ML19289C986
| ML19289C986 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 12/31/1978 |
| From: | Fleischaker D, Kristovich S, Phillips J CENTER FOR LAW IN THE PUBLIC INTEREST, FLEISCHAKER, D.S. |
| To: | |
| References | |
| NUDOCS 7901290080 | |
| Download: ML19289C986 (54) | |
Text
'
THIS DOCUMENT CONTAlHS
~
g pg y w ~ ~~
P00!? QUAUTY PAGES 4
j UNITED STATES OF AMERICK ~
ge
{
f NUCLEAR REGULATORY COMMISSION f
\\ gy Before The Atomic Safety & Licensing Appeal Board {
p$<k'
.,g L
c, 4
%/
a m
In t:1e Matter of:
)
)
PACIFIC GAS & ELECTRIC COMPANY
)
Docket Nos.50-27T LL (Diablo Canyon Nuclear Power
)
50-323 OL Plant, Units 1 & 2)
)
JOINT INTERVENORS' PIQUEST FOR DIRECTED CERTIFICATION Pursuant to 10 CFR 2.718 (i), Joint Intervenors, SCENIC SHORELINE PRESERVATION CONFERENCE, INC., SAN LUIS OBISPO MOTHERS FOR PEACE, SANDRA SILVER, ECOLOGY ACTION CLUB, and JOHN J.
FORSTER, request the Appeal Board to direct the certification of the followiAg legal issue for its considera-tion:
whether exceptional circumstances exist which would permit the issuance of subpoenas, pursuant to 10 CFR 2.720(h),
to Dr. Mihailo Trifunac and Dr. Enrique Luco, consultants to the Advisory Committee On Reactor Safeguards (ACRS) in that Committee's review of the operating license for the Diablo Canyon Nuclear Power Plant (DCNPP) in order to require their attendance and testimony a' the licensing proceedings before the Atomic Safety & Licensing Board (ASLB).
I.
On September 1, 1978, the Joint Intervenors requested that subpoenas be issued for two experts who consulted to 79012900[o h
2.
the ACRS in that Committee's review of the reanalysis of the seismic design of the Diablo Canyon Nuclear Power Plant.
Those two experts were Dr. Mihailo Trifunac, an associate professor of civil engineering at the University of Southern California, and Dr. Enrique Luco, an associate professor of applied mechanics at the University of California at San Diego.
Both the Nuclear Regulatory Commission Staff (Staff) and Pacific Gas & Electric Company (Applicant) opposed the Joint Intervenors' request.
On October 10, 1978, the ASLB issued an order deferring ruling on the Joint Intervenors' request to subpoena Drs.
Trifunac and Luco, noting that the Commission was in the process of establishing a general policy on the amenability of ACRS consultants to licensing board subpoenas.
At the same time, the ASLB informed the Office of General Counsel that this issue was pending in the proceeding and requested an expeditious resolution of the general policy question.
On October 20, 1978, the Commission asked the parties to this proceeding and one other, as well as the Office of Executive Legal Director, to submit comments on the general policy question regarding the amenability of ACRS consultants to. licensing board subpoenas.
The ACRS advise: the Commission of its position in a joint session with the Commission
5 g
3.
Y on November 3, 1978.
On November 29, 1978, the Commission issued a document entitled INTERPRETATIVE COMMISSION STATMENT ON AMENABILITY TO SUBPOENA OF CONSULTANTS TO u'HE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS UNDER 10 CFR 2.720.
There the Commission recognized that its decision must balance two legitimate and possibly conflicting considerations:
on the one hand, ACRS consultants represent an important source of scientific opinion useful to licensing boards in resolving technical issues; on the other hand, making those consultants amenable to subpoena might impair the ACRS' ability to obtain qualified consultants hesitant to expose themselves to the " hassle" of appearances before licensing boards.
The balance was struck
-1/
The CRS took the position that its consultants should enjo, the same degree of immunity from subpoena as the members of the ACRS and, in any case, should not be subpoenaed unless it could be shown that the consultants have unique possession of material facts, that is, f act s not known to witnesses made available t o the Executive Director of Operations.
The Applicant endorsed the ACRS' position.
The Office of Executive Legal Director took the position that ACRS consultants should be subpoenaed only where they have unique possession of materials facts.
The Joint Intervenors argued that ACRS consultants should be subpoenaed when their view would materially assist in the full development of a record, and in the alternative, if the exceptional circumstances test were to apply, it should act be limited to those instances where the potential witness had unique possession of material facts, as urged by the Office of Executive Legal Director.
4.
by providing the ACRS consultants the limited protection of the exceptional circumstances test.
The Commission directed licensing boards to administer the generally worded " exceptional circumstances" test on a case-by-case basis.
In addition, the Commission provided the following guidelines in its interpretative statement:
By the terms of the rule, " exceptional circumstances" are not limited to situa-tions in which the particular individual has unique knowledge of facts.
Further-more, the mere fact that a particular person may have been an ACRS consultant in a particular case does not mean that
" exceptional circumstances" must be shown as a predicate for a subpoena to that person in another case.
Th _ : limitation applies only in cases in whir.1 he has served as a consultant.
Finally, parties may not seek to probe the reasoning pre-cess underlying the collegial ACRS report through the device of a subpoena to a consultant.
2/
On December 6, 1978, the ASLB heard oral argument on the subpoena request in light of the Commission's recently issued policy statement. -3/
On December 8, 19 78, the AS LB, without providing any explanation for its decision, found that exceptional circum-stances had not been established and denied the Joint Intervenors' request for a subpoena.
Instead, as recommended by the Applicant, the ASLB incorporated into the record all written comments submitted by Drs. Trifunac and Luco to the ASLB.
2/
INTERPRETATTIVE CO.9:ISSION STATEPENT ON AMENABILITY TO SUBPOENA OF CONSULTANTS TO THE ADVISORY CO.91ITTEE ON REACTOR SAFEGUARDS UNDER 10 CFR 2.720, at 2-3.
3/
TRANSCRIPT, at 4273-4333.
5.
II.
The request denied by the ASLB is for the issuance of subpoenas to two acknowledged experts who have criticized strongly the Staff and Applicant reanalysis of the adequacy of the seismic design of the DCNPP.
That reanalysis is the central safety issue in the operating license proceeding.
The facts surrounding the reanalysis are set out below in order to place the question of the issuance of the subpoenas in proper perspective.
Construction on both units of the DCNPP was well underway when the existence of a geologic fault offshore and running within 2-1/2 miles of the DCNPP was confirmed. -4/
This fault -
the Hosgri fault - was initially d.iscovered by two exploration geologists working for Shell Oil Company in the late 1960's.
In 1971-1972, a consultant for the Applicant became aware of reports of the fault in the scientific literature.
By mid-1973, the existence of the Hosgri fault was disclosed to the NRC Staff in an amendment to the FSAR. -5/
At this time, the full significance of the Hosgri fault was not known; however, the NRC Staff requested the U.S. Geological Survey (USGS) to assess the earthquake potential of the fault.
By late 1975, after extensive study, the USGS concluded that the Hosgri fault could sustain a 7.5 magnitude earthquake. -6/
4/
A construction permit was issued for Unit 1 in 1968; for Unit 2, in 1970.
-5/
Application for operating licenses for both Units 1 and 2 were made in July, 1973.
-6/
Supplement No. 4 to the Safety Evaluation Report, Appendix C (May 1, 1976) [ cited hereinafter as SER].
7.
sufficient to bring the design up to a 0.6-0.7g capability are impractical.
The design reanalysis could take up to a year or two to completa. 9/
The Staff concerns were echoed in another internal memorandum, dated February 20, 1975:
2.
The as-built plant has significant margins in its des gn and it is capable of resisting c
seismic loads well in excess of those associated with the 0.4g seismic event assumed for design.
The staff is presently convinced that the design is " good" for a 0.5g event.
However, to convince the ACRS and others of this will require 3 -
6 months of applicant and staff effort.
If an event in the order of 0.6g need be con-sidered, it will be possible to show that many parts of the plant can safely resist such 1 sds; however, many parts will need to be r.)dified and very likely for some of these the modifications will not be practical.
The evaluations to accomplish such a task, with the rigor that will be required, will entail years of applicant and staff effort.
3.
The " tone" of the questions and comments from the ACRS Subcommittee members and the consultants indicated to me that they would find a 0.5g value acceptable and could be convinced that the as-built plant could adequately resist the increased loads associated with the event.
However, they are not likely to be convinced unless a rather complete analysis is performed for a 0.5g event and the use of our currently approved seismic design criteria.
To date no such analysis is available; the present 9/
See, Attachment 1
8.
positions of the applicant and the staff as to the ability of the plant to adequately resist a 0.5g event are based on qualitative assessments supported by a few typical cal-culations. 10/
In January, 1976, the Staff found itself in practically the same position.
The USGS was maintaining its position that the appropriate SSE value for the DCNPP was a 7.5 magni-tude earthquake on the Hosgri fault.
Neither the Staff nor the Applicant had performed an analysis to determine whether the safety features of DCNPP could withstand the effects of a 7.5 magnitude earthquake.
One thing had changed, however:
11/
~
~~
Unit 1 of the DCNPP was almost ready for fuel loading.
The NRC Staff was without a rationale to support issu-ance of an operating license.
An internal memorandum of January 12, 1978 sets forth the alternative courses that the Staff could pursue:
1.
Formally request that an independent review be conducted within USGS to confirm or modify the current USGS review team finding.
Thic request might be made by Chairman Anders to the Secretary of the Interior and could, if legal considerations demand, be based on generic West Coast concerns rather than the Diablo Canyon issue alone.
2.
Concurrently, form a team of consultants consisting of men of national stature in the fields of geology, seismology, and seismic design to evaluate the situation.
This team might pursue various paths, including:
_1_0/
See, Attachment 2
_1_1/
Supra
9.
a.
Assume that a " great" earthquake could occur as the USGS review implies and demonstrate by logic, evidence, and judge-ment that the energy transfer to the site would be limited to within the design capa-bilities of the nuclear plant.
- b. An independent evaluation of the evidence to arrive at a conclusion on the SSE that might be compared to those of the USGS review team and the panel of experts established by licensee.
- c. Assess the prospects for upgrading the plant design to higher SSE values by plastic analysis and/or structural modifications and testing.
3.
Concurrently, form a task force to review the current status in an attempt to deter-mine if a probabilistic basis can be estab-lished to license Unit 1 for an interim period of operation while the other reviews are being conducted.
4.
Concurrently, inform the licensee of the course of action we are pursuing and require him to pursue similar and/or alternative courses so that our final decisions may be made on the basis of our evaluation of his efforts supplemented by our independent assessments.
12/
As it turns out, the principle basis for the Staff position is an analysis that accepts a 7.5 M earthquake on the Hosgri fault, and then seeks to demonstrate that, for the most part, the facilities response will be limited to l_3/
within the original design capabilities.
_1_2/
See, Attachment 3 13/
SER, Supp.
4, Section 2.5, 2.8, App C.
With the exception of structural modification to the turbine building, the replacement of numerous pipe-hangers, modification to safety related structures systems and components has been limited.
SER, Supp.
7, Section 3.0-3.10.
10.
This analysis is principally the work of a single man -
Dr.
N.M.
Newmark.
On his advice, 0.75g was designated the
" effective acceleration" to be used as the zero period limit to scale the response spectra for the reanalysis.
The spectra were then reduced by varying amounts for those structures having large foundations (the so-called " tau effect"), e.g., the containment building, the turbine building, etc.
Finally, the reanalysis assumes 7% damping in the struc-tures.
The net result is response spectra with reduced zero period limits equivalent to 0.5g to 0.55g. - the values identified by the Staff in 1975 as the maximum values that would not neces-14/
sita e " impractical" modifications.
As the ACRS noted, both selection of 0.75g as the zero period limit for the design response spectra and the reduction for the effects of building size are based largely on Dr.
Newmark's " judgement" rather than empirical data and analysis.
Furthermore, use of the tau reduction effect is unique in nuclear power plant design.
Since there are expected to be significant dif-ferences between the nature of the ground motions close to a large earthquake and that at greater distances, to which most available data apply, the Staff relied heavily on the experience and judgement of its consultant, N.M. Newmark, an acknowledged expert in the field of earthquake engineering and structural design.
On his advice, an effective zero-period acceleration of 0.75g was used to determine the free-field response spectra to be used in engineering analyses.
These spectra were then reduced by varying amounts to obtain spectra for those structures
~~14/
The re ference is to more than one response spectrum be-cause there is one for each major structure.
10.
This analysis is principally the work of a single man -
Dr.
N.M.
Newmark.
On his advice, 0.75g was designated the
" effective acceleration" to be used as tne zero period limit to scale the response spectra for the reanalysis.
The spectra were then reduced by varying amounts for those structures having large foundations (the so-called " tau effect"),
e.g.,
the containment building, the turbine building, etc.
Finally, the reanalysis assumes 7% damping in the struc-tures.
The net result is response spectra with reduced zero period limits equivalent to 0.5g to 0.55g. - tha values identified by the Staff in 1975 as the maximum values that would not neces-sitate " impractical" modifications. --14/
As t'ne ACRS noted, both selection of 0.75g as the zero period limit for the design response spectra and the reduction for the effects of building size are based largely on Dr.
Newmark's " judgement" rather than empirical data and analysis.
Furthermore, use of the tau reduction effect is unique in nuclear power plant design.
Since there are expected to be significant dif-ferences between the nature of the ground motions close to a large earthquake and that at greater distances, to which most available data apply, the Staff relied heavily on the experience and judgement of its consultant, N.M. Newmark, an acknowledged expert in the field of earthquake engineering and structural design.
On his advice, an effective zero-period acceleration of 0.759 was used to determine the free-field response spectra to be used in engineering analyses.
These spectra were then reduced by varying amounts to obtain spectra for those structures
--14/
The reference is to more than one response spectrum be-cause there is one for each major structure.
I t
11.
in the plant having foundations extending over large areas.
This reduction for the effects of building size is also based largely on judge-ment and experience rather than on extensive observations and analyses and has not heretofore been applied in the design of nuclear power plants.
15/
rinally, use of a 7% damping factor, while permitted by Regulatory Guide 1.61, is higher and thus less conservative 16/
than the damping factors used in the original design. --
III.
Following disclosure of the existence of the Hosgri fault, the ACRS began reconsideration of the DCNPP seismic design in September, 1974.
Thereafter, over the course of 3-1/2 years, the ACRS held several meetings on the matter.
Drs. Trifunac and Luco were employed early-on by the ACRS 17/
to assist in the review of the Staff and Applicant reanalysis. --
18/
From the beginning, --
these two experts raised substantial
--15/
Letter from Stephan Lawroski, Chairman, ACRS, to Joseph M.
Hendrie, Chairman, NRC, dated July 14, 1978, SER Supp.
8, App. B [ cited hereinafter as SER, dapp.
9, App. B].
16/
Supra
--17/
The Applicant asserts that it conducted an independent reanalysis.
The Staff, however, has taken the position that the Newmark spectra are to control the reanalysis.
Furthermore, it was Dr. Newmark, not the Applicant's consultants, who selected the 0.75g value for the effec-tive acceleration, introduced the tau reduction effect,
' e 7% damping in the reanalysis.
and approved use of
--18/
Dr. Trifunac and Dr. Luco's criticisms and recommenda-tions are set forth in their first written comments to the ACRS, submitted in October and November, 1976.
Their criticisms remained essentially the same through-out the review.
11.
in the plant having foundations extending over large areas.
This reduction for the effects of building size is also based largely on judge-ment and experience rather than on extensive observations and analyses and has not heretofore been applied in the design of nuclear power plants.
15/
Finally, use of a 7% damping factor, while permitted by Regulatory Guide 1.61, ir higher and thus less conservative than the damping factors used in the original design. --16/
III.
Following disclosure of the existence of the Hosgri fault, the ACRS began reconsideration of the DCNPP seismic design in September, 1974.
Thereafter, over the course of 3-1/2 years, the ACRS held several meetings on the matter.
Drs. Trifunac and Luco were employed early-on by the ACRS to assist in the review of the Staff and Applicant reanalysis. --17/
18/
From the beginning, these two experts raised substantial
--15/
Letter from Stephan Lawroski, Chairman, ACRS, to Joseph M.
Hendrie, Chairman, NRC, dated July 14, 1978, SER Supp.
8, App. B [ cited hereinafter as SER, Supp.
8, App. B].
16/
Supra
--17/
The Applicant asserts that it conducted an independent reanalysis.
The Staff, hcwever, has taken the position that the Newmark spectra are to control the reanalysis.
Furthermore, it was Dr. Newmark, not the Applicant's censultancs, who selected the 0.75g value for the effec-tive acceleration, introduced the tau reduction effect, and approved use of the 7% damping in the reanalysis.
--18/
Dr. Trifunac and Dr. Luco's criticisms and recommenda-tions are set forth in their first written comments to the ACRS, submitted in October and Novemb ar,1976.
Their criticisms remained essentially the same through-out the review.
12.
questions regarding the adequacy of that reanalysis, and in particular, selection of 0.75g as the " effective acceleration" zero period limit, introduction of the tau reduction effect, and use of a 7% damping factor for the structure.
In essence Drs. Trifunac and Luco argued that the Staff and Applicant approach distorted the analysis and may, therefore, yield unreliable results.
With respect to use of an " effective acceleration value,"
Dr. Trifunac stated as follows:
The term ' effective peak acceleration' has been introdoced suggesting that the structure will see something smaller than actual peak accelerations.
Though such approach may be useful for earthquake-resistant design of ordinary structures by means of the response spectrum technique, the term ' effective peak acceleration' has not been defined in a way that would enable the derivation of consistent results by several different experts in the field.
Since the procedures for scaling Reg.
Guide 160 spectra are based on maximum vibra-tory acceleration, as defined in Appendix A, this departure from routine design practices makes it difficult to evaluate the number and the nature of the consequences which would result from such an approach, 19/
On the same subject, Dr. Luco made the following obser-vations:
A judgemental factor has been used to reduce the 1.15 peak acceleration recommended in USGS Circula 672 to a value of 0.75g.
This ill-defined factor has been used in the past to account for discrepancies in the level of damage.observeo as compared with the predic-tions of ordinary seismic analysis which do not account for the effects of soil-structure interaction are based on nominal values for damping and strength, assumed linear behavior
--19/
TRIFUNAC, COMME"TS ON SEISMIC DESIGN LEVELS FOR DIABLO CANYCN SITE IN CALIFORNIA, (April, 1978), at 2-3 [ cited hereinafter as TRIFUNAC] - Attachment 4.
13.
and do not include energy dissipation in parti-tions and other non-structural elements.
This
' catch-all' reduction factor has no place in the design of carefully analyzed structures, such as those in nuclear power plants. 20/
With respect to the tau reduction effect, Dr. Trifunac concludes that the approach adopted, (a) requires unrealistic assumptions; (b) results in a one-sided analysis considering only the reduction effects; and (c) requires reliance on assumptions inconsistent with those used to evaluate the
" effective acceleration."
Scattering and diffraction of high-frequency waves from the foundations of different plant structures has been proposed as a vehicle to justify reduction of high-frequency spectra amplitudes (tau effect).
The manner in which this reduction has been effected requires un-realistic assumptions, for example, that the foundation is rigid.
The manner in which this assumption is introduced into analysis is often one-sided and considers mainly only those con-sequences of the physical phenomena which lead to reduction of spectral amplitudes.
Other consequences of this phenomena, for example, torsional and rockinc excitation of foundations which may amplify the structural response have been so far either overlooked or treated inade-quately.
This has been achieved by utilization of dynamic models for analysis which are so defined that only an incomplete physics of the problem, i.e.,
seismic excitation and the structural response, can be considered.
Hypocentral rather ^aan distance closest to the fault have been ased to evaluate peak and effective peak acceleration.
This assumption implies certain angles of approach of seismic wave energy.
These angles of approach should then be consistent with the extent to which tau
~~20/
LUCO, REVIEW CF THE SEISMIC EVALUATION FOR POSTULATED 7.5 M HOSGRI EARTHQUAKE, UNITS 1& 2, DIABLO CANYON SITE (May 30, 1978), at 2-3 [ cited hereinafter as LUCO) -.
14.
effect is alloued to influence the spectral amplitude.
Little or no attention seems to hare been given to mutual consistency of these assump-tions, and in some cases, inconsistent assumptions have been utilized.
For example, hypocenter will increase the distance: at which peak acceleration is evaluated, thus reducing the estimated peak acceleration.
This would, however, also imply that waves arrive toward the foundation almost vertically.
In consideration of tau effect, however, horizontal dimensions of foundations appear to have been used implying horizontal incident waves. 21/
Dr. Luco similarly concludes that the Staff and Applicant approach for the tan reduction effect relys on the questionable assumption of horizontally incident waves and, further, that the approach fails to consider adequately the torsional excitation associated with horizontally incident waves.
L The high-feguency comp,nents of the free field motion have been reduced by the so-called
' tau filtering procedure' to account by the scattering of waves by the supposedly rigid foundation.
This correction amounts to a reduction of the Newmark free field design spectrum by 20s to 30% for frequencies higher than two cps, cycles per second.
Slightly lower reductions have been used in the Blume The correction for large foundation spectrum.
scattering effects is based on assumptions of a rigid foundation and horizontally propagating SH waves.
Although the assumption of a rigid foundation may be reasonable, it must be recog-nized that deviations from the assumption lead to localized higher stresses in the lower por-tions of the different structures.
The assump-tion of horizontally incident SH waves is highly questionable, considering that the epicentral distance is comparable with the focal length.
Under these conditions, the possibility of nearly vertically incident waves may not be ruled out.
For verticially incident waves, the scattering by the foundation is practically nonexistent, given the shallow embedment.
_2_1/
TRIFUNAC, at 2-3
15.
Assuming for the si.ke of the argument that t.t seismic excitation at the sites corresponds to horizontally incident SH waves, I find that the reductions proposed by Newmark and Blume are tr a high when compared with analytical solutions.
Far horizontally incident SH waves, the reduction of the translational componer 2, of motion is coupled with the existence of a marked torsion input to the structure.
The Applicant has included accidental eccen-tricities of five to seven percent to repre-sent these torsional effects, From the point of view of the analysis of the structural response, it does not seem adequate to introduce the torsional input through the use of accidental eccentricities.
Such procedure, which leads to the coupling of the torsional and translational response, in essentially symmetrical structures, distorts the response of the. natural frequencies of the system.
The effects of the torsional input may be significant for the turbine building in which the possibility of portion of the structure undergoing inelastic deformation may increase with eccentricity.
If it is shown that the seismic excitation of the site corresponds mainly to horizontally inci-dent waves, the reductions of the translational and torsional response should be evaluated on the basis of the more exact methods presently available.
To include an exaggerated reduction of the translational motion without incorporating the full torsional effects is improper. 22/
Jn the Staff and Applicant's reliance on a 7% damping factor, Dr. Trifunac concludes as follows:
_2_2/
LUCO, at 3-6
16.
The large damping equal to seven percent has been adopted for dynamic response calculations.
Though the apparent damping for the complete soil-structure system subjected to carthquake excitation may be larger than seven percent inadequate basis has been presented to justify seven percent damping in structural systems only.
Selection of too large structural damping coupled with only two dimensional or simple three dimen-sional analysis of soil-structure interaction can lead to unreliable response estimates. 23/
Finally, ov a. 2-s. Trifunac and Luco set out specific recommendations for the derivation of response spectra to be used in the reanalysis.
Their recommendations, almost iden-tical in content, are for an approach that, in their view, is
'24/
more reliable that the Staff and Applicant approach. --
IV.
Without having formally responded to the matters raised by Drs. Trifunac and Luco, the NRC Staff, in the Spring of 1977, gave the "go-ahead" to the Applicant to begin the reanalysis and, where necessary, the modification of the DCNPP using
_2_3/
TRIFUNAC, at 3-4 24/
The Applicant has sought to dismiss Dr. Trifunac's criti-cisms on the grounds that they are simply a matter of
" fine tuning," TRANSCRIPT, at 4312.
The Applicant argues that Dr. Trifunac fundamentally agrees with their posi-tion:
both Dr. Trifunac and the Applicant conclude that a 6.5 M earthquake should be designated as the SSE.
This argument is dissembling.
Designatio: of a magnitude for the SSE is only one of numerous steps that must be taken to derive response spectra for the reanalysis.
The bottom-line question then, is how would Dr. Trifunac's (cr Dr. Luco's) response spectra compare with those derived by the Applicant or Staff?
That comparison has never been made.
17.
the Newmark response spectra as the controlling design cri-teria. --25/
The ACRS continued its review of the facility, holding several meetings at which Drs. Trifunac and Luco con-tinued to question the Staff and Applicant methods.
At no time during the period of reanalysis and testimony did the ACRS issue any statement or document approving the criteria adopted by the Staff for the reanalysis.
On May 26, 1978, the Staff issued Supplement 7 to the Safety P74alus' ion Report, which reflected the Staff's conclu-sions resulting from two years of reanalysis and testing of the facility.
Essentially, the Staff concluded that DCNPP was ready to cperate. --26/ After holding a final set of hearings in June, 1978, the ACRS, on July 14, 1978, wrote to the Commission that "without endorsing all details of the NRC seismic design basis and criteria, the use of the Staff approach leads to an acceptable level of safety in this instance."
Nothwithstanding issuance of the ACRS sign-off, Dru.
Trifunac and Luco did not withdraw or change their comments regarding the defficiencies in the Staff and Applicant analysis.
25/
The Applicant was permitted to use the Blum (Applicant's consultant) spectra only where it exceeded the Newmark spectra.
Thus, Newmark's spectra set the minimum accep-table level of design.
--26/
Numerous items were left unresolved in Supp. 7.
The pendancy of these items, however, did not affect'the Staff's basic conclusicn that the DCNPP was ready to operate.
Many of the open items were l'ted " cleaned-up" in Supp.
8, issued in November, 197b three weeks before commencement of the evidentiary hearings.
17.
the Newmark response spectra as the controlling design cri-25/
teria. --
The ACRS continued its review of the facility, holding several meetings at which Drs. Trifunac and Luco con-tinued to question the Staff and Applicant methods.
At no time during the period of reanalysis and testimony did tne ACRS issue any statement or document approving the criteria adopted by the Staff for the reanalysis.
On May 26, 1978, the Staff issued Supplement 7 to the Safety Ev.4uation Report, which reflected the Staff's conclu-sions resulting from two years of reanalysis and testing of the facility.
Essentially, the Staff concluded that DCNPp was 26/
ready to operate.
After holding a final set of hearings in June, 1978, the ACRS, on July 14, 1978, wrote to the Commission that "without endorsing all details of the NRC seismic design basis and criteria, the use of the Staff approach leads to an acceptable level of safety in this instance."
Nothwithstanding issuance of the ACRS sign-off, Drs.
Trifunac and Luco did not withdraw or change their comnents regarding the defficiencies in the Staff and Applicant analysis.
25/
The Applicant was permitted to use the Blum (Applicant's consultant) spectra only where it exceeded the Newmark spectra.
Thus, Newmark's spectra set the minimum accep-table level of design.
26/
Numerous items were left unresolved in Supp. 7.
The pendancy of these items, however, did n.t affect'the Staff's basic conclusion that the DC"PP was ready to operate.
Many of the open items were lated " cleaned-up" in Supp.
8, issued in November, 1978, three weeks before commencement of the evidentiary hearings.
18 V.
The Appeal Board's authority to direct certification of a legal issue or a licensing board ruling is well established. --27/
It is equally well established that the authority will be used sparingly.
Normally, the Appeal Board will not monitor the day-to-day conduct of licensing proceedings through the directed certification of legal issues of licensing board
'28/
rulings. --
However, the Appeal Board will direct certifica-tion where a prompt decision is needed to prevent detriment to the public interest or to avoid unnecessary delay and expense.
Directed certification should be granted in this case.
In the first place, the public interest is best served by immediate review of the ASLB decision belot.
The central safety issue raised in this proceeding is the adequacy of DCNPP's seismic design.
This is an operating license pro-ceeding.
Commission rules make a licensing board decision, approving the application for an operating license, immediately effective. 10 CRF 2.764.
Unless a stay is later granted, this plant may well commence operating without due consideration 27/
See, 10 CFR 2. 718 (i) ;
Public Service Co. of New Hampshire (Seabrook Station, Units 1 & 21, ALAB-271, 1 NRC 478 (1975).
28/
Pacific Gas & Electric Co. (Diablo Canyon Nuclear Power Plant, Units 1 & 21, ALAB-514, 8 NRC (December 22, 1978);
Toledo Edison Co. (Davis Besse Nuclear" Power Station)
ALAB-300, 2 NRC 752 (1975);
Seabrook, supra, at 483.
18 V.
The. Appeal Board's authority to direct certification of a legal issue or a licensing board ruling is well established. --27/
It is equally well established that the authority will be used sparingly.
Normally, the Appeal Board will not monitor the day-to-day conduct of licensing proceedings through the directed certification of legal issues of licensing board rulings. ~~2S/
However, the Appeal Board will direct certifica-tion where a prc=pt decision is needed to prevent detriment to the public interest or to avoid unnecessary delay and expense.
Directed certification should be granted in this case.
In the first place, the public interest is best served by i==ediate review of the ASLB decision below.
The central safety issue raised in this proceeding is the adequacy of DCNPP's seismic design.
This is an operating license pro-ceeding.
Co==ission rules make a licensing board decision, approving the application for an operating license, immediately effective. 10 CRF 2.764.
Unless a stay is later granted, this plant may well co==ence operating without due consideration
--27/
See, 10 CFR 2.718 (i) ;
Public Service CO. of New Hampshire (Seabrook Station, Units 1 & 2), ALA3-271, 1 NRC 475 (1975).
--2S/
Pacific Gas & Electric Co. (Diabic Canycn Nuclear Power Plant, Units 1 & 21, ALAS-514, S NRC (Decenter 22, 1978);
Toledo Edisen Co. (Davis Besse Nuclear Power Station)
ALA3-300, 2 SRC 751 (1975);
Seabrock, supra, at 483.
19.
having been given to the opinion of two experts, who are as knowledgable as anyone on the critical safety issue facing this facility.
The instant the facility goes critical, large numbers of systems and components, essential to safety, become radioactive.
The contamination of those systems and components will figure in any future calculus relating to seismic design.
Despite paper assurances to the contrary, common experience teaches that radioactive contamination will brunt otherwise compelling arguments that design modifications are needed.
The ASLB's solution - incorporating into the record the written comments of Drs. Trifunac and Luco - does not satisfy the need to consider fully and fairly the opinions of these two experts. As the transcript makes clear, each party has its own interpretation as to the meaning of the Trifunac and Luco comments. --29/
There is no reason why, on the face of the matter, the ASLB should give more weight to one party's interpretation than to another's.
The only way to give fair and full consideration to the opinions of Drs. Trifunac and Luco is to subpoena them to the proceeding, permit all parties (and the ASLB) to examine them, and let them speak for them-selves.
29/
TRANSCRIPT, at 4312-14
20.
Second, immediate review of the ASLB decision may well avoid unnecessary delay and expense.
The arguments in favor of subpoenaing Drs. Trifunac and Luco are compelling.
- First, the DCNPP is to be licensed on the basis of an analysis that seeks to demonstrate that, with few modifications, the as-built plant can withstand an earthquake substantially larger than that to which it was originally designed and constructed.
Second, Drs. Trifunac and Luco, both intimately familiar with the facts surrounding the DNCPP seismic issue and both acknow-ledged experts in earthquake engineering, have criticized methods and assumptions essential to that analysis.
- Third, assuring the adequacy of the DCNPP seismic design is extremely important.
If'the Appeal Board refuses to review this matter now, it will likely face the issue later - most probably in the context of a motion to stay operation of the facility.
Should the Joint Intervenors position prevail, as we believe it must, the case will have to be remanded.
Susbstantial time and expense that could have been saved by early resolution of the issue, will, instead, have been lost.
We are mindful that the policy that discourages Appeal Board review of "interlocu-tory" rulings, " acknowledges and accepts the risks which attend
[upon] a deferral to the time of initial decision of the appellate review of procedural rulings made during the course of trial."
Diablo Canyon, supra, (slip opinion, at 4), citing
20.
Second, immediate review of the ASLB decision may well avoid unnecessary delay and expense.
The arguments in favor of subpoenaing Drs. Trifunac and Luco are compelling.
- First, the DCNPP is to be licensed on the basis of an analysis that seeks to demonstrate that, with few modifications, the as-built plant can withstand an earthquake substantially larger than that to which it was originally designed and constructed.
Second, Drs. Trifunac and Luco, both intimately familiar with the facts surrounding th.e DNCPP seismic issue and both acknow-ledged experts in earthquake engineering, have criticized methods and assumptions essential to that analysis.
- Third, assuring the adequacy of the DCNPP seismic design is extremely important.
If the Appeal Board refuses to review this matter now, it will likely face the issue later - most probably in the context of a motion to stay operation of the facility.
Should the Joint Intervenors position prevail, as we believe it must, the case will have to be remanded.
Susbstantial time and expense that could have been saved by early resolution of the issue, will, instead, have been lost.
We are mindful that the policy that discourages Appeal Board review of "interlocu-tory" rulings, " acknowledges and accepts the risks which attend
[upon) a deferral to the time of initial decision of the appellate review of procedural rulings made during the course of trial."
Diablo Canyon, supra, (slip opinion, at 4), citing
21.
Davis Besse,' supra, at 100.
In this case, however, the facts require the Appeal Board to take the exceptional step and direct certification of the issue set forth above.
Respectfully submitted, J
V David S. Fleischaker, Esq.
1025 Fifteenth Street, N.W.
Suite 502 Washington, D.C.
20005 (202) 638-6070 John R. Phillips, Esq.
Steven Kristovich, Esq.
CENTER FOR LAW IN THE PUBLIC INTEREST 13203 Santa Monica Boulevard Los Angeles, California 90067 Attorneys For Joint Intervenors SCENIC SHORLINE PRESERVATION CONFERENCE, INC.
SAN LUIS OBISPO MOTHERS FOR PEACE ECOLOGY ACTION CLUB SANDRA A.
SILVER JOHN J. FORSTER DECEMBER 31, 1978
tearse saasts asuCttast sittutAYcay couuttssost seasmesseres. a. a. asses February lie 1975 ptet 18ae 50-231 determined gl.at the aa-butta f acility would be able to withste d and s4-321 e=ch an accelerarten but with !!stle er me u.argte far easty elm e
of the deof ts.
Caelegleal Surv.y) espost voeld aet costli:t with its teetatiThe O
- s l
A. Clas.b4sse. Directet. Divistee of teactet L8cca11pg concInsion en t!.e 0 $g watue.
va January 28. 1971
%e Servey's' report was sacetwe3 en DI*3to CMYM AfVlf U g
of the Evivey on J4maery 31.1975and staf f representatives met with raptsser tati vas 3
baste of e.se avettable taferut'**, batteves thatThe of te8f f 1l The is11estas table provides he, datee esseeleted with the Diable t
e Canyon cases in sucese of 0.5 sa ocealstat ea 3 is tote appreptista fet the Distle Caeyan s *ce.
8 4'e have met internally em atte pas *,tas several stras Ja the scient l
y3M gM past up to the lastetant Direccer lessl.
CP A.*plicatime disemessene, my spiilee of the sitution esJ steps that ascJ to 1e O, the bassa et stra a CP Ibswed 11.4/67
$/29/48 seasidered is as fellows:
4/23/44 11/t/70 e
OL applisasten 7/10/7) 7/10/73 1.
The opplicant to octetre addettecal ref ers.ntsu to alter the ss rve/*s ope U,a anJ ear eensultants (U. 5. Ceolagteal Surtsy and U Gs Jette Swivey) eenewsred with she appliesete' telected ge l
. $. Coast scJ It espects to,euetit adJitieast isfasvatten abcus
. es.
on.l selsmlegical basen for desig.
eleggcol ne asesse.ier t of that tafermat ten by the leavey a,?* ara % 1. $$75 c easial s ent hts sneluded en 352 e t 0.4g.
b'e 2.s.0 ewt trail resula la se Ett Suppletamt abaset It<.y 8 sJ the. sta!!
eclacaeJ crite c ts for selente daign. Anel=Jang at.e desige s. estra (L M. Nestnah) concorred with the applicents' Febreaty la sas 19. bee scheduled e ase-day $s.hterminice esctf a3 at the site
, 2975. Tr a C t
- g at da. plat wale.6 s.e4 the s.et..eJs to be used for st*a design t'eveyss, the Ce:n' area wit:
e Escawse of tt.; e tste of the 5 ret taa 45.J the *M:s* isla t preSably not the current time Unit 1 constyvssten is ever tot goe.plete, i
At fus t le.J date fer th.itfiscl lend date is eet ts.arad to be e$ wet Noveaber of thf e year that ansat be a;.da, I so4t A a6.s sspate a e.ce nt me nn:
and the be a.cessaty two : caths ef ts
. :: S.
2 is estimated te be el.ous nine aanths later.
staff enn propere a ottant pels y+ type d.et str. a t t.u :::s t The the fit.t rect;a; a3. lass the e
Der C1. renew se p4 4rtog corplet ten.
* s in i. la er.y event. tu spp!sentsen se s rar,;!y seatasex
' af ss;eus use ses4cJ en octcher St.197LThe $tJI with & few notable anJ l wault a ticic.ste s'ut t%a Ptn wait b.e laur t?. n sta.* t<
ws sair.tss ac a<s6a of th.
The pg gnc ipal ee.isefoe.
=
at thich 4:ait 1 wit! be te:.ely te lead (wel.
gamlaty at.d evisselegy fe-the aire. Jis w g, %
,y lafesuttre La4 beroca evallable dwa tcg tt e uurse of our OL revi cory!ste at the air.e the stR mas tasseJ.ovatw ate.a.od that of the U. S. Ceeleg!cel Servey was met
- 2. ' he cwrrert *best twees* ef one geeleg)-salarete;y s* tff is th at
'8 y'
.ind est O.
ew the finct survey pensglam may well valu fees its presc ; stata prepared fes lessance en Janeaaty it.1935.Am 5FR Sof pleeant was Itas pi*d 88 paalttee would teamit 1% on Sat valu:
px J that, seest!dettag the new talert.atise evallable, an $5EThe statt baJ tentatively but will e-4t all.ely resalt in se 154 value 1.as th.a a'ae se 0 fe! c? act 0.7;)
c o ac l e.F.:
f
";l ve.lue of 0.5 T's eersos t **best guess" of east stretter21-nA49tstl staf f D O
3 vow 14 be approptiste for the ette. The'etaff had else
- Q h..5.y '7
. t.
be accapett!e f:that flee curtssit d*>ttcn vill s.es le able to be *cers t;ats f ta a sensate les41cs ta e :eens of 3.S;.
sa sr sat.nals v p r.
will prcb.bly shs.w that ver.e puts of tLa pl.;atectlpis caelJ tt u;.aattabr
(
- **"rQ
')', t,1 are c.W1. af
~.
en O m. 'm"'%
j.c 4,$
va s.,u..,9.y s,a y
e a
'i
,a i
s
,s,*
s s
to 4
s I
A. claabusse February it.1973 A. Cis-busse
. 4" Pshtaary 11,197) withstandiss wartscs leadiscs la excess of 0.4g, 0.5g, 0.4g, asJ 0.7g, but that othe vtt s! parts will not have such c.pability.
e.
Estabtf uhseat of a special policy a'dvisary group to a:d in the detistan-esking process. This etcht innlede 1.1* vise::
Scoe increase in espsbtlity is pcssible frea desigs changes that sight be undertaken but chunges suf ficient to Laing the design i
euch as Dr. Eeurs D. Kasth, R. Hinagua, etc.
up to a 0.6-0.7g capability are impractical. The design t [
sean413.s could take up to a year er tuo to escriste.
I believe the abave oveltaes the present situatten as we la l'it-1 view it.
I strongly recesstad yewr Isocdiate attentica to this prettaa.
3.
The staff to faceJ with a herrcadeus backfit dacision. The decisten will 11kaly be based on both technical and polie
- p* f cons:dcrattene.
L' hale the teshnical considerations may be altered
'?
by additier.s1 information that may develop dwalag the next month d#
'* *~)f
' er se, the degree of alteratten is not espected to be signifi-R. C. DeTowns, assYster.t Etractar cant. The re f en e, the basic proble=s that will esist and the basis far tight Fater Reacters Creep 1 dectstens that will need to be tide are kr.own at this tine.
Divistan of Reacto: Liccasitg These who iall be involveJ in the policy dccisions (the Ceamissten
=
cannot be involved according to T. Engl hstdt) she=14 becoma Enclosores laeuledgeable with the situatten as seen as practicabla. $$nce Heating Notice
,?
the peticy decistoms will be infloonced by the technical facts and practicalities devolved, early intelvement in the on-going I
Ec8 E Sa Boyd technical review esy be prudent. The eastiast and mast df l
- 0. D. Parr means et uaJerstanding the rechnical issuce is probably ti sudh I
>4r.' A111s an a:tsnJance at.ae yebruary 18-15, 1973 Subcocntttee coeting at T. Elrone the site. The subconmittee will consist of Dr. Gkrent and Dr. Bu:b (and perhaps L. you). In additica, op to eight ACRS f
consellants will participate. A copy of our Leetir.g s.etice and the agenda ter the cessing is at tached. Tha prepased attendance by Dr. Ceulter of the Cselegical Survey is indicative of the eerieu nas-. of sha Svevey's concern in sta natter.
4.
Consideratten eight aise wall be given tot l'
ps.%.
a.
Streegshening the legal contingent escagned to the case.
ej la adJ! sten, a review Light be v4Ja of the assigned Attg u) to assure the level of crtdentiale end expertsnes is coa.
- E' otatent with the task that is to La faced.
Sa lt!
b.
Str.httheatrg ti.e engf aecring staf f by providing for f !'j special censultieg advice f ren groups and individuals
,i auch as Newnark Aspeciates. Tranklin Institute, 1
kJ Dr. J. Hendrie, etc.
0-4 g
e s
W WP M8 e.m me e.e.
eeums.a.
- me
.a-me..ee.,,.,...,n...
ve.rse stans seuctran y.tcutatomy corcast-om g
wasweac t o=. e. c. nsis A. Clanksee Fshrwary 00. 1975 8
y bruary ?O.19 75 that, salass specit te bidance approprf ate to t%Is teatwa situatten as provided to the USGS geelectres asJ setsrelegists, they erill pseceed with their rev'ev bastig it.sren their Nde for A. ClarSerse stendard mathada and arrive at a ette 555 "g" vilme well in I
eteess of the 0.4g valve approved for the Cp and used for the tt! A40 Ccriott - SEI5filC IS$UES desi e of the striest completed plant.
t As ac25 $v.cesstatea resting to review the g.shle Caneen O!. app 1statten 9
2.
The es-built plant has significant margins in its design and s as ci.mtutt a l 13 San Lut s 051spe en February 18 and 19.1979.
4 it is capable of resisting selsste leads well in excess of The
- uusrnt s 'e consisted of Pre Ohr at (Chat reas) aad Dr. Swsb. In if
$t these asescisted with the f' 43 set *Pf e seet:t assa e.1 for destge.
wh a t.in, aven AC?
esaswitants were in attenJance the first day The staf f to presently convinced that the design is "t;ssi" for vi tra - s u,; 34+h a ss devated a! cost in its entirety to the salente "
e 0.5s event, tiowaver, to cor.vtace the ACit$ and athers of this 14o 1 s.. spec tile cas tes discussoJ lacluted gen. *y. selsrology, Will requira 3-6 nonths of applicant end staff effet t.
If a.
s *mic e.
.*.r.. selsMc tea t!n3. med settnic scraa. Om the basis
'l event in t;.e.ortar of 0.a _r.cf8t_bt r.o:uldasat., it will be t,ese ttle r
4 :v atti t.:ce at the r xting, anJ
- ste distu6stens with other staff I
to show th.s tiany parts of the plant can safely resist suc h e e. 'w r.,14 0. s os ee.atat ives an.1 erplic.nt an1 L'ec e nn: hoose partict-leads; besever, seiny parts will need to be modified and very
- - tv. I tee av.sy with the tallawing tre;iresstsns t
'l ljkaly fet ges e of these the redif f tat! ens will f.et be practical.
The evelwa'twne to eccer.plf ah sisct, a casa' with the siger that 1.
1he eve main renceres which wtll deternine the SSE "g" value will be regat red, will enteil years of arplicact and s ett f ef f ort.
ase ta) the gaelegical de fints ten (extent) of the Itosgri f ault sene, anJ (til the selsste evant that muer be ess6ard to eccer 3.
The "genc" of the questle,ns and cor:sents tre1 the Arts Sube m ittee en the of fsnore f awlt. I believe that the USGS geologists g
menbers and the consultante indicat$4 to se it.at they satald find (lud revirver - F. 'screr,ws) will ratatale their twtrently e
a 0.53 vatwo ecceptable sad could b eewinct.t that the ts4.ilt inJ;c ited peastf on.
11.e " nee" taietra6 ten d< scribed by the plant coute adeguataly resist the inctaaset la.its assertated m ilcant and se bs ter, ally !c.ci.wnted in the near future is with Lt.e ever.t.
Isewever, they are not 3:tely 's lae les.vir.ced athly to conyssee the USCS to alter its current findtag.
unless e a st'.or eergiere analysts is perfes sed 'er = 0.53 event n ot t Le t te re that t h "arv" tr.tep.at tee. will convince the USC5 act t%e wie of eve cdtrently e,pieroved setsnte design criteria, s', r si e Seli H.p.i t d.e 7.) urst uit.e ocevered en a transeerse To data n-a roch smalysts is avetlableg the preeant eceit tena f 3al t soJ. tL e s t.,re. nee d e. it be. es i.;a d..ra d e, ripsble of a'
of the atplicant and the staf f as to the sittity et the pitet e<< vet in*, en the ur ge t savlt. I belt.va that tLe LSC5 assessaients supperred by a few typic.! talen:stions.
to aJegwetsly resist a 0.5g event are based e twallti.ttea r e t. -,lectus (J. D-evice - I,44 revie er) wawld cene.lvde that
- tt )*
tt.e ensuptlete "3" value for the ette syvid be 0.$3 (f ther d
cow!' ant se that the fawlt length were ItatteJ as the appit-
' ls la es? mary, as e reswit of developruers during the pett weet. I teitte.u,, )',l 18
- tant centenJs and the 1937 eve.at occurred on a transverse feelt.
to believe akt, unless spectile guidange, ts.pset t arti dinctica is l'.a.ewi. If the f&olt langth is s'eter:uned en the b sts of the providad pre ptly by the opper ma3 age *tet levels at !"2C and U*CS te H p) g j
the "verktag" levels in the two organisattens, gesitieas t'lat 42 eet USC4 geole; tral laterpretat ten. than Ene set tels event t u r r e r.
th t sust be assued by the U!C$ setts. elegists. In ordar to be necessettly reflect the jud vet of apper-level vanJgmat will be
,O,, g t
eca.*atter with the ruthodology and for ti e $.n Onofra 2/3 ferais1.ted see dace;3en:ed to the extent that later stelficatie't vill.
- 5 oalu tto3, will tewit in
- 7. sit a "g" valve si:.ilar to thac g j',N * -
det.e ar.td f or the 5sn Onat.a site (0.6?g). Er is my optaten N <.
a i
4 t*
. C A
s
. s l
-o d
m.
- . #.-4
,4*k43 s,
- k
+.s.j A tP
~
- 1 N
e,,, p gl tu e'
b y
l
=
.e*
e
-.e~
_ ~
A'" TACHE 2 Page 2 of 2 1
I eng h
De e' '
eh e
6 e
- J 40
- e e
eG e
- C me to e he e6 e
e y
e w
e g
g g
.s.e U C t
8 G8*
e h
>s 8
eng et ek,g e. a$ e es es W
ee N-est e
Sh e
e V
ee**ee4 s u qs==
e3 J
h te 8's C
O G
3 e e e
we he e
a we a Se en en Eb O
g e
e 9
e et a be se E
e O
e e
O U
eC e u s
ee e
ee **
l e'M n
e.*
-es S
e et e e
e e
.. & M e
3
- e e
- E me e
p e
e e.
=
Oe S
ase De.4,.0 se o
.W w
g e
e e
e.
- e e
we aid 0 4
8 G*,. > #"
M e
se 9
5 T
- e ee se e w e se se 7
g g.
.e D
as se Q e we W
> 0e
- e 1
to e
Q es se et e
es W =eei I
- O u
.o e,
e L. *IO e
)
g eg
,e DP
- g te 9
as e
g es 3 et a c
e w
e s-e en en e.
4' 3
e.o n
u m
a
.e e
i 4
e em G
u e
Js e
og p.e E
es en en o
e e
en
,o em to as is a e o
y e
E e
46 a
O De to to 3
ee o
me eo e
e e
g en s
80 ee A
g
- 4 M
C en e
C.
e
" 9*
o na e
43 to a A S
to to pg h et a K es le e
es
- C e
ee1w W
5 mg h
e e
e as pu e
en se
=e to y e G eeO eq 6%
et e e **
M fi eC a O*
S ee M
e o &
I 0
.a.
3 9
8 e to es e
.d
& to =.e=
e.
ea ee e e.e a ene
- s.e.e e
>s o
% ev o
ee e.
g e.e.
= = = 2 m.an e
.e my 4
e==
we be es o,
O N
e
. b5 N
- INbEMNOO=
e C**
e - * - = * =
e M
ene esJ e m.. _m 0
se.
.o.a.
-e-..
e e- -...e.e w
Wee.>..eeesmem p ia ww w W -
Y b
o e
og en en e
e es we Se o
e es oe 9e e
es
- es e e to De S
se G
'*""=e w es A es e
A 3
e C en ne
> es o A
e C db e e ee 4 *s 3 e hm S
39-e de se e e It e&.
9 g e
'S J ee ees C ee a es es t e\\
e e
en e.s es==
e C f e9 4A EO
- e we"D D 8
e -e e
e e.e u y'.,e 3 es e 2 o
J e w on 5.
O f g
3 O e eese we es 4
% fe e ne de es ob ee 85 eme he W U
g e a se we se >0 M
$ e es ne h M
w e ab o 9
9
- Ste et e 3
- b C
eOe a
ww $ es f
See 8 I O tot as 6 wu C e
S Vb e en b3 ee 4.e e
b d 8
ge%&
O g e e. J ee4 > Ce n
e e
en as 6 st en E4 De se
- =as %
o t
r9 eO 3 eC%
C a e
emas es
- J as*to es es ee ne W Wh g
- g e
^e & me eJ J'.
as 3 aa a... 4sa 3
es d
.e u
9==
L$ senoW
=a e
$.e
- g Q llB $-
W 9
e-M J w.
9
,3 w le e4
- e e.4 m*S'Jm 4*
e.
o 4 4
e
t g
t '
\\
gJ YlY7,4
\\
4 FR008'Jf 70 t$7 A*t151 Ettif 70 t!CT'?! DTAff 0 Cx'vtre i
BtCRCROU'fD A':D FIQPo$r0 proc 4w j
Construesten peraf te were isetted for Diable Canyon 1 & 2 is Aptil ItC8
)
se
- Deceedser 19 /C.
respectively.
l' air 1 is nearly saryleted; Unit !
g is a year er se befied.
The 5 E for the site vae set by Dr. t;ccask, j
the t%$. and the 17. S. Coast & Ceedetic Survey. 1Se vsloe selecte4 was 0.4 g/8 M E O
g
/M A 3 and the plant has beei coastructed to that value.
The OL applicattees f or bath units wave decketed in Octe'aer ISF).
C r teview a
was esses.t tally eetpleted by Jamstry 19753 at ti.at tire we recognised f// M (jf,(/[#d!
ggf fM that en the basis of new evidence the $!' elevid I.. increased.
1he staf f believed that st e t*$C5 steeld centwr st en a site "g" vale of 6.Sg end we had (37e suf ficient were to confess sie that t'te ;1r.:t teund estely witharand ru:h an eerrhquate. On January 18.1*1$. se
}f" //// foWgf//)( M//[8 sere ser;.rlsed by a L'SC5 report that concluded that an ecceleration s f 0. 52 was in its estaten inadaquate fer the atte en tre bests of presrat laterretten.
gg7f,t/4-
/ / 2 /76 At ttist time Dr.I. sacemended th at tra appresches be tal.en te rerelve
< ** t "-e -
'u t. < t:a a t e<*ene octa d u.c n' 'e: e.<:*
t o teavf aca the Uf C5 of the adeq=scy of a 0.3g 35t. tecom'ly, secuts the CsCT, at tt's styf ewer level, eseelJ re.u ts adamant and seek t ther reane te tenf f!5 er stadd ry f ee fief teg. The esetetaat vet et sde to pursve egly the if rat apetoach. Tedsy, s e are resan;ta,lly sherv 'we were last Joiese rf except that the plant is alesst v.nady for twel itsdias. 0:sce asets we are faced seitt. fevelepist a piegs aet to ettablish a basis to petred s a decisiets to be e:de wie t!.e ts c.msing of bia' ale Car.ysn.
We believe is i.e=td be tenewJeas te ow; zrain purtwe a sfagle path baseJ ou acquirint ads'4ttegat tr.f atr.sti.-a to convince tha me:S s owiew team (a acl:.ttwaly f ew tr J'vidusts) te M>
1: edify its eurrent postrit.n.
'do again sacer.end a ::utti-f ac ated
$h appreath. TIo prepers we weeld propeee to parave is as fellever ft >
0 1.
Tsumahy ret sest that an inde;tsident teview be condw:ted t ithin H6 t:5CS to ae.itse er cc/ify the curr.me USCS reeice tea
!!.*1:.g.
t%y Thle reen.t t f::ht be t..de by C%:.f rman Apfers to the feer. tary o
ef ti.e 7sserint aad coulJ. If legal ceasiderarter.s dear.ed. be.
g 'q-based on scen ts Yest Coast esacetas rather thsa the D':.ble Canyae teetow alone.
g,,
- &~
e
.Concurrsetly, forts a teas of eeareltsats seneisting at t w of g
k*e*
- g' 2.
f t' mattenal stature in the flat,.!s of geologf, se:sestasy, and
- .fa %) v. ' ' = f. \\
,.t.,1. o.1, t. e.
,,t.
t
..t. art.s.
ih. t s..mt
..n possua vpricos patu. %clodingt pg *
.r 44 1
1.r F
- *1 5
l
{t S*1 @ N l.~
///
1
/
to
$1*
=
d e i
,g 3
i i.
8 Asswoe a " great" earthpake coute occwn a.
f r+14.s the *ae ty transfer to the att a u JHeat de:.:e.utsate by logic evidence. sad j 4 g
as the tf5CS veview that e MLiWt f.a
[]SoclaVED COSJ'!ELMIG.
v;tt.ta st.e deangs coadtlittet of u
aa t the nut? car n LirJ
)n deve lsy tna a P'* t'* ' I* I*)I*" "*
,,,,g y, b.
an independent evalua constJarstien skuld be given tot ghe g>s st61e C4-yss pretV
. nciv.i.. en tie is, tion of the evidence te at sve at a t i,a t 6,
the 16CS rewtew tou. sad the psael of.ie,t be cas,ared to t,.es..f I.
The sepact of M a th dies's caer y prctic' S Sd espouts establishet puettens. The leipset of
'I
,i.at 4,,re-d for e-potentist dutal for eytatin of a e.
Assest the pre.spect s f+v vegreding th where the taats for dental is in conneveny.
og smc4 a-g SSE vals:es ty plastle analysis and/or se plaat design to higher 4.
eattens and testing.
tivcturat radift*
The inaet of eer deaf 88*es on the asraterin hora,ra tha Califatuts 3.
tencersentir vet ers.
to deterais.e if a pas *S *b tlistic be tfera e task fotse to teview the terrows st e
en atterpt lished te licsnse tsatt atua la e $.* f t 'I 8' st.lle the othst seviews are belai ta's er. intesi,a perted eroperett'ena s tan be estab-ddcists is en the viabillt, et cr: tim'11 IFST4 h s conducted.
as 5an Ces,f r. P11 f 8 '. e tc.@ sitered sassr:ste<al Lf* a. Es.C.
4 4.
Cca wtrant'ly.1sform the ticannee of die 93, gn7,c. of est decistens 6:t N "MM ], {4
- ase purtwin cousses so ; hat eier 13a nt decisiere say be tsintist anJ/er alternative and sequt <e ht= to pi*tse.a eawrse of attlen we
.g,,1 arts whers it is ustr&ain that t
of our evaluetisa of kla ef fetta swpple eared bwJe en the besta 1
withstare attaged postgt baas.,n a
,; g, g,3 3,3, as6 essa.p t s.
g,g y our indepenJent
,ptag. vent struc turat destga. P$P. h!p tnsl,,,,,,g,.
pte.ntur. u rlo:n
?.
,,g,,,u ge.. v tve f ailures. etc.
L'd
.pgreaches laJteated 4 ave.are planr.ta; to t ana $*c.edtate actions to inttl t Spectifralty we plan tese e some of the 1.
foet wi t h
- .t' i
I" esu;.m to se t. win approval af s'ae ;sioeval app t.pJ a le er to e43.1 74 as e}7e+ 23) 74 5ept prettaa.
1 1.
Was tooetta e (at Chtssp. Illinot a) with (
to forawlste tre cars of eaasuttan:s di tr. Few -k to begla y't -)
and it's tad force disco.=J 1:a it. a f alpevescussed in staa 3 tbave.
- 6 )
42 o rice t f ot Gro Woaik will make a presenta:. At tne sa.ee f't say cir. gly wadertraad the basas and lis it1ea to we se that we N
r and anechar.tcal fattwrea.p.* pet en satsatt dasica margins and prehabilitiat tene of his recent
,- l '?
es of stavetural es :*
3.
e t:aa t base.
ste serre4r..ta v.f resta and et the progras tea i tIsser this week, sith the licensee to sdvise bl
e, e of puses.e ta est.ablit.h a basis (e.' a detition
- r. end te in vttw s f 16 e au Iowsness of the ptobido
'l Ra p,1 tory p.4n.sger.m: he 1 posed Jar.orttatel. It la essettial that strong coasulteelt es e the pr<.b4111ty task force.y (b *s.w. age" the team of be tue hic:.ast leven managers that se can p These runagers should s ad.
ractically assigra to the 8
t
i ATTACHMENT 4 Page 1 of 7 a.s 1 1%,n, f,
v s
f
/
f$
f"'Y N,
S$v.*-
7T 7
, g,i3 t,.
N " bY'
,h S
ae is
\\b, cg y C0!G ENTS ON SEISMIC DESIGN LEVELS FOR D] Alho CANYON SITE IN CALIF 0.%'IA J
U18 P
,y M.D. Trifunac 0
~.
~ April, 197S 4
l I
\\
I.
(
(
ATTACHMENT 4 Page 2 of 7 The follouing comments deal with scismic design criteria for the Diablo Canyon site in California and represent a brief summary of my observations and prelitinary conclusions which are based on misec11ancous written material and on a number of meetings during the period starting in the summer of 1977 and ending in April of 1978.
Information_which I had on certain aspects of this effort may be incomp]cte. The general picture and the summaries of the current status of this project neverthc1 css seem adequate for the following comments and recommendations.
Much has been written about detailed aspects of scismic design criteria for the Diablo Canyon site and it wculd be impractical to address again numerous points in detail and completely.
Rather, I will attempt to present an overall summary of what I belicyc to be unresolved problems at present, and what might be possible avenues to resolve them.
General Conments on the Current Inputs and Criteria for Scismic Design 1.
Generally accepted assumption appears to be that the SSE on llosgri fault opposite the plant site should be an M = 7.5 carth-quake. This magnitude, recommended by USGS, has been determined mainly on the basis of the possibic length of faulting on the
!!osgri fault system.
2.
Since M = 7.5 at a distance of 5 - 10 km from the site Icads to large peak acceleration (about 1 g) considerabic effort has been devoted to the analyses which are des'igned to show that
\\
i,
I
. i..
2 ATTACHMENT 4 Page 3 of 7 these large amplitudes can and may be reduced through considera-tion of the following phenomena:
a) Scattering and diffraction of high frequency waves from the foundations of different plant structures has been proposed as a vehicle to justify reduction of high frequency spectral amplitudes (t effect). The canner in which this reduction has been affected requires unrealistic assumptions, for ex-ampic, that foundation is rigid.
The manner in which this C
assemption is introduced into analysis if often one-sided and considers nainly only those consequences of the physical pheno =ena which'1 cad to reduction of spectral amplitudes.
Other consequences of this phenomenon, for ex =ple, torsional and rocking excitations of foundation which may amplify the structural response have been, so far, either overlooked or treated inadequately. This has been achieved by utili:ntion of dynamic nodels for analysis which are so defined that only an inco=picte physics of the probica, i.e.,
scis=ic excita-tion and the structural response, can be considered.
b) The term " effective peak acceleration" has been introduced suggesting that tIie structure will "sec" se=ething s=:11er than actual peak..ccelerntion. Though such approach may be useful for carthquake resistent design of ordinary structures by means of the response spectrum technique, the term " effec-tive peak acceleration" has not been defined in a way that would enahic the derivation of consistent results by several 9
- r..
-T
\\
i 3
e ATTACHMENT 4 Page 4 of 7 different experts in the field.,Since the procedures for scaling Regulatory Guide 1.60 spectra are based on " maximum vibratory ground acceleration" (as defined in Appendix A) this departure from routine design practices makes it diffi-cult to evaluate the number and the nature of the consequences which would result from such an approach.
c) !!ypocentral rather than distance closest to the fault has been used to evaluate peak and effective peak acceleration.
This assumption implies certain ang1cs of approach of scismic wave energy. These ang1cs of approach should then be con-sistent with the extent to which "T effect" is allowed to influence the spectral amplitudes.
Littic or no attention seems to have been given to mutual consistency of these assumptions anc in some cases, inconsistent assumptions have been utilized.
For example, deep hypocenter would increase the distance at which peak acceleration is evaluated, thus reducing the estimate of peak acceleration amplitudes. This would, however, also imply that the waves arrive towards the foundation almost verti.cally.
In consideration of "I cffcet" however, horizontal dimensions of foundations appear to have been used implying horizontal incidence of waves.
d) The large damping equal to 7'. has been adopted for dynamic 1esponse calculations. Though the apparent damping for the co.mplete soil-structure system, subjected to carthquake excitation may he much larger than 'M, inadequate basis has been presented to justify 7t damping in structural systems y -.
}
4 4
ATTAC1G".ENT 4 Page 5 of 7 only. Scicction of too large structural damping coupled with' only two-dimensional or simpic three-dimensional analysis of soil-structure interaction can Icad to unreliabic response estimates.
3.
At Icast three seismic risk studies have been prepared to estimate the probability of execeding the selected design criteria at the-Diablo Canyon site (Blume, Ang and Newmark, Anderson and Trifunac).
These studies have produced results which, in some cases, differ by as much as two orders of magnitude.
Concurrent with the com-parisons of these studies, considerabic effort has been devoted to different details in the nethodology emp,loyed in these calcula-tions.
Littic or no explicit effort and discussion has been de-voted to the modcIs of scismicity which are essential input into such calculations, even though this may represent the most impor-tant contribution to the discrepancies among the results of different studies.
In some extreme cases (e.g., report by Blume and Kircmidjian) claborate work has been carried out, apparently in vain, to show that a particular method fo.r scaling peak accel-cration (Trifunac,1976) supposedly 1 cads to "too large" estimates of peak acceleration irregardless of the fact that those results of Trifunac (1976) have never been used and do not represent a basis for the derivation of scismic risk models by Anderson and Tri funae.
In the report by Ang and Newmark, subs:antially smaller than average seismicity has been assumed near the site.
This may Icad to an underestimate of actual risk.
==
-S h
k
t
!.~.
S ATTACHMENT 4 Page 6 of 7 Recommendations
/
A.
Ground Motion:
1.
Deterministic approach based on tl:e assumption that an carthquake of magnitude M = 7.5 or gre.ater will occur oppo-site the plant site should be re-evaluated. This magnitude might be an indicator of the extent of geologic faulting phenomena but it is not necessarily the most reliable basis for evaluating the nature of strong shaking close to the fault. There are numerous examples in literature of sig-nificant differencer between M and M, for example, which g
3 are based on short and long period scismic waves, respectively.
~
Often studies have shown that larger earthquakes may be thought of as a sequence of several or many discrete events which can sequentially take place along a long fault.
- Finally, the largest recorded acccicration, so far, has resulted for M ~ 6.5 only.
For these reasons, and from the design view-g point, I would prefer to adopt M = 6.5 on !!osgri opposite the site and r.at M = 7.3.
2.
Near-f'.cid source theory (not a finite element or finite difference model of the source and its surroundings) could be used in conjunction with the spectral analysis of strong motions recorded elsewhere to evaluate the amplitudes of response spectra independent of peak acceleration estimates or of seismic risk analyses.
p--
\\
l.'
6 e
ATTACHMENT 4 Page 7 of 7 B.
Ilesnonse:
1.
Three-dimensional coil-structure interaction analysis should be carried out. This should be donc assuming that the frev-field response spectra f.r design result from incident Sil, SV or Rayleigh waves.
For Sil and SV cxcitation, horizontal, vertical and 45' incidence analysis s'iould be considered. This approach voeld offer the followint, advantages:
The "T cffect" if present will be accounted for correctly.
a.
b.
Torsional and rocking excit::tions will be included into the analysis correctly.
The proximity of the earthquake source and the fact that c.
the waves most likely arrive horizontally will be accounted for correctly.
d.
The radiation damping in the soil will be introduced into analysis properly so that the high value of 7'e for struc-tures would not be required.
Except for the fact that 7 ', damping is permissibic according to the regulating guide 1.61, this high structural danping recommended fer the seismic analysis at the Diablo Canyon site has not been justi-fled.
Forced vib:.. tion test (available in 11.:.. and Japan) data, where the effect of soil-structure interact ion and different mode of energy input into t he st ruc t ure during..n. experiment, relative to incident earthiguake waves is not accounted for, may be of little tu.e in c:.tabl i:.hing t he act ual damping. in st ructor. s and their co:aponent s for sei smic response calculation
\\
e j
(.
~
a ATTACHMENT 5
-j.
Page 1 of 19 s
REVIEW OF TIIE ' SEISMIC EVALUATION FOR POSTULATED 7.5M 110SCRI EART!!QUA):E, UNITS 1 AND 2, DIABLO CANYON SITE' by J. Enrique Luco I
g, tQ, p
\\
>A 6
n.
^% ' s"L.
~f ggt g
-2 g
p ).#j S
d r?i t
r/
~
A Report to the Advisory Com:nittee on Reactor Safeguards U. S. Nuc1 car Regulatory Commission.
s 30 May 197S W*
t I
e-I
\\
ATTACHMENT 5 I'.
Page 2 of*19 i
REVIDI AND REC 0!NENDATIO.'!S After detailed review of the report 'Scismic Evaluation for Postulated 7.5M !!osgri Earthquake' (Ref.1), I have the following comments and recommendations :
1.
Free-Fi eld Desi nn Spectrum.
In my opinion, the free-field design spectrum used for re-evaluation of the Diablo Canyon Nuc1 car Power Plant does not reflect the strong notion at the site for a 7.5 magnitude carthquake at an epicentral distance of 5 hilometers, but rather the motion for a 6.5M carthquake at that distance.
The free-field design spectrum developed by Newmark and adopted by NRC corresponds to a simplified version of the average cf the two Pacoima Dam spectra recorded during the 6.5M San Fernando carthquake with the high-frequency portion reduced by use of an 'cffective' peak acceleration (Fig. 1).
The Elume design spectrum developed for the applicant closely follows the Newmark spectrum.
The peak.accelerati,on, velocity and displace-
/
nont controlling the high, intc~rmediate and low frequency portions of the Newmark design spectrum are in agreement with the avernic (50; percentile) peak valdes obtained by Trifunac (Ref. 2) for a 6.5M carthquake'while falling short by 40 to 60 percent from the corresponding values for a 7.5M carthquake (Tabic 1).
The peak values consistent with the Neumark spectrum are also considerably louer than those suggested in USGS circular 672 (Ref. 3) as shown in Tabic 1.
In addition, compari:.en of the Newmark and Blume de-for sign spectra uith estimates of the average response spectrum 1.
g.
I 1.
\\
I i.
3 ATTACHMENT 5 Page 3 of 19
~
c 7.5M c'arthquake as obtained by Trifunac (Ref. 4) also shows differences of the order of 30 to 50 percent (Fig. 2).
The applicant has indicated that the thrust fault mechanism and the location of the Prcoima Dam instrument in the San Fernando carthquake may have increased the recorded peak accci-eration.
These possibic effcets are negligibic in view of the fact that the standard deviation for peak accelerations, which '
has not been considered, corresponds to a factor of 2.
Also, the records for the Ms=7.2 Gazli, Russia carthquake of 1976 indicate a epicentral distance of
- peak horizontal accaleration of 0.Sg at at 10 kilometers.
Correcting for attenuation using the Gutenberg's relation leads to a peak acceleration of 1.0g at 5 kilometers for the Ga:li carthquake in general agreement with the results of Trifunac and the USGS recommendation (Tabic 1).
In view of these facts, I must conclude that the Newmark and Blune design spectra do not, correspond to the ground motion for a 7.5M carthquake at an epicentraf distance of 5 kilometers.
I pro-pose that the estimate of the. average responsc s'pectrum for M=7.5, 5 kilometers, epicentral distance and rock sites of Trifunac (Ref.
- 4) be used as design spectrum.
This spectrum is consistent with the only records availabic for large magnitude and short epicen-tral distances (San Fernando, Koyna and Gaali) as well as with the USGS circular 672 recomracndations.
2.
'Eficctive' Peak Acceleration.
A judgnental factor has been used to reduce the 1.15g peak acceleration recoe.:nended in USGS circular 672 to a value of 0.75p.
This ill-defined factor
~
2.
W..
i
- f.,**4*'
\\
I e
=
2 s.
ATTACHMENT 5 Page 4 of 19 has bec,n used in the past to account for discrepancies on the Icyc1 of damage observed as compared with the predictions of ordinary scismic analyses which do no,t account for the effects of soil-structure interaction, are based on nominal values for damp-ing and strength, assume linear behaviour and do not include the energy dissipation in partitions and other non-structural cle-nents.
This catch-all reduction factor has no place in the de-
. sign of carefully analyzed structures such as those in nuclear power plants.
Factors which may reduce the response or the level of damage should be identified and properly included in the struc-turni models.
In the case of Diablo Canyon. m.'ny of these factors have already been incorporated in the analysis:
use of test strength rather than nominni values, use of higher than common danping va1ues, reduction by scattering of waves by large founda-tions and pos-ibic inclusion of ductility.
The arbitrary reductier.
of the high-frequency components of motion affects the response piping and equipment.
I recon ='end the. climination of this reduc-tion of the input motion.
3.
On the Effect of Scattering of Waves by Rigid Foundation <.
The high-frequency components of the free-field motion have been reduced by the so-called tau-filtering procedure to account by the scattering of waves by the supposedly rigid foundations.
This correction amounts to a reduction of the Newmark free-field design spectrum by 20 to 30 percent for frequencies higher than 2 cps.
Slightly )cuer reductions have been used in the Illuac 's spectrum.
The correction for foundation scattering ef fects is based on the 9
3.
I
(
\\
.l
(
ATTACHMENT 5 i-Page 5 of 19 i
ynssumption of a rigid foundation and horizontally propagating S!!
Although the assumpt' ion of a rigid foundation may be ren-waves.
sonabic, it must be recognized that deviations from the assumption
~
1 cad to localized higher stresses in the lower portions of the different structures.
The assumption of horizontally incident SH waves is highly questionable considering that the epicentral dis-tance is comparabic with the focal depth.
Under these conditions, the possibility of nearly vertically incident. waves may not be ruled out.
For vertically incident waves the scattering by the foundations is practically nonexistent given the shallow embed-ment.
. Assuming for the sake of the argument that the scisnic exci-tation at the site corresponds tv horizontally incident SH waves, I find that the reductions proposed by Newmark and Blume are too high when compared with analytical solutions.
For horizontally incident S11 waves the reduction of the translational components of motion is coupled with the yxistence of a marked torsional input to the structure-(for details refer to the attached papers, Tlle applicant has included ' accidental' cecentricitics of 5 and 7 pcycent to represent these torsional effects., The use of an eccentricity of 5 percent corresponds to the use of a peak tor-
' he base of the containment of the order sional acceleration at t
2 as may be inferred from Tabic 4-5 of Ref.
1.
of 0.025 rad /sec This torsional acceleration corresponds to a tangential acceler-ntion at the base of the containment exterior.of 0.025 x 70/32=~
0.05g.
The results of Ray and Jhaveri of URS/nlume presented in 4.
r
A"'TACHMENT 5 Page 6 of 19 Fig. 56 of Appendix D30A, but not used in the an:1ysis, show that a peak torsional acceleration cf the order of 0.1 rad /sec corres-the base of the con-ponding to a peak tangential accelerstion at tainment exterier of 0.2: would be nore :ppropriate.
It cay be the use of a 5 percent eccentricity underestimates concluded that This observation is con-the torsional input by a factor of four.
sistent with the original vork of Newmark (Ref. 5) which indicates
. that an eccentricity of the order of 25 percent would be necessary the torsional effects induced by horizontally propa-to represent gating 511 vaves.
It must be centioned that the increase in pech acceleration of 0.2g bcsed on a core reclistic esti= ate of the torsional input nore than co=pensates for the reduction by tau-exterior.
filtering fron 0.75g to 0.67; for the cont. innent
?
Fron the peint of view of the nn:1ysis of the structural re-it does not seem adequate to introduce the torsional input
~.
- sponse, Such procedure through the use of ' accidental' eccentricitics.
which 1 cads to the coupling of the tors,ional and trans1ction 1 re-sponse in essentially synectric structures distorts the respons e The effects of the ter-and the n:tur:1 frequencios of the systen.
for the turbine building in which sional input may be significant the possibility of portions of the structure undergoing inciastic deformations may increase the eccentricity.
5f it is shown that the sciscic excitation :: the site cor-responds ncinly to horizontally incident unves, t! e reductions of cvaluated on the translaticnal and torsional response should b To the basis of the cere exact methods presently av ilabic.
5.
7..,_
t
\\
I g
s ATTACHMENT 5 l
Page 7 of 19 include an exaggerated reduction of the translational motion with-out incorporating the full torsional effects is improper. '
4.
Soil-Structure Interaction.
In Appendix D-LL3A of Ref.1, the applicant presr a a comparison of the results obtained by the fixed base analysis of the axisymmetric containment model with tae-filtered spectra as input (F.B.Axisym.) with those obtained from a soil-stIucture interaction finite cicment model with the Newmark free-field motion (without tau-filtering) used as surface control
' motion (PLUSli-SSI).
Eased o>.
the results shown in Fig. 3A-1 cf Appendix D-LL3A, the applicant concluded that 'the use of tau-filtered inputs with fixed base models as used for scismic analyses of Diablo Canyon ~ structures is -conservative. '
This comparison is not valid, and the c'enclusion is not warranted by the ana2ysis.
For a valid comparison, we must require that the fixed base axis-ymmetric analysis and the fixed base PLUSil analysis give essentia11; the same response everyuhcre except at high frequencies where the fixed base PLUSil results not.ideluding the tau-filtering should be slightly higher.
This is not the case as shown in Fig. 3 of this report obtained from results shoun in Figs. 3A-1 and 3B-5 of Appendices D-LL3A and D-LL3B.
Since the fixed base PLUS!! model is inconsistent with the fixed base axisymmetric model, no valid conclusiory as to the effects of soil-structure interaction can be obtained by comparisons of the type shown in Fig. 3A-1.
It must be mentioned that it has been shoun that two-dimensional models such as PLUS!! may underestir. tate the response at the top of the structure by 30 to 50 percent.
6.
P
\\
J
.. (
j.,
ATTACHMENT 5 l
i Page 8 of 19 l
In Appendix D-LL3D, comparisons are presented of the. response for a fixed base and an SSI model both computed using PLUS!! and the Neumark free field spectrum (uithout tau-fil tering) as control notion on the frec-surf ace.
Assuming that the results presented arc internally consistent, it is possibic to draw some tentative conclusions.
Fig. 3D-2 of Appendix 5 ~ ~ ~3 indicates that the peak acccierations on the containment exterior obtained including the SSI cffects are approximately 10 percent lower than those obtained on a rigid base.
Since the SSI results automatically include the effects of scattering of vaves by the foundation as well as the effects of radiation damping into the soil, it may be concluded that the reduction of 20 percent (0.75g to 0.6g)' by tau-effect proposed by Newmark and a sinilar reduction used by Blume are not conservative.
Figs. 3D-3 and 3B-4 of the same Appendix indicate that the story shcar forces and overturning moments on the contain-nent exterior obtained incipding the SSI arc equal or slightly In higher than those obtained for the rigid base PLUS!! model.
this case, any reduction of the fixed base results by tau-filterir.g woul'd underestimate the stresses in the structure.
Assuming that the PLUS!! results are correct and consistent, it may be concluded that the tau reduction proposed by Newmark and Blume overestimates the reduction effects of wave scattering and soil-structure interaction for vertically incirlent shcar unves.
In particular, the s tresses computed on the basis of spectra re-duced by tau-filtering would underestimate the stresses that re-sult from the SSI PLU31! analysis by at Icast 20 percent.
7.
T'.
z-e
-VOM 9
9
ATTACHMENT 5 j,'
Page 9 of 19 s-
/
The applicant has indic*atcJ that the shcar wave velocity at the site exceeds 3600 f t/sec.
The low-strain and iterated (or strain dependent) shear waves velocitics used in the PLUS!! SSI mode) are not reported.
I request that this information be made availabic.
In Appendix DLL-15 -(Amendment S3), a uniform shear unvc velocity of 3500 ft/sec. *is used.
I rec.,mmend that the tau-filtering approach be clininated three-dimensionsi soil-structure an:1ys'is for and that a complett.
vertical and horizontally incident S!! waves be undertaken.
This approach will have the advantage of producing realistic estimates of the wave scattering and torsional effects.
The peak spectral response'for the PLUS!! fixed base analysis occurs at a frequency of S.3 cps chile the corresponding frequency for the axisymmetric fixed base analysis is 4.5 cps, indicating a difference of IS percent.
If this difference reficcts the accur-acy with which the fixed base fundamental frequency can be compu-ted, then it would seem that the peak widening of the floor re-sponse spectra of S percent on the high frequency side nay be in-ThePLUSilSShresonant frequency is IS percent lower sufficient.
than the PLUSil fixed base ' frequency.
This again sects to indicate that the 15 percent peak videning of floor response spectra on the Iow frequei. ',-. uc is not sufficient.
S.
Scisnic nisl: Analyses.
Several scismic risk analyses for the Diablo Canyon site have been perforned.
The estimates cbteined for the probability of excedance of the !!os gri design spect rum di f-fer by two orders of nagnitude.
The applicant (Appendix D-LL 11) 8.
h 5
E
\\
.w j
l s,
. ATTACHMENT 5 Page 10 of 19 esticates that the probability of excacding an 'offective' accci-cration of 0.75g in 50 years is 0.1 percent.
Anderson and Trifunuc (P.cf. 5) estimate that the probability of exceeding the high-frequency portion ci the llosgri design spectrum in 50 years varies from 10 to 20 percent, depending on the scismicity model considered The difference corresponding to a factor of 100 to 200 can be ana-ly:cd by considering the following factors:
(i)
The applicant considers the probability of excedance of an 'cffective' acceleration of 0.75g while Ander-son and Trifunac use as a basis of reference the O.75g llosgri design spectrum.
The use by the ap-plicant of an 'cffective' rather than ' instrumental' acceleration of 0.75g reduces the probability of ex-cedance by a factor of four.
(ii)
The use of Blume's SAM-IV and SAN-V attenuation re-lations as opposed to the use of the Trifunac's re-lations leads to re_ duction of the probability of excedance by a factor of ten.
(iii)
The rest of th,e differences corresponding to a fac-tor of 2.5-4 can be attributed to the different scismicity nodels considereu.
!!aving isolated the causes of the discrepancies in rish es ti-cation, I will discuss thca in detail.
I have indicated that the
' reduction of the peak acce]cration to an ' effective' icyc1 should not be used in the analysis of nucicar power pla::ts.
For the pur-pot.c of es tin:ating the risF. of exceeding the !!os tri design spec t ru:::,
9.
\\
T-a e n-p
\\
s-i 4
ATTACHMENT 5 Page 11 of 19 the anchor acceleration of 4.75 should be treated as actuol peak
~
acceleration.
In this case, the probability of excedance in 50 years as obtained by Blune's analyr.is would be of the order of 0.4 percent (refer to Tabic 11.8, D-LL 11) rather than 0.1 percent.
The main source of differences in scismic risk estimates can be associated with the type of acceleration-cagnitude-distance relation used.
The applicant's risk analysis is based on the use of the Bluce 's SAM-IV and SAM-V procedure.
In ny opinion, this' procedure Icads to accelcrations which da not reficct the strong notion in the near source region of large magnitude carthquakes.
If one considers.the three largest carthquakes for which records were obtained in the 'nca. source region, one finds that the ob-served peak accelerations are three to ten times larger than thos e predicted by the SAM IV-V procedure (Tabic 2).
Since the standard deviation for peak accelerations corresponds approximately to a factor of two, it may be cojn cluded that the SAM procedure is not valid in the near source region'of large carthquakes.
Tabic 2 indicates that Trifunac's relations 1 cad to accurate estimates of the observed peak accclcrations (the average ratio of observed to predicted peak acceleration is 1.07).
Fig. 41-I of Appendix D-LL 41 shous that the use of the SAM procedurc Icads to probabilitics that are 10 times louer than those obtained on the basis of the Trifunac's.rclations for the same sciscicity model.
The scisni-city nodel described in Appendix D-LL 11 Icads then to a probabil-ity of exceeding a peak acceleration of 0.75g in 50 years of the order of 4 percent.
e 10.
_.a
\\
ATTACHMENT 5 Page 12 of 19 i-I f
The seisnicity model used in Appendix D-LL 11 is based on the scisnic recurrence relation obtained by Smith for Central Coastal California (Appendix D-LL 11A).
These recurrence relations arc based on the scismicity during the period 1930-1975 and do not in-clude the 7.2M 1927 carthquake in the region.
The recurrence curves as shoun in Fig. 11A-2 of Appendix D-LL 11A underestinate the number of carthquakes with nagnitudes larger than six, and are based on a nominal value for the parameter b of 0.92.
Additional study by Srith (Appendix D-LL 45A) indicates that a more appropri-ate value for b would be 0.8S6.
The parancter b which controls the relative contribution of the high nagnitude carthquakes to the total scisnicity has n' strong effect on the calculated risk.
The use of b=0.CS6 would increase the calculated probabilitics by a factor of two (refer to Tabic 4 5.3 of Appendix D-LL 4 5).
The scisnicity model considered in Appendix D-LL 11 is consis-tent with the scismicity obtained in Appendix D-LL 41 using the a=2.S0 geologic record of fault dislo, cation (a,=3.12 in D-LL 11, 4
6 based on 10 years record and a= 3.20 based on 20 x 10 yc373 record in D-LL 41).
The scisnicity calculated on the basis of the geologic record of lateral fault slip does not include the scisai-liarailte n city associated with vertical slip along the !!osgri fault.
(Appendix D-LL 41A) quotes a report by Earth Science Associates in-dicating that the 'Jateral slip was probably subordinate to vertical the scisnicity should be increase?
novement.'
If this is the case, to account for vertical slip.
Conr.idering all the factors mentioned, it scens that the 11.
s
i s
ATTACHMENT 5 Page 13 of 19 i
l probability of 10 to 20 pertent in 50 years obtained by Anderson and Trifun c properly reficcts the scismic rish of excedence of t5c licsgri design spectrum.
o O
O e
0 e
e e
G 9
_p h
e 9
e S
4 9
12.
e k
- )
ATTACHMENT 5 Page 14 of 19 i
REFERENCES Scismic Evaluation for Postulated 7.5!f !!osgri Earthquake, 1.
Units 1 nnd 2, Diablo Canyon Site, Pacific Cas and
- Elcetric Company.
Trifunac, If.D., " Preliminary Analysis cf the Peaks of Strong
. i.
Earthquake Itotion-Dependence of Peaks on Earthquake ;iagni-
.tude, Epicentral Distance and Recording Site Conditions,"
Bull. Scism. Soc. of Ancr., Vol. 66, pp.189-219 (19 75).
Page, R. A., D.M. Boore.1'.B. Joyner, and II.W. Coul ter,
3.
Ground Motion Values for Use in the Scismic Design of the Trans-Alaska Pipeline System, U.S. Geological Survey Circular 672, 1972.
T-ifunac, M,D., " Forecasting the Spectral Amplitudes of Stron; 4
arthquake Ground Motion," Sixth Korld Conference on Earth-3 quake Engineering, New Delhi, India, 1977.
SJ
- Newnark, N.M., " Torsion in Symactrical Buildings," Fourth World Conference on Earthquake Engineering, Vol. II, A-3, Santiago, Chile, 1969.
Anderson, J.G., and !!.D. Trifunac, Uniform Risk Absolute 6.
Acceleration Spectra for the Diablo Canyon Site, California, to the Advisory Committee on Reactor Safeguards, A Report U.S. Nuc1 car Regulatory Commis sion, December,19 76.
j e
g 13.
i
\\
1 ATTACHMENT 5*
j-Page 15 of.19, TAnLE1.
COMPARISON OF MAX 1 MUM CROUND MOTIONS Penh value s M = 6.5 V = 7.5 uned by N ewma r k*
Trif u na c" US GS' Trifunac" USJ No. 672 No.,
0.75
,0.69(1.29) 0.90 1.07.(2.00) max (g) a-24 23(48) 39 39(84) 5*
max (in/sec) v 8
C(19) 16 12(30)
Zi dmax (in)
- Newmark, N..f., "A Rationale for Ocvelopment of Design Spectra for Diablo Canyon Rer.ctor Facility," Appendix C, Supplement No. 5, SER, Diablo Canyon Nuclear Power Station Units I and 2, NRC,1976.
Average (average + standard deviatign) peak motion for rock at an epicentral distance R = 7.5 km based on Trife.nac,' M. D., "Preliminar y Analpis of the Peaks of Strong Earthquake Ground Motion - Dependence of Peaks on Earthquake Magnitude, Epicentral Distance and Recording Site Condi-tion s, "
B. S. S. A., M, 169-219 (197 5).
Page, R. A., et al., "Cround Motion Values for Use in the Scismic Design
. of the Trans-Alacha Pipeline System," Ccological Survey Circu:ar 672, 1972.
~
\\
a gro.o* *.e t.
\\
L TABLE 2.
'Cor.parison of Recorded and Predicted Peak Accelerations.
Trifunac(b)
Recorded Predicted Ratio Predicted Ratio Peak Peak Observed /
Peak Observed /
Acccl.
Accel.
Predicted Accel.
Predicted cn to -i 9u zo 1971 Pacoima(1) 1.25g 0.124g 10.08 0.839g 1.49 Ng 1967 Koyna(2) 0.63g 0.213g 2.96 0.766g 0.82 k,
1976 Ga:li(3) 0.80g 0.190E 4.21 0.000g 0.89 5.75' l.07 (1) M-6.5, epicentral distance 3 km, focal depth 15 km.
(2) M-6.5, epicentral distance 5 km, focal depth 5 km (assumed).
- (3) M,=7. 2, cpicentral dis tance 10 km, focal depth 25 km.
i (4)jsVs=12,000 E-2.04,3 -o (5) s=2, p=0.50, u
/
e w wpe
_.y
.. a.
L3 I
i
.I I
I I
l l
l 1
'b.
.q,
>*0 s<nw r,
,.?
g@
f go 6
h s
&e
/.
~
+
/
c' I$ s-c
+
9 m
o ac
/
/
ma o
9 D
\\
4 g g4 p
l
- v. 0 c
4 Nt~
b 10
~
\\,
ma x
q)
.I Tf um o
E-8 ra "U y e
\\
a: p.
f S
e
$3,*c'-
\\
9, 2
'os t
'\\
o s
\\
\\
n,p.
no
) 'O o
r
'o vs t
,s' s
v g.O-
,r I
D l
l I
f I
- C't\\
l I
o.o r 0.or o.t o.2 a.C I
2.
5 to a
So no Frcquency, cps FIG.1 - DESIGN SPECTRUM COMPARED WITH PACOIMA DAM SPECTRA, 2 PERCENT DAMPING.
1 4.
s 1.
I, _.
4 8
'10 ATTACHMENT 5 Page 18 of 19 in
, r~
tt i
11 m
i 2
E.
Q 44 y
P*
e O
W Y
s es
_m a
2 t,
na s
^
- 2 JJ r
v v I.
v i]'
I c
g
~
r.
t-s.
g
' C 13 y&
v 0v f
O j
L I
W 3
v, j
.M p.
g s
.e >
/
11 o
- t...t L
.p G
t')
)
U
- 5' 0
=
s 8
.h l
i C
2 y
W n
e
(
d g
D O
f U
C l
tl h'
d 3
h 7
/
w e I-c
..<0'
/
6 s
z c
/
v.
E
./
./D L
l b
/
c I
(
\\
~
\\
g N
d E
u.
.y
\\
\\
~
N N
o'
~
w.
s
.s
. e....... k
..:,'. N.
O i
O.
In O,
m o
in O
in ci d
6 (s B s w.
t
/
J w...
e k
s ATTACHMENT 5 Page 19 of 19 O.
O b.
l' y.
~
'4
.T.
1 m
9 E
,1
.h f
a to o
j l
j g
.t s
.. f s
./
I~ q y,
s.
'21-Os M
' J','
m o
qe
(
OO t t!O
[
I C
(
qu s q
.l O
i e
o kl _ i...H.
. _.i. _.. '
!.. j) _.___.
qr i
Q 4
_ ',....._.tf._.. i _.m._..__ -
..__...<.)..._._.
<o<
.h_._g___(o._....,'....
r Q
g.
g Q,
. _.f.
O
~
/
- O "I4
~ ~~ -~-~ ~~
4'*-"
x x
- l ' " '" ~~'
O
' '.,.__ =.= 4
.= u-
- =, f...Q' W - - - -.
a.
__.../__._-
o
,4
___L.._
7 g '
._, _ _ z.
}-
1 u.__. -
.g 4
o u
c e
.p. _
.i,4 W
l
.i..
- j. vi
,.. E.
\\.
./. -
qs t~
A
._[._.
.......... _.n..)
O.
t-N b
L.
- . p~f ys n sn.-
.- ~<
Q
.N Ar X s
t-g.
e i
u.. n c 4
E, 1.
].
o i
. _m _. g...
q q
._.[
J......___p_..
i n
.i_.. -
r
........ ~.-_
\\-.
e
.._._....__a..._.--
- g..
. _1.
a.
..s 1
_..mm
.e..
4 I
O W
O O-O m
e,o w
l 6 ' tv o u w s, a c o v s i.n,o.r v y 4
i
\\
6 t
t.
UNITED STATES OF AMERICA NUCLEAR REGULATORY COM3,;SSION Before The Atomic Safety & Licensing Appeal I oard In the Matter of:
}
}
PACIFIC GAS & ELECTRIC COMPANY
)
Docket Nos. 50-275 OL (Diablo Canyon Nuclear Power
)
50-323 OL Plant, Units 1 & 2)
]
CERTIFICATE OF SERVICE I hereby certify that I have this 31st day of December, 1978, served copies of the foregoing JOINT INTERVENORS' REQUEST FOR DIRECTED CERTIFICATION upon all of the parties listed below by depositing copies thereof in the U.S. Mails, first-class, postage prepaid.
Elizabeth S. Bowers, Esq.
Dr. William E. Martin Chairman Battelle Memorial Institute Atomic Safety & Licensing Columbus, Ohio 43201 Board Panel U.S.
Nuclear Regulatory James R. Tourtellotte, Esq.
Commission L. Dow Davis, Esq.
Mail Drop East West 45L Richard Goddard, Esq.
Washington, D.C.
20555 Lawrence Brenner, Esq.
Office of Executive Mr. Glenn Bright Legal Director Atomic Safety & Licensing BETH 042 Board Panel U.S. Nuclear Regulatory U.S.
Nuclear Regulatory Commission Commission Washington, D.C.
20555 Mail Drop East West 450 Washington, D.C.
20555 Mrs. Elizabeth Apfelberg 1415 Cozadero Docket & Service Section San Luis Obispo, CA 93401 Office of the Secretary U.S.
Nuclear Regulatory Mr. Frederick Eissler Commission Scenic Shoreline Preserva-Wasl ington, D.C.
20555 tion Conference, Inc.
4623 More Mesa Drive Santa Barbara, CA 93105
<>r 2.
Sandra A.
& Gordon Silver Arthur C.
Gehr, Esq.
1792 Conejo Avenue Snell & Wilmer San Luis Obispo, CA 93401 3100 Valley Center Phoenix, Arizona 85073 John Phillips, Esq.
Center For Law In The Bruce Norton, Esq.
Public Interest 3216 N. Third Stree.t 10203 Santa Monica Boulevard Suite 202 Los Angeles, CA 90067 Phoenix, Arizona 85012 Paul C. Valentine, Esq.
Mr. Yale I. Jones, Esq.
321 Lytton Avenue 100 Van Ness Avenue Palo Alto, CA 94302 19th Floor San Francisco, CA 94102 Janice E.
Kerr, Esq.
Lawrence Q. Garcia, Esq.
Mrs. Raye Fleming J.
Calvin Sime on, Esq.
1920 Mattie Road California Pub _ic Utilities Shell Beach, CA 93449 Commission 5246 State Building F21B Technical Associates 350 McAllister Street 366 California Avenue San Francisco, CA 94102 Suite 6 Palo Alto, CA 94306 John C. Morrissey, Esq.
Philip A.
Crane, Esq.
Carl Neiburger James O.
Schuyler, Nuclear Telegram Tribune P ojects Engineer P.O. Box 112 Pacific Gas & Electric Company San Luis Obispo, CA 93402 77 Beale Street, Room 3127 San Francisco, CA 94106 kN 3 bfd David S.'Fleischaker, Esq.
1025 Fifteenth Street. N.U.
Suite 602 Washington, D.C.
20005