IR 05000293/1988012

Insp Rept 50-293/88-12 on 880306-0417.Violations Noted. Major Areas Inspected:Radiation Protection,Physical Security,Plant Events,Maint,Surveillance,Outage Activites & Repts to NRC

ML20197F270
Person / Time
Site:	Pilgrim
Issue date:	05/31/1988
From:	Blough A NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I)
To:
Shared Package
ML20197F263	List: IR 05000293/1988012 ML20197F258 ML20197F265
References
50-293-88-12, NUDOCS 8806100256
Download: ML20197F270 (46)

Text

_x i - .

/ +

O k\ <

,*

g

<,

.

.\,> ;It ( '

<

3' ,

',s I

!(t

,d >'

c,

.

/

/ U. S. NUCLEAR REGULATORY COMMISSION

' '

REGION I,

'

Docket / Report No.: 50-293/88-12 e

Licensee: Boston Edison Company 800 Boylston Street s' , (' -

Boston, Massachusetts 02199 i i

'

Facility: Pilgrim Nuclear Power Station

>

j) 1 t ation: Plymouth, Massachusetts

'Datas: Marcn 6, 1988 - April 17, 1988 Irspectors: C. Warren,SeniorResidsti!nspector j s, J. Lyash, Resident Inspector T. Kim, Resident Inspector s J. Golla, Reactor Enfineer i Approved by:

A. h ndy Blo g Chfef

. , _ _ _ b 3/~ @ 7

.,

.

?

/ Date Reactor Projects Section'No. 3B s Division of Reactor Projects \,

,

. inspection Summary:

'

Areas inspected: Roatine resident inspe'ction of pirnt operations, radiation proteqt i on , physical s2curity, plant events, maintenance, sur7 eillance, outage activinc ies, and reports to the NR Principal T Rensee manager.;ent representa-tives ccntacted are listed in Attachment I. :op'e s ( d' handouts used by the \,

iicensee during their March 10, 1988 presentation on the Direct Torus Vent '

Modification are included as Attachnent I Results: ,

i Violation: Inadequate des 4gn control and review were evidenced in the incor-rect installation of the reactor water level gauges. Also, weaknesses in ade-quate test procedures and teriinical reviews vsre i(:entified in the preopera-tional tests performed on these instruments. (il10j?9 42-02 Section 3.c)

i  ;'

Unresolved Itecn: An error in the liccnsee's procedure for calculation of bat- ,

tery capacity was not identifie:1 duricig a performance test or post-test revievs. ' '

Other battery testing and maptenance weaknesses were! also identified. (UNR 88-12-01, Section 3.b) -;

,

I

'

'

( t ,, l

't

,

f i

l , l

'

800S100256 880531

>PDR

'O DCD

.

lADOCK05000293 /

_ . _ _ . .

, ,g . ., .

._ _ .

,

.7 ,

,

-

.

, > >

_

,y

'pq ,

+

,

, , Ol _,

t Inspection Summary (Continued) 2 l

a ',.

Strengths:

'

'

--

, The .-licensee's approach to problem investigation and -root cause analysis was prompt and . positive. Eve 4 critiques led by the Operations Section Manager and root cause analyses performed:. by the onsite Systens Engineer-ing Group ' appeared ~ to be thorough and aggressive. (Sections 2 ~and,,3.c); The levels of detail, technical accurac/,- and 'the overall quelity 'of' LERs

have improved during the last six rao73hs. (Section 5) , ,

,, 7

/

a

'

,

'h

a I

t if

1. ' #

] p a  ! r f

-

,

d

,

3 _ r

_ _ _ _

,

,

. l

. ,

,

.

TABLE OF CONTENTS

,

Page Summary of Facility ana NRC Activities........ ................ 1 Followup on Previous Inspection Findings (Modules 92701 and 92702)....... ................................. ............. 2 Routine Periodic Inspections (Modules 71707, 61726, 62703, 71709 and 71881)............................................. 4 General Plant Tour Observations........................... 5 Region I Temporary Instruction 87-07, Storage Battery Adequacy Audit..... .... .............. ............. .. 5 Plant Maintenance and Outage Activities................... 8 December 1987 Containment Integrated Leak Rate Test (CILRT) Results Evaluation............................ . 10

. Review of Plant Events (Modules 71707 and 62703).... .......... 12 Reactor Water Cleanup System Spurious Isolation........... 12 Spurious Secondary Containment Isolatio .... ...... . . 13 Review of Licensee Event Reports (LERs) (Modules 90712). .. . . 13 Management Meetings (Module 30703)............................. 16 Attachment I - Persons Contacted Attachment II - Direct Torus Vent Presentation Handouts c

I i

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

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

., .,' .

.

.

DETAILS 1.0 Summary of Facility and NRC Activities The plant was shutdown on April 12, 1986 for unscheduled maintenance. On July 25, 1986, Boston Edison announced that the outage would be -extended to include refueling and completion of certain modifications. The reactor core was defueled on February 13, 198 The licensee completed fuel re-load . on October 14, 1987. Reinstallation of the reactor vessel internal components and the vessel head was followed by completion of the reactor vessel hydrostatic tes The primary containment integrated leak rate test was also completed during the week of December 21, 198 During this report period, the licensee continued with the post modifica-tion / maintenance testing of plant equipment. Effective on March 30, 1988, Mr. Roy A. Anderson, former Planning and Outage Manager at Pilgrim, relieved Mr. Robert J. Barratt as the Plant Manager. The licensee has not yet named a permanent replacement for Mr. Anderso It was announced on March 29, 1988 that Mr. F. N. Famulari, former Deputy Quality Assurance Department Manager, has replaced Mr. D. L. Gillespie a; the Quality Assurance Department Manage Also, effective on April 1,1988, Mr. C. J. Gannon lef t his position of the Radiation Protection Manager to become the Planning and Outage- Services Section Manager. Mr. W. Muliins was named the Acting Radiation Protection Manager / Chief Radiological Engineer until a permanent Radiation Protection Manager is selecte NRC inspection activities during the report period included: 1) evalua-tion of the licensee's revised Emergency Operating Procedures during the week of March 14,1988,2) review of the licensee's radioactive waste pro-cessing systems and effluent monitoring during the week of April 4,1988, 3) evaluation of the licensee's security program effectiveness during the week of April 11, 1988, and 4) review of previous inspection findings dur-ing the week of April 11,1988. -0n March 9,1988, Mr. William T. Russell, Regional Administrator, Region I, toured the station with the resident inspector Also on March 10, 1988, Dr. Thomas E. Murley, Director, Of fice of Nuclear Reactor Regulation (i. ".A), Mr. William T. Russell, and other management representatives from both NRR and Region I toured the site, inspected the plant areas where the direct torut vent system has been installed, and interviewed licensee engineering department personnel regarding the syste _ _ _ _ - _ _ _ - - _ _ _ _ _

,

.

. .,' .

.

.

2.0 Followup on Previous Inspection Findings (Update) Unresolved Item (UNR 87-53-05), Part 3, Review the results of licensee inspection of the Emergency Diesel. Generator (EDG) lube oil

- filters and strainers. During the November 12, 1987 loss of offsite powe event the prelube pump for the "B" EDG failed to restart on demand. It was identified that - metal chips had become lodged between pump internal

~

components, causing the pump to bind. The licensee agreed _to perform inspections of other lube oil system components to determine if additional foreign material was present. The "B" -EDG lube oil system was subse-quently drained. The inspector witnessed licensee inspection of the tube oil sump, filters, and straine No foreign material which could iave caused the prelube pump failure was identified. Similar inspections were later performed on the "A" EDG during a routine equipment outage. No sig-nificant foreign material was foun These ' inspections indicate that externally generated material was not the cause of the pump failur On December 13, 1987 the "B" EDG prelube pump again failed to start when energized. Disassembly revealed a failure identical to that observed in November 1987. The licensee sent parts from one of the failed pumps to a materials laboratory for analysis. Results from this analysis indicated that a loss of internal clearance caused severe idler gear chafing against the pump head to the point of spot-melting. This allowed chips to tear from the pump head, weld to the idler gear, and cause the failures. The licensee's system engineering group believes that the loss of internal clearance is caused by an inability of the pump to absorb thermal growt ' The pump shaft is direct couoled to the motor. The drive gear is slip fit to this pump shaf Each pump was found .to have rust between the shaft and the drive gear, eliminating any drive gear movement during thermal *

growt The licensee has contacted both the pump and diesel vendors regarding the results of the evalbation. In addition, the licensee engi-neering department is reviewing the current application to determine if pump replacement with an alternate design is warranted. The prelube pumps t.re not considered safety-relate The licensee's root cause evaluation appears to have been thorough. The pumps are routinely shutdown and restarted during biweekly EDG surveillance The licensee is pursuing available replacement options. The inspector had no further question This portion of UNR 87-53-05 is considered ciosed. Remaining portions of this item will be reviewed during future inspection (Closed) Inspector Follow Item (86-29-03), review licensee analysis of standby gas treatment system (SGTS) single failures. This item was last updated in inspection report 50-293/88-07. The inspector expressed con-cern that the 2000 Cubic Feet per Mincte (CFM) setpoint foi the SGTS dis-charge flow monitor would not ensure a nega +.i ve secondary containment pressure under all circumstance If the operating train were degraded and flow remained just above the 2000 CFM trip point, the standby train would not automatically start and the secondary containment function could

.- _

. - . .. . _ .-

'

- -

. ,

.

.

be lost. The licensee provided the inspector with Field Revision Notice (FRN) 86-70-270 dated September 27, 1987. This FRN raised the flow set-point to'3350 CFM. Based on review of the safety evaluation included with the FRN, and recent SGTS performance testing results, it appears that this change adequately addresses the ine;,ector's concer The inspector also quest ioned che need to perfbrm a post-work' flow test -

to verify proper sizing of the newly installed crosstie' cooling orific Licensee calculations, construction documentation and .a description of controls' applied to final inspection and system closecut were provided by the-licensee. The licensee stated that the above items constitute - suf-ficient basis to conclude that the orifice is correctly sized and instal-le Periodic performance tests of the system, required by- technical specifications, would also indicate any substantial blockage or- design error. The inspector had no further questions. Based on the above, this item is considered close (Closed) Violation (87-45-03). In response to discovery of non-job related reading materials and a card playing machine in the control room the licensee took steps to identify the source of the materials and whether they had been used by on-shift personne Personal interviews were conducted by the Senior Vice President-Nuclear with members of the Operations Department. The resu't' of these inter-views established that the materials were brought to the control rrom -by members of the operations staf The licensee's investigation did nat identify cases where the material had been used in the control roo In aodition to the interviews, the licensee took addition.1 actions to pre-clude recurrence. These additional steps included meetings between senior management and the operations staf f, assignment of management personnel to

, observe backshift and weekend control room conduct, and prohibition of non-work related reading material and entertainment devices from any '

process buildin Based on the *esults of the licensee's investigation and control room observations ano interviews conducted by NRC personnel this item is close (Closed) Unresolved Item 87-57-02, incorrectly installed reactor vessel level gauges. The details of this item are discussed in Section 3.c of this repor i l

.

l

i i

!

- *

- . - . .r--. _ . -4 -r --i,.---,------***rv- r ~wv------5+-=---e,v,---v-v,y ---giv+-*-ee-*-r-t- --**g-w-*-t--*--vvn --~%*=*e---

. - - - - . .. . - --

.

. ., .

.

4 3.0 Routine Periodic Inspections The inspectors routinely toured the facility during normal. and sackshift

,

hours to assess general plant and equipment conditions, housekeeping, and adherence to fire protection, security and radiological control measure

Inspections were conducted between 10:00 p.m. and 6:00 a.m. on April 7 and April 16, 1988 for a total of four hours and during the weekends of March 26, -April 9 and April 10, 1988 for a total of nine hours. Ongoing work activities were monitored to verify that they were being conducted in accordance with approved administrative and technical procedures, and

,

that proper communications with the control room staff had been estab-lished. The inspector observed valve, instrument and electrical equipment lineups in the field to ensure that they were consistent with -system operability requirements and operating procedure During tours of the control room the inspectors verified proper staffing, access control and operator attentivenes Adherence to procedures and

limiting conditions for operations was evaluated. The inspectors examined equipment lineup and operability, instrument traces and status of control room annunciators. Various control room logs and other available licensee documentation were reviewe The ir,spector observed and reviewed outage, maintenance and problem inves-tigation activities to verify compliance with regulations, procedures, codes and standard Involvement of QA/QC, safety tag use, personnel !

j qualifications, fire protection precautions, retest requirements, and

,

reportability were assesse The inspector observed tests to verify performance in accordance with ;

approved procedures and LCO's, collection of valid test results, reti.wal ,

and restoration of equipment, and deficiency review and resolutio Radiological controls were observed on a routine basis during the report-

'

ing period. Standard industry radiological work practices, conformance to radiological control procedures and 10 CFR part 20 requirements were observed. Independent surveys of radiological boundaries and randem sur-veys of nonradiological points throughout the facility were taken t y the ;

inspector, i

,

'

Check; were made to determine whether security conditions met regulatory I requirements, the physical security plan, and approved procedures. Those 1 checks included security staffing, protected and vital area barriers, personnel identification, access control, badging, and compensatory i measures when require !

,

,

r

.. _-

.

• -

. .

'

a. General Plant Tour Observations On March 10, 1988, Dr. Thomas Murley, Director of the NRC Office of Nuclear Reactor Regulation (NRR), and Mr. William Russell, Adminis-trator of NRC - Region I, toured the plant with the resident inspec-tors. In addition to a general plant tour, installed portions of the licensee's direct torus vent (DTV) modification were examined. Sub-sequently, Dr. Murley, Mr. Russell and members of the NRC technical staff interviewed licensee staff and received a presentation describ-ing the development and design basis of the DTV. Copies of handouts used by the licensee during the presentation are included as- Attach-ment I b. Region I Temporary Instruction 87-07, Storage Battery Adequacy Audit Region I Temporary Instruction (RTI) 87-07 was performed to determine if the licensee has established a program to ensure ' storage battery !

operability, in accordance with the current licensing basis. The safety-related DC power system at Pilgrim includes three class IE, seismically qualified, lead-calcium type storage batteries. Two d; visional, sixty cell,125 VDC batteries supply safety-related con-trol power and some motor operated valve (MOV) loads. A single 120 cell 250 VDC lead-calcium battery supplies mctive power to the high pressure coolant injection system MOVs. Each of these three_ batter-ies is equipped with a dedicated charger. A single backup charger is shared between the two 125 VDC batteries, and a second backup charger is provided for the 250 VDC battery. In addition to these three safety-related batteries, Technical Specifications require an oper-

. able 125 VDC battery to serve switchyard and transformer protective i

relaying circuits. Two sixty cell lead-calcium oatteries located in i the relay house are installed to fulfill this requirement. Technical

.

Specifications also require the operability of the 24 VDC battery and

.

l charger associated with the diesel fire pump. Other station batter- l ies not addressed by Technical Specifications include two neutron monitoring system 24 VDC batteries; one 125 VDC security battery, diesel generator air compressor 12 VDC battery, and emergency light-ing batterie i Several previcus NRC inspections have focused on design, maintenance i and testing of storage batteries at Pilgrim. NRC Performance Assess-ment Team (PAT) 50-293/85-30 performed a detailed review of ~ the mod-ification package for replact: ment of the 250 VDC batter The PAT reviewed design specifications, manufacturer's duty test results, !

licensee periodic battery performance testing and battery operability I e

criteri Specialist inspections 50-293/87-09 and 87-21 reviewed the j technical adequacy of battery performance test procedures, test j i

l

1 l

_l

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

'

.

. . ., .

.

results, battery storage area housekeeping, and physical condition of the cells, intertie bars and storage racks. Special NRC Electrical Team Inspection 50-293/88-08 also performed inspections of battery ,

and storage rack physical condition. Inspections 87-09, 87-21 and --

88-08 all identified indicators of poor battery maintenance prac-tices. Inspection 88-08 contains a notice of violation addressing this are During the current period the inspector toured each of 'the battery areas and . examined general housekeeping, physical location and arrangement of the area, the existing condition of the batteries and racks, and verified the operability of storage area ventilation sys--

tems. Cleanliness and housekeeping conditions had improved somewhat from those noted in NRC inspection 50-293/88-08. The three safety-related storage batteries along with the neutron monitoring batteries are located in dedicated, locked rooms in the turbine building. A ventilation system is provided to maintain acceptable room tempera-tures. The ventilation exhaust is withdrawn from the area high point .

to prevent gas buildu Associated fans, dampers and duct work appeared to be in good conditio The licensee presently monitors pilot cell condition weekly, specific gravity and voltage of all cells quarterly, and performs a discharge test once each operating cycle. This testing is applied to all sta-tion batteries. The procedures appeared to be technically adequate and consistent with the vendor manual and industry standards. Appro-

' priate battery and storage rack physical inspection and maintenance

'-

instructions were inc uded to ensure continuing seismic' qualifica-tio Precautions regarding proper ventilation and protection from ;

ignition hazards were containe The licensee maintains a battery '

charger maintenance and calibration procedur The inspector reviewed the licensee's DC load profile and its basis.

,

Current DC system configuration was reviewed to determine consistency with the assumed loads. Both safety-related 125 VDC batteries and the 250 VDC battery were replaced in 1980 and 1981 respectivel Results of the most recent discharge tests were evaluated and indi-cate that sufficient capacity, under worst case circumstances, still-exists. Typical test results indicate capacities in excess of 95

,

]

!

j percent of original values. An overall minimum capacity of 80 per-

,

cent has been established as the battery electrical end of life. The

licensee plans to conduct a duty cycle test during the next refueling outag I

< .

f

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

'

• *

, ..

9 The inspector noted that the formula used to calculate battery capacity in procedure 8.9.8, Battery Rated Load Discharge Test, is incorrec Application of the temperature correction factor as specified yields invalid result In completed discharge tests reviewed, it appears that' personnel completing the calculation did not apply the correction factor because of this formula erro In

, addition the discrepancy was not identified during the licensee's post-test resalts revie The effect on the results however, is minimal, less than two percent. The licensee committed to review the procedures and to make appropriate corrections. Procedure 8.9.8 also

, includes steps for recharging of the battery after completion of testin However, only a single set of specific gravity and cell voltage readings are taken to verify acceptable battery recharg No followup readings are included to verify that the battery par-ameters are stabilized. The licensee statec that in practice, oper-ations personnel do taks followup readings. The licensee maintenance i manager stated that the need to formalize this practice by incorpora-tion into procedure 8.9.8 would be evaluated. The inspector request-ed the results of the most recent battery charger calibration and maintenance activity. The licensee was not able to provide documen-tation of this work prior to the close of the inspection period. The licensee committed to identify the last time the procedure had been performed and to supply the results and next scheculed performance date to the inspecto During the inspection period the operations department performed sur-veillance procedure 8.C.16, Quarterly Battery Cell Surveillanc Specific gravity readings for a large number of cells on several of the batteries were found out of specification. The licensee later identified that the wrong type of hydrometer had been used to take the readings. Use of the correct hydrometer resulted in acceptable results. The inspector noted that the hydrometer usually used in performance of the test is maintained by the operations department and is not a controlled or calibrated instrumen The Operations Section Manager stated that a controlled instrument would be used in future testing. The inspector also noted that water used for addi-tion to batteries was stored in bottles in the battery rooms. This does not represent positive control of water quality. The Operations Section. Manager stated that the bottles would be removed and water for makeup to the batteries would be obtained directiv from the chemistry departmen Licensee actions to correct deficiencies in procedure 8.9.8, ensure use of a controlled instrument for measuring specific gravities, provide better control of water quality and to provide the results of battery charger surveillances will be evaluated in a future inspec-tion under unresolved item 88-12-0 I

' *

. ,

.

c. Plant Maintenance and Outage Activities Followup of HFA Relay Failures On January 17, 1988, a spurious reactor scram signal was generated during a routine instruinent calibration. Following the actuation the licensee identified that a secondary containment isolation had not resulted as designed. Investigation revealed that a contact on a General Electric (GE) HFA relay had not fully closed when the relay was deenergize Failure of the contacts to close prevented the isolation signal from going to completio This incident was de-scribed in paragraph 4.d of resident inspection report 50-293/87-5 The failed relay was removed and shipped to GE for testing. Both GE and licensee analyses concluded that the relay had been improperly adjusted during installatio Individual contact fingers must be adjusted to provide adequate contact wipe on closure. The failed relay contact was found to have very little wipe. When the relay was deenergized and the contact closed it continued to display high re-sistance. The relay was adjusted and installed just prior to refuel-ing by a licensee electrical technicia This technician installed only one additional relay at that tim The licensee removed the second relay and found indications of similar misadjustments. The procedures used by the technician appeared adequate. The training provided prior to the relay replacement however, may have been wea '

On March 16, 1988, the Watch Engineer noted that automatic scram re-lay 5A-K14C was chattering loudly. As a precaution a manual half scram was inserted and the relay deenergized. Licensee investigation identified that relay SA-K14C was chattering due to insufficient voltage supply to the coi GE HFA relay SA-K4C contacts in the power supply circuit for the SA-K14C coil exhibited high resistance, causing the observed voltage loss. Relay 5A-K4C was removed, examined i and found to have inadequate contact wipe resulting frcm misadjust- '

ment. This misadjustment was similar to the condition found on the j secondary containment isolation relays described above. During the l current outage the licensee replaced about 180 HFA relays. A special relay setup and replacement team was formed and extensive training was conducted. In addition 100 percent quality control coverage of the activities was maintained. Improperly adjusted relay SA-K4C was installed by this tea The licensee's systems engineering group is l developing a matrix of personnel versus relay replacements in which l they were involved A temporary procedure will be written to sample I the 180 relays repl. md this outage using a sample plan based on the '

matrix. The licenst.= . also reevaluating the training provided and the procedures used. The inspector will continue to monitor licensee followup in this are . , - . _ .- - .

., ,- .

.

Incorrectly installed Reactor Level Gauges During a followup investigation to an inadvertent reactor scram sig-nal on January 17,1988, the licensee identified two reactor vessel level instruments (LI 263-59. A&B) with ~ incorrectly connacted sensing lines. The instruments were recently installed under Plant Design Change (PDC) 85-07 and had not been turned over to 'the operations

,

department. These new Barton gauges would only be used for~ local indication if reactor shutdown f*om outside the control room was needed. The incorrect installation was due to an error in the con-figuration drawings which were issued as a part of Field Revision Notice (FRN) No. 62 to PDC 85-07. The initial PDC 85-07 package was reviewed by the plant Operations Review Committee (ORC) for its im- ,

pact on plant safety and also for its adequac FRN 62 however was

not considered as a major FRN and thus bypassed the ORC revie The .

i licensee initiated a Potential Condition Adverse to Quality Report to :

track this concern. Their engineering department is reviewing the requirements and guidelines for determining major versus minor FR Based on review of licensee records and interviews with licensee ,

personnel, the inspector determined that the pre-operational testing of the instruments was inadequate. The pre-operational testing pro-cedure TP 87-86 did not prove that the instruments tracked actual water level as required by the PDC 85-07. Instead, TP 87-66 appeared

_

to be a simple instrument calibration. The inspector also reviewed the f aily operator surveillance records, Station Procedure 2.1.15, .

'

and noted that the instruments LI 263-59 A&B have been checked with readings recorded as pegged high. The operators interviewed indi-cated that they had not raised any questions about the abnormal gauge reatings since both gauges were tagged out of service and had .not been turned over to Operation The inspector noted that these instruments (LI 263-59 A&B) are not included anywhere in the Tech-nical Specifications (TS). The licensee is reviewing the regulatory ;

requirements and licensee commitments to determine if the instruments should be in the T i The licensee's investigation concluded that the cause of the scram was the particular method used to calibrate the instruments. The calibration was performed with the instruments isolated from the pro-cess line and drained cf all the water. The test equipment ~was at-l tached to the instruments and air pressure was used to simulate the

.

'

differential pressure from the process. When the instruments were returned to service, the air pockets released into the linas caused pressure fluctuation at several instruments served by the saa4 lines,

'

resulting in the reactor scram. The licensee indicated to the in-i spector that the proposed corrective actions include revising the '

) Procedu-e 8.M.2-2.1.2 to require "wet" calibration and to evaluate j the adequacy of other instrument calibration procedures. The inspec- ,

i tor will review the licensee actions in this area in a future inspectio !

i l

. . _

_ ~

., ,- .

f I

.

The licensee's investigation and root cause analysis led.by the on-rite Systems Engineering Group were aggressive and thorough. The licensee is currently reviewing the adequacy of other safety-related :

protective instruments installed during this outage. Thus far, the problems associated with the instruments (LI 263-59 A&B) appeared to be an isolated case. However, the design control deficiency at, evi-denced in the FRN 85-07-62 is in violation 'of 10 CFR 50, Appendix B, Criterion III, and the Boston Edison Company Quality Assurance Manual (BEQAM). BEQAM Section 3, Design Control, requires that measures be established for the control of design activit,ies to assure appropri-ate quality standards and design review Further, BEQAM Section .

8.3.2.8 requires that methods for verifying design changes, such as design revieres and qualification testing are properl/ chosen and fol-lowed; the most adverse design conditions are specified for test pro-grams used to verify the adequacy of designs. Contrary to the above on January 19, 1988, it was determined that parts of the Plant Design Change (PDC) 85-07 for installation of new reactor water level-gauges had not been properly reviewed and released in that the configuration ,

drawings were incorrect, which resulted in incorrect installation of the gauges. The FRN 85-07-62 was released on December 12, 1986 and the implementation of the FRN 85-07-62 was completed on April 22, 198 It was also determined that the . design verification testing for the installed reactor water level gauges, Temporary Procedure '

87-66, Pre-operational Test of the New Barton Indicating Units LI l 263-59A and LI 263-598, completed on June 10, 1987, did-not meet the :

requirements of the BEQAM, Section 3.3.2.8 in that the testing failed ;

' to verify the design adequacy (VIO 88-12-02). Failure to establish adequate test procedures and to perform adequate technical review I during the blackout diesel generator testing and the plant process computer point tie-in activities were the subject of a previous vio-lation as documented in the inspection report 50-293/88-07, d. December 1987 Containment Integrated leak Rate Test (CILRT) Results Evaluation The inspector reviewed the licensee's December 1987 CILRT results documented in accordance with the requirements of 10 CFR 50 Appendix 1 J paragraph V.B.3. These results were summarized in a technical i document entitled "Reactor Containment Building Integrated Leakage !

Rate Test" and attached to the licensee's letter dated March 15,~1988

-

!

to the NRC. The report contains a test summary and general test description, presentation of test results for the Type A (CILRT) and

.

'

Types B & C (Local Leak Rate Tests, LLRT), and a description of the licensee's efforts to improve containment integrity (ILRT/LLRT Bet-terment Program). Bcth Mass point and Total Time calculational met-hods were employed for the December 1987 CILRT. The Total Time met-hod of ANSI N45.4-1972 is consistent with the requirements of the current version of 10 CFR 50 Appendix J and is the method of record for the test.

.

. - . . _ - - - .

.

.. .,' .

- .

l

The purpose of .the test was to demonstrate that leakages through the primary containment building and systems penetrating containment do not exceed that allowed by plant technical. specifications. The test-was conducted with containment isolation valves (CIV's) and contain- !

ment pressure boundaries (CPB's) in an "As-Left" condition. The con-tainment.could not' meet the leakage criteria in the "As-Found" condi- '

tion due to excessive . local leakage. This has been acknowledged by

,

!

the licensee and reported per the requirements of 10 CFR 50.73,

.Licensec Event Report (LER) system. The test was witnessed by an NRC i regional inspector' during a routine safety inspection and was fol-lowed by a successful verification test. Inspection findings are documented in USNRC Region ~ I Inspection Report No. 50-293/87-5 Pertinent test parameters and results are presented below: , Type "A" Test parameters and Acceptance Criteria a Test Method Absolute

, , Calculational Methods . Total Time (per ANSI N45.4-1972) '

Mass Point (per ANSI /ANS 56.8-1987)

'

4 Test Duration:

i Stabilization Period 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> Data Gathering for Leakage 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> Calculation

'

Verification Leak Rate Test 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> li Test Pressure 59.69 psia (full pressure test) Maximum Allowable 0.750 wt. %/ day leak Rate at upper bound of 95% confidence limit Test Results Wt. %/ Day

)

)  !

Acceptance, maximum 0.750

allowable leak rate Measured Leak Rate, Lam 0.189 for Total Time Method

} l

!

.- -.

.. . ... . .- .. - - . . . - - - - - - .

7, . ,,' n I

. .

l Test Results Wt. %/ Day Leak Rate at the Upper Bound -0.240 g of the 95% Confidence. interval l

'

Total Corrections 0.010 (Type 8 & C penalties and water levels) l Total Type "A" "As-Left" 0.250 Leak Rate, Total Time i Method I

Conclusion Acceptable in "As-Left"  !

condition i

The inspector concludes that, based on a review of the results, the l

! containment has passed its acceptance criteria in the "As-Left" con- i ditio Failure in the "As-Found" condition has been - acknowledged and reported by the license .0 Review of Plant Events

.

The inspectors followed up on events occurring during the period to deter-mine if licensee response was thorough and effective. Independent reviews

of the events were conducted to verify the accuracy and completeness of

' licensee information. During this period, the licensee made 'the following reports to the NRC pursuant to 10 CFR 50.72:

I Reactor Water Cleanup System Spurious Isolation '

On March 11,1988, at 10:20 p.m. , tha licensee experienced an auto-matic closure of the inboard primary containment isolation valve on the reactor water cleanup (RWCU) system suction line. Investigation by the licensee indicated that the technicians performing a surveil-lance on the electrical portions of the system inadvertently grounded a wire which had been 'if ted during the surveillance test. Grounding the wire resu'ted in a blown logic power fuse, and deenergization of this portion of the logic caused the valve to automatically clos The fuse was replaced and the test subsequently complete The licensee's investigation concluded that the cause of the actua-tion was non-licensed utility technician personnel error. An Instru-ment and Control (I&C) technician was removing an area high tempera-titre switch in the RWCU logic circuit for a routine calibration in

i accordance with the Procedure 8.M.2-1.2.2, "Reactor Water Cleanup  !

Area High Temperature". Factors contributing to the errnr were the J

!

. ._ . . .

.. .,' .

!

.

.

type of electrical connections involved with the work, and the gloves worn by the technician to perform the work. The gloves (i.e., inner cotton lining gloves and outer rubber gloves) affected the dexterity of the technician during the removal of a screw from the lug oi a temperature switch lea A similar inadvertent isolation of the RWCU system occurred on

!

'

December 17, 1987 during the performance of the Procedure 8.M.2-1.2.2. '

As a corrective action a revision to the Procedure 8.M.2-1.2.2 was under consideration to incorporate a request for removing the RWCU system from service when the area high temperature switches were to be calibrated. The licensee decision on this proposed revision had not been finalized at the time of this event. Since the revised pro- !

cedure involves removing a normally operating system from service, it is considered only a short term corrective actio The licensee has initiated an Engineering Service Request to review possible change to the frequency for calibration of the area high temperature switches, a'id a possible modification of the temperature switches or tempera- ,

ture switch connections. The inspector had no further question t Spurious Secondary Containment Isolatioa On March 31, 1988, at 12:42 p.m. an inadvertent secondary containment isolation and an automatic start of the "A" and "B" standby gas

'

treatment trains occurre A licensed operator performing a daily surveillance test of the refueling floor high radiation monitors failed to properly reset the downscale trip for two of the channels.

.( When an upscale trip was inserted in a third channel the full isola-i tion signal was generated, resulting in the actuations. The trips i were reset and the system returned to normal a short time later. The NRC was informed of the actuations via ENS at 1:55 p.m. Licensee ,

investigation identified that personnel error was the primary caus Poor enmmunications between control room personnel and a weak proced- ,

ure nere also found to be contributors. The licensee has counseled i the operator involved. Control room communications is the subject of j an ongoing licensee training program. In addition a review of the routine daily surveillance test procedure was initiated. The inspec-

+or had no further questions.,

5.0 Review of Licensee Event Reports (LERs)

LERs submitted to NRC:RI were reviewed to verify that the details were

!

clearly reported, including accuracy of the description of cause and ade-

-

quacy of corrective actio The inspector determined whether further information was required from the licensee, whether generic implications were indicated, and whether the event warranted onsite followup. The fol-

'

icwing LER's were reviewed:

l

., ,

.

.

LER N Event Date Subject 87-007-00 10/18/87 Automatic Actuation of the Reactor Pro-tection System Due to a Personnel Erro /15/87 Unplanned isolation of shutdown cooling 87-008-01 during implementation of a modifica-tio Immediate inspector followup of this actuation is described in inspec-tion report 87-4 During inspection 87-50, the inspector noted that the LER identified personnel error as the pri-mary root caus Based on the licen-see's own root cause evaluation the root cause was found to be procedural deficiency. The licensee committed to issue an updated LER. Subsequently, LER 87-008-001 was submitte /23/87 Seismic Class I conduit routed through Class II area of the circulating water intake structure due to an original design deficiency. Existing unresolved item 87-34-01 was established to track NRC followup to this problem, and licensee evaluation of other Class II

, structure /2/87 Full reactor scram signal due to a spurious trip of average power range monitor ( APRM) "E".

Immediate inspector followup of this !

actuation is described in inspection report 87-2 The actuation was not initially reported by the licensee as required by 10 CFR 50.7 This was identified by the inspectors and tracked as unresolved item 87-45-0 The licensee submitted LER 87-010 on November 20, 198 I l

l

- _ _ - _

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

-

, .

, ,

-

i 15 1

-

i l

LER N Event Date Subject 87-011-00 7/7/87 Full reactor scram signal due to a failed logic car Inspector followup of this actuation

, is described in inspection report 87-2 An LER was not initially submitted by the licensee. LER 87-011 was subsequently issued in response to unresolved item 87-45-05 on November 24, 198 /28/87 Full reactor scram signals due to spik- ,

ing of intermediate range monitor '

Inspector followup of these actuations

, is described in inspection report  !

! -87-4 A ~eview conducted by the '

licensee in response to unresolved item

"

, 87-45-05 identified that the required  ;

LER was not submitted. LER 87-012 was '

subsequently issued on December 7, 198 '

) 1 87-013-00 11/8/87 Breaching of a security vital area boundar Violation 87-50-02 is pending enforce-ment action in this are /12/87 Loss of Offsite Powe ;

An Augmented Inspection Team was dis-patched in response to this even Inspection results are documeated in report 87-5 /7/87 Unplanned isolation of shutdown cooling

'

during installation.

Inspector followup of this actuation is described in inspection report 87-57.

!

i

.. .,' -

LER N Event Date Subject 87-016-00 11/24/87 Unplanned actuations of prima ry con-tainment, secondary containment and standby gas treatment system Notice of Violation 87-50-07 was issued as a result of followup to this even /25/87 Reclassification of a plant area as a security vital are Unresolved Item 87-50-03 was opened to monitor licensee action in this are The inspector noted that the levels of detail, technical accuracy and the overall quality of LERs have improved during the last six month .0 Management Meetings At periodic intervals during the course of the inspection period, meetings were held with senior facility management to discuss the inspection scope and preliminary findings. A final exit interview was conducted by the resident inspectors to convey final inspection results and findings on May 9, 198 No written material not already available to the public was provided to the licensee by the inspector. The inspector confirmed during

~

the exit interview that no proprietary information was supplied by the licensee during the perio l l

l

.

. ,,'

.

Attachment I to Inspection Report No. 50-293/88-12 Persons Contacted R. Bird, Senior Vice President - Nuclear

• K. Highfill, Station Director R. Anderson, Plant Manager E. Kraft, Plant Support Manager F. Famulari, Quality Assurance Manager A. Morisi, Planning and Outage Manager (Acting)

D. Swanson, Nuclear Engineering Department Manager J. Alexander, Operations Section Manager J. Jens, Radiological Protection Section Manager J. Seery, Technical Section Manager R. Grazio, Field Engineering Section Manager P. Mastrangelo, Chief Operating Engineer R. Sherry, Chief Maintenance Engineer W. Mullins, Chief Radiological Engineer D. Long, Security Section Manager F. Wozniak, Fire Protection Division Manager

• Senior licensee representative present at the exit meetin :

.

ATTACHMENT II

_

Direct Torus Vent Presentation llandouts DIRECT TORUS VENT SYSTEM MEETS NRC REQUIREMENTS FOR SEALED CLOSED ISOLATION VALVE

- NO EFFECT ON DESIGN BASIS ACCIDENTS NO CHANGE TO TECHNICAL SPECIFICATIONS USE FULLY CONFORMS TO NRC APPROVED EPGs

,

SIGNIFICANT IMPROVEMENT RELATIVE TO EXISTING VENT CAPABILITY

!

l l

\ _ _ _ _ _ _ _ _ _ _ _ _ ___. _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

.

DIRECT TORUS VENT SYSTEM FIGURE 3.2 - I T ^

-

/

t REMOTE ( MANUAL i

I sw

. i j l I I RUPTURE i DIS K 1 . DRY 3 x u

$\

y WEL I

+y; ; ; ;R2 !;  ; ,;; : :  : : :  : 6 zy <  ; w J -

@ /

-

2o , lQ

, ; m -

w /

/

'

/ -~ x ipH -tv SBGT o '

i /

50428 504A --p /} --b/24 20'. 7 flPE DUCT FC

"a ~

_ DUCT F Pt ;

WET WELL EXISTING

/ / // NEW PIPE

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

-.

.

. .

.

-

.

.

Basic Strategy of the EPGs o

Defense in depth

.

o identify appropriate actions and limits in advanc o Provide a graduated response keyed to certain important plant operating parameters.

,

^ o Prevent as damage to either core or containment as long .

possibl I-o Maximize the time available to recover system i o

Mitigate core damag P

f

i i

i f

I s

., ,-

Basic Primary Containment Control POOL DRYWELL CONTAINMENT CONTAINMENT TEMPERATURE TEMPERATURE PRESSURE WATER LEVEL 0 ERATE NORM ODERATE NORM OPERATE NORMAL LEVEL PDOL COOLING DW COOLING SBGTS CONTROL l l l t v ir v

1. 3 f 3 I '

LCO LCO SCSIP f LCO 3 EXCEEDED? j q EXCEEDED? j APPROACHED?j EXCEEDED?

l I l l v v v v C ER ATE ALL OPERATE ALL OPERATE REDUCE EX 3C

ODL COOLIN3 DW COOLING WW SPRAYS RDV INJECTtDN I

t v l v v HCTL I I ' I '

MAX DW TEMP ' SCSIP STPLL A:: ROACHED?j (APPROACHED?j q EXCEEDED? j (ADPROACHED?j l

v 1r 1r CE::ESSUR1ZE m OPERATE Rev ' DEPRESSUR ZE

, DW SPRAYS RPV

/ v 1r

~CTL MAX DW TEMP I PSP I STPLL ( E):E:DED? t EXCEEDED? j q REACHED? j EXCEEDED?

1r m

'

SLOW DOWN m

'

.

.

RPV 1r I ' Y PCPL A

LCO Limiting Concition for Operation k PPROACHED?J nCTL Heat Capacity Temperature Limit 1r PCPL Pnmay Containment Pressure Limit VENT 3

,SP F Pressure Suppression Pressure CONTAINMENT SCSlo Suppression Chamber Spray Initiation Pressure ,

STPLL SRV Tait Pipe Leve! Limit

?

.,..... ..

-

.

\

.

Primary containment venting is required for pressure control only:

o When all other decay heat removal mechanisms combined are inadequat o When primary containment pressure is well beyond that calculated for any design basis acciden e When the structural capability of the containment is

threatened, directly or indirectl '

When plant conditions have so degraded, the operating crew cannot reasonably rely upon a fortuitous turn of events to reverse the situatio :

I

~

Not venting will, lacking the fortuitous turn of events,

!

result in primary containment failu,e and most probably loss of adequate core cooling and core damag i Venting will result in preservation of primary containment

)

integrity for as long as possible and most probably

!

continued adequate core cooling without core damag l l

i l

l l

_ _

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

.. . . .

.. .,- .

. .

.

Plant Conditions Which Must Exist Before Venting '

,

o Pool cooling unavaliable/ insufficient o Drywell cooling unavailable / insufficient e Wetwell sprays unavailable /insu4icient

,

o Drywell sprays unavailable / insufficient

'

e Main condenser unavailable '

o RPV depressurized

o Shutdown cooling unavailable / insufficient

,

L o

Primary containment pressure in excess of that calculated for any design basis accident

5

.

I d l l

i l

!

I

!

J

_

.. .. .

Potential Negative impacts of Containment Venting:

Du p NPSH reduction e Subsequent de inertion o 'nability to reciose the vent path e

Reactor building habitability degradation e

inadvertent venting o

Premature venting

l

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

. , ,,

. ]i\pJ i *  !

-

I w n l //, z a Nk

= < a = = , - - r>

I e /s I .g t: w i; > O( E 6*

~ ~  ;

,

.

, 5 fat a

, 2-a R"

a )

9 i e - j - - -

l .

d w-K > > ' ~< -

ls Y- > >

,

%

s =

W X '

E j g > .-

.X 5 K- =

si \

-

0~ . . . -

5" d g*

@

-

> 5

~

5 ~ ' ~ L E u

- - - -

,

, .\ , 1 i i A. A

• '

% *

n '

a ,3 ll

,

! , j

- ,

-

.

a3:

,

i

'

l

! '

'

w=Z

'

ym

'

,. IS3 I i .I

.

t I i '

~ll. .

g

'

l $U' i

._

I f

I l ,

i

-

g i .' , )  !

I4

,

l 3'

kWd :I ' ,

gi' s a

• g .

g i -

, -

6* * *

E-1 w m3 3 I .I, ,

c C i F5 f l 4 0

-

__

_ ,

'

-y l .i ., . 1 5 * i i m

. . ,

t 23 o .3 j

,

j , :/, f

., i . '

,K' "y l >i

.

i r

. b i

-

'

i j4 *~

} I

• zt

,

.e l l i

t e

i

E i>. ' , 't-V 3 a, { l l

i'

Y l- s -

'

~

A

~

W's L. W ,

i t ~e : .

- -

1 IW In E l1 j s oZ

~~ }

w~

.

b 0 l

.

~x

,

'T,i g_ l ',

i l au l ec l $$ ' 3 1

{

,

I UN

. w

y'i l

!

I w-w i

. l

ed e-e '

.Wi I'

S E 2 I '

@7 e ew * < l , --

E >:.2 8* .. .

>

'

,

@f J .e 3= g I C

• U q 6 HO l t b Q ,

,  ; --

i

[

H *

s M N Y G L 6

'

• >zf z 3

'

l $3 *"

w

@

.

b

\ *l-a t

25 - w

W5 j Y

,

i

-

l w 4 g z 5

'

g5 .i

,

1 fg

g d  :

is zia-Y  :

. s T I M

h470013-035. ;;

~.GBh )

l J

. . - . . --

.

- -

.

'

-

.

=

PRLl_ IMINARY -

~

QORE_ DAMAGE _f REOUENC.Y_

.

'

_

PILERIM IPE RESULTS m

N Containment Contain.nent Overpressure

" 'A Overpressure , ,y)

-.

(1W) 21% _ Hi Heactor Pressure 3 7 *'-

(TOUX) \ " """ ' ' #'"'"

\ (TOUX)

n a/. 5 ~'* 3 0 AIWS

\ '

-

Low Heactor

'

o2% ~ P'OSSU'e (TOUV)

3 9 =<- { j (DCA

.

23 % ATWS Sb ~~'

L 10 %

.

o4%

ICIA' $% Low Heactor Pressure (TdllV)-

-

-

^ .

WITHOUT CONTAINMENT VENTING CDF:4 Se-5/yr WITH CONTAINMENT VENTING CDF=2.7E-5/yr

.

_

.

""'

w'

' __ _ ,

_

-

e" g -

.

-

PRELIMINARi Iable 1  :

PitCRIM DIRLC1 10R!a5 VINI SENSIIIVI;Y 5tilllY-StistARY OF REstiliS -

Irequency of Preventive Ventip_q f requergy_ i Mitigative Venting 5equence containment IW 5equentes Class Pressurect) 11., C t l

<.91-4/Yr ifjuX S.91-9/V' 9.lf-Il/Yr 580 I.2I-6 2.3I-8 IljtlV 2.81 -7 1. II -8 AV 2. SI -8 IE-9 1.5I-6/Yr I.JE-7/Yr

_ Core Damage Probability Containment Release Probability Dose Consecuences_

5p}uence V e., t No Vent Vent No Vent Vent flo Vent II)tlX 9.lE-6/Yr 9.lE-6/Yr 1.4E-7/Yr 1.3E 7/Yr 1.3 R/Yr 1.1 R/Yr SPil 2.3E-5 2.3E-6 1.2E-6 5.5E-7 If)llV 1.lE-6 .lE-6 2.8E-7 1.6F -7 . fi IWf}ilV 2.10-6 2.lE-S 2 .11'- 6 2.lE-5 63 630 AV IE-7 10-7 2.4E-8 1.6E-8 .01 .16 Ollier II-5 11 - 5 11- 5 1[-5 200 200 2. 5L 'i/Yr 4.4E-5/Yr 1.4L-5/Yr .t.2t S/Yr 260 R/Yr 84P .-

• A TW us9tions as to3E&7R dose consequences by accident class (reference if1COR 111.1 f or Peat.h Bottom).

vent 4E+5R ATWS 7t*7R All others IE+7R bl33

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

. . .

.

Preliminary FREQUENCY of DECLARING GENERAL EMERGENCY 9 E-4 Once /1000 yr General Emergency Declared Because of EAL's Associated With Precursort, to '!enting But No Venting Occurs Because of Successful Recovery l

!

?

b

& 1

'

8-e E-4 Once /3000 yr Containment

'

Venting Wita j

Steam Environment '

r 2 E-6 Once /500,000 yr \

Containment Venting With Steam and Noble Gas Environment

-__ -

.

--

_ .. .

.

, T -

PreliminarV -

TRANSIENT TIME FRAMES for

.

SEQUENCES LEADING to VENTING

-

~ Containment Station Blackout __ LOCA with

,

Heat Removal [SBO) Iniection Fai'ure

(TW) IA)Q Core Uncovered 3 hr ~0hr.

Vessel Failure 5.3 hr 1 hr.

i Vent Pressure Pressure 3 2 h hr 2 0 h r.

'

Ultimate 53 h h hr.

.

,

.

!

!

u_______________._____.._______._________ _ _ _________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

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

.

.

Venting under the plant conditions specified in the EOPs offers the following advantages over permitting containment failure:

Slower energy release into areas where vent paths are directed minimizing environmental effects (Direct Torus Vent avoids steam release to reactor building altogether)

Containment pressure controllimiting NPSH concerns

>

Controlled release rates to the environment retaining containment atmosphere to maximum extent practical and permitting termination once repair is effectiive.

,

- Maximize fission product scrubbing

,

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

. .

8 .

.

Competing Risks associated with containment venting:

-Earlier releases provide less time for repair activities however,

-Time available between venting presssure and contain:nent i s not as large as that already available to effect repair Increased likelyhood of recovery during this period is small.

'

-Ability of plant personnel to continue repairs in vicinity of containment once containment exceeds design pressure or core damage has occured is small,.

Earlier releases provide less time for Emergency Plan implementation Declaration of General Emergency and recommendations for protective actions are expected early in events which may lead to containment venting i

J

!

i

-

.

-

EFFECT of SEP MODIFICATIONS I

High Rx SBO Low Rx Hi Con ATWS S. A. Con Pres Pres Press SEP improvements TOUX _TQUV TWQUY Response EOP's X X X X X

_

Direct Torus Vent X Fire Protection Sy X X X

_

X 3rd Diesel Gen X Backup N2 Supply X Containment Spray X X X ADS Logic X X RCIC Turbine setpt X ATWSMod Enriched Boron X

RPT Reliability y Feedwater Trip X

TRACG X

!

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

.. ..-

.

50.59 Considerations Venting has been approved undar previous versions of the E0Ps. The direct terus vent is initiated by procedure under conditions specified by the E0Ps and therefore no new accident is created by the installation or use of the direct torus ven A rupture disk set as low as 2.5psig (Group II isolation setpoint) is sufficient to assure that the direct torus vent will have no effect on the l probability of occurrence or consequence of previously analyzed design l basis event In addition, tnat the outboard valve is sealed closed '

provides additional assurance that no previously analyzed event is affecte The outboard valve for the direct torus vent meets the definition of a sealed closed valve and therefore no changes to the Pilgrim Technical Specifications are require Direct Torus Vent Description The direct torus vent is hard piped to the stack bypassing the the ductwork of the St'.ndby Gas Treatment System. The oneumatic supply to the valves is nitrogen thrn" 9C operated solenoid valve To actuate ..e system Symptoms as presented in Rev 4 of the BWROG EPGs must be present:

- Containment pressure in excess of that expectad for any design basis event

- Substantial hydrogen generation during a period when the containment is deinerted Operator must take multiple deliberate actions

- Jumper isolation signal to inboard valve (A050428)

- Install fuses in circuitry and acute keylock switch for !

outboard valve (A05025) l EPG Accident Management Philosophies l Provide guidance to the operator as to the appropriate actions for any mechanically possible sequence of events regardless of likelihood or j whether or not it is a part of the design basis i Provide a graded approach to protection of the core, containment.and plant equipment as log as possible maximizing time for operator action and repair activities Provide guidance for the purpose of core damage mitigation S-470013-068 1 LTR88A l

l i

ir

' -

.c ..

.

Emergency Procedure Guidance with respect to Containment Venting Venting for containment pressure control requires that Main condenser must be out of service Suppression pool cooling be unavailable or insufficient Drywell and wetwell sprays must be unavailable or inadequate

Primary containment pressure must be greater than that expected for

,

any design basis event As a result, venting occurs only when other means of containment heat removal are inadequate and the structural integrity of containment is threatene Venting is initiated at this point to preclude more serious failures which are expected to occur if the containment is permitted to fai As an example, venting under the conditions specified in the Emergency Procedures prevents core damage by precluding failure of the containment into the reactor building preserving the operabili ty of core cooling equipment located in these area Potential Negative Effects of Venting Pump NPSH reduction Subsequent deinertion Inability to close vent paths Degradation of reactor building environment Inadvertent venting Premature venting It should be noted that for the most part these negative effects are also associated with containment failure and that an important aspect of venting is minimizing these effects to the maximum extent practica The only significant impact of the hard piped vent over other venting systems is in its ability to preclude degradation of the reactor building environment altogethe Inadvertent actuation of the the vent is is minimized by the deliberate multiple actions required for initiation and by the rupture dis S-470013-068 2 LTR88A l l

!

I

, _ _

l

- .. .

..- .

l

.

Quantitative Evaluation of Venting

'

The preliminary results of the Pilgrim IPE were modified to evaluate the competing risks associated with containment venting. The general structure of the IPE event trees is shown and contains the following functional headings:

Reactivity control (reactor trip)

Reactor pressure control (safety valves)

Reactor inventory control (high and low pressure injection)

Containment control (containment temperature and pressure control)

Inventory makeup (reactor injection following containment failure)

Containment venting plays a role in the quantification of the containment pressure control heading. Other systems included in this heading are the main condenser, RHR, and containment sprays. The principal effect of venting is to reduce the frequency of sequences associated with containment pressure control failure (sequences TWQUV & TQUWQUV on the event tree diagram).

A secondary effect of containment venting is to mitigate offsite releases following core damage and coincidental failure of containment heat removal system The event tree diagram was modified to include a branch at the containment heat removal heading for core damage sequences TQUX and TQUV, In this way the IPE models could be used to evaluate not only the effect of containment venting on core damage, but on the potential for significant releases as wel In this way, these negative effects are small as compared to the beneficial effects of venting in preserving adequate core coolin S-470013-068 3 LTR88A l

--

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

.. ..-

.

Overall effect of venting on core damage frequency

'

The primary purpose of venting as an accident management strategy is to protect core cooling equipment from the effects of uncontrolled failure of the containment into the reactor buildin In this manner, venting can preserve core cooling during sequences in which containment pressure control systems are unavailable for an extended perio To demonstrate this benefit graphically, a summary of the effects of venting on the Pilgrim core damage probability as estimated by the preliminary results of the IPE was presented in pie chart for Without the capability to vent, containment heat removal failure sequences make up nearly half of the overall core damage probability. By permitting venting under the conditions specified by the Pilgrim Emergency Procedures, the frequency of core damage associated ,

with containment heat removal failure sequences drops by an order of magnitude l and the overall core damage probability is reduced by approximately 40%.

]

In this regard, containment venting as specified by the Emergency Procedures has a largc beneficial effect on plant risk by effectively eliminating containment heat removal failure as an accident clas Quantitative evaluation of competing risks A more detailed quantitative breakdown of the competing risks associated with venting was presented in the form of Table 1 (Pilgrim Direct Torus Vent Sensitivity Study). Again, the preliminary results of the Pilgrim IPE were used as a basis for this analysis. The analysis presents information essociated with the expected frequency of venting as well as the change in risk with and without the ability to ven The analysis includes comparisons of competing risks not only for the purpose of preventing core damage, but evaluates the consequences of venting during post core damage conditions as wel The following outlines several assumptions that were made which are important to to the outcome of the analysis. For the purpose of this analysis, conservative assumptions are defined as those which enhance the benefits of postponing or prohibiting containment venting, potentially noncenservative assumpticos are those which favor venting. A discussion of the effects of the more important of these assumptions is presented as the results are examine Potentially Conservative Assumptions

- Repair and recovery activities are assumed to occur even after containment pressure rises above design or core damage occurs (in fact for personnel safety reasons, these activities may be terminated under these conditions whether or not venting is initiated)

- Credit is taken for use of core cooling systems located external to the reactor building following containment failure (this minimized the importance of the vent during containment heat ,

'

removal failure sequences)

S-470013-068 4 LTR88A l

.

.. ..

- A relatively high failure rate for the vent is assumed minimizing

• its effectiveness (this failure rate, .1/ demand, is based on i subjective evaluations which include reluctance of the operator I to initiate venting because of steam and radionuclide release to I the reactor building and environment or due to pressures from I external sources such as the NRC and emergency response l organizations) I

- Consequence analyses assume complete depressurization of the containment on actuation of the vent, taking no credit for maintaining containment pressure with the vent (in accordance with Emergency Procedures) or terminating the vent cnce repair is effectiv Potentially nonconservative assumptions

- Little credit for repair of the main condenser is taken for l sequences in which the main condenser fails randomly or is the !

initiating event (these sequences amount to 1/3 to 2/3 of containment heat removal failure sequences)

The upper half of Table 1 presents the expected frequency of containment venting for preventive (precluding core damage) and mitigative (post core damage)

reason The results of the analysis indicate that venting for either reason is a rare event (on the order of 1/3000yr) and that venting under conditions in I which significant radionuclides would be in the containment is extremely rare ]

(less than 1/500,000yr). Mitigative venting is presented for the two purposes outlined in the Emergency Procedures, containment pressure control and combustible gas control. The frequency of venting for pressure control purposes is assumed to require failure of containment heat removal systems in addition to those system failures which lead to inadequate core cooling. For this reas )n, these frequencies are less than the frequency of core damage. Venting for hydrogen control purposes would be initiated only if core damage occurred coincidentally with the containment being deinerte These frequencies therefore reflect the relatively small amount of time that the plant operates with the containment deinerted (approximately 1%). While the actual frequency of venting might be a factor of three or more less than suggested by this 1 analysis (due to assumptions regarding recovery of the main condenser), these frequencies indicate that principal purpose of venting is for preventative !

reasons under conditions in which little or no core damage had occurre The bottom half of Table 1 presents a comparison of competing risks with and l without containment venting as an accident management strategy. These risks I include comparisons of core damage probability, the potential for significant releases from the containment and dose consequence The core damage probabilities are the same as those presented in the pie chart Containment heat removal sequenc'es (those labeled TWQUV) are the only sequences in which core damage probability is affected by containment venting. This is because other accident classes consider core damage for reasons other than S-470013-068 5 LTR88A

. ..

containment heat removal failyre cnd generally occur prior to containment overpressure due to decay heat generation. The likelihood of inadequate core cooling for these sequences is therefore uruffected by the presence or absence of containment pressure control systems. The core damage probability for containment heat removal failure sequences is substantially less than the frequency of venting for several reason First, containment failure is assumed to occur at tl.e ultimate strength of the untainment which is approximately twice the containment design pressur As it takes longer to raise containment pressure to its ultimate capacity, there is more time for repair activities and hence a higher likelihood of recovering failed heat removal systems. The additional time for repair accounts for approximately a factor of two reduction in the core damage frequency as compared to the venting frequency. Second, it is assumed that even though the containment may fail due to loss of heat removal, there is a possibility that systems outside the reactor building will cor.tinue to be available to provide core cooling. Thest systems include the feedwater and fire systems and account for the majority of the difference between the venting frecuency and the probability of core damage due to containment overpressure failur If venting is successful, the assumption is made that containment failure and core damage can be avoided altogethe Therefore, the frequency of containment heat removal failure sequences shown in the column in which venting is permitted is not the core damage probability associated with venting but represents those sequences in which venting failed or was not initiated. The it is a factor of 10 less than the frsquency of core damage in the no venting column reflects the assumption that the operator will be able to initiate venting successfully with a 90% likelihood of succes As noted in the discussion associated with the pie charts, the implementation of a venting strategy has a relatively sienificant effect on the reduction in core damage probability (approximately 40d by effectively removing containment heat removal failure sequences as a risk contributo The center column in the bottom half of Table 1 represents the effects of venting on the potential for significant releases following an accident. For the purposes of this evaluation, the term significant release is defined as the release of fission products from the containment in the form of particles, volatiles or noble gases generated as a result of a core damage event. Given this definition, a significant release is assumed under any of three conditions; containment failure resulting in core damage (such as containment heat removal failure events, TWQUV), containment failure occurs follcwing core damage or ventingSBO, TQUX, is initiated TQUV andfollowing coreclasses AV accident damag)e (the this

. With latterdefinition, two represented by the the risks associated with venting earlier than if the containment were permitted to fail can be weighed against taking advantage of additional time to effect repair and '

recover A simple model was applied to these accident sequences to evaluate I the potential for repai The model is exponential, assumes a mean time to l recovery of 19 hours2.199074e-4 days <br />0.00528 hours <br />3.141534e-5 weeks <br />7.2295e-6 months <br /> and is similar to models derived in WASH-1400 for recovery of mechanical equipment. As expected, the effect of taking advantage of additional time for repair is a reduction in the potential for releases during some sequences. (It should be noted that one of the assumptions made in this analysis is that repair activities can continue even following events in which core damage has occurre A more realistic assumption would be that much of this repair activity would have to be terminated for personnel safety reasoas S-470013-068 6 LTR88A

__

l ' ..

l l

l l which would limit the difference between the vent and no-vent frequency for the

. pctential of significant releases.) llowever, the accident class which reflects an improvement from implementation of venting is once again that which represents the containment heat removal failure scenarios which benefit by effectively elininating the potential for core damage if venting is successfully initiated, As a result, while the frequency of release for some accident classes may improve by postponement of venting, the overall net effect of allowing venting to occur is a reduction in the poter.tial for significant releases by preventing core damage during scenarios in which containment heat removal equipment failures might lead to containment failur The last column in Table 1 compares the risks associated with venting in terms of offsite consequence . For the purpose of this analysis, offsite dose was chosen as a measure of these consequence A simple frequency times consequences analysis was performed to derive the values presented in this section of the table. The frequency of occurrence used was that derived in the potential for significant release section of the tabl The dose consequences for each accident class are shown in the footnote in the lower left corner of the table. It should be noted that site specific dose consequences for tne various accident classes have not been derived for Pilgrim. As a result, the values shown in Table 1 were borrowed from work performed as a part of IDCOR activities atsociated with resolving the Severe Accident Policy. While not derived for the Pilgrim Plant specifically, the relative difference in the dose for each of the accident classes should still provide a reasonable measure of the competing risks associated with venting. For this analysis, venting is assumed to occur through the suppression pool. For sequences in which the containment is challenged but not vented, containment failure is assumed to occur in the drywell . Sequences in which containment failure leads to core damage TWQUV events) have a higher dose consequence than other sequences because (core damage occurs within a containment which is already assu failed. Comparing the consequences of permitting containment to fail as opposed I to releasing through a vent path, most accident classes show a reduction in offsite consequences if venting is initiated to preclude containment failur This is a result of the scrubbing which occurs through the containment vent path .

!

(which reduces rsleases to noble gases if it occurs through the suppression pool I as preferentially directed by Emergem:y Procedures). The beneficial effects of scrubbing more than offset the perceived benefits of postponing releases in favor of repair activities for post core damage venting sequence More importantly, once again, is the effect of venting on containment heat removal failure sequences. Venting effectively eliminates theses sequences as an accident class providing a significant reduction in the potential for offsite consequence ;

Several assumptions were iilade in the analysis which may have an effect on the conclusions drawr. as a result of the study. Among these assumptians were repair and recovery model characteristics, and postulated locations for containment failure and the vent path. Sensitivity studies were performec to determine the i

!

significance of these assumptions on the results. With respect to recovery models, it was determined that a mean time to recovery ~as short as several hours was required to balance the risks associated with venting early during post (. ore damage events with the benefits derived by scrubbing. This unrealisticly short S-470013-068 7 LTR88A

. . . . - _ - ..

.

.. < .,- .

l l

time for repair of major mechanical systems suggests that the results are !

relatively insensitive to assumptions made with respect to recovery models. A sensitivity analysis was also performed on the magnitude of reduction in offsite dose derived by venting. It was determined that dose reductions as small as a 1 factor of two still resulted in a reduction in offsite consequences during post l core damage events over letting the containment fai Combined with the l elimination of core damage sequences associated with containment heat removal failure, this suggests that even if the location of the containment failure happens to occur from an area such as the suppression pool airspace or if !

venting is initiated by paths such as from the drywell, it is still appropriate !

to vent as opposed to permitting containment failur Effect of Venting on Emergency Response  :

Declaration of a General Emergency is expected early in an event in which venting might be initiated. This is a result of the Emergency Action Levels associated with the Pilgrim Emergency Plan implementing procedure Symptoms associated with these EALs include a loss of the ultimate heat sink and

,

'

anticipation of containment pressure rising to above design pressure. Either of these symptoms would lead to a General Emergency whether or not offsite dose projections were in excess of protective action guideline From the Pilgrim venting sensitivity analysis, the frequency of implementing the Emergency Plan was estimated for sequences in which venting might be initiate These estimates indicated that a General Emergency would be declared on the order of 1/1000yr, that during a large fraction of these sequences recovery of containment heat removal equipment would preclude the need to vent (venting would occur on the order of only 1/3000yr) and that only the smallest fraction of those sequences would involve venting with fission products in the containment atmosphere as a result of a core damage situation (approximately 1/500,000yr) .

i

!

, ,

S-470013-068 8 LTR88A

,

l

_ __ . __ _

.,

es . 9 .

.

Conclusions The concept of venting is an accident management strategy available to a plant operating staff to protect the core and the containment under circumstences in which the structural integrity of the containment is threatened due to overpressure or the presence of combustible gase Venting is a rare event requiring the occurrence of multiple failures coincident with an accident or transient and, as a result, is not expecteo to occur over the lifetime of any given plant. Implemented in accordance with the guidance provided in the BWR Owners Group EPGs, venting will occur only after other means of controlling conditions within the containment have been unsuccessful and and failure of the containment effectively appears to be inevitabl Venting in these circumstances offers the following advantages over permitting the containment to fail, o Venting permits a slower energy release into areas in which vent paths are directed. This permits milder environments in areas such as the reactor building providing more assurance that equipment located in these areas will continue to operate then if the containment were permitted to fail in an uncontrolled manne Venting with the direct torus vent precludes releases to the reactor building altogether, o Vencing allows the operator to control containment pressure. When performed in accordance with the instructions of the EPGs, use of the containment vent allows the operator to maintain pressure below the prima ry containment pressure limit. This is as opposed to the potential for uncontrolled depressurization of the containment if containment failure were to be permitted. By controlling containment pressure, the operator can minimize NPSH problems to the maximum degree that is practical providing more assurance that core cooling systems which are taking suction from the suppression pool can continue to operate, o Venting permits controlled releases to the environment. Because the operator is able to maintain containment pressure through the use of venting equipment, release rates to the environment will be slower and over a longer duration than might occur if uncontrolled containment failure were permitted. This allows the retention of as much of the containment atmosphere as possible for as long as possible prior to its release. Controlling releases thrcugh venting equipment also has the advantage of being able to terminate releases altogether once repair activities are effective in returning failed containment control equipment to servic o Venting maxinizes the amount of fission product scrubbing that occurs along release paths. Venting hardware includes equipment associated systems such as containment purge and vent and the atmospheric control sy:tems.- These paths are fairly restrictive and provide additional surfaces on which filtering and plateout of vol.atile and nonvolatile fission products can occur. This effectively reduces the size of the

,

!

radionuclide release which might occur if the containment were ;

i

l S-470013-068 9 -

LTR88A -

l

.. a.,- .

permitted to fail uncontrolled from an undefined locatio When performed in accordance with Emergency Procedure Guidelines, venting is preferentially performed through the suppression pool. Controlling

the release path in this manner provides a substantial degree of scrubbing, effectively limiting releases to noble gase While venting as opposed to permitting the containment to fail appears to have a number of advantages, it is recognized that there are competing risks associated with venting as an accident mitigation tool. One element of containment venting criteria as it is directed in the Emergency Procedures is that it be initiated ;

prior to reaching the ultimate strength of the containment. Because there is !

margin between the criteria at which procedures suggest that venting be initiated and the ultimate capacity of the containment, there exists a limited '

number of scenarios in which repair activities may have resulted in recovery of )

failed systems and equipment after venting was initiated, but prior to the time ;

that containment failure would have occurred. In this regard, there is a small likelihood that had venting not been initiated, releases would have been avoided and any fission products that were released to the containment during the course of the accident would have been retained. To some extent then, it would appear advantageous to postpone venting utilizing the additional time for repair of l containment heat removal equipment or emergency plannin l

o Postponing containment venting to take advantage of additional time j for repair activities may not enhance the likelihood of successfully I recovering necessary equipment. This is particularly true if the systems under repair are located in areas in the vicinity of containment such as the reactor building. Conditions in these areas associated with the accident or as the contain.nent exceeded design conditions would not necessarily permit the operating staff to occupy these areas for personnel safety reasons, limiting time for repair and recovery whether or not venting was initiate In addition, analysis of these scenarios indicates that the amount of time which is gained by postponing venting is less than that which is already available to perform repair activities prior to reaching the venting criteri The increase in the likelihood that equipment repair will be successful in the time frame subsequent to reaching venting criteria is considered to be small given that repairs were l unsuccessful up to this point. This suggests that the benefits of postponing venting are limited when compared to the advantages I outlined abov I o With respect to Emergency Plan activation it is likely that conditions leading to the need to vent the containment will result in declaration of a General Emergency early in the event. This is a result of Emergency Plan implementing procedures which have been developed in accordance with NUREG-065 These procedures require declaration of an emergency based on certain symptoms and combinations of equipment failures which would occur during a transient in which venting might ultimately be required. Such events include LOCAs with ECCS failure, LOCAs with unsuccessful containment perfonnance which could threaten S-470013-068 10 LTR88A . _ , -

_- - - _ _ _ _

~<e.,. ,  !

ECCS functions and' loss of heat removal capabilities following shutdown. Because of the structure of Emergency Plan procedures in .

this regard, notification of the public and recomendations for l protective action are expected early in an event in which venting l might be initiated, i l

In summary, containment venting is an accident management strategy intended to l prevent or minimize the consequences of transient and 3ccident events in which '

containment conditions are approaching limits at which failure is anticipate Under these rare circumstances, venting provides an additional opportunity for the operating staff to intervene in the course of the accident, control the location and rate at which any releases might occur, and terminate those releases once repair a'.tivities are effecte The advantages associated with venting in this manner outweigh the potential disadvantages of releasing the containment atmosphere in a time frame slightly earlier than if the containment were permitted to fai While not expected to be required over the lifetime of of the plant, venting is the appropriate course of action under conditions specified by the Emergency Operating Procedures to reduce the potential for core damage and to minimize offsite consequences. Venting can be initiated by several means at the rHigrim Plant. Venting from the suppression pool with the direct torus vent mimizes scrubbing and limits any offsite releases to the maximum extent practica Use of the hard piped vent also limits uncertainties assuciated with releases of steam to areas outside containment associated with other vcnt path l I

. .

l S-470013-068 11 LTR88A l

- - -- - - - - -

A

• , 4 .08 g e .

We - - , - at

.

t f

,

,

l l

l I

i i

I l

i I

l

,

, .

l 8 l :I 'i J C I J.'l',l [031

_ . _ , , . , - - .

, f[U,ja d*bd/ tiOddd l

J