 100 Guidelines. ("Also the maximum time for valves
c' ' closure should not exceed five seconds to assure that the purge valves would be closed before the onset of fuel failure following following a LOCA"). Simil6r concerns apply to vent designs. This has been interpreted by the writer as specifying that the large 42 inch valves should be closed except in Modes 1, 2, 3, and 4: And if purging is necessary in these same Modes 1, 2, 3, and 4, then smaller lines (8 inch and 10 inch) should be considered and the source to be use for evaluating offsite dose is that of R.G. 1.4 In fact, the Zion FSAR does not provide for operation of these smaller lines known as the Mini-Purge System in Modes 1, 2, 3, and 4 as they are only licensed to be j used when radioactive contamination inside the containment during use of the large purge lines in Modes 5 and 6 is excessive. In respect of the 5 sec. closing time criteria of SRP 6.2.4, BTP CSB 6 8 the writer has shown that this is less conservative than the Reg. Guide 1.4.C.I.a criterion that the related source term "should be immediately available for leakage from the primary containment," and the SRP 4.2 criteria that release of the related source. term occurs on DNBR, at 1/10 sec into the event: And that the most non-conservative proposition which appears nowhere in current regulatory ) guidance is that that the related fission product release occurs on rupture of fuel elements calculated according to Appendix r. requirements which do not l include any specific mention of calculating time for fission product release, let alone a necessarily conservative estimate of such timing, but which calculates times to rupture of some 35 seconds. Closing times of up to 15 sec's have also been proposed and this had been addressed by the writer in his DP0/DPV material and shown to be invalid for the particular circumstances of the LOCA event. The writer also draws attention to the very significant fact that if the valves are open on a LOCA the regulatory requirenents and practice for
19 nultiple barrier protection against the release of high concentrations of fission products is infringed by having only one uncertain barrier between the catastrophic levels of gap activity and the environment whilst the valves so ren.ain open, end is thereby in violation of necessary regulatory requirements. Also, that because of these same physical circumstances, the non conservatism of any assumption involving release of fission products into an open containment is dramatically increased simply by having these valves open as a sterting condition as already shown by the by the writer's offsite dose calculations; Ref. 3. i There is a provision in the SRP for allowing offsite doses greater than 10 CFP. 100.11 limits, providing that the valves are open no more than i total of 90 hrs. This is an obvious violation of 10 CFR 50.36 Requirements for Technical Specifications which require that the plant always be in a fully protected condition. As a consequence, it is also in violation of the Safety Goal Policy Ref. 9, as also discussed elsewhere in this response, it has to be recognized that the current design basis and the current licensing basis of the Zion does not allow these large purge valves to remain open in Modes 1, 2, 3, and 4. Furthermore that there has not been any application approved for any amendment to such operation based on offsite dose considerations: and therefore any such operation which may have been taken by the facility would be in violation of their license. The facility does provide engineering systems which ultimately allow necessary entries into the containment without infringing containment integrity by the need to open these large valves and thereby that arguments relevant to the release of fission ~
s ,g9, product in the early phases of a LOCA are moot and that potential risks of the related uncertainties by allowing them to remain open are completely unacceptable: For Zion Units 162, the following sumarizes the specific Licensing Design Bases for Containment Purge, Containment Ventilation, Containment Pressure and Vacuum Relief, and Personnel Access to Containment as Limited by Airborne and Direct Radiation. For Zion 1&2, full details of the licensing basis provisions for, and constraints upon accessing containment whilst maintaining containment integrity are described in the related UFSAP Section 9.10. These include the Containment Purge System Section 9.10.1-1 to assure safe continuous access to containment within 3 hours after a planned or unplanned reactor shutdown, and the mini-purge system for use when the containment activity is too high to use the large purge valves. For operations at Rated Power, a containment ventilation system is provided to maintain environmental temperatures to a maximum of 120"F through a Reactor Containment Fan Cooler System (Section 9.10.3.3.1), and related Containment Activated Charcoal Filter Units located i insidecontainment(Section 9.10.3.3.2) are provided for fission product cleanup of the atmosphere prior to personnel access; additionally a Pressure and Vacuum Relief System (Section 9.10.3.3.7) is provided to handle the nonnal pressure changes in the containment which result from containment air temperature changes, barometric pressure changes, instrument air bleeds, and inleafage from the penetration pressurization system. Further in respect of protection against direct radiation, Section 11.2.2.1 of the UFSAR provides that shielding is designed for two hours per week access to the cyl,indrical annular region between the crane wall and the outer wall of the building. At
a' 21 i elevation 616 feet, directly above this annular region, the access time is one hour per week. Areas near the edge of the refueling pool above the reactor vessel and within the primary system equipment compartment are high radiation areas which are nomally inaccessible during reactor power operation. Therefore, if there is a need to purge centainment during power operations in spite of the above provisions, it must be concluded that their is a design and/or operating deficiency in one or other of these normally operating systems which must be corrected, instead of exposing Public Health and Safety to potentially serious offsite dose consequences in excess of 10 CFR 100 limits. Further, any deficiency in the Peactor Containment Fan Cooler System is ceuse for additional concern as it is also a Safety Related System for mitigating pressure rise in the containment in the early phase of the LOCA which could thereby also be adversely impacted. This infonnation confims the position that the purge valves for Zion 1&2 are to remain closed in Modes 1 l 2, 3, and 4, and that provision has already been made for access to containment, l and those other features of plant operations, nomally proposed for their use. Summarily, the conclusion by the DP0 review connitteei of confusion, is not an uncommon regulatory experience deriving from multiple criteria setting different safety limits from which a conservative approach to containment purge valves based on regulatory requirements for fission product release does reveal the most limitirg set of conditions which must be applied, and which l l. has been evaluated and presented by the writer. linfortunately, much of the l SRP on containment purge valve isolation tines and thereby extended use into l 1 l Modes 1, 2, 3, and 4 has been predicted r,n fission product and related release l criteria which are much less conservative then the Reg. Guide 1.4 criteria, l and also SRP 4.2 criteria for fuel clad hermiticity, without any appropriate
r s .s s' 27 safety evaluation for departure therefrom. Furthermore, I also confinn by detailed reference that Zion was designed and licensed so that the containment purce valves are to remain closed in Modes 1, 2, 3, and 4 and that for plant operations at 100% rated power provision has already been made for access to containment, and those other features of plant operations, nonnally proposed i as a reason to violate this requirement; and it is the writer's experience that this may also apply to other Westinghouse facilities. Thereby, the writer proposes that the current related uncertainties in fission product release and the potentially excessive offsite doses arising therefrom, could never justify an alternate proposal to allow them to remain open at this time,
7.0 CONCLUSION
The Zion reactor with a licensed maximum power rating for LOCA calculations of 17.9 hi/ft and fuel designed to be at a clad pressure of 1400 to 1700 psig at normal full power conditions, gives calculated temperatures over the first seven seconds of a LOCA blowdown, of 1500 - 1700'F, and " evaluated" pressure differences across the clad which would result in fuel ruptures. Furthermore, the overwhelming importance of fuel clad failure, both by DNBR infringement and related loss of hermeticity, and or clad rupture, and it's related fission product release to offsite dose during the blowdown phase of a LOCA provides l no margin for any uncertainty about surety of centainment to prevent it's release to the environnent. The writer proposes that for Zion there is a sufficient probability that fuel failure will occur to the extent necessary to justify his primary position that the containment purge valves l should thereby remain closed in operational Modes 1 (power), 2 (startup), 3 (hot standby), and 4 (hot shutdown). Furthermore, it has been confirmed by detailed reference that the Zion facility was designed and licensed so that
i 23 i i the containment purge valves are to remain closed in Modes 1, 2, 3, and 4 and that provision has already been inade for access to containment, and those l other features of plant operations being currently proposed as a reason to amend this requirement at an unacceptable risk to Public Health and Safety. ) l l l l 1 l i i I l l l l l i t
List of Peforencer, 2. Memorandum to Dr. T. Murley from R. Licciardo dated May 11, 1989,
Subject:
Differing Professional View Concerning a) Issuance of SER to 7 ton 1/2 Allowing Full Power Operation with Open 42* Conta'inment Isolation Valves, b) Methodology Used for Celculating Related Offsite Doses. 2. Memorandurn to Dr. J. Taylor from R. Licciardo dated October 19, 1989, Sub.iect: Differing Professional Opinion Concerning a) Zion 1/2 Containment Isoletion Yalves, and b) Methodology Used for Calculating Offsite Doses. ] 3. Pemorandum to F. Miraglia from R. Licciardo dated July 20, 1989,
Subject:
Differing Professional View Concerning Containment Isolation Valves at Zion, with enclosure entitled: "An Evaluation of the Criterion for and the Calculation of Offsite Doses Deriving from Open Containment Purge Ya,1ves During a LOCA at Zion Units 1/2, July 20,1989. 4. Memorandum to R. Licciardo from Dr. T. E. Murley dated September 13, 1989,
Subject:
Differing Professional View. 5. Memorandum to R. Licciardo from Dr. J. M. Taylor dated January 2, 1990,
Subject:
Disposition of Differing Professional Opinion - An Independent. Outside. Qualified Review. 6. L. J. Stefken, (personal communication to G. N. Lauben), dated July 18, 1989
Subject:
Calculation of Response of Fuel Rod in Zion Rea'ctor During targe Break LOCA.
7. N. Zuber, et, al., Date: December,1989,
Subject:
Quantifying Reactor Safety Margins: Application of Code Scaling Applicability, and Uncertainty Evaluation Methodology to a Large Break Loss of Coolant Accident, NUREG/CR-5249. EGG-2552. 8. Memorandum to W. Hodges from G. N. Lauben dated August 21, 1909,
Subject:
Conynents on a DPV Concerning Early Blowdown Cladding Rupture During a large Breat LOCA. 9. Page 28044 Federal Register, Volume 51, No. 149, Monday August 4, 1986, Rules and Regulations Nuclear Regulatory Comission 10 CFR Part 50, Safety Goals for the Operations of Nuclear Power Plants: Policy Statement.
- 10. OCED LOFT-T-3703: Best Estimate Prevention for OECD LOFT Project:
Fission Product Experiment LP-FP-1.
- 11. Janes M. Broughton, et. al., PET LOCA Test Series, Tests LOC-3 and LOC-5 Fuel Behavior Report, NUREG/CR-2073, June 1981, 12.
J. M. Broughton, et al., *PBF LOCA Test LOC-6, Fuel Behavior Report, " NUREG/CR-3184. April 1983. 13. L. J. Stefken, " Development Assessment of FRAP-T6" Interim Report No. EGG-CDAP-5439, May 1981.
, i 14. D. L. Hagrwan, Zircaloy Cladding Shape At failure (BALON-?), EGG-CDAP-5379, July, 1981. 15, Memorandum to Reactor Systems Branch Engineers Division of Systems i Technology, from Robert Jones Acting Chief,
Subject:
Branch Review Guidance: Review of Fuel Pin Parameters for LC0A Analyes.
- 16. Charles L. Nalezny, Sumary of Nuclear Regulatory Comission's LOFT Program Research Findings, NUREG/CR-3005, Jur.e.1983
- 17. Memorandum to Dr. T. Murley from Dr. J. Taylor dated January 8,1990,
Subject:
Operational Usage of Large Purge System Yalves - PWRs; Re-exaniination of NRR's Safety Policy and Practices. l __}}