ML20237K194
| ML20237K194 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 08/10/1987 |
| From: | David Ferguson TEXAS UTILITIES ELECTRIC CO. (TU ELECTRIC) |
| To: | NRC - COMANCHE PEAK PROJECT (TECHNICAL REVIEW TEAM) |
| References | |
| CON-#387-4232 OL, NUDOCS 8708190131 | |
| Download: ML20237K194 (14) | |
Text
_ _ - _
1 REl.ATED CORRESP,0,NDf39fa
.f
.??
Filed: August i 1987."
UNITED STATES OF AMERICA NUCLEAR REGULATORY COMMISSION
}]((g m
before the ATOMIC SAFETY AND LICENSING BOARD 4
)
In the Matter of -
)
Docket Nos. 50-445-OL
)
50-446-OL TEXAS UTILITIES ELECTRIC
)
COMPANY et al.
)
)
(Application for an (Comanche Peak Steam Electric
)
Operating License)
Station, Units I and 2
)
)
FURTIIER ANSWERS TO CPRT INTERROGATORIES (Set No.12)
Interrogatory No.1 Interrogatory:
a.
Is it Applicants' position that they know of no way tosquantify the level of detection of the reinspection program for random errors?
b.
Is it Applicants' position that they know of no way to quantify the level of detection of the original inspection program for random errors?
c.
Is it therefore Applicants' position that Applicants' know of no way to calibrate one program to the other program (e.g..
the reinspection program to the original inspection program)?
d.
If Applicants' positing is different what is indicated in a.,
b., or c.,
above, please explain in precise detail how Applicants' position differs.
Answer:
a.
In the absence of any context or qualification the applicants do not agree with the statement. (See answer to Interrogatory Id.)
b.
Within the absolute sense of the question, the applicants do not agree with this statement. (See answer to Interrogatory Id.)
l
'gDRB708190131 870s10 40)
~
ADOCK 05000445 PDR
c.
Again, within the absolute sense of the question, the applicants do not agree with this statement. (See answer to Interrogatory Id.)
d.
For the purpose of this response to the interrogatories concerned, we define a " random error" to mean an error made in completing a +1 inspection check list by a qualified QC inspector, who has been certified and properly trained to perform the inspection and who is performing the inspection to a procedure adequately detailed with appropriate directions.
We have noted in interrogatories la, b, and c that our answers depend on the context of the question and on the degree of absoluteness intended by the questions.
Considering question la, it is our view that it might be possible to set up an experiment or a type of qualifying test to judge the level of detection for each third-party inspector more precisely than is established by the inspec-tor's satisfaction of applicable qualification, certification and training standards.-
The practicality and usefulness of such an experiment (or qualifying test) is questionable, when one considers the myriad of deviation types, deviation magnitudes and inspection conditions an inspector must confront in the field and for which he or she ' would have to be tested in order to determine specific inspection reliability.
Therefore, while we cannot agree with the stated assertion 1
of interrogatory la in the absolute sense, we know of no practical way to quantify with high precision the detection level of the reinspection effort.
With regard to interrogatory I b, to the extent that the concern is with regard to inspection efficiency or reliability, the Applicants probably do not agree with the stated assertion.
However, short of going back to the original inspection of items and identifying all the deviations that were found during this period, where the defective hardware item was repaired or reworked or replaced and found to be acceptable, and then somehow determining the number of 2
Y
deviations that were not origirially identified during the original inspection, assuming 100% reliability of the latter effort the task cannot be accomplished.
Even could it be accomplished, it would be of limited scope, siace only the accessible and recreateable att~ibutes are now capable of being reinspected.
Finally, it is questionable what one would do with the data, were it obtained, beyond the sort of things that CPRT and TU Electric are already doing with the data obtained from the CPRT Pr(gram.
Our answer to interrogatory Ic depends upon the usage of the word calibrate.
Basically, the weld " calibrate" means to standardize a device or a program that is in question against a controlled (standardized and accepted) device or mogram.
To the extent that either the historical inspections or the CPRT inspections are accepte d as a standard, it is not possible to calibrate one to the other with numeric al precision unless both inspee. ion efforts can be quantified numerically.
Eve i if it were possible to quantify them, it is not clear what the purpose of such a calibration would be.
On the other hand, if
" calibrate" is used to in :lude the concept of qualitative romparison, then qualitative comparisons can be made and the Applicants would disagree with the stated assertion.
Given the SRT's ability to judge the adequacy of CPRT inspector training, certifi:ation and qualifications, and the adequacy of the procedures, instructions ard checklists used by the inspectors, it follows that the CPRT inspectors are cc.mparable in effectiveness to the industry norm of acceptance inspections us ing qualified and properly trained people working to l
adequate procedures.
I
~
~
?
i
[
+
i The CPRT has tested the " process" that Additional Board Questlom created the as-found state of the plant as of the time of the CPRT reinspection effort.
The " process" was inclusive of all those actions taken by the Project prior. to CPRT, The. end result of the CPRT investigation will be a set of conclusions.
In some cases, the conclusion will be that the all that went into the " process" worked well enough prior to the CPRT effort to have provided
, reasonable assurance that all defects that would have prevented the plant from being capable of operation in conformance with NRC regulation were detected and corrected (i.e.,
no construction deficiencies, adverse trends, unclassified deviations, unclassified trends, or special cases identified during reinspection).
In other cases, when the " process" did not work well enough to have l
provided reasonable assurance (i.e., CDs, ATs, UDs, UTs or SCs identified during l
reinspection),. a corrective action program will be recommended that when implemented will provide reasonable assurance that all defects that would have prevented the plant from being capable of operation in conformance with NRC regulations were detected and corrected.
Much of the CPRT investigation is based on a 95/5 sampling screen for deficiencies.
But, in fact, the sampling also provides a more rigorous screen since hardware deviations are evaluated for their potential to indicate or implicate adverse coaditions in the unreinspected portion of the plant.
The
" process" tested by the CPRT sampling effort includes sub-processes, each of which may have contributed to the as-found state of the plant.
These sub-processes include the effectiveness of engineering to provide design margins up 1"[E)xplain clearly the methodological basis and reasoning for stating that
'The SRT does believe it to be so, and demonstrable, that the level of reasonable assurance afforded by the CPRT Program, when completed, will be comparable to that that would be expected to result from the implementation of a 10 CFR Part 50, Appendix B, program.'" ASLB Memerandum and Order 3/16/87, at 2.
to and beyond code allowables, the effectiveness of construction to follow design requirements within tolerances, tim effectiveness of the original (prior to CPRT) corrective action programs to correct deviations, the effectiveness of other programs such as testing, operations, maintenance, etc., to detect and correct deviations and not to create new ones, and the effectiveness of audits and overviews to identify problems with any of these programs and to cause effective correction.
The CPRT Quality of Construction (QOC) reinspection program does not 4
replace the 10 CFR Part Part 50, Appendix B program for CPSES, nor is it designed, of itself, to answer questions about the original CPSES Appendix B program.
On the other hand, it was designed to test the adequacy of the end j
product of the construction and quality assurance process (i.e.,
the installed hardware).
With regard to the quality of construction relative to the original design, it is clear that a well conceived and executed QA/QC program can compensate for a poorly executed craft / construction program.
It is also clear that a good craft / construction program, in the absence of an effective QA/QC program, could also produce an acceptable end product.
In fact, there are a number of combinations of QA/QC effectiveness and construction performance that would produce an acceptable end product.
The CPRT in essence is testing the effectiveness of the total process that existed at CPSES prior to CPRT by reinspecting the end product.
Although the CPRT findings are not being directly compared to those at other plants that are now operating and that were i
determined to have compliant Appendix B programs, the findings on the end j
product are being compared by experienced engineers to the standard of reason-able assurance that all construction defects that would have prevented the facility from being capable of operation in conformance with NRC regulations
~
1
c O
have been detected and will be corrected by appropriate corrective action.
Finally, it should be understood that Appendix B programs do not supply
" reasonable assuiance" on the basis of assumed perfection.
To the contrary, such programs will pass errors, as will all human programs.
" Reasonable assurance," rather, comes from the existence of a program with a reasor: ably high probability of detecting failures in other aspects of the process, alerting someone to the need to investigate and eradicate the cause and the product (s) of the failure. The CPRT program functions in precisely the same way (except that it operates after the fact).
2 Additional Board Question:
If by ' errors" by Board means deviations, the CPRT program will not directly predict the number of residual deviations that existed just prior to implementing the reinspection.
If by " errors" the Board means those deviations with potential safety significance (deficiencies), the CPRT program will identify those cases in which the level of residual deficiencies that existed prior to implementing the reinspection is high enough that the samples do not pass the 95/5 statistical screen.
it will also identify those cases in which the level and severity of residual deviations are high enough that the data collected from the samples do not pass the even more demanding adverse trend and root cause/ generic implication screens.
Reworded Interrogatory No. 3(b)(2)
Interrogatory:
[ identify] any documents that were in fact relied on by [the Applicants) in producing their answer to [ Interrogatory 3(b)(1)].
2" Discuss whether it is true that by inspecting the hardware Applicants will j
obtain information concerning the number of errors left in the plant after QA/QC and other events (such as testing? reconstruction?) have occurred, prior f
to the implementation of the reinspection program."
ASLB Memorandum and i
l Order 3/16/87, at 2.
l.
I Answer:
The only documents relied upon in the ' development of the response to Interrogatory 3(b)(1) are those cited directly ([1] Dixon and Massey, introduction to Statistical Analysis, [2} Box and Tiao, Bayesian Inference in Statistical Analysis, [3} ' Boissonnade, 'CPRT Sampling Plans -- Addendum), plus (4) Bowker and Lieberman, Engineering Statistics 2d ed., [5] Lieberman and Owen, Tables of the Hypergeometic Probability Distribution, and [6} Benjamin and Cornell, Probability, Statistics and Decision for Civil Engineers, which are referenced in either the CPRT Program Plan, Appendix D, or in the documents cited in the response to Interrogatory 3(b)(1).
References [1], [2] and [4]-[6] are in the public domain. Reference [3] is attached.
Interrogatory No. 3(c)
Interrogatory:
(1)
For enh ISAP, which possible strata have been identified by the Applicants?
(2) Where is such identification contained in the Applicants' documentation? (3) Provide it if it has not already been provided.
Answer:
The extent of the CPRT knowledge prior to execution of the Action Plans I
with regard to the existence of strata is detailed in each Action Plan.
By the nature of most of the TRT Action Plans, one or more strata are defined by the particular problem identified in the Action Plan (e.g., unauthorized plug welds in pipe-support base plates and cable tray hanger supports, ISAP V.d).
The self-initiated Action Plan, Vll.c, has focussed on a set of strata based on construc-tion work categories (CWCs), homogenous work activities (HWAs) and safety-significant attributes.
A preliminary list of CWCs is given in the CPRT Program Plan.
The final list of CWCs, HWAs and the attributes reinspected will be contained either the Action Plan VII.c Results Report or in the Working Files i.
associated with that Action Plan.
Interrogatory No. 6(g)
Interrogatory:
(i)
[ Identify any document that was relied on in the course of developing CPRT Program Plan Appendix D.]8 (iii)
If such documents are not available....
Answer:
(i)
Appendix D is based on the following references:
1.
Webster, " Developing Sampling Plans for TRT Issues," CPRT File, 03/12/85.
2.
Boissonnade, "CPRT Sampling Plans -- Addendum," CPRT File, 12/27/85.
3.
Webster, " Sample Plans for Finite Populations 100, " CPRT File, 8/15/85.
4.
Bowker and Lieberman, Engineering Statistics,2d ed., Prentice Hall,1972.
5.
Owen, Handbook of Statistical Tables, Addison-Wesley,1962.
6.
Lieberman and Owen, Tables of the Hypergeometric Probability Distribution, Stanford University Press,1961.
i 7.
Rand Corporation, A Million Random Digits, Free Press,1955.
8.
Benjamin and Cornell, Probability Statistics and Decision for Civil Engin-eers, McGraw-Hill Book Company,1970.
9.
Ang and Tang, Probability Concepts in Engineering Planning and Design.
John Wiley & Sons,1975.
10.
Box and Tiao, Bayesian inference in Statistical Analysis, Addison-Wesley, 1972.
References 1,
2 and 3 are attached, references 4-10 are in the public domain.
(iii)
Not applicable.
- See ASLB Memorandum and Order 3/l6/87, at 4. --- J
'4 4
Interrogatory No.-6(h) and (i)
.g Question:
Identify all individuals, in addition to Mr. F.
A. Webster, who had any
-involvement in the planning and formation stages of Appendix D or the CPRT Plan, including the original and all revisions, and in any changes to Appendix D.
For each. individual listed in' your response ' to question h. above, answer the following questions (i)
Describe all involvements the individual had in the planning-and formation states of Appendix D. Answer for the original and each revision.
(ii) ' Identify all activities which the individual' performed or participated in j
regarding Appendix D (the original and each revision).
l (iii)'
(See ASLB Memorandum and Order 3/16/87 at S.)
(iv) Provide a resume for each individual...
Answer:
h:
Revision 0:
y.
' Messrs.' Webster and John Reed, Jack Benjamin Associates.
Messrs. Don Davis, Howard Levin, John Guibert and Frank Dougherty, TERA Corp.
Mr. Terry Tyler, ENERGEX.
Mr. John Hansel, ERC.
The SRT Members (Messrs. Beck, French, Buhl and Guibert).
Revision 1:
Messrs. Webster, Reed and Auguste Boissonnade, JBA.
Messrs. Davis and Levin, TERA.
Mr. Robert Lee, Leaders in Management, Inc.
Mr. Terry Tyler, TUEC.
The SRT Members (Messrs. Beck, Buck, Nyer, Buhl, Guibert and French).
i(i) and (ii).
The individuals listed under Revision 0 were involved in t
4 discussions in early 1985 regarding the philosophy and appropriateness of the sampling program.
Based on these discussion a draft of a sampling philosophy was developed, which eventually became the first three sections of Revision 0 of Appendix D.
The remaining sections and attachments were developed by Mr.
j I
Webster with various reviews by the individuals listed.
Changes and revisions to Appendix D were based on various discussions among the individuals listed above.
All technical changes were generated by Mr. Webster.
(iii)
With the exception of Messrs. Beck (SRT Chairman) and Tyler (formerly CPRT Program Director), all of the persons listed meet the CPRT Program Plan Objectivity Criteria.
(iv) The Applicants believe that CASE has previously obtained a resume, to l
the extent available, for each of the persons listed.
If CASE believes otherwise, it should identify the person (s) for which it does not believe it has already received a resume, and the Applicants will determine if one exisu.
Interrogatory No. 7 Question:
(See ASLB Memorandum and Order 3/16/87, at 6.}
Answer:
With respect to Action Plan Vll.c: Only those hardware items that were accessible and QC-accepted at the time of initiation of the CPRT reinspection i
effort were included in the sampling program.
To the extent that items may have been reworked and QC-reaccepted as a result of TRT or other investigation efforts, these were included in the CPRT program as potential samples.
The sampling program represents a " snap shot" of the accessible, QC-accepted I
hardware at a particular point in time (approximately August-September,1985).
j l
l _
._______________b
s
,e.
There is a possibility that a few of the items that were sampled have been deleted since the reinspection, but the number, if any, is small.
Changes to hardware that have occurred or will occur as a result of DAP findings or TU Electric
- Corrective Action Program" redesign efforts (or, indeed, any other cause other than CPRT-recommended ' corrective actions) are not a part 3
)
of the sampling program.
Changes to hardware that have occurred or will occur as a result of CPRT-recommended corrective actions will be overviewed to the extent and in the manner set forth in Revision 4 of the Program Plan, Appendix H.
The Action Plan statistical statements are designed to prove that the population in question does or does not pass the statistic screen.
Such statist-ical statements are valid and apply to the populations from which the samples were obtained.
With respect to Action Plans other than VII.c: What is being sampled varies from Action Plan to Action Plan.
In general, populations being sampled are narrowly defined in order to address the specific existing or possible hypothe-sized problems.
For example, Acticn Plan III.b is concerned with compliance with FSAR pre-operational test commitments.
Thus, the sample was obtained from the complete list of FSAR pre-operational test commitments.
Action Plan II.b is concerned with surface hardness of exposed safety-related (Category I) concrete placed between January, 1976, and February, 1977.
Therefore, the sample was obtained from the testable concrete in this time frame.
In other words, what is sampled depends upon the issue, as stated in the Action Plan.
With respect to items possibly deleted and the effect of reworking, see the foregoing regarding Action Plan VII.c.
~
e 4
SIGNATURES I, Donald R. Ferguson, being first duly sworn, do. depose and say that I am the CPRT Program Director, that I am familiar with the information contained in the CPRT files and available to CPRT third-party personnel, that I have assisted in thei preparation of the foregoing answers, and that the fore 8olag answers are true, except insofar as they are based on information that is available to Texas Utilities or the CPRT (third-party personnel) but not w!!hin my personal knowledge, as to which I, based on such information, believe them to be true, be>ald $.
ym Donald R. Fergusof Swora to before me this 7 day of August,1987:
'AU
' G_A otary Public l '
j
/9g My Commission expires:
l'O / a 4/i As to Objections:
"l id (k ~
Thomas G. Diln'an( Jr.
R. K. Gad III William S, Eggeling Kathryn A. Selleck Ropes & Gray 225 Franklin $treet Boston, Massachusetts 02110 (617) 423-6100 Attorneys for Texas Utilitics Electric Company l
l l
l.
- TOTAL PAGE.02 **
I en w nns
-m nu i m e ninw::Nm m st wm
]
,-J DCf.ki: Ti;!
~
UviF C CERTIFICATE OF SERVICE
'87 A3314 A10 :34 I,
R'.
K. Gad III,. hereby certify that on August 10, 1987, I r
o%
v made service of the "Further Answers to CPREUENtyprog'idr'ies'(Set a
~
No. 12)" by. mailing copies thereof, postage prepaid, to:
Peter B.
Bloch, Esquire Mr. James E.
Cummins j
Chairman Resident Inspector j
Administrative Judge Comanche Peak S.E.S.
i Atomic Safety and Licensing c/o U.S. Nuclear Regulatory Board Commission U.S. Nuclear Regulatory.
P.O.
Box 38 Commission Glen Rose, Texas 76043 Washington, D.C.
20555 Dr. Walter H. Jordan Ms. Billie Pirner Garde Administrative Judge GAP-Midwest Office 881 W.
Outer Drive 104 E. Wisconsin Ave.
-B
. Oak Ridge, Tennessee 37830 Appleton, WI 54911-4897 Chairman Chairman
' Atomic Safety and Licensing Atomic Safety and Licensing Appeal Panel Board Panel U.S. Nuclear Regulatory U.S. Nuclear Regulatory Commission Commission Washington, D.C.
20555
' Washington, D.C.
20555 Lawrence J.
Chandler, Esquire Mrs. Juanita Ellis Office of the Executive President, CASE Legal Director 1426 S.
Polk Street U.S.. Nuclear Regulatory Dallas, Texas 75224 Commission Washington, D.C.
20555 1
i i
2 Renea Hicks, Esquire-Ellen Ginsberg, Esquire Assistant Attorney General Atomic Safety and Licensing Environmental Protection Division Board Panel P.O.
Box 12548, Capitol Station U.S. Nuclear Regulatory Commission
' Austin, Texas 78711 Washington, D.C.
20555 Anthony Roisman, Esquire Mr. Lanny A.
SinkinL Suite 600 Christic' Institute
.1401 New York Avenue, N.W.
1324 North Capitol Streek Washington, D.C.
20005 Washington, D.C.
20002 Dr. Kenneth A. McCollom Mr. Robert D. Martin Administrative Judge Regional Administrator
'1107 West Knapp Region IV Stillwater, Oklahoma 74075 U.S. Nuclear Regulatory Commission Suite 1000 611 Ryan Plaza Drive Arlington, Texas 76011 Elizabeth B.
Johnson Geary S. Mizuno, Esq.
Administrative Judge Office of the Executive Oak Ridge National Laboratory.
Legal Director P.O.
Box X, Building 3500 U.S. Nuclear Regulatory Commission Oak Ridge, Tennessee 37830 Washington, D.C.
20555 Nancy H. Williams 2121 N.
California Blvd.
Suite 390 Walnut Creek, CA 94596
)s :.
x z: -
.,q R.
K.
Ga /III l
l l
_ _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ -