ML20129F556

Insp Rept 50-382/96-20 on 960827-29 & 961003.Apparent Violations Being Considered for Escalated Ea.Major Areas Inspected:Maint & Engineering

ML20129F556
Person / Time
Site:	Waterford
Issue date:	10/24/1996
From:	NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
To:
Shared Package
ML20129F533	List: ML20129F526 ML20129F556
References
50-382-96-20, NUDOCS 9610290157
Download: ML20129F556 (20)

See also: IR 05000382/1996020

Text

,

ENCLOSURE

U.S. NUCLEAR REGULATORY COMMISSION

REGION IV

Docket No.: 50-382

License No.: NPF-38

Report No.: 50-382/96-20

Licensee: Entergy Operations, Inc.

Facility: Waterford Steam Electric Station, Unit 3

Location: Hwy.18

Killona, Louisiana

Dates: August 27-29 with in-office review until October 3,1996

Inspectors: L. E. Ellershaw, Reactor Inspector

J. Kudrick, Senior Reactor Systems Engineer

G. E. Werner, Project Engineer

Approved By: Dr. Dale A. Powers, Chief, Maintenance Branch

Division of Reactor Safety

'

ATTACHMENTS:

Attachment 1: Partial List of Persons Contacted

List of Inspection Procedures Used

List of items Opened, Closed, and Discussed

Attachment 2: List of Documents Reviewed

i

,

I

k

9610290157 961024

PDR ADOCK 05000382

0 PDR u

...~.-..-.---.--.-.-. . - - . _ ._- . - - - . - - . . -- .

• ,

,

'

,

% i

! .

l -2-

!

i i

' EXECUTIVE SUMMARY

Waterford Steam Electric Station, Unit 3 (

j NRC Inspection Report 50-382/96-20

1

e

! This inspection was performed using the guidance of NRC Inspection Procedures 73756,

.

4 " Inservice Testing Of Pumps And Valves," dated July 27,1995; 92902, " Followup - t

'

Maintenance " dated March 14,1994; and 92903," Followup - Engineering," dated March .

l 14,1994, to determine whether the inservice testing program had been properly

established and implemented in accordance with Technical Specifications, the applicable

ASME Code, correspondence between NRC and the licensee concerning requests for relief l

l and requirements imposed by NRC/ industry initiatives.

)

i

l Maintenance  !

!

.a

l * Pump ;.est requirements had been established in accordance with the 1980 Edition

)

2

of the ASME Code through the Winter 1981 Addenda (applicable to Waterford 3), 1

j and were being satisfactorily implemented (Section M1.1).

i

i * An inspection followup item was identified to review and evaluate data used to

support the licensee's position that safety-related pumps had not degraded to the

point where they would not meet system design requirements, and that there were

no operability concerns (Section M1.1).

• The failure to shut Containment Spray Valve CS-118A following completion of Train

A containment spray pump testing, as required by Step 7.1.23 in Procedure

OP-903-035, was an apparent violation of Technical Specification 6.8.1.a (Section

M 8.1 ) .

• The disposition (over a five month period) of a condition report, which led to the

identification of a 10 CFR 50.72(b)(2)(iii) required notification to NRC, was untimely

(Section M8.1).

I

Enaineerina

• The f ailure to perform the required written safety evaluation to provide the bases for

determining that the difference between the facility design configuration and the "

Final Safety Analysis Report did not involve an unreviewed safety question was an

apparent violation of 10 CFR 50.59 (Section E8.2).

!

- - . . . . - . . .- _- _ . _ . .

_ . _ - ._. ._ - . - - _ . . . - - ..- _ - -

i

1

I

,

I

\

, -3- '

l

i

l Reoort Details l

1

<

Summarv of Plant Status i

4

'

i Waterford Steam Electric Station, Unit 3, was at full power operation during this

j inspection.  :

!

ll Maintenance

1

-

M1 Conduct of Maintenance 1

1

M 1.1 ASME Code Section XI inservice Testina

s

a. Insoection Scope (73756)

,

1 The inspectors reviewed the most recent ASME Code Section XI inservice test data

a

regarding operability verification for selected Train B pumps given in the following

procedures:

• " Emergency Diesel Generator Fuel Oil Transfer Pump," Surveillance

Procedure OP-903-117, Revision 1;

! * " Component Cooling Water Pump," Surveillance Procedure OP-903-050,

Revision 11;

.

• " Boric Acid Pump," Surveillance Procedure OP-903-004, Revision 9;

s

* " Auxiliary Component Cooling Water Pump," Surveillance

Procedure OP-903-050, Revision 11;

,

• " Charging Pump," Surveillance Procedure OP-903-003, Revision 8;

!

.

• " Containment Spray Pump," Surveillance Procedure OP-903-035, Revision 8;

i

,

• " Safety injection Pump," Surveillance Procedure OP-903-030, Revision 10; l

. and

!

l

l

!

• " Chilled Water Pump", Surveillance Procedure OP-903-063, Revision 9. .

. l

4 b. Observations and Findinas )

i

j The inspectors determined from their review of the licensee's records that pump

test parameters and acceptance criteria had been established in accordance with

j ASME Code requirements, and that the performance of all pumps met the

acceptance criteria.

J

.

,

f

. . - - - . - - - - . _ - -

l *,

!

-4-

l

l

l

l During their review of the surveillance test package for Component Cooling Water

Pump B, the inspectors identified four change notices applicable to the surveillance

procedure. One change notice reduced the test flow rate from 6000 to 5600 gpm

(Change Notice 1 dated September 18,1995), and another change notice reduced

the test flow rate from 5600 to 4800 gpm (Change Notice 2 dated February 7,

1996). The change from 6000 to 5600 gpm was initiated to provide consistent

test conditions, since the surveillance procedure for Component Cooling Water

l Pump AB specified a flow rate of 6000 gpm, while the surveillance procedures for

pumps A and B specified 5600 gpm. The change from 5600 to 4800 gpm was

recommended by system engineering to preclude the diversion of component

cooling water to the reactor coolant pumps which could cause low flow alarms and

increased reactor coolant pump cooler temperatures.

Section IWP-3110 in Section XI of the ASME Code allows a change to reference

values, as long as an inservice test at the conditions of an existing set of reference

values is performed and the results analyzed. When these are verified as being

satisfactory, then a second test is to be performed at the new reference conditions,

and these results are then to be used to establish the new set of reference values.

The licensee performed these actions and documented the results in accordance

with the requirements of Section IWP-3112 in Section XI of the ASME Code.

l

Component Cooling Water Pumps A, B, and AB were each tested using the initial

'

reference values, followed by a test using the new reference conditions. The tests

were documented as follows: Component Cooling Water Pump A, Work

Authorization 01144519, performed on February 24,1996; Component Cooling

Water Pump B, Work Authorization 01144513, performed on March 15,1996; and

Component Cooling Water Pump AB, Work Authorization 01143621, performed on

February 8 and 9,1996.

The inspectors also learned that licensee personnel had initiated Condition Report

CR-96-0414 on March 20,1996, to address a concern regarding component

cooling water pump inservice test acceptance criteria. Licensee personnel had

identified that the inservice test acceptance criteria permitted component cooling

water pump performance to degrade below the design basis accident performance

point without requiring action to restore pump performance. The design flow rate

for the component cooling water system was 6554 gpm. The certified pump curve

showed that the pump should develop 151 feet of head at 6554 gpm, or 148 feet

of head at 6800 gpm. This difference translated to 2 percent of head margin or

3.8 percent of flow margin. However, the inservice test acceptance criteria, which

was established in accordance with ASME Code requirements, allowed for (a)

7 percent head degradation and 6 percent pump flow degradation below the

! reference point, respectively, before the alert range limit was reached, and (b)

4

10 percent head degradation and 10 percent pump flow degradation before the

action range limit was reached. Consequently, this amount of pump degradation

__ _ _ . . - . _ _ . _ _ . . _ _ _ _ _ . __ _ . _ . _ . _ _ _ . _

4

4 /

.s.

would result in the system performing below the design basis flow rate before the

inservice test trend data indicated any problem. The licensee's review of the

condition report that discussed this susceptibility resulted in a number of corrective

actions, some of which were still open at the end of the onsite inspection.

The licensee performed a review of all other safety-related pumps for similar

conditions. Subsequent to the onsite inspection, the inspectors were informed that

the review determined that inservice test acceptance criteria, with the exception of

the component cooling water pumps and Auxiliary Component Cooling Water Pump

A, were set sufficiently high such that pump degradation would be detected before

the design basis accident required flow rate was encroached on.

An associated root-cause analysis report dated May 14,1996, stated that the

component cooling water pumps had been, or were, operating in a condition where

the pumps would not have provided 148 feet of head at 6554 gpm. It also stated

that a previous analysis associated with another condition report (CR 95-0955)

determined that even with flow as low as 6000 gpm, the design basis heat load

could still be removed by the component cooling water system. In addition,

previous special flow balance tests had been performed using Special Test

Procedure 01150154,"CCW System Flow Balance Test." The results from those

special flow balance tests showed that the total flow rate was approximately 6300

gpm, which was below the design basis accident flow rate of 6554 gpm.

Because of the lower-than-expected flow rates, engineering performed a 10 CFR

50.59 evaluation to assess the impact of the lower flow rates on heat transfer

during a design basis accident. The evaluation, approved on October 20,1995,

assumed a 6000 gpm component cooling water system flow rate, which bounded

the design flow rate of 6554 gpm and the actual measured flow rate of 6300 gpm.

New analyses were performed for the design basis accidents using the assumed

lower flow rates. These analyses demonstrated that the component cooling water j

system heat removal performance with the lower total flow rate was greater than '

that assumed in the accident analyses. The evaluation concluded that, even with i

flow rates as low as 6000 gpm, the component cooling water system would still '

perform its safety function and all applicable acceptance limits would be met.

During their review of the root-cause analysis report, the inspectors noted that the

issue of pumps degrading below the design basis accident flow rate before the

inservice test program would detect the problem, had been identified at Arkansas 1

Nuclear One as early as 1990. It appeared that in February 1994, Waterford I

personnel became aware of the issue, and on March 20,1996, the condition report

was issued on this subject. While all corrective action tasks have not been

completed, licensee personnel informed the inspectors that their review of all

safety-related pump test data for the past year showed that the pumps exceeded

the design basis accident flow rates established for their respective systems.

__

s

,

6-

The inspectors did not review and evaluate the data which was used to support the

licensee's position that the pumps had not degraded to the point where they would

not meet system design requirements, and that there were no operability concerns.

.

This matter will be reviewed as an inspection followup item during a future

inspection (50-382/9620-01).

M8 Miscellaneous Maintenance issues (92902)

M8.1 (Closed) License Event Reoort 50-382/96-012: Containment Spray Valve CS-118A

Found Partially Open.

a. Inspection Scope

The inspectors reviewed Licensee Event Notification 30902 dated August 21,1996,

and Licensee Event Report 50-382/96-012 dated September 20,1996, which

informed the NRC that Containment Spray Valve CS-118A was found partially open

(approximately one and one-half turns) on November 11,1995, subsequent to

testing performed on the Train A containment spray pump. Valve CS-118A serves

as a flow test line isolation valve. The licensee event notification also identified that

during a design basis loss-of-coolant accident while in the recirculation mode,

leakage through this valve back to the refueling water storage pool could have

exceeded the 8 gpm limit established in Updated Final Safety Analysis Report

Table 15.6-19 and, subsequently, could have overexposed the control room

operators.

b. Observations and Findinas

Following performance of Surveillance Procedure OP-903-094,"ESFAS Subgroup

Relay Testing- Operating," Revision 8, licensee personnel observed that containment

spray riser level had dropped approximately 55 ft. Investigation revealed that Valve

CS-118A, an Anchor Darling,4 in, normally locked closed, manual gate valve, was

not fully closed and was found to be one and one-half handwheel turns open.

Condition Report CR 95-1165,which was initiated on November 11,1995,

concluded that containment spray pump operability was not affected since

backleakage to the refueling water storage pool was quantified by engineering

judgement to be less than the acceptance criterion of 60 gpm.

The system engineering superintendent reviewed Condition Report CR 95-1165 and

identified that the condition report f ailed to address Valve CS-118A backleakage

radiological consequences. Consequently, licensing personnel initiated Condition

l Report CR 96-0287 (March 2,1996) to address the new issue.

'

The licensee identified that Valve CS-118A had last been operated on

September 19,1995, during performance of Surveillance Procedure OP-903-035,

" Containment Spray Pump Operability Check," Revision 8, Section 7.1,

" Containment Spray Pump A." During the surveillance, Valve CS-118A was opened

l

u

s

5

-7-

and then subsequently locked closed; however, discussions with operations

personnel revealed that the valve apparently had not been completely closed. This

was determined during a subsequent check (as a result of the containment spray

riser level drop) on November 11,1995, when personnel used a " valve operator" (a

procedurally allowed leverage bar), and closed the valve an additional one and

one-half handwheel turns. Coridition Report 951165 identified the difficulty in

operating the valve as a cause for the valve not being fully closed.

Based upon the above dates, Valve CS-118A was partially open for'approximately

53 days; however, the plant was in a refueling outage during this period for 44 days

(September 22 to November 5,1995). In accordance with Technical

Specification 3.6.2.1, the containment spray system was not required to be

operable in Mode 4 (below 400 psia), and in Modes 5 and 6. The inspectors review

of control room logs showed that the plant reached Mode 4 and 350 psia at 0600

on September 23,1995. Subsequently, during power ascension, the plant reached

Mode 4 and 375 psia at 0600 on October 31,1995. Therefore, the valve was

partially open when it was required to be closed between September 19 and 23,

and between October 31 and November 11,1995. l

Technical Specification 6.8.1.a, requires that written procedures are to be l

implemented for those activities referenced in Appendix A of Regulatory Guide l

1.33, one of which is surveillances. Surveillance Procedure OP-903-035, used to

satisfy that requirement, contained Step 7.1.23, which required that Containment

Spray Valve CS-118A be closed and locked following completion of Train A

Containment Spray Pump testing. The performance and verification of this step

was documented by initials on Step 23 in Attachment 10.1 to the procedure on

September 19,1995. The failure to shut Valve CS-118A as required by

Surveillance Procedure OP-903-035 was an apparent violation of Technical

Specification 6.8.1.a (50-382/9620-02).

Valve CS-118A was one of the 13 valves previously identified by the licensee as

not having been included in the inservice test program and, therefore, not tested

(see NRC Inspection Report 50-382/9609). The licensee had placed the valve in the

inservice test program, and on March 2,1996,the valve was tested for leakage.

Test results showed no leakage (i.e.,0.0 gpm).

.

However, from March to June 1996, the licensee attempted to bound the potential

leakage problem that existed during September through November 1995, through

design calculations, bench testing of a similar valve, and performing a special test

through partially opened Valve CS-118A. However, none of the licensee's efforts

were successful on quantifying the leakage.

I

-

- - .- - -

s

,

8

On August 21,1996, engineering personnel determined they could not quantify

refueling water storage pool backleakage at low flow rates. Based upon this

information, licensing personnel, in accordance with 10 CFR 50.72(b)(2)(iii), made a

4-hour notification on August 21,1996. The inspectors reviewed the reportability

aspects of this issue and found them acceptable.

Subsequent to the onsite inspection, the licensee submitted Licensee Event

Report 50-382/96-012 dated September 20,1996. In the report, the licensee

discussed a special test that was performed on September 7,1996,in another

attempt to quantify the leak rate through Valve CS-118A. While there were no

consequences associated with the actual event, the test determined that the

maximum leak rate during a design basis loss-of-coolant accident would have been

approximately 11.8 gpm, which exceeded the 8 gpm backleakage limit specified in

Table 15.6-19 in the Updated Final Safety Analysis Report.

The backleakeage limit was established to maintain the control room thyroid dose

limits specified in Criterion 19 of Appendix A to 10 CFR Part 50. This number

(11.8 gpm) was arrived at by using actual flow rates from three different valve l

settings, and by perforrning valve leakage calculations using expected values during

an accident for elevation head, containment spray pump head, valve and line losses, ,

and differential pressure across Valve CS-118A. The test also determined that the

backleakage would not have caused off-site dose limits (10 CFR Part 100) to be

exceeded. The licensee determined that a leak rate greater than 12.5 gpm during i

the design basis loss-of-coolant accident would be required to exceed the off-site  !

dose limits. The licensee performed additional control room dose limit calculations.

In addition to the excessive backleakage through Valve CS-118A, the licensee

considered the number of hours personnel would be in the control room during an

accident, and included other possible leakage paths. The Licensee Event Report

concluded that the dose attributed to these conditions would not exceed the thyroid

dose limit for the control room staff. l

c. Conclusions

Pump test requirements had been established and were being implemented in

accordance with the 1980 Edition of the ASME Code through the Winter 1981

Addenda (applicable to Waterford 3). An apparent violation of Technical l

Specification 6.8.1.a occurred when licensee personnel f ailed to shut Valve

CS-118A following a containment spray pump test. The licensee's disposition (over l

a five month period) of Condition Report 96-0287 (March 2,1996), which led to i

the identification of a required report (August 21,1996) to NRC, was untimely. i

l

. .- - -. . ._.- .. - --..--. - _ ._. - - -. - - . - - . - -- . . - --.

,

i a,

j

f

-9-

!

!

lll. Enaineerina

E2 Engineering Support of Facilities and Equipment

i

i E2.2 Review of Updated Final Safety Analysis Report Commitments

i

A recent discovery of a licensee operating their facility in a manner contrary to the

Updated Final Safety Analysis Report description highlighted the need for a special

l focused review that compares plant practices, procedures, and/or parameters to the

Updated Final Safety Analysis Report description. While performing the inspections

l discussed in this report, the inspectors reviewed the applicable sections of the

l Updated Final Safety Analysis Report that related to the areas inspected.

3

i in addition, the inspectors reviewed the licensee's 10 CFR 50.59 Safety Evaluation

j Report, "LDCR 97-0047, Location of Cabinets C-3A(B) and C-4 Outside the CVAS

d Boundary, Rev.1," dated July 28,1996. This report was initiated on July 22,

a

1996, because the licensee became aware that their justification, in response to  ;

'

Final Safety Analysis Report Question 480.36, for not conducting 10 CFR Part 50,

Appendix J, Type C leak tests on the valves in the instrument lines through  ;

Containment Penetrations 53 and 65, was not correct (Appendix J, Type C leak

'-

tests are pneumatic tests used to measure containment isolation valve leakage

{' rates). This resulted in errors between the wording of the Updated Final Safety

Analysis Report and the current plant configuration regarding the containment

vacuum relief lines.

l'

The 10 CFR 50.59 report identified the following Updated Final Safety Analysis

sections and tables requiring changes: Section 1.9.37; Section 7.1.2.7;

Table 3.9-9; Table 6.2-32, and Table 6.2-43. In addition, the following tables in the

Technical Requirements Manual were also identified as requiring changes

l Tables 3.6-1 and 3.6 2. Licensee personnel stated that the Updated Final Safety

l Analysis Report would be changed to reflect the actual plant configuration, and to

I revise leak rate test requirements.

!

'

Additional details and information pertaining to the potential consequences of these

l discrepancies are located below in Sections E8.1 and E8.2.

1 E8 Miscellaneous Engineering issues (92903)

i

l E8.1 (Closed) Unresolved item 50-382/9609-04: the determination of whether there

{ were active safety functions for Excess Flow Check Valves CVR-302A(B),

! " Containment / Annulus Differential Pressure Sample Line Check Valve."

i Containment Penetrations 53 and 65 each contain two containment vacuum relief

I instrument lines. One instrument line senses differential pressure across the

containment vessel and provides a signal to actuate the containment vacuum relief

,

j

i

5

, - - -

- - -

w n-,

=

• ,

- 10-

system. The other instrurnent line monitors the same differential pressure and

provides an input to the plant computer.

Excess Flow Check Valves CVR-302A(B) were installed in the containment vacuum

relief sensing instrument lines to meet the guidance specified in Regulatory

Guide 1.11. This was committed to in Section 6.2.4.2.2 of the Updated Final

Safety Analysis Report. Regulatory Guide 1.11 states, in part, "In the event of a

rupture downstream of the valve, the valve should close automatically or be capable

of being closed during normal reactor operation and under accident conditions."

In Condition Report 96-0272, dated March 1,1996, the licensee identified

"

that Valves CVR-302A(B) had a closed safety function. They were added to the

l inservice test plan as ASME Code Category C valves, but with a cold shutdown test

J frequency justification'. The condition report further stated that these valves were

listed in Table 3.6-2 of the Technical Requirements Manual which identified them as

I containment isolation valves and subject to Technical Specification 3.6.3, and

10 CFR Part 50, Appendix J testing.

The licensee's representatives indicated that the valves had been tested during

} previous refueling outages in the closed direction and were capable of meeting their

, closed safety function. However, these penetrations were only tested during the

10 CFR Part 50, Appendix J, Type A tests. (Appendix J, Type A tests measure the

containment system overall integrated leakage rate under conditions representing

'

, design basis loss-of-coolant accident containment peak pressure). No individual

l testing of the valves was performed. Licensee personnelinformed the inspectors

that closed-system leak tight integrity tests had been cor. ducted on the closed

systems during refueling outages; however, this required that the valves be open in

j order to verify system integrity.

Subsequently, licensee personnel stated that, following additional reviews, they

j identified that the excess flow check valves did not have a closed safety function

and did not have to be tested in the closed direction. The licensee supported this

position by referencing Final Safety Analysis Report Question and Response 480.36

and Updated Final Safety Analysis Report, Table 6.2-32, which showed that

10 CFR Part 50, Appendix J, Type C containment isolation valve leak testing was

not required for these penetrations. In addition, Table 6.2-43, showed that

Penetrations 53 and 65 were tested during Appendix J, Type A tests and were not

required to have a separate Appendix J, Type C test. However, NRC acceptance of

that position, as stated in the Safety Evaluation Report, "Waterford Steam Electric

'They also were part of the subject of apparent violation 50-382/9609-02, regarding a

f ailure to include valves in the inservice test program that were recuired to be tested in

accordance with Section XI of the ASME Code. )

_ ___ _ _ . __ __

• ,

-11-

Station, Unit 3, NUREG-0787" dated July 1981, did not imply that no testing was

acceptable. It simply indicated that an Appendix J, Type C leak rate test was not

necessary.

On July 20,1996, while the plant was in Mode 5, Cold Shutdown, the licensee

performed an inservice stroke test on Valve CVR-302B, and the valve failed to

close. The valve was declared inoperable, and Condition Report CR-96-1103 was

initiated. On July 21,1996, a stroke test was performed on Valve CVii-302A, with

similar results. The f ailure of Valve CVR-302A was incorporated into the existing

condition report.

l

The f ailed valves were removed from the containment vacuum relief sensing lines.  ;

The replacement valves were tested and found to be acceptable. The licensee's I

evaluation concluded that had these valves f ailed during plant operations, there

would not have been a safety concern. This evaluation was based on the fact that  ;

the tubing downstream of these valves was safety related (i.e., fabricated to ASME l

Code, Class 3 requirements, and classified as Seismic Category 1), and constituted j

a closed system outside of containment which would be assumed to fulfillits i

intended function (i.e., maintain its integrity) under accident conditions. The

inspectors agreed with the licensee's conclusion.

During the assessment of Condition Report CR 96-1103,the NRC inspector

discussed the issue with licensee personnel, who indicated that the instrument lines

did not terminate in the controlled ventilation area system, and that a potential for

bypass leakage existed. The inspectors, on July 21,1996, questioned licensing

personnel about what appeared to be a discrepancy, in that part of the licensing

basis was predicated on the lines terminating in the controlled ventilation area

system. On July 22,1996, further inspector questioning caused licensee personnel

to conduct a review of their response to Final Safety Analysis Report Question

480.36 regarding justification for not conducting Appendix J, Type C leak tests on

the valves in the instrument lines through Containment Penetrations 53 and 65.

Through this review, the licensee became aware that an incorrect response had

been provided to the NRC. This caused the initiation of Condition Report

CR 96-1123 dated July 23,1996, which documented this discrepancy. On August

26,1996, Licensee Event Report 50-382/96-009,which is discussed below, was

issued to address the issues identified during the licensee's review. Subsequently,

an August 12,1996, predecisional enforcement conference was postponed pending

review of the issues that were raised by the licensee's report.

E8.2 (Closed) LER 50-382/96-009: failure of Containment Vacuum Relief

Valves-402A(B) (excess flow check valves). On July 22,1996, the licensee

became aware that the cabinets for the instrumentation associated with the

containment vacuum relief system (sensing and monitoring lines) were not within

the controlled ventilation area system, and that this discrepancy was inconsistent

with the licensing basis. Previously, on July 5,1996, the licensee determined that

!

_. _ ._

I

l-

l

..

l

l

l

-12-

l

!

the required Technical Specification testing had not been performed on Valves

CVR-401 A(B) and CVR-402A(B) which were located in the containment vacuum

relief monitoring lines.

During initiallicensing of Waterford Steam Electric Station, the licensee responded

to Final Safety Analysis Report Question 480.36 (dealing with testing of certain

containment penetrations) by stating that the containment vacuum relief sensing

lines and the containment vacuum relief monitoring lines each formed a closed

system outside of containment, were seismically qualified, and terminated within an

area exhausted by the controlled ventilation area system (a filtered ventilation

system). Based on this rationale, the NRC staff determined that the design and

isolation provisions of Containment Penetrations 53 and 65 would be acceptable to

meet General Design Criterion 56 of Appendix A to 10 CFR Part 50. This

determination was included in Section 6.2.4 of Safety Evaluation Report,

"Waterford Steam Electric Station, Unit 3, NUREG-0787" dated July 1981.

Further licensee review of their response revealed that, not only did the instrument

lines not terminate in the controlled ventilation area system, but the monitoring lines

did not meet the design criteria for a closed system outside of containment and they

were not seismically qualified.

10 CFR 50.59(a)(1) allows the holder of a license to make changes to the facility as

described in the safety analysis report unless the proposed change involves an

unreviewed safety question. 10 CFR 50.59(b)(1) requires the licensee to maintain

records of changes in the facility, to the extent that these changes constitute '

changes in the facility as described in the safety analysis report. The records must

include a written safety evaluation which provides the bases for the determination

that the change does not involve an unreviewed safety question.

The actual design configuration of the containment vacuum relief system was

different from that described in the licensee's response to Final Safety Analysis

Report Question 480.36, and until July 28,1996, the licensee failed to perform the

required written safety evaluation to provide the bases for a determination that the

deviation from the response to Final Safety Analysis Report Question 480.36 did

I not involve an unreviewed safety question. This failure was an apparent violation of

10 CFR 50.59 (50-382/9602-03).

Criterion 56 of Appendix A to 10 CFR Part 50 requires each line that connects

directly to the containment atmosphere and penetrates primary reactor containment

i shall be provided with two containment isolation valves unless it can be

l demonstrated that the containment isolation provisions for a specific class of lines,

such as instrument lines, are acceptable on some other defined basis. In addition,

the use of a simple check valve as the automatic isolation valve outside

4

containment is not alloweo.

I

l

,

s

4

+.

- 13-

Regulatory Guide 1.11, which the licensee committed to in Section 6.2.4.2.2 of the

Updated Final Safety Analysis Report, provided suitable bases for demonstrating

acceptable alternative containment isolation provisions. However, in order to use

the provisions of Regulatory Guide 1.11, the valving provided for each instrument

line penetrating containment must reflect the importance of two safety features: (1)

the function the line performs, and (2) the need to maintain containment leaktight

integrity.

With respect to the sensing lines, the importance of the safety function was

recognized and application of Regulatory Guide 1.11 was justified. The lack of a

safety function for the monitoring line, however, established that use of Regulatory

Guide 1.11 was not an appropriate alternative with respect to the isolation

provisions as described in General Design Criterion 56, and the use of an excess

flow check valve was not an acceptable isolation barrier. The failure to implement

General Design Criterion 56 requirements by not providing the specified

containment isolation barriers in the containment vacuum relief monitoring lines was

a result of the licensee failing to recognize that the actual design configuration was

different from their response to Final Safety Analysis Report Question 480.36.

The inspectors also noted that during the operating license review, the NRC staff

asked for additional justification for not including Containment Penetrations 53 and

65 in an Appendix J, Type C leak testing program as discussed in Question 480.36.

The licensee response reiterated the isolation barriers as described above, but also

indicated that Solenoid Operated Valves CVR-401 A(B), which would close on a

containment isolation signal, would be added to the containment vacuum relief

monitoring lines during the first refueling outage. There was no mention of any

testing requirements associated with those valves since the acceptance of the

penetrations was based on the premise that closed systems existed in conjunction

with the excess flow check valves. The licensee's representatives stated that the

valves would be open for each Appendix J, Type A test so that the integrity of the

closed system would be tested. In addition, licensee personnelinformed the

inspectors during this inspection, that a local leak tight integrity test had been

conducted on all the closed systems during those refueling outages when the

Appendix J, Type A test was not performed.

After identifying that the containment vacuum relief monitoring lines did not

constitute a closed system, the licensee isolated these lines by closing the upstream

Solenoid Isolation Valves CVR 401 A(B). On July 26,1996, Valves CVR-401 A(B)

and CVR-402A(B) were leakrate tested under Work Authorization 01149602 with

the following results, as given in Table 1.

's

.

..

-14-

TABLE 1

VALVE AS-FOUND AS-LEFT

LEAKRATE (sccm) LEAKRATE(sccm)

CVR-401 A 20.0 20.0

CVR 4018 21.6 21.6

CVR-402A 129,000 44500*

CVR-402B 242,500 44500*

• Assigned design leakage - actual measured leakage was less

than this value.

Excess Flow Check Valves CVR-402A(B) failed to close during the initial leakrate

test and were removed. It appeared that these valves failed because of excessive

manufacturing roughness in the valve body bore. The replacement valve body

bores were smoothed prior to installation and were successfully tested as shown

above. Technical Specification 3.6.1.2 limited combined bypass leakage to a rate

of less than or equal to 0.06 L,. The Solenoid Isolation Valves CVR-401 A(B) were

left shut since the combined bypass leakage through these lines could exceed

0.06 L,if both solenoid valves failed to shut. This leakage was determined by

licensee calculation to be equal to 63,069 scem and was incorporated into

Surveillance Procedure STA-001-006," Leak Rate Testing," Revision 2, Attachment

10.9. As of July 26,1996, with Valves CVR-401 A(B) shut, the combined bypass

leakage was 14,924.7 scem.  !

l

l

Technical Specification Surveillance Requirement 4.6.1.2.d required that bypass l

flow be measured at least once per 24 months by either performing a Type B (not

applicable to containment isolation valves) or Type C test on individual penetrations

that can be tested. The failure of the licensee to test Penetrations 53 and 65 prior

to July 26,1996,in accordance with Technical Specification Surveillance l

Requirement 4.6.1.2.d, was another result of the licensee not recognizing the  !

discrepancy between the actual design configuration of the containment vacuum

relief system and their response to Final Safety Analysis Report Question 480.36.

The licensee determined that both control room doses and offsite doses could have

been exceeded had a design basis loss-of-coolant accident occurred assuming

rupture of the nonessential monitoring line and failure of either Valve CVR-401 A or

CVR-401B with Valves CVR-402A or CVR-4028 failed in the open position

(as-found condition). Licensee calculated control room and offsite doses are shown

in Table 2.

.s

.

-15-

I

TABLE 2

LOCATION THYROID DOSE WHOLE BODY SKIN DOSE

(rem) DOSE (rem) (rem)

2 Hr Exclusion 405.82(300)' 13.43(25)" N/A

Area Boundary

30 Day Low 162.85 (300)~ 3.323(25)* N/A

Population

Zone

30 Day Contro! 45.71(30)* 1.36 (5)* 31.51 (30)*

Room ,

l

1

• The values within parentheses are regulatory limits from the standard review plan.

On August 21,1996, the licensee submitted License Amendment Request W3F1-

96-01441," Discrepancy Regarding the Design and Testing of Instrument Sensing

Lines Penetrating the Primary Containment," to resolve the design and testing

discrepancies associated with both the containment vacuum relief sensing and

containment vacuum relief monitoring lines. j

The inspectors reviewed the Updated Final Safety Analysis Report for other

instrumentation lines penetrating containment and reviewed the construction of

those lines. The inspectors verified that no other containment penetration lines

utilized excess flow check valves. Also, the licensee searched their parts data base

and identified that the excess flow check valves were used exclusively on the

Containment Vacuum Relief System (Penetrations 53 and 65). Updated Final Safety

Analysis Report Section 6.2.4.2.2, " Instrument i.ines," identified only one additional

instrument line (Penetration 54) that penetrated containment.

This instrument line was used for the wide range containment pressure

instrumentation, and was depicted on Drawing B-430, Sheet P-77. The Updated

Final Safety Analysis Report described the containment wide range pressure

instrumentation as consisting of a sealed, liquid-filled system with a bellows,

constructed in accordance with ANSI N271-1976," Containment isolation

Provisions for Fluid Systems." ANSI N271-1976 was used to satisf'y the

requirement of 10 CFR Part 50, Appendix A, Criterion 56.

The inspectors performed a walkdown of the piping and instrumentation outside

containment for Penetration 54 and identified that the transmitter was not enclosed

in protective shielding (protection from missiles and water jets) as required by ANSI

N271-1976. Followup discussion with the licensee identified that the piping and

transmitter wl:'.; located in an area that had been analyzed for jet impingement and

_- - -

-

l

-16

no jet impingement concerns were identified. In addition, other components located

in the same area were all seismically qualified; therefore, no missiles would be 1

associated with their failure. Based upon this information, the inspectors l

determined that Penetration 54 hardware were adequate to provide the required

safety function.

V. Manaaement Meetinas

X1 Exit Meeting Summary

Subsequent to the conclusion of the inspection on October 3,1996, the inspector

telephonically presented the inspection results to members of licensee management on ,

October 9,1996. The licensee acknowledged the findings presented. l

The inspector asked the licensee whether any materials examined during the inspection

should be considered proprietary. No proprietary information was identified.

I

l

l

I

-

o. l

7

ATTACHMENT 1

PARTIAL LIST OF PERSONS CONTACTED

Licensee

P. Caropino, Licensing Coordinator

C. Dugger, General Manager, Plant Operations l

J. Fisicaro, Director, Nuclear Safety

C. Fugate, Shift Supervisor, Operations

l

T. Gaudet, Manager, Licensing

l

P. Gropp, Supervisor, Mechanical Specialties, Design Engineering  !

J. Holman, Manager, Safety Analysis

J. Houghtaling, Technical Assistant, Design Engineering

J. Howard, Manager, Procurement / Programs Engineering

P. Melancon, inservice Test Engineer

G. Robin, Supervisor, Programs Engineering

L. Rushing, Manager, Mechanical / Civil Design Engineering

M. Selman, Vice President, Operations

P. Stanton, Design Engineer, Mechanical Systems

C. Thomas, Supervisor, Licensing

R. Thweatt, Supervisor, Design Engineering, Mechanical

D. Urciuoli, Licensing Engineer I

D. Vinci, Manager, Systems Engineering

K. Walsh, Lead Senior Engineer, Operations i

A. Wrape Ill, Director, Design Engineering

.

l

NRC

!

K. Brockman, Acting Director, Division of Reactor Safety

'

D. Powers, Chief, Maintenance Branch i

L. Keller, Senior Resident inspector '

T. Pruett, Resident inspector

LIST OF INSPECTION PROCEDURES USED

IP 73756 Inservice Testing of Pumps and Valves

IP 92902 Followup - Maintenance '

IP 92903 Followup - Engineering

- _ . - - - _ - . - . - - - . . - . . . _ . . . - - _ - - . . - - - . - . . - .

s

!

9

!

!

-2- i

i

i

,

!

LIST OF ITEMS OPENED, CLOSED, AND DISCUSSED

Opened

50-382/9620-01 IFl An inspection followup item was identified to review and  :

evaluate data used to support the licensee's position that

safety-related pumps had not degraded to the point where thc,'

would not meet system design requirements and that theie

were no operability concerns. i

i

i >

! 50-382/9620-02 eel The failure to shut Containment Spray Valve CS-118A -

l

following completion of Train A containment spray pump

l testing, as required by Step 7.1.23 in Procedure OP-903-035,

'

was an apparent violation of Technical Specification 6.8.1.a. >

!

l

i

l 50-382/9620-03 eel An apparent violation regarding the failure to perform the i

required written safety evaluation to_ provide the bases for a l

determination that the difference between the facility design {

configuration and the Final Safety Analysis Report was not an  !

unreviewed safety question. This resulted in a failure to meet

General Design Criterion 56 requirements with respect to

providing the specified conteir. ment isolation barriers in the t

l containment vacuum relief system, and a failure to test

i Penetrations 53 and 65 pnor to July 26,1996,in accordance

with Technical Specif cation Surveillance Requirements.

Closed

i

50-382/9609-04 URI This item pertained to determination of active safety functions

for excess flow check valves CVR-302 A(B).

.

50-382/96-009 LER This item dealt with failure of Containment Vacuum Relief C

Valves-402A(B)(excess flow check valves). [

50-382/96-012 LER This item dealt with the failure to close Containment Spray

Valve CS-118A following pump testing.  ;

,  !

! i

i

1

l

I

.

i

l

I

.-n. , -- -~ e- - ,

o.

A

,

t.

ATTACHMENT 2

LIST OF DOCUMENTS REVIEWED

Desian Documents

"Waterford Quality Assurance Program Manual," Revision 5

" Technical Specifications," Amendment 112

" Updated Final Safety Analysis Report," Revision 9

"Waterford 111 Pump and Valve inservice Test Plan," Revision 8, Change 1

"Waterford ill Pump and Valve Inservice Test Plan," Revision 7, Change 10

W3-DBD-024," Inservice Testing Basis Document," Revision 0 and

Revision 1

W3-DBD-014, " Safety-Rr.iated Air-Operated Valves," Revision O

W3-DBD-042, " Sampling Sy. item," Revision 0

Procedures

Surveillance Procedure OP-903-035," Containment Spray Pump Operability Check,"

Revision 8

Surveillance Procedure OP-903-117," Emergency Diesel Generator Fuel Oil Transfer

Pump," Revision 1

Surveillance Procedure OP-903-050," Component Cooling Water Pump," Revision 11

Surveillance Procedure OP-903-004," Boric Acid Pump," Revision 9

Surveillance Procedure OP-903-050," Auxiliary Component Cooling Water Pump,"

Revision 11

Surveillance Procedure OP-903-003, " Charging Pump," Revision 8

Surveillance Procedure OP-903-035," Containment Spray Pump," Revision 8

Surveillance Procedure OP-903-030," Safety injection Pump," Revision 10

l

Surveillance Procedure OP-903-063," Chilled Water Pump," Revision 9

Surveillance Procedure STA-001-006," Leak Rate Testing," Revision 2

i

Administrative Procedure UNT-006-010," Event Notification and Reporting," Revision 14

I

' o.

.

.

I.

2-

Administrative Procedure, UNT-006-011," Condition Report," Revision 4

Administrative Procedure, UNT-006-021," Pump And Valve Inservice Testing," Revision 2

Site Directive, W2.302, "10 CFR 50.59 Safety and Environmental Impact Screening and

Evaluation"

Condition Reports

CR 95-1165

CR 96-0287

CR 96-0272

CR 96-0414

CR 96-1123

Work Authorizations

0114960

01144519

01144513

01143621

<