ML20236D162
| ML20236D162 | |
| Person / Time | |
|---|---|
| Site: | Oconee, 05000000 |
| Issue date: | 10/20/1987 |
| From: | Shao L Office of Nuclear Reactor Regulation |
| To: | Varga S Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20195F761 | List:
|
| References | |
| FOIA-87-714 NUDOCS 8710280049 | |
| Download: ML20236D162 (3) | |
Text
--
[
'o UNITED STATES g
8 NUCLEAR REGULATORY COMMISSION e
f W ASHINGTON, D. C. 20665
% *..,* /
October 20, 1987 I
MEMORANDUM FOR:
Steven A. Varga, Director Division of Reactor Projects FROM:
Lawrence Shao, Director Division of Engineering and Systems Technology
SUBJECT:
ALLEGED BACKFIT CONCERNING TWO STAFF POSITIONS INVOLVING CONTAINMENT INTEGRATED LEAKAGE RATE TESTING i
By letter dated September 1, 1987 Duke Power Company (DPC) asserted that the staff's interpretations of two Appendix J items constituted backfitted require-ments as defined by 10 CFR 50.109. The two items were: (1)thestaff'sreluc-tance to allow short-duration tests using the Mass-Plot (or Ness Point) analysis a
2 method, and (2) the staff's requirement that as-found (Type B and C) leakage be included in Type A test results. The purpose of this letter is to respond to these assertions in accordance with NRR Office Letter No. 52 and NRC Manual 4
Chapter 0514.
J We have reviewed the beses provided to support DPC's conclusion that staff's interpretations in the two areas constituted backfits as defined by 10 CFR 50.109.
Based on this review, we find that the first item does not meet the criteria outlined in the regulation to define the issue as a backfit issue. For the second item, we find that the' staff interpretation was in place before the effective date of the backfit rule. Therefore, the question of backfit is beyond tha scope of 10 CFR 50.109 regulation with respect to the second item.
Ir. sumary, the staff has found that the two items identified in DPC's September 1, 1987 letter (copy enclosed) should not be subjected to the regulatory analysis requirec for a backfit under 10 CFR 50.109. A discussion of DPC's arguments is presented below.
SHORT-DURATION TESTING USING THE MASS POINT METHOD Tr.e first issue raised by DPC concerns the minimum period of time the Type A integrated leak rate test (ILRT) should be conducted and the test methodology
~utilizec. DPC's assertion is that the staff has supplemented this test time requirement of Appendix J through staff positions which become backfitted requirements when enforced. DPC bases this position on two points: (1)the i
language in the standard - " those responsible for the acceptance of the containment structure"-- means the licensee and not the NRC; and (2) the NRC's l
recent unwillingness to accept short duration testing with Mass Point method l
represents a change in staff position.
The duration is not exp11 city defined in Appendix J.
However, 10 CFR 50, Appen-dix J, II.D states that the leakage rate for test purposes is that leakage which occurs in a unit of time, stated as a percentage of weight of the original con-tent of containment air at the leakage rate test pressure that escapes to the j
N outside atmosphere during a 24-hour test period. Further,AppendixJ,III.A.3.(a)
I
~;sw-
^
i c 1 1
)
states that 611 Type A tests shall be conducted in accordance with the provi-f sions of the American National Standard N45.4-1972. The test time is discussed in American National Standard ANSI N45.4-1972, " Leakage Rate Testing of Contain-ment Structures for Nuclear Reactors," dated March 16, 1972. ANSI N45.4-1972 states that leak tests shall be conducted for a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> period.
Exceptions to the test duration are permitted provided that the exceptions are reviewed and approved by "those respoi.sible" for the acceptance of the containment structure, j
DRC states in its September 1, 1987 letter that the licensee is responsible for l
acceptance of the containment structure and, therefore, DPC is responsible for l
both the test method and test duration and that the staff's failure to accept l
this position constitutes a backfit.
I The NRC's position regarding the responsible party for the determination of acceptability of short-duration tests has been well established. The respon-sibile party is the hRC and not the licensee as asserted by Duke. This position was discussed in a memorandum from G.C. Laines (NRR) to A.F. Gioson (Region II),
" Duration of the Integrated Leak Rate Test for Containment," Jated July 8,1987 I
(Enclosure 2 ).
Additionally, the duration of Type A tests and the need for NRC l
approval of methodologies of shorter curation is discussed in memorandum from D. Davis to Regional Directors dated March 21, 1974 (Enclosure 5). Therefore, the staff position that Type A tests are to be conducted for a test period of 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> is a longstanding position and not a backfit.
The licensee asserts (September 1,1987 letter Enclosure 1, page 3) that a staff position was established when an NRC inspector witnessed a test of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> duration and did not identify any violations, thereby, giving tacit appreva' to short-duration testing using Mass Point. Concerning DPC's assertion that the staff has recently changed a long standing position relative to the acceptance of short-duration testing using the Mass Point method, DPC stated that this original position was in effect since the inception of Appendix J.
As evidence of these conclusions, Duke cites two NRC inspection reports (50-270/
B3-30 and 50-287/81-04) for Oconee Units 2 and 3, in which the NPC " tacitly" accepted two short-duration ILRTs that also used the Mass Point method. These tests were conducted in 1981 and 1983.
As defined in NRR Office Letter No. 52 and NRC Manual Chapter 0514, the staff position is not an interpretation of the more general regulations as would be contained in the SRP or Regulatory Guides.
In the prcsent case, the staff posi-l tion is a legal requirement, deriving from explicit regulations. The legal l
requirements incorporated by reference into the regulation from ANSI N45.4-1972 l
as discusseo above, is that test duration shall be at least 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The stan-l dard al.so allows for shorter tests with NRC approval, but such approval must be l
formal and explicit, insofar as it represents a deviation from the basic explicit l
Tacit approval in inspection reports does not suffice.
In l
legal requirement.
i fact, action by a single region is also insufficient, because any exception to i
the basic legal requirement should apply to the entire industry. Therefore, only formal, explicit action by staff sufficies as NRC approval.
i h
i
a A In summary, the staff does not agree with DPC's assertions. The staff's position is a long-standing policy requiring all tests to have at least a 24 hour2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> duration, unless the test is performed in accordance with the provisions of the Bechtel Topical Report BN-TOP-1, Revision 1, "Testino Criteria for Integrated Leakage Rate TE. sting of Primary Containment Structures for P!uclear Power Plants," dated November 1, 1972. One provision of this approach is the requirement of the Total Time method for calculating the leakage rates. This staff position has been in place since October, 1973 (Enclosure 5). Although this is an internal NRC memorandum, our position has been widely disseminated to the public, both orally and through documents available to the public. Industry awareness of our position can be further demonstrated by reference to an EPRI l
report on ILRTs (Enclosure 7) published in 1982.
In this report, the staff's position is correctly described and discussed on pages 3-32 and 3-33.
"AS FOUNO" ILRT LEAKAGE PATE The staff's positien recuiring "as found" ILPT leakage rates to be determined and a test failure to be declared if the "as foubd" result exceeds the acceptance criterion given by Appendix J was formally sent to industry by information Notice 85-71, dated August ??, 1985. However, the backfit. regulation.
l 10 CFR 50.109, states that a regulatory analysis for backfit is not required for actions taken before October 21,1985. Therefore, this particular staff position is rot sub.iect to the provisions of the referenced hackfit regulation.
J J
[ l /[';,' f W c %. l h wrence Shao, Director
- diivision of Engineering end Systens Technology
(
Enclosures-4 < tater I
J.
DPC letter of Septernber 1,1987
?
Memorandum from G. Laines to A. Gibson Jub 8,1987 3.
Inspection Report 50-270/83-35, dated December 15, 1983 4..
Inspection Report 50-269/81-04, dated April 6,1981 S.
Memorandum from Davis to Regional Directors dated Parch 21, 1974 6.
IEIN No. 85-71 7.
EPRI NP-2726, Final Peport, " Containment Integrated Leak-Rate Testing Improvements," November 198?,
l pages 3-3? and 3.13 i
i
Nec.fy JP %I'S], dw !* ;
DUKE POWER GOMPm Encl 05ure I p.o. sox 33 80
{
cnAarorTz, N.c. 282 42 q
['['
l HA1. B. TL CKE R p
l pC g3 g,,/ /b i
September 1, 1987 f
f k
i krM/kC U.S. Nuclear Regulatory Commission Document Control Desk i
L2ashington, D.C.
20555 l
Subject:
Oconee Nuclear Station i
Docket Nos. 50.-269, -270, -289
)
McGuire Nuclear Station
]
Docket Nos. 50-369, -370 q
Catawba Nuclear Station Docket Nos. 50-413, -414 j
Containment Integrated Leak Rate Testing l
i Imposition of Backfit by Virtue of Change in Staff Position Gentlemen l
l
SUMMARY
l In recent sonths the NRC Staf f has adopted various interpretations of 10CFR50 Appendix J vhich appear to be contrary to previous Staf f Positions and which do not appear to be soundly based upon the text of Appendix J.
Two items of par-ticular concern to Duke Power are 1) the Staf f's reluctance to allow short-dur-crion tests using the Mass-Plot analysis method, and 2) the 3taff's requirement that as-found (Type B and C) leakage be included in Type A test reeults.
The purpose of this letter is to assert Duke's position that these interpretations constitute backfit requirements, as defined by 10CTR50.109, and as such should be subjected to the Regulatory Analysis required by 550.109.
DISCUSSION In 1973,10CPR50 Appendix J went into effect, referencing ANSI N45.4-1972 (Leak-ege-Rate Testing of Containment Structures for Nuclear Reactors) as the operative machod for performing Leak-Rate Testing at nuclear stations.
Since that time Duke has performed leak-rate testing according to Appendix J, with some exemptions.
As time and technology have progressed, the NRC Staff has supplemented the re-q quirements of Appendix J through Staff Positions, both tacit and explicit, which 1
bacome backfitted requirements when enforced.
For example in 1977 the Staff, absent any revision to Appendix J, began requiring that Integrated Leak Rate Test (ILRT) acceptance criteria include a 95 percr.nt Upper Confidence Limit (UCL),
i rather than actual measured leakage.
Inspection of Appendix J and ANSI N45.4-1972 rsveals no mention of UCL.
This requirement was presumably based on the then-draft revision to the standard, which was issued in 1981 as ANSI /ANS-56.8.
Cloarly, had the backfit rule been in effect in 1977 the imposition of the more atringent UCL requirement would have been considered as such.
Y Ftf 4 j p g 1.-
t
.A Document Control Desk September 1, 1987 Page 2 Another example of a changed Staff position as it af fects Appendix J testing relates to the method used to analyze ILRT data.
ANSI N45.4-1972 requires that either the " total time" or the " point-to-point" method be used to analyze ILRT data. However, starting in 1976 the Staff endorsed, again absent any revision to
- Appendix J, the, Mass-Plot (Mass-Point) mothed.
As late as 1986 (reference: Staff R, view of 14ak-rate Methodology, Deputy Director, NRR to Director, Div. of Inspec-tion Programs, IE, April 1, 1986) the Staf f has recognized the acceptability of the Mass-Plot method.
On August 1,
1986 Duke was notified by the Staff that Mass-Plot was not an acceptable analysis method because Mass-Plot was not provided for to Appendix J.
Duke had already obtained exemptions to allow use of Mass-Plot et the McGuire and Catawba stations, and subsequently obtained an exemption for Oconee.
The preceding two examples illustrate instancas in which "Staf f Position" has achieved virtually regulatory status without being accorded the due process required for rulemaking.
Fortuitously, neither has to date resulted in signi-ficant adverse impact. There are, however, two issues related to ILRTs which loom es significant contributors to increased probability of test failure and increased outage time.
These issues are short-duration Mass-Plot testing and inclusion of as-found leakage from Type B and C testing in Type A test results.
Appendix J, through ANSI-N45.4 requires that ILRTs be performed for a 24-hour period, except "if it can be demonstrated to the satisfaction of those responsible for the acceptance of the containment _ structure that the leakage rate can be accurately determined during a shorter test period, the agreed-upon shorter period may be used."
(ANSI-N45.4, paragraph 7.6).
There are two important points in this exception.
First, the exception is made without reference to the method of test analysis contained in the standard.
Second, the exception refers to "those responsible for the acetptance of the containment structure." The NRC Staff, in its April 1, 1986 review of leak rate testing methodology stated that the only test of less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> which is acceptable to the NRC is the method specified in the Bechtel Corporation Topical, BN-T OP-1 (the total time and point-to-point methods).
The Staff further contends that "those responsible for acceptance of the containment structure" refers to the NRC.
In Inspection Report Number 30-269/86-13 (June 30,1986) the Staf f states " ANSI-N45.4, paragraph 7.6 requires a 24-hour test unless a test of shorter duration has been agreed upon by NRC."
The Report also states "The Region believes that the second sentence [of paragraph 7.6) is singly a statement of the obvious; specifically, that with the review and approval of the regulatory body that initially approved the rule, in this case the NRC, an acceptable alternative to the requirements of that rule may be implement-cd."
It is not obvious at all that "those responsible for.the acceptance of the con-tainment structure" and "the regulatory body that at roved the rule" are one and the same. The licensee is responsible for De operat.zon, maintenance, and overal.1 l
quality of the nuclear station.
Acceptance of the containment structure is one espect of the licensee's responsibility to assure that the health and safety of the public is not endangered.
Tne ability to verify this at.ceptability in less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, using Mass-Plot, has been explicitly recognized in the revised a
standard ANSI /ANS-56.8-1981, " Containment System Leakage Testing Requirements."
It is implicit in the current standard, N45.4-1972, that if the acceptance cri-teria are met, the test duration is irrelevant.
l J.
Document Control Desk S2ptember 1, 1987 Pcge 3 The recent unwillingness of the Staff to accept short duration Mass-Plot testing represents a change in Staff position.
The table below lists two Inspection R3 ports in which the NRC Inspector witnessed and, by signing the report with nc related violations or identified items, tacitly approved short-duration tests using Maes-Plot.
MINIMUM TEST DURATION REPORT NO.
REPORT DATE TEST DURATION PER PROCEDURE 50-270/83-35 De c. 15, 1983 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br /> 25 min.
6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> 50-287/81-04 April 6, 1981 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br /> 45 min.
Not in procedure Note that the inspector (s) reviewed at least one procedure which specified a ninimum 6-hour test.
As noted, the provision in the standard to allow short duration testing did not make reference to any of the analysia methods included in the standard.
In cddition to the total time and point-to-point methcds (two options in the broader category of the absolute method) the Standard describes testing by the. Reference-Vsssel Method.
The Reference-Vessel Method is distinctly.different from the cbsolute method; nevertheless, the Standard, in paragraph 7.10, reaf firms that "If it can be demonstrated to the satisfaction of those responsible for the acceptance of the containment that the leakage rate can be accurately determined during a shorter test period, the agreed upon shorter period may be used".
It follows that if two diverse methods have the capability to satisf actorily determine leakage rates in less than 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, then the duration of the test is not as critical a perameter as the April, 1986 Staff position seems to indicate.
In accordance with the backfit policy guidance presented in Chapter 0514 of the NRC Manual, the elements of imposition of a backfit are satisfied.
The Staf f's original position of acceptance of short duration testing is documented in the cbove inspection reports.
The Staff's new position is documented in the April, 1986 Staff review of leakrate methodology.
Actual imposition of the backfit has been accomplished by the need to perform 24-hour tests or risk ILRT failure. The previous Staf f position had been in ef fect since the inception of Appendix J.
The issue of as-f ound leakage has already proven significant at McGuire Nuclear
)
Station, where one unit was ruled to have failed an ILRT by virtue of having not
~
included as-found leakage in Type A test results.
The Staf f's position, estab-(
lished in Information Notice 85-71, was identified too close in time to oched-l uled ILRT to allow resolution in a timely manner.
Information Notice 63-/1 was dated August 22, 1985; before the effective date of 10 CTR 50.109 (October 21, t
I 1985).
However, the backfit policy identified in Generic latter 84-08, which is cimilar to $50.109, is applicable to this issue.
f Appendix J appears to have been predicated upon the intent of assuring containment integrity for the period of operation following the test. The acceptance criteria for the test require that the measured leakage (LTM) rate be 75% of the allowable l
leakage (LT) rate.
Thus, assuming a nominal degradation (25%) of containment l
l integrity over the subsequent period, the leakage *:11; still be within the allow-cble limits at the conclusion of the period.
l l
s 4
Document Control Desk September 1, 1987 Pcge 4 Tha inclusion of as-found le akage in Type A test results thus causes a double-counting of normal (anticipated) containment integrity degradation.
The.75 La cceeptance criterion of Appendix J concedes an acceptable amount of expected d: gradation.
To further require that the actual degradation of the containment (i.e., the as-Tound leakage) result in a total leakage of less than.75 La, rather
~
than 14, is beyond the intent of the rule.
Any other interpretation must refute
' ths definition of La as an acceptable leakage and.75 La as an acceptance criter-icn.
Tha periodic testing and maintenance which is performed only serves to reinforce the probability of sound containment integrity.
The Staff apparently accepted this philosophy for many years; as indicated by the fact the NRC inspectors have reviewed test procedures and witnessed ILRTs, and have not (until recently) identified any deviations or violations.
There appears to be a shift in the philosophy of the intent of Appendix J from ensuring future operability to veri-fication of past operability. It may be valid that future operability may best be expected based upon past history of containment leakage.
It may also be that this nsv philosophy of verifying past operability results in penalization for utili-ties' valve maintenance programs by increasing the time required to perform Type B cnd C t esting, increasing the possibility of Type A test failure, and thus in-l creasing the Typ e A test frequency.
In fact, the valve and penetration main-l tenance program will accomplish the same goal as the ILRT program, without sub-I jocting the utility to the economic risks (i.e., increased outage time) associated l
with ILRT failure.
An example of the Staff's previous position that Type C is-found leakages need not l
bo included in ILRT test results can be found in Information Report 50-269/90-06, transmitted to Euke by letter, R.C. Lewis to W.O. Parker, dated March 20, 1980.
The inspector noted that 14 of 60 penetrations were not aligned as required for ILkT. As a result, the inspector re' quired that the as-found leakage from those 14 volves be included in the ILRT results. As-found leakage from the 46 valves which i
were properly aligned were not required to be included.
This specific exclusion of valves frem the as-found testing requirement serve s as documentation of a previous Staff position that as found valve leakage need not be included in ILRT results.
Again, the Staff position presented in Information Notice 85-71 repre-sonts a departure from that position and should receive the appropriate analysis to determine that there is a substantial increase in the overall protection of the public health and safety from the imposition of this backfit.
CONCLUSION The Staff's original positions on short duration Mass-Plot Testing and as-found leakage are documented in the various inspection reports referenced elsewhere in i
this letter. By approving the inspection reports with no applicable violations to dsviations identified, the Staf f has tacitly approved short-duration Mass-Plot Tosting and ILRTs without as-found leakage. More recent Staff documents, notably the April 1,1986 Staff review of leakage methodology and Inspection Notice 85-71, change Staf f position in such a way as to place additional burden on Licensees without demonstrating that a significant increase in overall protection of public hoalth and safety will be achieved. Duke considers that these issues are backfits end should receive the appropriate regulatory analysis.
This analysis should ecmpare the cost of the increased outage time associated with extended leak rate
L, I5 Document Control Desk l
l September 1, 1987 l
Page 5 l
l testing to the significance of the overall increase in protection of the hetith cnd safety of the public.
The analysis should also state the Staf f's interpre-l_
tation of the intent of Appendix J relative to verification of past or insuring future integrity of the containment structure.
Vary "truly yours,
)
l
/.'., / r b? IA {g*'
/
mi
/
xsd s y
Hal B. Tucker i
SAG /84/j gc Attachments xc:
Dr. J. Nelson Grace, Regional Administrator U.S. Nuclear Regulatory Commission - Region II 101 Marietta Street, Suite 2900 Atlanta, Georgia 30323 Executive Director for Operations U. S. Nuclear Regulatory Commission Washington, D. C.
20555 Dr. K.N. 'Jabbour, Project Manager Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington. D.C.
20555 Ms. Helen Pastis, Project Manager Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C.
20555 Mr. Darl Hood, Project Manager Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C.
20555 Mr. P.K. Van Doorn NRC Resident Inspector Catawba Nuclear Station Mr. W.T. Orders
{
NRC Resident inspector McGuire Nuclear Station Mr. J.C. Bryant NRC Resident Inspector Oconee Nuclear Station
,