IR 05000354/1988023

Safety Insp Rept 50-354/88-23 on 880830-1011.Major Areas Inspected:Operations,Radiological Controls,Surveillance Testing,Maint,Emergency Preparedness,Security,Ler & Engineering/Technical Support

ML20206C547
Person / Time
Site:	Hope Creek
Issue date:	11/01/1988
From:	Swetland P NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20206C527	List: IR 05000354/1988023 ML20206C526
References
50-354-88-23, NUDOCS 8811160318
Download: ML20206C547 (14)

Text

, __

.

.

"

.

.

.

U. S. NUCLEAR REGULATORY COMMISSION

REGION I

Report No.

50-354/88-23 License NPF-57 Licensee:

Public Service Electric and Gas Conpany P. O. Box 236 Hancocks Bridge, New Jersey 08038 Facility:

Hope Creek Generating Station Dates:

August 30, 1988 - October 11, 1988 Inspectors:

Glenn W. Meyer, Senior Resident Inspector David K. Allsopp, Resident Inspector Approved:

r

// / ff

,

Paul D. Swetlan'd,' Chief, Projects Section 28

' Da'te Inspection Summary:

Inspection 50-354/88-23 on August 30, 1988 to Oc,tober 11, 1988 Areas Inspected:

Resident safety inspection of the following areas:

operations, radioic71 cal controls, surveillance testing, maintenance, emergency preparede ss, security, engineering / technical support, safety assessment /cssurance of quality, and Licensee Event Report and open item followup.

Results: An Executive Summary follows.

Overall, PSE&G found problems regarding transmitter setpoint errors for feedwater flow, a missed surveillance test, and both trains of Control Room Emergency Filtration (CREF) being inoperable due to inadequate post-modification testing of a ventilation system adjacent to the control room.

PSE&G exhibited a good ability to detect, evaluate, and correct these problems.

8811160310 001103 PDR ADOCK 05000354 Q

PNV

- _ _ -

_ _ _.

_

__

.

.

'

.

.

EXECUTIVE SUMMARY Hope Creek Inspection Report 50-354/88-23 August 30, 1988 to October 11, 1988 Operations: Licensed operattrs raised concerns during a CREF surveillance test, which led to identifying problems with the required control room differential pressure and the operability of CREF. A quarterly surveillance for draining water from diesel fuel storage tanks was missed; this was considered a licensee identified violation.

Radiological Control.

The inspector found routine surveys of radiologically controlled areas to be effective.

Maintenance / Surveillance: An incorrect lead was lift (d during a surveillance test despite good procedural guidance and independent verification. The consequences of the error on the surveillance test were minimal, but the potential consequences of other such errors could be significant.

Drywell temperature monitoring was evaluated and found to be acceptable, but temperature anomalies in the readings are being further evaluated.

Emergency Preparedness: Emergency Drill H88-03 was reviewed.

The inspector concluded that exercising of C Team personnel was effectively achieved and that problem identification during the drill was good.

The computer simulation of a SPOS display in the control room was judged to be a good training aid.

Security: An open item regarding guard inattentiveness was closed out.

Engineering / Technical Support:

PSE&G identified that the reactor had been operated marginally above rated thermal power due to errors in the feedwater flow transmitter setpoints regarding transmitter static pressure compensation and venturi thermal expansion adjustment.

This area remains unresolved pending NRC review of PSE&G and GE evaluations.

The post-modification testing of the Unit 2 control room ventilation system was revealed to have been inadequate, but was judged to be a licensee identified violation of post modification test requirements. The inadequate testing permitted CREF to become inoperable.

Safety Assessment / Assurance of Quality:

The questioning attitude of personnel, which resulted in identifying problems regarding CREF, feedwater transmitter calculational errors, and a missed surveillance, was judged to be an asset.

The technical persistence exhibited regarding GE calculational errors was exemplary.

Supervisory oversight of field work was commendabl.- _- _ _ _ _ - - _ _

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

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

- _ _ _

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

_ _ _ _ _ _ _ _ _ - _

.

.

'

.

.

.

<

Oetails

,

1.

SUMMARY OF CPERATIONS The unit remained <' power during the entire report period.

Reductions in reactor power below full power included a two week period while calculating and verifying full power feedwater flow setpoints. Also, on September 30 an Unusual Event was declared and terminated twelve minutes later due to a shutdown initiated due to Technical Specification requirements (both trains

'

of Control Room Emerpency Filtration (CREF) System being inoperable).

2.

OPERATIONS (71707)

i 2.1 Inspection Activities On a daily basis the inspectors verified that the facility was operated safely and in conformance with regulatory requirements.

Public Service Electric and Gas (PSE&G) Company management control was evaluated by direct observation of activities, tours of the facility, interviews and discussions with personnel, independent verification of safety system l

status and Limiting Conditions for Operation, and review of facility records.

These inspection activities were conducted in accordance with NRC inspection procedure 71707 and included weekend inspection on September 24 and deep backshif t inspection on September 23.

,

2.2 Inspection Findings and Significant Plant Events

A.

On September 30 a shutdown was initiated due to Technical i

Specification (TS) requirements, and an Unusual Event was declared.

!

Twelve minutes later the shutdown and the Unusual Event were

terminated.

Technical Specifications require that both trains of the

'

i Control Room Emergency Filtration (CREF) System be operable.

However

!

i on September 30, PSE&G determined that the March 1938 surveillance l

tests previously used to establish the operability of CREF no longer

j properly tested the differential pressure between the control room

'

and adjacent areas.

The problem had been created by a recently i

installed Unit 2 ventilation system. The differential pressure had

been measured in only one area, not all adjacent areas. After the i

,

J Unit 2 modification the pressure measured in the control room was no longer representative of all areas adjacent to the control room.

Accordingly, both CREF trains were declared inoperable, and a proper test of CREF was initiated utilizing manometers to measure all adjacent areas.

Technical Specification 3.0.3 requires that at least one CREF i

train be returned to service within one hour or a shutdown be initiated.

'

When the testing exceeded one hour, the shutdown was initiated and was terminated twelve minutes later when the first CREF train had t

been properly tested and had been declared operable.

The reduction

!

in reactor power was negligible.

,

!

- -

.

.

. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _. _ _ _ _ _ _ _

.

.

-

,

.

-4-The inspector noted that concerns about the differential pressure were initially ratsad by licensed operators following routine surveillance testing of CREF, during which the differential pressure was not normally measured.

(The monthly test operates the CREF traine for an extended period to dry out any moisture in the filters.)

However, the operators noted that readings on the installed differential pressure gauge were abnormal and corrected the problem by turning off l

a newly installed ventilation system in an adjacent area.

The inspector

.

judged that the operators' findings represented a good practice in confirming expected readings on available instrumentation.

The problem was initially corrected by shutting off the newly installed Unit 2 ventilation system.

Later, ventilation system oalancing permitted the ventilation system to be returned to service.

The error associated with the ventilation system adjacent to the control room is further discussed in Section 8.8.

B.

On October 6 PSE&G determined that a required TS surveillance had been missed.

TS 4.8.1.1.2.d specifies that at least once per 92 days the accunulated water be removed from the diesel fuel oil storage tanks.

Although the tank draining procedure was issued to the operating shift in early September, the draining was not performed because a procedure revision was requested.

However, the procedure revision was not timely, and the tank draining was forgotten.

On October 6, a routine review of surveillance test completion found that the test was overdue (including grace period) on October 2 and that it had been overlooked. The procedure was immediately performed.

The resultant tank draining determined that very minimal amounts of water existed in the fuel oil storage tanks.

The corrective actions for the missed tes', included revising the tracking of routine tests to ensure that postponed tests would not be forgotten by instituting a monthly schedule listing on which deferred tests will be highlighted and test completion will be tracked.

Also, disciplinary action was taken against the person who had set the procedure aside and the person who had reviewed the routine test completion status too late.

Further, the computer 3,cking system for surveillance tests is being revised to list tests as overdue prior to their actual TS overdue date.

The inspector concluded that these corrective actions were appropriate and acceptable.

The inspector determined tnat the missed surveillance test was a licensee identified violation (354/88-23-01) which meets the criteria of 10 CFR 2, Appendix C, so that no citation would be issued.

C.

(Closed) Inspector Followup Item (354/87-10-01); Regulatory Guide (RG) 1.97 labeling. A RG 1.97 inspection identified unlabeled category 2 instruments mounted en main control room panels.

PSE&G corrected this deficiency, and the inspt:ctor verified all category 1 and 2 instruments on main control room panels were correctly labeled.

This item is close ___--__ - ___________ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _

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

- _ _ _.

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

_ _ _ _ _ _ _ _

.

.

'

.

-5-D.

The resident inspectors participated in a team inspection of the Emergency Operating Procedures (EOPs) during SeptemLer 6 to September 16. The results of the inspection will be documented in NRC Inspection t

Repert 50-354/88-200.

3.

RADIOLOGICAL CONTROLS (71707, 33526)

PSE&G's compliance with the radiological veotection program was verified on a periodic basis.

These inspection utivities were conducted in

accordant with NRC inspection procedsre /1707. The radiological control i

activities inspected were effective with respect to meeting the objectives of the radiological protection program.

,

In addition, the inspector reviewed the routine sin 7111ance of

!

centamination within radiologically controlled areas.

The inspector

!

"

reviewed procedures RP-Ap.ZZ-102, Radiological Surveillance, and

,

RP-GP.ZZ-001, Radiological Surveys, accompanied a technician during the

"

,

I weekly survey of the 54' and 77' elevations of the Reactor Build:ng, and reviewed the routine survey log.

The inspector concluded that surveys are

'

being performed in an effective manner to monitor and limit the spread of contamination.

4.

SURVEILLANCE TESTING (61726)

4.1 Inspection Activity

j During this inspection period the inspector performed detailed technical

!

'

procedure reviews, witnessed in progress surveillance testing, and reviewed

'

completed surveillance packages.

The inspector verified that the i

arvetIlance tests were performed in accordance with Technical Specifica-I tions, approved procedures, and NRC regulations. These inspection activities were conducted in accordance with NRC inspection procedure 61726.

The following surveillance tests were reviewed, with portions witnessed e

l by the inspector:

IC-FT.SP-053 Functional test of Main Steam Radiation Monitor C

-

IC-FT. BB-037 Functional test of B Channel Drywell High Pressure j

-

IC-CC.GS-001 Calibration of Hydrogen /0xygen Analyzer Channel A (

-

,

r IC-DC.ZZ-212 C 'ibration of RACS flooding detector l

-

i

.

OP-IS.BC-002 Inservice test of C RHR pump

-

'

l OP-15.EG-004 Inservice test of D SACS pump

-

,

P l

!

!

!

- - - - - - - - -

- - - - -

. _. ___

.

'

.

-3 l

OP-IS.BD-001 Inservice test o' RCIC pump

-

OP-ST.GU-001 Monthly operability test of FRVS

-

4.2 Inspection Findings F

A.

An incorrect alectrical lead was lifted during performance of the functional test on the C Main Steam Line R4diation Monitor. To ensure the proper electrical lead was disconnected, the lifted lead

,

procedural step had previously been hisnlighted with a note, required

'

an indenendent verification, and spec.fied the proper lead color.

The steps were checked off as having been accomplished correctly, but an incorrect lead (the field wire,.iot the cabinet wire as required

!

by the procedure) was lifted from the specified terminal. The consequences of the incorrect li ted lead were minimal, in that no c

equipment actions resulted

>..o the testing was only slightly delayed.

Specifically, when an anexpected instrument response occurred later

,

durinq the test, the problem was quickly identified, and the test

<

succ.nsfully completed. However, in discussions v ith plant management the inspector noted the potential for similar personnel errors in other circumstances to have significant, adverse effects.

The inspector noted that the error occurred despite detailed procedural guidance and independent verification.

Plant management stated that this error would be factored into ongoing corrective actions to avoid personnel errors.

The multiple personnel errors are considered to be a license? identified failure to follow procedures (TS6.8) for which no citation will be written.

(354/88-23-04)

B.

Temporary Instruction 2515/98 - Containment Temperature Profiles The adequacy and representativeness of containment temperature monitoring were evaluated and found to be accentable.

The inspector reviewed drywell air flow and temperature element locations, containment temperature profile, and the method of determining drywell average air temperature.

The temperature elements are uniformly located axially and radially throughou the drywell and are located to minimize the effects of the dr,well unit air coolers.

The calculational methodology complies with Technical Specifications (TS),

in that the average drywell temperature is a volumetrically weighted average of the TS specified temoeratures.

The average drywell air temperatures in June, July, and August of 1988 were 94,101, and 97 degrees F, respectively.

These temperatures were well below the TS maximum temperature of 135 degrees.

Although all average temperatures were acceptable, the inspector noted unexpectedly large variations in cay-to-day temperature readings.

In the worst case, a 30 degrees F. temperature swing occurred between successive days with no apparent cause.

PSE&G is evaluating this discrepancy, and the inspector will review the resolution in a future inspection report.

-

_ _ _ _ ___. - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ____ -_-_

_ _ _ _ _ _ _ _

_

___-

_ _ _ _ _ _ _ _

.

.

,

.

\\

'

\\

-7-

!

!

l C.

During observation of the drywell high pressure functional test, the

'

inspector noticed that the cover for the Rosemount trip card file did i

,

not properly lock into place. Ar.

camination of the 618 panel in the

,

lower relay room revealed that the covers on irip card files E21-238,

-

-248, -25B, and -268 were not properly locked.

In discussions with l

the inspector, the maintenance manager stated that Hope Creek intends l

that the trip card file covers be locked, although this is not l

,

required for system qualification or v erability, and that a problem

~

with the locks was being repaired to permit this.

Also, during the functional test the inspector examined the recent

!

'

modifications to the ECCS logic tester.

The inspector noted that

!

with the new toggle switches, which replaced keylock switches, the switch position was much easier to determine.

Accordingly, the toggle switches should prevent switch status errors similar to previous errors.

However, the inspector noted that due to their extended i

handle, the toggle switches are more susceptible to accidental repositioning, i

i 5.

hiAINTCNANCC (62703)

l

During this inspection period the inspector observed selected maintenance activities on safety related equipment to ascertain

that these activities were conducted in accordance with app.ovad

procedures, Technical Specific 2ticns, and appropriate industr!al codes and standards.

These inspections were conducted in accordance with NRC inspection procedure 62703.

Portions of the following activities were observed by the inspector:

Work Order Procedure Description

-

i MD-pM.EA-002 Service water silt survey i

871125117 MD-CM.EG-001 Repair of D SACS pump

-

.i

'

910414034 1BFSV-139 Pilot valve replacement on l

HCU 34-59

'

!

880906086 Ir-GP.ZZ-008 Standby 8.iquid Control optical i

switch mpair 871107045 MD-GP.ZZ-003 Repacking of RWCU valve 146 881005098 MD-CM,KJ-001 Torqu'?g of exhaust valves on i

D dic:

generator TN maintenance activities inspected n -..ffective with respect to meeting

,

the safety objectives of the maintenant program.

The inspectors noted l

good coverage of the field work by the first line supervisors.

L r

I

.

_

-

--

.

.- -_

.

.

.

-_

_

. - _.

_ - _ -

.

.

'

.

.g.

.

6.

EMERGENCY PREPAREDNESS (82301).

t The inspector reviewed the performance of PSE&G personnel during Emergency

.

Drill H88-03, which was held on September 28. The drill involved full

"

exercising of ;,tation personnel and included declaration of a General Emergency and a simulated offsite relea-One of the drill's objectives was to exercise the C Team of emergency.,sponse personnel, which 0 'uded some people performing roles for the first time.

The inspector noted that the drill referees included experienced response personnel, who provided guidance for the less experienced personnel on an appropriate basis.

The drill also exercised the Hope Creek Operational Support Center (OSC) for the first time u its new location.

During the drill numerous minor problems were encountered, e.g., communication systems for the OSC malfunctioned, a Post-Accident Sampling System (PASS) liquid sample could not be taken on one of several sample locations, procedural errors occurred,

,

l etc. The inspector concluded that the PSE&G observers had appropriately

identified the shortcomings and that the objective of exercising the C Team had been achieved. The inspector noted that a real time simubtion of the

<

drill scenario was provided on a temporary display of the Safety Parameter Display System (SPDS) in the control room.

The inspector concluded that

-

this simulation improved the training for the control room operators

i regarding SPDS * id the use of the Emergency Operating Procedures (EOPs),

'

7.

SECURITY (71707)

l

,

A.

PSE&G's compliance w'th the security program was verified on a

periodic basis, including adequacy of staffing, entry control, alarm i

stations, and physical boundarie:.

The inspectors concluded that the i

security activities inspected were effective with respect to meeting the objectives of the security program.

I l

j B.

(Closed) Violation (354/87-22-01); Inattentive security guard. A i

violation was issued against an inattentive guard, who was posted as

'

a compensatory measure on the protected area fence line and was discovered by an NRC inspector.

The PSE&G response dated December i

21, 1987 was reviewed and found to be acceptable.

PSE&G conducted

training, which included it. formation on diet and exercise to prevent

'

i drowsiness, and improved individual security force member i

accountability. Based on the acceptable corrective actions and the

{

!

good performance of the security force regarding attentiveness over

the last six months, this item is closed.

!

!

8.

ENGINEERING / TECHNICAL. SUPPORT (37702, 61706)

i l

l A.

Feedwater Flow Errors

!

Summary:

l

t l

PSE&G found two errors in the measurement and calculation of feedwater j

flow, which resulted in ac+.ual reactor power being above the ca'culated I

i

!

!

l

,

.

-

__. _ _ _ _

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

________ _

_____________-__ _______ ___- _

_

_

__ -

.

.

.

t-9-

.

.

,!

t reactor power (calorimetric) and potentially above the rated reactor thermal power.

The errors involved the lack of a correction to both feedwater differential pressure transmitters to account for known feedwater static pressure effects, and inaccuracies in the thermal tapansion factor for both feedwater venturis.

These errors resulted

in the transmitters' full scale setpoints (differential pressure)

'

'

being approximately 2.8% high, which results in calculated feedwater

!

flow and reacte gower errors of about 1.4% high.

The actual errors as calculated irom transmitter adjustments and their effect on the instrument operating range were approximately 1.23% and 1.18% on differential pressures of the A and B feedwater lines, respectively.

This corresponds to feedwater flow and reactor power errors of.61%

and.56%.

Following identification of this problem, the Hope Creek t

,

reactor was operated at 98N power until all factors associated with

-

calculating reactor power were verified to be correct.

On October 7 Hope Creek returned to full power when these verification efforts ware completed.

PSE&G will submit a written report, which will

'

address the maximum possible error and its effect on the transient analyses of the reactor.

The potential for enforcement in this area will be evaluated under Unresolved Item 354/88-23-02.

,

Scenario:

On September 21, 1988, reactor engineering informed the operating shift of the determination that a nonconservative error existed in the setpoint calculations for the feedwater flow transmitters, because

.

"

the static pressure correction had not been applied.

The error was found concurrently by evaluations of P0tential improvements in l

'

electric generator megawatt output, and by engineering verifications

'

of setpoint calculations on nonsafety-related transmitters.

This I

feedwater flow error was estimated to be approximately.8% and resulted

,

in a reactor power error of the same magnitude.

The reactor power

,

was reduced to 984 to compensate for this error, while the revised

'

setpoints were calculated and the transmitters properly calibrated.

!

On September 22 the NRC Operations Center was notified that the rated j

thermal power of the operating license had potentially been exceeded due to the transmitter setpoint error.

On September 23, a design change package (DCP) was approved for the revised transmitter setpoints, the transmitters were recalibrated to

,

the revised setpoints, and the Average Power Range Monitors (APRMs)

were calibrated based upon the revised feedwater flow calorimetric.

The reactor power remained at 98% power while additional concerns regarding the calculation of reactor thermal power were resolved.

,

These concerns had been identified by PSE&G while verifying the

accuracy of calculations for feedwater flow from measured differential pressure and for thermal power.

.

On September 26, General Electric (GE) assured PSE&G that the identified concerns regarding the feedwater flow and power calculations for Hope Creek were invalid, and that the calculations were accurate.

PSE&G increased reactor power to 100%.

'

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

-

.- - - -

-

.

y

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

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

.

.

~

.

-10-

!

.

On September 28 GE informed PSE&G that other nonconservative errors had been in the feedwater flow and reactor power calculations.

The errors involved the adjustment of the calculation for temperature effects on the feedwater venturis.

PSE&G immediately reduced reactor power to 98*4.

Hope Creek informed the NRC Operations Center that additional nonconservative errors had existed in the thermal power calculations, which could have caused the rated thermal power to have been exceeded. On September 30, personnel from GE arrived onsite to aid in verification efforts on thermal power calculations.

On October 4. GE issued Rapid Information Communication Services Information Letter (RICSIL) 30, which identified the venturi thermal expansion error and its potential applicability to other boiling water reactors (BWRs).

The inspector transmitted RICSIL 30 to NRC Region I and Headquarters offices for generic review.

On October 7, PSE&G and GE verifications were completed and reviewed by Hope Creek plant management, and the reactor was returned to full power.

Technical Background:

Hope Creek has two 24 inch feedwater lines, each of which has a flow venturi.

The measured differential pressure across the venturis is used to determine feedwater flow for the calorimetric calculation of reactor thermal power and for input to control circuits regarding recirculation system flow. None of these functions are designated as safety-related.

(Different venturfs in the three feedwater pump discharge lines are used to measure feedwater flow for vessel level control.) The full scale setpoints on the differential pressure transmitters are calculated based on the laboratory measurements of the venturis srior to their installation.

The process computer calculates reactor thermal power from the measurements of the heat energy in the fluids going into and out of the reactor vessel during a steady state period, i.e., a calorimetric calibration. This computed reactor power is used to adjust the power level of the reactor to full rated output and to calibrate the nuclear instruments, i.e., the APRMs, which provide the reactor powe-inputs to the Reactor Protection System (RPS).

Although there are other small adjustments, the computed reactor power is approximately equal to the feedwater flow times the energy difference between the steam out and the feedwater in. Accordingly, errors in feedwater flow cause approximately equal errors in calculated power.

Pressure Effect on Transmitter:

The differential pressure transmitters which measure feedwater flow are Rosemount model 1151 and have a known bias due to static pressure.

Specifically, the full scale setpoint calculated for this transmitter must be adjusted based on the static pressure at which the transmitter

_ _ _ _ _ _ _

-

_ _ _ _ _ _ _ _ _ -

__.

_ _ _ _ _ _ _ _

._ _ _ _ _ _ _ _ _ _ _ _ _ _

-_

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

.

.

.

I-11-

.

measures differential presse-by reducing the setpoint 1.45% per 1000 psig of static pressure. At Se normal, full power feedwater pressure of 1150 psig, a reduction in differential pressure setpoint of 1.6675%

is needed.

PSE&G concluded that this static pressure correction had not been made.

Because flow is proportional to the square root of differential pressure, the 1.6675% differential pressure error resulted in a calculated feedwater flow error of approximately 0.83%.

This flow error was nonconservative regarding reactor power, i.e., actual power is above calculated power, but was conservative regarding the recir-culation system flow control. However, when the transmitters were recalibrated to the revised setpoints, the transmitter outputs were increased by lower amounts, 0.50% on the A feed line and 0.43% on the B feed line.

PSE&G concluded that the smaller recalibrations were due to acceptable instrument drift.

Thermal Effect on Venturi:

i After finding the transmitter setpoint calculation error, PSE&G and i

GE performed confirmatory calculations on the equations, parameters, and instruments used to calculate the reactor thermal power.

This I

confirmatory effort found discrepancies in the setpoint calculations I

of the feedwater flow transmitters due to lack of compensation for i

thermal expansion of the venturis.

Specifically, for any given flow l

through the venturis, the measured differential pressure will vary

'

according to the steady state temperature of the venturis, because the venturis expand with increasing temperature and the inner diameters of the venturis change.

This adjustment via a thermal expansion factor is used in the calculation of the transmitter full scale setpoints and in the feedwater flow calculation, which is computed by the process computer.

When PSE&G found the discrepancies in the thermal expansion factor, GE initially assured PSE&G that the GE feedwater flow calculations were accurate, but following continued questioning by a PSE&G engineer, GE later confirmed that errors existed.

Specifically, there was no thermal expansion adjustment in the setpoint calculations; and the calculations did not account for the difference between the venturi's laboratory measurement at room temperature and the 420 degree normal operating temperature. Also, the process computer assumed expansion coefficients for carbon steel venturis instead of the stainless steel venturis installed at Hope Creek.

The errors in the thermal expansion factor resulted in reductions to the transmitter full scale setpoints of 1.1%.

Safety Significance The inspector calculated that the actual transmitter adjustments to correct transmitter pressure and venturi temperature errors resulted

.

.

.

-12-in combined increases in full scale transmitter output (differential pressure) of 1.54% and 1.47% for the A and B feedwater lines, respectively. This results in increases of calculated feedwater flow of.77% and.73%, respectively.

These feedwater flow errors would have caused reactor power errors of about the same magnitude.

Based on the corrections to the transmitter setpoints, the inspector calculated higher errors in rated feedwater flow of 1.40% in both lines.

PSE&G and GE have confirmed similar errors in setpoints and have begun analyses to determine the magnitude of the errors, including the statistical variation of the measured parameter. Also, the effect of the maximum possible error on the transient analyses will determined and evaluated.

PSE&G stated that the results of the analyses would be included in the 30 day event report to be submitted.

PSE&G and GE noted that the initial conditions for the reactor transient analyses included feedwater flow at 105.0% of rated flow and thermal power at 104.3% of rated power, as documented in Fina'

Safety Analysis Report (FSAR) Table 15.0-3.

Based on these init*41 conditions, PSE&G stated that there appeared to be sufficient margin to provide for safe reactor operation at the nonconservatively higher feedwater flows and reactor powers which resulted from the errors, but that the error analyses and transient analyses would determine the actual conclusions.

Inspection The inspector teparately evaluated the process computer's determination of core thermal power in accordance with NRC inspection procedure 61706 and found it to be acceptable.

The process computer input parameters were utilized by the inspector to manually verify portions of the process computer output.

The inspector reviewed the core thermal power evaluation procedure, procedure irplementation, and a recent procedure revision correcting a feedwater flow transmitter calibration error. Also, the inspector reviewed the revised setpoint calculations for the feedwater flow transmitter.

Corrective Actions Following the discovery of the second error on September 28, PSE&G committed that the reactor power would not be increased from 98% to 100% power until all instruments, parc. meters, and calculations had been verified to be correct by PSE&G and GE.

PSE&G and GE established a detailed action plan to address all aspects of this verificaticn effort.to ensure proper completion.

This verification effort was completed on October 7, and full power was achieved the same day.

In addition, PSE&G added additional manpower to the ongoing engineering evaluation of setpoint calculations on nonsafety-related instruments

in order to accelerate the evaluation's progress.

PSE&G stated that

<

the corrective actions would be addressed in the event report to be

submitted by October 21, 1988.

!

__

, _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

_

__

_ _ _ _ _ _ _ _

___ _ _ _.

.

.

~

.

-13-Conclusions NRC conclusions regarding the effect of the feedwater flow errors will be be made following review of PSE&G's written report, including the worst case errors, effect on transient analyses, determination of root cause, and review of GE evaluations of applicability to other reactors. The followup on this area will be under Unresolved Item 354/88-23-02.

B.

Modification Effects on CREF As noted in Section 2.2.A, the control room operators noted a concern regarding the ability of the Control Room Emergency Filtration (CREF)

System to develop the TS required positive differential pressure above adjacent areas.

Evaluation by technical personnel concluded that a recent modification to the abandoned Unit 2 control room area had adversely affected the Unit 1 control room, and had been undetected by post-modification testing.

Specifically, the control room for Unit 2 (abandoned) was modified to provide office space for operations and reactor engineering personnel via Design Change Package (DCP)

4EC-1026, which was completed and occupied in September 1988.

PSE&G later concluded that the ventilation system for the office space had inadvertently pressurized the office space, such that the required positive differential pressure of the Unit 1 control room to this adjacent area could not be provided by CREF. After this determination, the newly installed ventilation system was shut down to restore CREF's operability.

Although the DCP noted that the control room boundary was involved (a door in the boundary was removed and compensation was made to maintain the boundary), no post-modification testing was performed to verify that the required positive differential pressure for the Unit I control room was maintained.

The DCP had been performed by offsite engineering and had been reviewed by both onsite and offsite engineering personnel. The inspector concluded that this lack of adequate post-modification testing represented an error by the design and review personnel but was not a programmatic shortcoming.

Based on licensee identification of the problem and meeting the criteria of 10 CFR 2. Appendix C, this violation (354/88-23-03) will not be cited.

Additional PSEC.G evaluation of the oversight will be contained in a Licensee Event Report (LER) to oe submitted.

9.

SAFETY ASSESSMENT / QUALITY VERIFICATION The inspector observed that the questioning attitude of plant personnel had been fruitful in identifying underlying problems.

Specifically, the identification of the CREF differential pressure problem and the feedwater transmitter setpoint errors represented good followup on potential problems.

Also, regarding the venturi thermal expansion errnr, an engineer continued to pursue his questions after assurances from GE that the concerns were invalid. This technical persistence was exemplar,

-

-

.

..

'

.

-14-

,

.

As noted in Section 5, the inspector observed good coverage of field activities by the first line supervisors. On some occasions the supervisory licensed operator was also in the field for equipment testing in addition to the maintenance supervisor. The inspector noted that this commendable practice occurred during a period when some maintenance supervisors were loaned to Salem 2 during its refueling outage.

10.

LICENSEE E' VENT REPORT (LER) AND OPEN ITEM FOLLOWUP (92700)

A.

PSE&G subnitted the following event reports and periodic reports, which were revi,$ wad for accuracy and timely submission.

Monthly Operating Report for August 1983 LER 88-22-00 Main Turbine Trip During Weekly Surveillance Testing of the Thrust Bearing Wear Detector Resulting in A Reactor Scram-Equipment Failure; discussed in Section 2.2.A of Inspection Report 50-354/88-22.

LER 88-23-00 A Reactor Water Cleanup Pump Seal Failure RWCU Isolation-Equipment Failure The inspector also reviewed LERs 87-49-01, 88-003-01, and 88-013-01, which supplemented previously reviewed LERs.

B.

The following previous inspection items were followed up during this inspection and are tabulated below for cross reference purposes.

Closed 87-10-01 Section 2.2.C.

.

Closed 87-22-01 Section 7.B.

No further concerns were identified during these reviews.

11.

EXIT INTERVIEW (30703)

The inspectors met with Mr. J, Hagan and other PSE&G personnel periodically and at the end of the inspection report period to summarize the scope and findings of their inspection activities.

Based on Region I review and discussions with PSE&G, it was determined that this report does not contain information subject to 10 CFR 2 restrictions.

_

e O