Arizona Public Service Company, et al.; (Palo Verde Nuclear Generating Station, Unit No. 1), Exemption

NUCLEAR REGULATORY COMMISSION

[Docket No. STN 50-528]

Arizona Public Service Company, et al.; (Palo Verde Nuclear

Generating Station, Unit No. 1), Exemption

I

The Arizona Public Service Company, et al. (APS or the licensee) is

the holder of Facility Operating License No. NPF-41, which authorizes

operation of the Palo Verde Nuclear Generating Station, Unit No. 1

(PVNGS-1). The license provides, among other things, that PVNGS-1 is

subject to all rules, regulations, and orders of the U.S. Nuclear

Regulatory Commission (the Commission or NRC) now or hereafter in

effect. The PVNGS-1 facility is a pressurized water reactor located at

the licensee's site in Maricopa County, Arizona.

II

Section III.D.1.(a) of Appendix J to 10 CFR Part 50 requires the

performance of three Type A containment integrated leakage rate tests

(ILRTs) at approximately equal intervals during each 10-year service

period of the primary containment. The third test of each set shall be

conducted when the plant is shut down for the 10-year inservice

inspection.

III

By letter dated December 28, 1994, the licensee requested temporary

relief from the requirement to perform a set of three Type A tests at

approximately equal intervals during each 10-year service period of the

primary containment. The requested exemption would permit a one-time

interval extension of the third Type A test by approximately 20 months

(from the 1995 refueling outage, which begins in May 1995, to the sixth

refueling outage (1R6), currently scheduled for September 1996) and

would permit the third Type A test of the 10-year inservice inspection

period not to correspond with the end of the inservice inspection

interval.

The licensee's request concluded that the proposed changes for

PVNGS-1, a one-time extension of the interval between the second and

third ILRTs and a decoupling of the third test from the outage

corresponding to the end of the 10-year inservice inspection period, is

justified for the following reasons:

The previous testing history at PVNGS-1 provides substantial

justification for the proposed test interval extension. Type A testing

is performed to determine that the total leakage from primary

containment does not exceed the maximum allowable leakage rate

(La) as specified in the PVNGS-1 technical specifications (TS).

The primary containment maximum allowable leakage rate provides an

input assumption to the calculation required to ensure that the maximum

potential offsite dose during a design basis accident does not result

in a dose in excess of that specified in 10 CFR 100. The allowable

La for PVNGS-1 is 0.10 percent by weight of the containment air

per 24 hours at Pa, where Pa is defined as the calculated

peak internal containment pressure related to the design basis

accident, specified in the PVNGS-1 TS as 49.5 psig. The acceptance

criteria for the Type A test is 75 percent of La or 0.075 percent

by weight of the containment air per 24 hours at Pa.

In each of the two previous periodic ILRTs at PVNGS-1 (the results

were 0.066 percent and 0.067 percent by weight of the containment air

per 24 hours at Pa, respectively), the results obtained were below

the test acceptance criteria of 75 percent of La or 0.075 percent

by weight of the containment air per 24 hours at Pa, thereby,

demonstrating that PVNGS-1 is a low-leakage containment.

The licensee performed a plant-specific study concluding that the

extension of the Type A test has a negligible impact on overall risk.

This study relied heavily on the existing Type B and C testing program

which is not affected by this exemption, and will continue to

effectively detect containment leakage.

Additionally, the licensee stated that its exemption request meets

the requirements of 10 CFR 50.12, paragraphs (a)(2)(ii) (the underlying

purpose of the regulation is achieved), and (a)(2)(iii) (compliance

would result in undue hardship or other costs that are significantly in

excess of those contemplated when the regulation was adopted), for the

following reasons:

The licensee categorized mechanisms that could cause degradation of

the containment into two types: (1) Degradation due to work which is

performed as part of a modification or maintenance activity on a

component or system (activity based); or (2) degradation resulting from

a time based failure mechanism (i.e., deterioration of the containment

structure due to pressure, temperature, radiation, chemical or other

such effects). To address the potential degradation due to an activity

based mechanism, the licensee reviewed containment system related

modifications performed since the last Type A test. The licensee

concluded that the modifications performed did not impact containment

integrity, or the modifications have, or will be, tested adequately to

ensure that there is no degradation from an activity based mechanism.

In addition, the licensee maintains administrative controls which

ensure that an appropriate retest, including local leak rate testing,

if applicable, is specified for maintenance activities which affect

primary containment integrity.

Regarding time based failure mechanisms, the licensee concluded

that risk of a non-detectable increase in the primary containment

leakage is considered negligible due to the 10 CFR Part 50, Appendix J,

Type B and C testing program. The licensee stated that without actual

accident conditions, structural deterioration is a gradual phenomenon

which requires periods of time well in excess of the proposed 81-month

test interval which would result by performing the third periodic Type

A test during the sixth refueling outage in Unit 1. Other than accident

conditions, the only external mechanism inducing stress of the

containment structure is the test itself. The licensee maintains that

the longer test interval would, therefore, lessen the frequency of

stressing the containment.

Additionally, the licensee has performed the general inspections of

the accessible interior and exterior surfaces of the containment

structures and components prior to the previous Type A tests, as

required by 10 CFR Part 50, Appendix J, Section V.A. These inspections

are intended to uncover any evidence of structural deterioration which

may affect either the containment structural integrity or leak

tightness. At PVNGS-1, there has been no evidence of structural

deterioration that would impact structural integrity or leak tightness.

In a phone conversation with the licensee on March 23, 1995, the staff

noted that these inspections, though limited in scope, provide an

important added level of confidence. The licensee committed to perform

the general containment civil inspection during the upcoming refueling

outage (1R5).

The 10 CFR Part 50, Appendix J, Type B tests are intended to detect

local leaks and to measure leakage across pressure containing or

leakage limiting-boundaries other than valves, such as containment

penetrations incorporating [[Page 17830]] resilient seals, gaskets,

doors, hatches, etc. The 10 CFR Part 50, Appendix J, Type C tests are

intended to measure reactor system primary containment isolation valve

leakage rates. The frequency and scope of Type B and C testing is not

being altered by this proposed exemption request. The acceptance

criterion for Type B and C testing is 0.6 La. This acceptance

criterion is for the sum of all valves and penetrations subject to Type

B and C testing and represents a considerable portion of the Type A

test allowable leakage. The results of the as-left combined Type B and

C leakage measured since the last Unit 1 Type A test are 0.054 La,

0.06 La, and 0.13 La (for the February 1991, May 1992, and

November 1993 outages, respectively). The licensee maintains that these

test results are substantially below the acceptance criterion of 0.60

La and demonstrate a good historic performance of containment

integrity.

The proposed schedular exemption would allow the third Type A

leakage rate test in Unit 1 to be performed during the Fall 1996 (IR6)

refueling outage, which meets the 10 CFR Part 50, Appendix J,

requirement of performing three tests in a 10-year time period. The

performance of a fourth Type A test during the Unit 1 seventh refueling

outage, in order to coincide with the outage at the completion of the

extended 10-year ISI interval, is not deemed to be appropriate, as it

would result in additional radiation exposure to personnel, increased

length of the refueling outage and significant additional cost.

Omitting the test will result in dose savings by eliminating

contamination and by reducing radiation exposure from the venting and

draining of piping penetrations necessary to establish the appropriate

test conditions. There would also be dose savings from eliminating the

need to install and remove the temporary instrumentation necessary to

perform the Type A test. Performing a fourth Type A test would also

increase the duration of the affected outage by approximately 3 days

and result in additional costs associated with this increase.

A PVNGS-1 plant-specific analysis was performed to evaluate the

potential for extending the Type A test frequency. The PVNGS-1 plant-

specific analysis considered the extension of the interval to as much

as 240 months. The conclusion of the analysis was that the extension of

the Type A test interval has a negligible impact on overall risk. The

licensee's exemption request does not alter the frequency for

performance of Type A testing (i.e., it still maintains a frequency of

3 tests per 10 years). However, the licensee maintains that the data

from this study support the requested exemption from the requirement of

10 CFR Part 50, Appendix J, regarding ``approximately equal

intervals.'' The interval between the second and third Type A tests

would be 81 months with this exemption. The PVNGS-1 plant-specific

analysis supports the use of a 240-month interval with a negligible

impact on overall risk.

The licensee referenced 10 CFR 50.12(a)(2)(ii) as a basis for this

exemption. This section defines such a circumstance where ``application

of the regulation in the particular circumstances would not serve the

underlying purpose of the rule or is not necessary to achieve the

underlying purpose of the rule * * *.'' The underlying purpose of 10

CFR Part 50, Appendix J, Section III.D.1.(a), is to establish and

maintain a level of confidence that any primary containment leakage,

during a hypothetical design basis accident, will remain less than or

equal to the maximum allowable value, La, by performing periodic

Type A testing. Compliance with the ``approximately equal intervals''

clause of Appendix J is not necessary to achieve the underlying purpose

of the rule, as explained in the above technical justification.

The licensee also referenced 10 CFR 50.12(a)(2)(iii) in its

submittal, which states the NRC may grant exemptions from requirements

of 10 CFR Part 50 when ``compliance would result in undue hardship or

other costs that are significantly in excess of those contemplated when

the regulation was adopted, or that are significantly in excess of

those incurred by others similarly situated * * *.'' The current PVNGS-

1 Type A test schedule would require that four Type A tests be

performed in an extended ISI interval. This current schedule would

result in unnecessary additional radiation exposure in order to perform

the test and unnecessary costs associated with the performance of the

test and the costs associated with the increase in the length of the

refueling outage. Regarding the impact of this exemption on overall

risk, it is the staff's experience that risk is insensitive to the Type

A test frequency at values of leakage close to La. Therefore,

while the staff agrees with the licensee's conclusion that the risk

increase resulting from granting this exemption is small, the time

interval has no particular significance. Additionally, the staff has

previously discussed with the licensee that its scheduling of

containment ILRTs early in the ISI interval is largely responsible for

the necessity of performing an additional test, and would not

constitute a hardship that was not anticipated at the time the rule was

written. Therefore, the staff has reviewed this exemption request

against the criteria of 10 CFR 50.12(a)(2)(ii).

Section III.D.1.(a) of Appendix J to 10 CFR Part 50 states that a

set of three Type A leakage rate tests shall be performed at

approximately equal intervals during each 10-year service period.

The licensee proposes an exemption to this section which would

provide a one-time interval extension for the Type A test by

approximately 20 months. Additionally, for schedular reasons, the final

Type A test of the 10-year inservice inspection period is proposed to

be decoupled from the requirement to perform it during the same outage

(the final Type A test would be performed the outage prior (1R6) to the

end of the inservice inspection period).

The Commission has determined that, pursuant to 10 CFR 50.12(a)(1),

this exemption is authorized by law, will not present an undue risk to

the public health and safety, and is consistent with the common defense

and security. The Commission further determined, for the reasons

discussed below, that special circumstances, as provided in 10 CFR

50.12(a)(2)(ii), are present justifying the exemption; namely, that

application of the regulation in the particular circumstances is not

necessary to achieve the underlying purpose of the rule. The underlying

purpose of the requirement to perform Type A containment leak rate

tests at intervals during the 10-year service period, is to ensure that

any potential leakage pathways through the containment boundary are

identified within a time span that prevents significant degradation

from continuing or becoming unknown. The NRC staff has reviewed the

basis and supporting information provided by the licensee in the

exemption request. The NRC staff has noted that the licensee has a good

record of ensuring a leak-tight containment. All Type A tests have

passed with adequate margin. The licensee has also noted that the

results of the Type A testing have been confirmatory of the Type B and

C tests (which will continue to be performed). Additionally, the

licensee has committed to perform the general containment civil

inspection during the upcoming refueling outage (1R5), thereby

providing an added level of confidence in the continued integrity of

the containment boundary. [[Page 17831]]

The NRC staff has also made use of a draft staff report, NUREG-

1493, which provides the technical justification for the present

Appendix J rulemaking effort which also includes a 10-year test

interval for Type A tests. The integrated leakage rate test, or Type A

test, measures overall containment leakage. However, operating

experience with all types of containments used in this country

demonstrates that essentially all containment leakage can be detected

by local leakage rate tests (Type B and C). According to results given

in NUREG-1493, out of 180 ILRT reports covering 110 individual reactors

and approximately 770 years of operating history, only 5 ILRT failures

were found which local leakage rate testing could not detect. This is

three percent of all failures. This study agrees with previous NRC

staff studies which show that Type B and C testing can detect a very

large percentage of containment leaks. The PVNGS-1 experience has also

been consistent with this.

The Nuclear Management and Resources Council (NUMARC), now the

Nuclear Energy Institute (NEI), collected and provided the NRC staff

with summaries of data to assist in the Appendix J rulemaking effort.

NUMARC collected results of 144 ILRTs from 33 units; 23 ILRTs exceeded

1.0 La. Of these, only nine were not due to Type B or C leakage

penalties. The NEI data also added another perspective. The NEI data

show that in about one-third of the cases exceeding allowable leakage,

the as-found leakage was less than 2 La; in one case the leakage

was found to be approximately 2 La; in one case the as-found

leakage was less then 3 La; one case approached 10 La; and in

one case the leakage was found to be approximately 21 La. For

about half of the failed ILRTs, the as-found leakage was not

quantified. These data show that, for those ILRTs for which the leakage

was quantified, the leakage values are small in comparison to the

leakage value at which the risk to the public starts to increase over

the value of risk corresponding the La (approximately 200 La,

as discussed in NUREG-1493).

Based on generic and plant-specific data, the NRC staff finds the

licensee's proposed one-time exemption to permit a schedular extension

of one cycle for the performance of the Appendix Type A test, and the

decoupling of the third test to be performed coincident with the

completion of the inservice inspection period, to be acceptable.

Pursuant to 10 CFR 51.32, the Commission has determined that

granting this exemption will not have a significant impact on the human

environment (60 FR 16180).

This exemption is effective upon issuance and shall expire at the

completion of the 1R7 refueling outage.

For the Nuclear Regulatory Commission.

Dated at Rockville, Maryland, this 31st day of March 1995.

Elinor G. Adensam,

Acting Director, Division of Reactor Projects--III/IV, Office of

Nuclear Reactor Regulation.

[FR Doc. 95-8585 Filed 4-6-95; 8:45 am]

BILLING CODE 7590-01-M