ML20211A467

Insp Repts 50-220/86-26 & 50-410/86-65 on 861117-870104. Violations Noted:Ineffective Corrective Action for Scram Discharge Vol High Level Trip & Momentary Breach of Reactor Bldg Integrity While in Power Operating Mode

ML20211A467
Person / Time
Site:	Nine Mile Point
Issue date:	02/06/1987
From:	Linville J NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20211A317	List: ML20211A312 ML20211A397 ML20211A467
References
50-220-86-26, 50-410-86-65, NUDOCS 8702190175
Download: ML20211A467 (15)

See also: IR 05000220/1986026

Text

. .

U.S. NUCLEAR REGULATORY COMMISSION

REGION I

Report No. 86-26/86-65

Docket No. 50-220/50-410

License No. DPR-63/NPF-54 Category B

Licensee: Niagara Mohawk Power Corporation

301 Plainfield Road

Syracuse, New York 13212

Facility: Nine Mile Point, Units 1 and 2

Location: Scriba, New York

Dates: November 17, 1986 to January 4, 1987

Inspectors: W.A. Cook, Senior Resident Inspector

J.E. Kaucher, Resident Inspector, Limerick 2

C.S. Marschall, Resident Inspector

G.W. Meyer, Project Engineer

W.L. Schmidt, Resident Inspector

Approved by: A . . [

r#Tinift e,/ Chief, Reactor Dat'e

'

rojects S tion 2C, DRP

Inspection Summary:

Inspection on November 17, 1986 to January 4, 1987

(Report No. 50-2'20/86-26 and 50/410/86-65

Areas Inspected: Routine inspection by resident inspectors of station activi-

ties (including Unit I operations and Unit 2 MSIV progress), plant tours,

surveillarce testing, safety system walkdowns, Licensee Event Reports (LERs)

review, allegation followup, review of Unit 1 Reactor Building Closed Loop

Cooling heat exchanger concerns, and a review of a 10CFR50.55(e) report. This

inspection involved 354 hours0.0041 days <br />0.0983 hours <br />5.853175e-4 weeks <br />1.34697e-4 months <br /> by the inspectors. Two violations were identi-

fied.

Results: Unit 2 MSIV progress is discussed in section 2.a and Standby Gas

Treatment System Problems are discussed in section 2.b. Details of a violation

involving ineffective corrective action for a SDV High Level scram are

discussed in section 2.1. Unit 1 and 2 control room activities are discussed

in section 3. A violation was identified at Unit I which concerns a momentary

breach of Reactor Building Integrity while in the power operating mode.

Details are provided in section 3. Unit 2 allegations concerning MSIV actuator

hydraulic fluid and Control Room HVAC are discussed in section 7.

An unresolved item regarding the Unit 1 RBCLC heat exchanger No. 13 is

discussed in section 8.

$

0

S

l0

. -.

.

-

,

DETAILS

1. Persons Contacted

The inspectors interviewed and discussed station activities with various

licensee representatives and contractor personnel.

2. Summary of Plant Events

UNIT 1

The plant operated at full power throughout the report period with power '

reductions for weekly control rod exercising and control rod pattern

adjustments.

UNIT 2

a. Main Steam Isolation Valve Review

The resident inspectors continued to closely monitor licensee's progress

in resolving Main Steam Isolation Valve (MSIV) problems. A region based

inspection was conducted the week of December 1, 1986, the results of

which are documented in Inspection Report 50-410/86-67. Resident

inspector coverage and licensee progress are noted below:

(1) Repairs to the damaged body of MSIV 6D were completed the week of

November 17, 1986 and the valve was reassembled and subsequently leak

tested unsatisfactorily. When dissassembled, the ball showed signs

of smearing of the stellite seating surfaces and scratching of the

tungsten carbide coating. The ball stem also showed signs of

abrasion damage. The licensee attributed both the damage to the stem

and coating to misalignment of the valve bonnet.

(2) On November 20, a newly coated ball was installed in the 60

MSIV body and the. leak test results were again unsatisfactory. A

different type of packing was used for this valve reassembly and it

became suspect as the cause of the leakage. Crosby was consulted and

a different type of packing was obtained and used. The valve was

reassembled and it passed the Type C leak rate test. The new Crosby

recommended packing was used in all subsequently reassembled valves.

(3) By the first week in December, all eight valves were reassembled and

the actuator modifications were completed. Wiring and logic testing

was then conducted which verified proper circuit response. On

December 3, a condition was discovered in the circuitry which

resulted in the paralleling of both Reactor Protective System

electrical power supplies. The event which lead to this discovery is

discussed in further detail in section 2.f.

. .

2 .

-

e

(4) Preoperational and surveillance testing was performed the week of

December 15 to verify proper valve operation. Except for some minor

limit switch adjustments, all valves operated properly. All MSIVs

met the three to five second closing time requirement.

(5) On December 20, the official Type C leak rate testing was performed

on all eight valves. MSIV 6B was the only failure, with a leak rate

of 8.9 SCFH. The valve was dissassembled and the ball showed

scratching and flaking of the tungsten carbide coating, similar to

the initial failure condition, although less severe. The bonnets

were removed from valves 6C and 6D to allow inspection of the balls.

-

Only minor scratching of the 6C and 60 ball coatings was observed.

The licensee conducted additional testing of the 6B MSIV ball in the

70 valve body to better quantify the failure in terms of leakage,

b. During this inspection period, several problems have occurred which

have either caused the Standby Gas Treatment (SBGT) System to auto-

matica11y initiate or to function improperly.

(1) On November 25, the SBGT system train A automatically started

due to a low flow condition in the Normal Reactor Building

Ventilation System. At the time the event occurred, prepara-

tions were being made to conduct the monthly surveillance on the

Reactor Building Ventilation Effluent Radiation Monitor. A

jumper installed on a ventilation damper control circuit was not

securely attached and fell off. When the jumper fell off, it

shorted to ground and caused the Reactor Building Ventilation

System to isolate. Licensee Event Report (LER) No. 86-09,

submitted on December 23, documented this event.

(2) On November 27 and December 8, SBGT automatically started due to

spiking of the Reactor Building Below the Refuel Floor Effluent

Monitor. LER No. 86-11, submitted on December 26, documented

this event.

(3) On November 28, SBGT automatically started due to a low flow

condition in the Normal Reactor Building Ventilation System.

The cause of the low flow condition was not clear. The

initiating event was a trip of the reactor building ventilation

supply fan. The operator who responded to the local ventilation

control panel did not thoroughly review the alarms on the panel

prior to resetting those alarms. The operator's observations

would have been particularly helpful since the progression of

events was uncertain.

The licensee concluded two possible causes for the low flow

condition: 1) the flow from one supply fan with two exhaust fans

running reduces the exhaust flow to below the trip setpoint, or,

2) the two running exhaust fans drew down the reactor building

pressure to 3" WG below atmospheric and then both tripped on

high reactor building differential pressure. Review of the

. .

3

alarm printer indicated that only the ventilation low flow

condition existed prior to the fan trips. LER No. 86-12,

submitted on December 23, documented this event.

(4) On December 7, during the performance of a surveillance on SBGT

System train B, operators observed that the heaters were not

energized, as designed. The heaters are controlled by an

internal system flow switch. During the first phase of system

operation, reactor building air is removed to create the

required negative pressure. When proper reactor building

pressure is established, the system enters a recirculation mode

to maintain that pressure. During the drawdown stage, the

heaters were energized. In the recirculation mode, the heaters

deenergized due to low flow sensed by the flow switch. The SBGT

System train A exhibited the same heater control / flow switch

problem. Licensee engineering evaluation resulted in a change

to the flow switch pressure sensing points to provide a better

indication of actual flow through the SBGT train.

(5) On December 31, a Reactor Building Ventilation Effluent Monitor

(HVR-RE14A) failed a surveillance test and was declared inoper-

able. To allow maintenance on the monitor, the monitor was

taken out of service and train A of SBGT System was started.

Train B of SBGT, which was left in standby mode, automatically

started due to a low Reactor Ballding to atmosphere differential

pressure. The licensee determined that the low reactor building

differential pressure appeared to have been caused by wind

effects on the atmospheric pressure sensing line.

(6) On December 31, while observing operation of the SBGT System,

the inspector noted that the instrument used to monitor the

differential pressure across one of the charcoal filter units

was pegged high (greater than 2" WG). The inspector reported

his observation to the control room and determined that this

problem had been previously identified by shift operators and a

station Problem Report was generated on December 24.

All of the above automatic SBGT System actuations were reported to

the Headquarters Duty Officer via the ENS. The resident inspectors

will review itcensee resolution of the above stated problems in a

subsequent inspection period.

c. On 11/17, the licensee determined that some field run concrete

embedded conduit, for lighting and communications cabling, did not

contain the required fire seals. Control Rod Drive testing was

temporarily suspended, pending an engineering evaluation of the

potential impact on secondary containment integrity. Drawings were

reviewed and inspections were conducted of the affected areas. No

secondary containment barriers were found to contain field run conduit

and control rod drive testing was then resumed. The inspector observed

some of the conduits in the Control Building and concluded that due

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

. .

t

'

4

s

.

to the small diameter, the cover plates in the opennings, and the lack

of straight through penetrations, these conduits were unlikely to be a

technical problem. The inspector reviewed licensee corrective action

and determined it was both adequate and timely in addressing this ,

deficiency. LER No. 86-08, submitted on December 17, documented '

this event..

d. On November 20, two consecutive half scrams occurred due to a loss of

power to Reactor Protection System (RPS) Channel B. The power was

lost from the Uninterrupted Power Supply (UPS) 3B which supplies RPS

Channel B logic and trip circuits. The licensee found that in both

instances the Electrical Protection Assemblies (EPAs) downstream of

UPS 3B had tripped. The licensee is still investigating the cause

of the EPA trips. All trip settings were tested and found within the

acceptable range. LER No. 86-07, submitted on December 19,

documented this event. The resident inspectors will review licensee

final resolution of this event in a subsequent inspection period.

e. On November 23, with all rods fully inserted, a Group 2 control rod

scram occurred. The Group 2 scram was a result of a loss of power to

the Group 2 RPS Channel A scram pilot solenoids, concurrent with a

half scram on RPS Channel 8 due to an Average Power Range Monitor

(APRM) surveillance test in progress. The licensee suspects the loss i

of power to the Group 2 Channel A scram pilot solenoids was caused by r

the inadvertant removal of fuses to the Group 2 scram pilot solenoid

relays, while hanging a tagout on adjacent relay fuses. The licensee

notified the NRC HQ Duty Officer of this scram via the ENS. The LER

has not been issued as of the end of this inspection period. I

f. On December 3, while performing energized logic checks of the 70 MSIV

actuator circuits, a full scram occurred. Both the 3A and 3B UPSs

lost power, causing both RPS scram sensor busses and all scram pilot

valves to deenergize. The licensee traced the cause of the UPS power

loss to technicians performing logic checks of the MSIV trip solenoid

power supply transfer circuits. MSIV trip solenoids are normally

powered from either the 3A or 3B UPS. When the trip solenoid's normal

power supply is lost, a transfer circuit energizes the trip solenoid

from the other UPS, thereby, preventing MSIV closure on the loss of

one UPS. During logic testing, a condition simulating the loss of

power to the trip solenoid, without actual deenergization of the UPS,

was initiated. The transfer circuit switched power to the alternate

UPS, but for an instant the two UPSs were paralleled. Because of

another unrelated problem, both power supplies were momentarily

paralleled out of phase. This transient caused both UPSs to experience

voltage and current surges which resulted in both power supplies being

electrically isolated by their respective protective devices. The

licensee has subsequently made a modification to the MSIV trip solenoid  !

circuit which they consider prevents the possibility of paralleling

UPSs. This modification is currently under review by the NRC staff.

LER No. 86-15, submitted on December 23, documented this event. l

- _ __ - __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

. .

5

. .

.

g. On December 10 and 11, three automatic High Pressure Core Spray

(HPCS) system initiations occurred. The HPCS pump started and went

into recirculation and the Division III diesel generator started and

ran unloaded, as designed. Water was not injected into the core

because of the high (actual) water level inje: tion valve interlock

prevented the injection valves from opening. The licensee initially

attributed the actuation signal to bumping the reactor vessel level

transmitters. Further evaluation by the licensee determined that the

suspect reactor vessel level transmitters' sensing lines have a

flexible tubing section just outside the containment penetration.

Bomping this flexible section of sensing line duplicated the low

reactor water level HPCS start signal. The licensee has roped off

the area around the flexible sections of sensing lines tb prevent any

further mechanical agitation. The potential for air entrapment in

the sensing lines was discussed with licensee representatives who

committed to review this as a potentially contributing factor. The

inspectors will review licensee action in a subsequent report period.

n. On December 16, the A train of Group 8 of the Primary Containment

Isolation System (PCIS) actuated when a testing error resulted in a

blown fuse in the power supply to the PCIS Group 8 train A relays.

During wiring checks of the Off Normal Status Panel of PCIS, a SWEC

technician erroneously connected a 125 VDC circuit to a 125 VAC

circuit. The Group 8 systems which isolated included: cooling flow

to the recirculation pumps; containment atmospheric monitoring;

drywell instrument air; and drywell equipment drains. The systems

responded as designed. Immediate corrective action included

replacement of the fuse, restoration of the isolated systems, and

suspension of further wirinq checks. The inspectors will review

licensee long term correctDe action in a subsequent report period.

1. On December 15, the Scram Discharge Volume (SDV) High Level Trip was

not bypassed prior to resetting a manually inserted scram. The mode

switch was moved from REFUEL to SHUT 00WN and the expected scram

signal was received. This scram was reset and approximately one

minute later a second scram was received due to the SDV High Level

Trip.

This event is similar to an event which occurred on November 5, when

the SDV High Level Trip was not bypassed prior to resetting a scram

and a second scram due to 50V High Level subsequently occurred. The

November 5 event was documented in Inspection Report 50-410/86-56 and

by the licensee in LER No. 86-01. Licensee corrective action taken

in response to the November 5 event was apparently ineffective in

preventing a recurrence. This is contrary to the requirements of

NMPC Quality Assurance Topical Report, Section 16, Corrective Action,

and the requirements of 10CFR50, Appendix B, Criterion XVI, and is a

violation. VIOLATION (50-410/86-65-01)

i . .

,

6

,

J. On December 18, both Division I, Low Pressure Core Spray

(LPCS) and Low Pressure Core Injection (LPCI), Emergency

Core Cooling Systems initiated and injected water into the

Reactor Vessel. The initiation signal was a spurious high

drywell pressure caused when a relay was being replaced in

the Reactor Core Isolation Cooling System. The Division I

. diesel generator started, as designed. One of the two

!

diesel room fans did not start automatically and required

manual starting from the control room. When the injection

was verified to be from a spurious signal, all systems were

returned to their normal standby status. The inspectors ,

observed no deficiences.

!

k. On December 22, due to the simultaneous draining of all three fuel

oil day tanks, all three Emergency Diesel Generators (EDGs) were

automatically disabled. Each day tank was partially drained to its

respective storage tank in preparation to run its associated fuel oil

i transfer pump. The transfer pump had to be run in order to draw a

representative sample for chemical analysis. When the day tank low

l level alarm was received, its respective EDG start logic locked out

l any automatic start signal. Control room operators immediately

identified the problem and had the day tank levels restored to their

l normal level. The licensee notified the Headquarters Duty Officer,

'

via the ENS, that the EDGs had been momentarily disabled. The

inspectors will review licensee analysis of this event and their

corrective action in a subsequent reporting period.

During review of this event, the inspectors noted that the logs in

each Emergency Diesel room did not indicate that a problem had

existed or that the chemistry surveillance had been performed. The

resident inspectors discussed the adequacy of the EDG logs with the

! licensee and will monitor the logs for improvement in subsequent

reporting periods.

1. On December 26, one of the twenty-six fire detectors which makeup the

Reactor Building 2625W Detection Zone System was found inoperable.

The detector was out of service for approximately 63 hours7.291667e-4 days <br />0.0175 hours <br />1.041667e-4 weeks <br />2.39715e-5 months <br /> and

compensatory action was not taken. This was contrary to Technical

l

Specification (TS) 3.3.7.8, which states, in part, that if a detector

is inoperable for more than one hour, an hourly fire patrol of the

'

affected area is required. As provided for in 10 CFR 2, Appendix C,

Section V, a Notice of Violation is not being issued for this

licensee identified TS violation. This TS violation was promptly

reported to the NRC by the licensee. The violation is of minor

safety significance, in that, the loss of the single detector did not

seriously degrade the overall effectiveness of the detection system.

The licensee corrective action was timely and positive, in that, upon

identification of the inoperable fire zone detector, a fire patrol

was immediately established. In addition, the licensee is conducting

retraining of all fire brigade personnel on the proper control of

_ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ - _

. .

7

-

.

,

. -

fire protection equipment and has initiated procedural revisions to

improve fire protection system administrative controls. There have

been no previous events of this nature at Unit 2.

m. On January 1, Division I and II batteries were declared inoperable,

based on an unsatisfactory surveillance test. During the performance

of the surveillance test, corrosion was noted on several terminals

and a high resistance was found on one connection. The licensee removed

the corrosion from the battery terminals and the one high resistance

connection. Subsequent resistance checks were completed satisfac-

torily. The inspector observed that the terminals appeared to have

varying amounts of corrosion inhibiting grease applied to them. The

inspector discussed this inconsistency in grease application with

licensee representatives and the vendor. The inspector determined

that the amount of lubricant is inconsequential as long as the

terminals and connectors are completely coated.

3. Plant Inspection Tours

During this reporting period, the inspectors made frequent tours of the

Unit I and 2 control rooms and accessible plant areas to monitor station

activities and to make an independent assessment of equipment status,

radiological conditions, safety and adherence to regulatory requirements.

The following was observed:

Unit 1

The Unit I control room atmosphere was evaluated on a frequent basis

during this reporting period. Few distractions were present and operator

knowledge of plant status was generally good, as demonstrated by an

infrequent need to refer to other sources for an accurate plant status.

As appropriate for an operational control room, noise level was low, only

personnel with official business were present, and the control room was

clean and free of extraneous material.

On December 10, 1986, during full power operations, the resident inspector

observed a security guard momentarily breach secondary containment. The

guard, who was responding from inside the reactor building to an alarm

condition for the reactor building outer airlock door, waited three to

four minutes for the outer airlock door to close. Assuming that the outer

airlock door was slightly ajar or that a malfunction had occurred in the

red light indicating that the outer door was open, the guard opened the

inner airlock door to investigate the cause of the alarm. When the guard

discovered that there were people in the airlock and that the outer door

was still open, he quickly closed the inner airlock door. Technical

Specification 3.4.0 requires secondary containment integrity during

reactor power operations. This is a violation. (50-220/86-26-01).

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

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _

. .

8

.

.

When station management was notified of this breach of secondary contain-

ment integrity, the licensee immediately counseled the responsible

individual concerning the correct actions to be taken when responding to

an alarm at an air lock door, if the red light prevents entry. On

December 12, 1986, the licensee drafted a memo to clarify the procedure

for assessment of alarms generated at air lock doors and the proper

actions to take if entry through an air lock door is prohibited. This

memo received wide distribution in the security department. This correc-

tive action was acceptable.

Unit 2

The control room atmosphere continues to be closely evaluated. During the

initial fuel load and open vessel testing phase, the number of operator

distractions and unnecessary traffic in the control room showed improve-

ment. Since the completion of Control Rod Drive testing, this trend has

not continued. The number of people not conducting official business and

traffic through the Control Room have increased.

The inspectors observed that all channels of the Unit 2 public address

system (HEAR-HERE) are announced in the control room, regardless of their

applicability to personnel in the control room. These non essential

announcements add to the noise level in the control room and appear to be

unnecessary. This observation was discussed with licensee management, who

acknowledged the concern, but did not consider the miscellaneous

announcements to be overly distractive.

Plant readiness for initial criticality, with respect to housekeeping, has

shown some improvement.

4. Surveillance Testing Review

The inspectors observed portions of the surveillance test procedures

listed below to verify that the test instrumentation was properly cali-

brated, approved procedures were used, the work was performed by qualified

personnel, limiting conditions for operation were met, and the system was

correctly restored following the testing.

-

N2-OSP-GTS-M001, Standby Gas Treatment System Functional Test,

revision 0, August 1986, performed on December 22, 1986.

-

Control Rod Drive System Testing performed in accordance

with N2-SVT-5-0V, during the week of November 17, 1986.

-

MSIV circuit and logic checks and preoperational testing at

Unit 2 was periodically monitored this inspection period.

No violations were identified.

. .

-

9

5. . Safety System Operability Verification

On a sample basis, the inspectors directly examined selected safety system

trains to verify that the systems were properly aligned in the standby

mode. The following systems were examined:

Unit 2

-

High and Low Pressure Core Spray Systems

-

Emergency Diesel Generators

-

Standby Gas Treatment System

No violations were observed.

6. Review of Licensee Event Reports (LERs)

The LERs submitted to the NRC were reviewed to determine whether the

details were clearly reported, including accuracy of the description of

the cause and adequacy of the corrective action. The inspectors also

determined whether the assessment of potential safety consequences had

been properly evaluated, whether generic implications were indicated,

whether the event warranted on site follow-up and whether the reporting

requirements of 10 CFR 50.72, where applicable, and 10 CFR 50.73 had been

met.

During this inspection, the following LERs were reviewed:

UNIT 1

LER ,#, Event Date Subject

86-23 8/1/86 Failure to Perform Testing Within

,

Required Interval - Fire Protection

System and Radiation Protection

86-29 9/10/86 Failure to Perform Testing Within

Required Interval - Torus

Temperature Monitoring System

86-34 12/6/86 Loss of Stack Sample Flow Due to

Software Problem

LER 86-23 and LER 86-29 document events associated with Technical Spect-

fication violations due to failure to perform a surveillance within the

required interval. Each of these failures was identified as a result of

! the licensee's implementation of a computer based program to control

! scheduling of surveillances. Since the violations were licensee identi-

,

fled, of minor significance, reported as required, immediately corrected,

could not have reasonably been prevented by corrective action for previous

I violations, and recurrence is prevented by the computer based program

,

which ideritified them, no Notice of Violation is issued for these events.

(

!

l

. .

.

10

,

LER 86-34 describes a violation of Technical Specification 3.16.14 b which

requires continuous monitoring of radioactive gaseous process and

effluent. LER 86-02 reported a similar event and identified a Technical

Specification conflict which requires a unit shutdown be initiated when a

loss of stack gas effluent monitoring occurs. In the December 6, 1986

event, a technician made a judgmental error in deciding not to report the

loss of stack gas monitoring, since the actual loss of monitoring was less

than an hour in duration. When licensee management became aware of the

event, immediate action was taken to counsel the technician regarding his

responsibility to notify the Station Shift Supervisor of changes in plant

equipment status and a standing order was issued to all chemistry techni-

cians to emphasize this requirement. In addition, the licensee has

included review of this event in the requalification training for

chemistry technicians. The amendment to Technical Specification 3.16.14b, which the licensee identified as a corrective action in

LER 86-02, revision 1 dated July 1, 1986, has not been submitted to the

NRC for review as of the end of this inspection period. Since this unique

event is of minor safety significance, was identified and reported by the

licensee, and adequate action has been taken to correct the technician's

error and prevent recurrence, no Notice of Violation is issued for this

event.

UNIT 2

LER # Event Date Subject

86-01 11/5/86 Scram due to IRM "D"

Upscale Trip and SDV High Level

86-02 11/4/86 All SRM Downscale

Channels Jumpered for Two hours

86-03 11/8/86 Partial Loss of Secondary

Containment Isolation Actuation

Instrumentation

86-04 11/9/86 Scram due to APRM Upscale

Trip

86-05 11/5/86 Loading of fuel in the quadrant

with SRM Channel C bypassed

The violations and events described in the LERs listed above were

previously reviewed and documented in Inspection Report

50-220/86-21 and 50-410/86-56, Section 2. An Enforcement Conference was

held on January 8, 1987 to discuss the violations.

A meeting was held with station management to discuss their overall

approach to LER preparation and review, on December 17. The inspectors

discussed three specific LERs and the problems found during their review

to illustrate their findings. In general, the LERs had adequate details

.. .

11

of the events, but the corrective actions specified lacked a proper root

cause analysis. Some technical errors were identified in the LERs which

should have been identified and corrected during the licensee's review

process, prior to submittal. The licensee plans to correct and submit

revised reports for the three LERs discussed (LER No's 86-02, 86-04, and

86-05). In addition, to address the general concern for proper LER

preparation and review, the licensee plans to revise their LER control

procedure.

,

7. Allegation Followup

During the inspection period, the inspectors conducted interviews and

inspections in response to allegations presented to the NRC. The

inspector and licensee actions resulting from these allegations are noted

below:

UNIT 2

Allegation 86-A-121: On October 8, 1986, control room heating and

ventilation design concerns were brought to the attention of the resident

inspectors by a licensee contract employee. The employee's concerns had

been previously identified to licensee station management, however, the

alleger was not satisfied with the licensee's response and desired to have

the NRC independently review the concerns. The contract employee's

concerns were formally presented to licensee management in a NRC Region I

letter, dated October 16, 1986. The licensee responded to the control

room ventilation concerns in a letter, dated October 24, 1986,

(NMP2L-0927). The inspector has reviewed the licensee's responses to

these allegations and found that, in general, the licensee's response

adequately resolved the concerns.

The following items refer directly to the concerns in the contract

employee's August 1,1986 letter to station management and provide the

inspector's conclusions regarding these concerns.

(a) The current design of the Nine Mile Point Unit 2 Control Room

environmental envelope meets the requirements of Regulatory Guide

(RG) 1.52, Section C, paragraph 2.g.

(b) The inspector conducted an independent review of the FSAR treatment

of hazardous chemicals and the effects of their release on control

room operators. The inspector concluded that the present design

adequately addresses all credible releases of toxic chemicals and

meets the requirements of RG 1.78.

(c) The inspector reviewed the completed preoperational test and deter-

mined that it has sufficiently demonstrated the adequacy of the

design to maintain the required positive pressure in the control room

to prevent infiltration, periodic testing, as required by Technical

Specifications, will verify continued system integrity.

. .

12

-

.

,

(d) The inspector concurred with the licensee's determination

that area differential pressure monitor 2HVC-PDI147 is not

required by RG 1.97 to be safety related.

(e) The inspector reviewed the preoperational test procedure

(N2-P0T-53-3, Rev.1) which was conducted to verify that the

maximum outside air flow does not exceed 1500 CFM and that

control room pressure can be maintained as required in the

FSAR. The inspector concluded that the current design

adequately achieves and maintains these design parameters.

(f) The inspector concluded that the existing design of the

control room HVAC system, coupled with the lower Technical

Specification limit of 90 degrees F in the control room,

will not result in exceeding equipment operating temperature

limits or endanger control room operators from a habitabilty

standpoint.

Allegation 86-A-143: An allegation was received concerning the

cleanliness of the hydraulic oil being used in the Main Steam

Isolation Valve (MSIV) actuators. The alleger stated that

controls on the replacement of the hydraulic oil were inadequate

and that the oil in the actuator sumps was contaminated.

MSIV actuators required modification to meet the Technical

Specification MSIV closure time requirements. To perform the

modification, the oil was drained from the actuators. After the

completion of the necessary modifications, the hydraulic oil was

replaced with new oil. The resident inspectors reviewed the

licensee's procedural controls for the MSIV hydraulic oil

replacement and discussed the most e xen+ oil change with

licensee representatives involved. The inspectors determined

that the licensee had previously addressed the quality of the

hydraulic fluid with the vendor. Oil quality specifications were

obtained from the vendor for onsite chemical analysis.

Replacement oil was sampled and found satisfactory prior to use.

Stone & Webster construction personnel, involved with the

hydraulic oil replacement, were interviewed and it was determined

that clear plastic five gallon containers were used to transport

the hydraulic fluid from the 55 gallon storage drums in the

warehouse to the Reactor Building for addition to the actuator

sumps. The inspector found this to be acceptable.

After the licensee became aware of this allegation, each MSIV actuator

hydraulic oil sump was sampled through a sump drain valve. All samples

were satisfactory and the analysis results were documented in the

licensee's Chemistry Maintenance Procedure. Inspector review of this

procedure identified no specification for visual clarity / suspended solids.

The lack of a visual clarity check was addressed with the licensee. NMPC

had concluded that some discoloration of the oil occurs normally during

usage, and that this discoloration was acceptable. It was determined that

. .

13

..

the hydraulic fluid system filter (40 microns) is designed to remove any

particulates and prevent any mechanical fouling of the hydraulic

components. One of the filters recently removed was opened for inspector

review. The inspector found the filter to be in good condition, consider-

ing it had never previously been replaced.

Inspector review of the licensee's actuator maintenance procedure identi-

fied that the procedure had not been revised to reflect the current

actuator design. Via discussion with responsible maintenance personnel,

the inspector determined that the proposed revision will also include: how

to add oil to the actuators, if required; how to check the oil level; and,

how to remove, inspect and clean the system filter. It was also deter-

mined that chemistry procedures will be revised to address the proper

hydraulic fluid sample frequency requirements.

Based of the inspectors' review, the concern could not be substantiated

for the MSIV actuators as currently assembled. The pending revisions to

the maintenance and chemistry procedures will be reviewed by the resident

inspectors in a subsequent inspection report. [

No violations were identified.

8. Failure to Identify Inoperable Component - Unit 1

On November 21, 1986, the licensee notified the NRC, via the ENS, of

a potential loss of the Unit 1 High Pressure Coolant Injection (HPCI)

and Control Room Emergency Ventilation (CREV) due to the failure of

Reactor Building Closed Loop Cooling (RBCLC) heat exchanger No.

13. Through wall leakage had been discovered during a

hydrostatic test conducted on May 20, 1986 (during the 1986

refueling outage). The Inservice Inspection (ISI) group which

conducted the test, requested a disposition of the heat exchanger r

from corporate engineering on June 16, 1986. Corporate

engineering evaluated the heat exchanger, which had been leaking

since 1978, as functionally operable and recommended a method of

repair. However, no evaluation was done of the effects of a r

seismic event on the heat exchanger or of the ability of the '

Condensate Transfer System to supply makeup to RBCLC. In August

1986, after completion of the refueling outage, a potential 10

CFR 21 report reveiw was initiated. The review identified that,

in a seismic event, the end of the heat exchanger could fall off

causing a complete loss of RBCLC. This catastrophic failure

would render HPCI and CREV inoperable due to lack of cooling

water and seal water supplied by RBCLC. However, this concern  ;

was not realized by station personnel until the week of November L

9, 1986. The heat exchanger was isolated on November 14, 1986.  !

i

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

. ..

.

.

, .. .

. .

14

. .

At the end of the reporting period, the licensee had completed an analysis

to demonstrate that failure during seismic loading was not a concern. An

analysis, to demonstrate the ability of the Condensate Transfer System to

supply adequate makeup to RBCLC, is in progress. The licensee has identi-

fled their failure to recognize RBCLC as a system required to support the

operability of safety-related systems and is addressing this concern. An

additional concern, identified to the licensee by the inspectors, was the

apparent oversight by station management in promptly resolving the heat

exchanger operability question. The heat exchanger failed its hydrostatic

test in May 1986, was placed in service, inspite of this failure, and was

not properly addressed by the licensee until November 1986. This concern

remains unresolved and will be addressed in a future inspection.

Unresolved Item (50-220/86-26-02)

9. Construction Deficiency Report Review - Unit 2

(Closed) Construction Deficiency Report (CDR) (86-00-21): Inadvertent

closure of automatic fire suppression control valve. On August 23, 1986,

three fire suppression systems for the Reactor Building were found ~

inoperable due to a filled common drain header, which had closed one

automatic deluge valve and would have closed the other two deluge valves

had there been actuation signals. The drain line had a closed valve in

it, and the line had become full from the water draining into it from

numerous previous actuations of the deluge valves. For the deluge valves

to function, an open, vented path must exist for water providing control

pressure to be continuously drained. NMPC concluded that the proper means

to ensure operation of these deluge valves was to cut the control pressure

drain lines from the deluge valves and allow the water to drain onto the

floor. Further, NMPC reviewed the control pressure drain lines on all

other NMP-2 fire protection systems and found the drains to be open,

vented, and without valves. The inspector reviewed the NMPC final report

dated October 23, 1986, inspected the cut drain lines, and inspected a

sample of other deluge valves to ensure unobstructed paths. The inspector

concluded that the NMPC resolution was acceptable and that the CDR was

closed.

10. Exit Meetinn

At periodic intervals and at the conclusion of the inspection, meetings

were held with senior station management to discuss the scope and findings

of this in.pection. Based on the NRC Region I review of this report and

discussions held with licensee representatives, it was determined that

this report does not contain Safeguards or 10 CFR 2.790 information.

,