ML20045C002
| ML20045C002 | |
| Person / Time | |
|---|---|
| Site: | River Bend  |
| Issue date: | 06/14/1993 |
| From: | Booker J, Lorfing D GULF STATES UTILITIES CO. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| LER-93-003, LER-93-3, RBG-38627, NUDOCS 9306210375 | |
| Download: ML20045C002 (14) | |
Text
. ._.. . - . - . - _ , -
9 4
GULF STATES UTILITIES COM.PANY PMR h! %D 5IABCN POST OfiIG IVJK 720 57. FPANCISVitif LOUf 51ANA 70775 ARE A CODE. 504 635 SOM 346 SES)
June 14, 1993 RBG- 38627 File Nos. G9.5, G9.25.1.3 U.S. Nuclear Regulatory Commission Document Control Desk Washington, D.C. 20555 Gentlemen:
River Bend Station - Unit 1 ,
Docket No. 50-458 Please find enclosed Supplement I to Licemee Event Report No.93-003 for '
River Bend Station - Unit 1. This supplement a submitted to provide the results of GSU's additionalinvestigation of this event.
Sincerely, JdWMYLO 9%-.
J. E. Booker Manager - Safety Assessment and Quality Verification River Bend Nuclear Group EkoRhuhfee LAE/JPS/ H/FRC/MA$/DCH/JJ / L /kym
%10031 o
9306210375 930614 PDR ADOCK 05000458 E
n ;l s PDR j t
t cc: U.S. Nuclear Regulatory Commission 611 Ryan Plaza Drive, Suite 400 -
Arlington, TX 76011-NRC Resident Inspector P.O. Box 1051 St. Francisville, LA 70775 ,
INPO Records Center 1100 Circle 75 Parkway Atlanta, GA 30339-3064 Mr. C.R. Oberg Public Utility Commission of Texas 7800 Shoal Creek Blvd., Suite 400 North Austin, TX 78757 Depanment of Environmental Quality Radiation Protection Division P.O. Box 82135 Baton Rouge, LA 70884-2135 A'ITN: Administator P
i
NRC FORM '
366 U.S. NUCLEAR REGULATORY COMMISSION APPROVED BY OMB NO. 3150-0104 (5 92) EXPlRES 5/31/95 l ES1! MATED BURDEN PER RESPONSE TO COMPLY WITH THIS l LICENSEE EVENT REPORT (LER) $2"M" o'E"ES"UnD#ELJSOENrM"N2 AND RECORDS MANAGEMENT BRANCH (MNBB 7714 U S. NUCLEAR HEGULATORY COWISSION, WASHINGTON DC 20555 0001. AND TO THE PAPERWORK REDUCTION PROJECT (3150-0104), OFFICE Or (See reverse for required number of di0its/ characters for eacn block) MANAcevENT AND suDoET, WASHINGTON, DC 20503.
f ACluTV NAME (1) DOCKET NUMBER (2)
RIVER BEND STATION 05000 458 PAGE 1 OF L (3) 1 j i
7"'5 m DEFLATION OF REACTOR AIRLOCK DOOR SEALS COINCIDENT WITH OUTER AIRLOCK l DOOR IMPROPERLY SEATED IN ITS FRAME EVENT DATE (5) LER NUMBER (6' REPORT NUMBER (7) OTHER FACILITIES INVOLVED (8)
MOtCH DAV YEAR YE AA MON'M DAY YEAR NUMBER NUM3ER 05000 F ACiUTY NAMi: DOCKET NUMBER 02 25 93 93 ~ 003 01 OG 14 93 05000 OPERATING THIS REPORT IS SUBMITTED PURSUANT TO THE REQUIREMENTS OF 10 CFR 1: (Check one or moren (11)
MODE (9) 1 20 402(b) 20.405(c) 50.73(a)(2)(iv) 73.71(b)
POWER 20 405(a)(1)(i) 50.36(c)(1) X 50.73(a)(2)(v) 73.71(c)
LEVEL (10) 100 20 405(a)(1)(a) 50.36(c)(2) 50.73(a)(2)(vii) OTHER 20 405(aH1)(iii) 50.73(a)(2)(i) 50 73(a)(2)(vin)(A) PP84 'n Absuet 20 405(a)(1)(rv) 50.73(a)(2)(u) 50.73(a)(2)(vin)(B) ,sf,4f 20 405(a)(1)(v) 50 73(a)(2)(ui) 50.73(a)(2)(x)
LICENSEE CONTACT FOR THIS LER (12)
NAMk IEL{ PHONE NUMHi H (incknie Ares Goce)
D.N. LORFING, SUPERVISOR - NUCLEAR LICENSING (504) 318-4157 COMPLETE ONE LINE FOR EACH COMPONENT FAILURE DESCRIBED IN THIS REPORT (13)
CAUSE SvFLM COMPONE NT MANUF ACTURER O P , CAUSE SYSTE M COMPONE NT MANUFACTURER SUPPLEMENTAL REPORT EXPECTED (14) EXPECTED M*H DAf RAR vts SUBMISSION y yn. ceMet* t WEC'EO SUBVMiDN DATE)
DATE (15)
ABSTRACT (Limit to 1400 spaces, i e., approximately 15 single-spaced typewntien lines) (16)
On February 25,1993, it was discovered that two incidents occurred which apparently caused the interlock mechanism in the upper containment airlock at elevation 171' to operate improperly. In one incident the seals on one airlock door were partially deflated while the other airlock door was not completely closed. In the other incident, the reactor building door staned to equalize while the outer door was not secured and sealed. Both of these cases constituted breaches of containment.
Therefore, this repon is submitted pursuant to 10CFR50.73(a)(2)(v) as a condition that alone could have prevented the fulfillment of a safety function.
Upon identifying the first incident, STP-057-0401 " Primary Containment Airlock Door Interlock Test" was performed to verify operation of the mechanical interlocks. A maintenance work order was genemted to inspect the interlock malfunction following the second incident. A prompt modification request has been completed so that the handwheel lock solenoid energizes directly from the airlock door open/close limit switches during manual opemtion of the door, with power supplied to the airlock. This will properly interlock the doors to prevent misoperation from causing a containment breach.
ve rOru m m ea l
+
REQUlftED NUMBER OF DIGITS / CHARACTERS FOR EACH BLOCK '
BLOCK NUMBER OF NUMBER TITI E DIGITS / CHARACTERS 1 UP TO 46 FACIUTY NAME 2 DOCKET NUMBER 3 IN ADDITION TO O5000 3 VARIES PAGE NUMBER ,
4 UP TO 76 TITLE ,
6 5
g pER BLOCK WEM DATE 7 TOTAL 2 FOR YEAR 6 LER NUMBER 3 FOR SEQUENTML NUMBER 2 FOR REVISION NUMBER 7 REPORT DATE 2 PER BL CK UP TO 18 - F ACILITY NAME 8
8 TOTAL - DOCKET NUMBER OTHER FACMES MOMD ;
3 IN ADDITION TO 05000 9 1 OPERATING MODE 10 3 POWER LEVEL
" CHECK BOX THAT APPUES REQUIREMENTS OF 10 CFR 'I 12 TO 50 M NAME '
LICENSEE CONTACT 14 FOR TELEPHONE CAUSE VARIES 2 FOR SYSTEM I 13 4 FOR COMPONENT EACH COMPONENT FAILURE 4 FOR MANUFACTURER ,
NPRDS VARIES
'4 CHECK E3X THAT APPLIES S EMAL REN EMCTED 15 D SUBWSSION DATE 2 PER BLOCK
~
A
,ca - , ym p-- g- y-- +p. ,-
. l NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSION APPROVED BY OMB NO. 3150-0104 m s2s EXPIRES 5/31/95 ESTiuATED BURDEN PER RESPONSE TO COMPLY WTH TH'S LICENSEE EVENT REPORT (LER) CiffS% foEERo"EEIJ?oO M E'8 Text cmmNUABON $$$"J8 ",'MQ"S"778[Ed"$l THE PAPERWOR< REDUCTION PRO.ECT (3'5401%. OFFICE OF MANAGEMENT AND BuoGET, ww ASHINGTON, DC 20SC3 7 ACILITY NAME p) DOCFET NUMBf R (2) LER NUMBER (S) PAGE p)
SEQUENTiAt EmiON NUMBEA NUVBER 05000 OF ;
RIVER BEND STATION 458 93 003 01 2 11 a m m ..c. m,y., .. .- m . ., wc ,- = o n .
REPORTED CONDITION On Febmary 25,1993, with the unit in Operational Condition 1 (Power Opemtion), it was discovered that two incidents occurred which apparently caused the interlock mechanism in the upper containment airlock (*AL*) (lJRB*DRAl) at elevation 171' to operate improperly In one incident the seals on one airlock door were partially deflated while the other airlock door was not ,
completely closed. In the other incident, the reactor door started to equalize while the outer door was not secured and sealed. Both of these cases constituted breaches of containment (*NG*).
Therefore, this repon is submitted pursuant to 10CFR50.73(a)(2)(v) as a condition that alone could have prevented the fulfillment of a safety function.
INVESTIGATION The first incident occurred at 1350 on February 25,1993. An individual entered the containment airlock from the outer door, auxiliary building side. He attempted to manually close the outer airlock door from inside the airkick. He rotated the handwheel towards the "close" position, unaware that the airlock door was not seated completely in its frame. Movement of the handwheel :
into the " seal" position caused the seals to inflate while the airlock door was still outside its frame.
Additionally, the door's latch pins were extended. Movement of the handwheel to the closed position released the mechanical interlock in the reactor door. However, due to the door's position, the latch pins were not in their keepers. The individual then proceeded to attempt to open the reactor ;
^
door. The reactor door handwheel was rotated towards the "open" position which initiated depressurization of the seals.
Having the seals deflated on the reactor door, while the outer door was slightly ajar, caused a momentary breach of the containment. The individual inside the airlock promptly closed and secunxi the reactor door upon being informed by individuals on the auxiliary building side of the <
airlock that the oute.r door was not sealed properly. He opened and then properly closed and secured the outer door. STP-0574)401 " Primary Containment Airlock Door interlock Test" was performed immediately after this incident to verify operation of the mechanical interlocks.
The second incident occurred at approximately 1940 on February 25,1993. An individual entered the upper containment airk)ck from the auxiliary building to check radiological postings within the airlock. As he entered the airlock, the outer door swung shut but the handwheel was not rotated to the closed position. The individual was in the airlock only momentarily when he heard equalizing air fmm the reactor door and noticed the reactor door handwheel moving as an individual was attempting to enter the airlock from the containment. This caused a momentary breach of the b
NRO FDAV St4A 15 9h
l l
l
. I NRC FCRM 366A, U.S. NUCLEAR REGULATORY CoMMISSloN APPROVED BY OMB No. 3150-0104 s et EXPlRES 5/31/95 NATSON NE . Sc 0 RS O A LICENSEE EVENT REPORT (LER) Couens ReGARowG sunceN EsTuATE To T E ivoaviTo..
TEXT CONTINUATION E'M5"^7us"[w!7.5"g$"
e ewe $,"sh'$
THE PAPERVIOR4 REDUCTON PH3)ECT (3'S>0104 OFFJCE OF MANAGEuENT AND 8JDGET, WASHINGTON, DC 20$03
~
DOC ET NUMBER W LIR NUMBEM (5) PAGE p)
F ACILi1Y NAME m LE QdNhAL RELSON NUMBER NUMBE R 05000 OF RIVER BEND STATION 458 93 - 003 01 3 11 im - w.s.,,,.yy ... ~ ..m u m m containment through the airlock equalizing valve. The person inside the containment physically stopped the movement of the handwheel from inside the airlock immediately after he heard the reactor door start to equalize. He then rotated the handwheel to the closed position. He secured the outer door by rotating the handwheel from the open to the closed position. The reactor door was ,
then opened to allow the other individual to enter the airlock and the control mom was immediately contacted. Maintenance work order No. R174728 was generated by the Shift Supervisor to inspect and correct the interlock malfunction following the second incident.
As identified in the vendor's operation and maintenance manual (3219.711-056-001), the two doors, one at each end of the airlock, are designed to be mechanically interlocked so that one door cannot be opened unless the other is completely closed and sealed. When the handwheel on one door is rotated to the open position, the mechanical interlock mechanism is designed to prevent opening of the opposite airlock door so that the containment cannot be breached during access through the airlock.
When opemting the doors manually, as identified in the vendor manual, special care must be taken by the individual when the door is in the open position. Operating instructions are posted on each airlock door identifying the proper operation of the doors in the manual and automatic mode.
In addition to the mechanical interlock, the airlock is designed with a solenoit activated handwheel locking mechanism. The purpose of the handwheel locking mechanism is tw. told. Under complete automatic operation, with power supplied to the airlock, the handle lock solucid mechanism functions as part of the differential pressure monitoring system. If a differential pressure of more than 0.5 psi exists across the airlock door, the handwheel locking mechanism is energized and for personnel safety prevents the continuation of the opening sequence until the pressure is equalized.
In addition to being used as a pan of the differential pressure monitoring system, under automatic conditions, the handwheel locking mechanism energizes when the door is opened. This prevents rotation of the handwheel while the door is in the open position. The handle locking mechanism that prevents the handwheel from rotating when the door leaves its frame, will not function when the door is operated in the manual mode (with control power off). Movement of the handwheel to the
" scal" position de-activates the mechanical interlocks. If this occurs with the door not fully closed, the seals will inflate, potentially damaging the seals and creating a personnel hazard. l The handwheel locking mechanism, which would have prevented the handwheel from rotating when the door was in the open position, was de-energized prior to the two incidents described above.
l l
ARC FOAV 36f.A 15 90 l
}
u.S. NUCLEAR REGutATORY COMMISSION APPROVED BY OMB Ho. 3150-0104 NRC FORM 366A.
ts sa EXPJRES 5/31/95 EST)MAUD BURDEN PER RESPONSE TO ColFLY W TH?S LICENSEE EVENT REPORT (LER) 'U"Es%SE'82E"nlEN $$.rUo7E us' C
Text cormfWABOf4 @@eM$*O'[Ts"*$"d$$39l THE PAPERWORK NEDUCTsON PRCLIECT (31 Soc 104), OFFCE OF MANAGE #ENT AND BJDGET.WASHtNGTON DC 20603 f ACILITY NAME II) DOCF ET NUMBER (2) LER NUMBER (4) PAQt 13)
$EQ4NT.AL RE=40N NUVBER NUMBER 05000 OF RIVER BEND STATION 458 93 003 01 4 11 un n .wa. w.a -.m n .,ewc w ns on Originally, the airlock doors were designed to operate by a pushbutton which actuated a sequencing system to automatically deflate / innate the door seals and hydraulically open/close the door. A review of modification requests (MRs) and condition reports (CRs) regarding the airlock interlocks indicates that since August 1985 repetitive problems have been occurring with the handle lock solenoid and the hydraulic operating system. The corrective actions implemented did not substantially reduce the airlock problems.
In August 1986, CR No. 86-1296 was generated to address a condition in the upper containment airlock (lJRB*DRAl) in which the outer airlock door was ajar with the seals inflated and the inner door was opened. This condition existed for approximately two minutes. As a result of this incident, MR No. 86-1438 was generated to relocate the door's open/close limit switch from the door hinge to above the door frame. Note that this limit switch controls the actuation of the handwheel locking mechanism. The purpose of this MR was to ensure the reliable operation of the handwheel locking mechanism. Modification request (MR) 86-1438 stated that the testing of the handwheel locking mechanism (electrical interlock) shall be included in STP-057-0401, " Primary Containment Door Interlock Test". The STP was revised in accordance with the MR; however, it indicated that the testing of the handwheel locking mechanism is not Technical Speci0 cation -
related and is only applicable when power is supplied to the airlock. The unreviewed safety question determination (USQD) generated for the MR indicated that the handwheel locking mechanism was not safety related. In addition, the USQD did not indicate that the handwheel locking mechanism was required for the safe and reliable operation of the airlock interlock system to ensure that the Technical Specincation requirements for the interlock are met. The 1986 review of the design requirements versus the licensir.g basis was inadequate.
In October 1990 four people were trapped inside the Upper Containment Airlock (lJRB*DRAl) after the airlock doors were locked up in the closed position and the handle lock solenoid started smoking and burning (Ref. CR 90-0934). As a result of this incident, a decision was made to disconnect the electrical power as well as the automatic hydraulic operating system and operate the doors manuaUy. The solenoid actinted handwheel locking mechanism which would have prevented the handwheel from rotating when the door was in the open position was de-energized when the electrical power was disconnected.
Repetitive operational and maintenance problems associated with the handle lock solenoid and hydraulic system led to the decision to disconnect the electrical power supply and operate the doors manually. The basis for this decision was as follows:
The design of the doors enabled them to be operated in the manual mode with the w ,Onv uA m
NRC FORPd 366A, U S. NUCLEAR REGULATORY COMMISSloN APPROVED BY oMB NO. 3150-0104 (5 er) . EXPIRES 5/31/95
'*[c"RuYToN N EEECTo"N"EEs{ 'sSo "n7 %$
LICENSEE EVENT REPORT (LER) Couutes AcGARom BunoEN estuATo To THc woRuAtoN TEXT CONTINUATION 7$"70iR*c "o^fu%"oM",S"gf"d'S'f05,"$'$
THE PAPERWORK REDUCTON PR3)ECT (3160-0104), OFFICE OF MANAGEMENT AND BdDGET, WASHINGTON, DC 20$03.
F ACiLITY N AME (t) DOCM ET NUMBER (2) LER NUMBE R (5) PAGE (3)
YEAR NUVBE R NUVBER 05000 OF RIVER BEND STATION 458 93 003 01 5 11 1LK1 pt more space os reewre1 use entiones copoes of NRC 1orm %%, (17p power on or off.
The airlock doors were signed to be mechanically interlocked so that one door could not be opened unless the other is completely closed and sealed. When one door is open, the mechanical interlock mechanism is designed to lock the opposite airlock door so that the containment cannot be breached during access through the airlock.
Regardless of the condition of the power supply to the airlock and the mode of opemtion the mechanical interlock system was designed to prevent breaching of the containment. The mechanical interlock relies on proper manual operation of the door.
The electrical equipment including the handwheel locking mechanism is non-safety related.
As pan of the disposition of CR 90-0934, the design requirements were reviewed against the licensing basis. The USAR indicated that both doors are mechanically latched and hydraulically swung. The USAR also stated that the doors are interlocked such that, in the event one door is open, the other cannot be actuated. It also stated that in the event of a power failure, it is possible to operate the airlocks manually.
Consistent with the design, the engineer interpreted the USAR to mean that the airlocks could be manually operated during nonnal plant operation and not just during a power failure. He believed that the USAR sections did not prohibit or restrict the 4 operation of the doors in a manual mode. However, NSAC 125 " Guidelines for 10 CFR 50.59 Safety Evaluations" indicates that changing the function of a stmeture, ;
system or component from automatic to manual operation could be considered as a ;
change to the facility as described in the Safety Analysis Repon. Therefore, the 10CFR50.59 evaluation (ISEE) for CR 90-0934 was not perfonned in accordance ,
with the guidance of NSAC 125 as identified in engineering procedu.e ENG-3-004, l
" Safety and Environmental Evaluations."
Based on the above the mechanical interlock alone was considered sufficient for the continued l operation of the airlock and maintaining containment integrity while the airlock was being used in either an automatic or manual mode.
I After the decision was made to eliminate the electrical power and operate the airlock doors manually l a modincation request was initiated to remove the components exclusively associated with automatic operation. Since the airlocks were designed to opente in the manual mode, the modification request was not considered to be essential for the continued operation of the airlocks in NRC FO8R4 366A (S 9a
NRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSION APPROVED BY OMB NO. 3150-0104 is ori EXPlRES 5/31/95 EST:MATD BURDEN PER RESPONSE TO COMPLY WITH TH'S LICENSEE EVENT REPORT (LER) 1"vTE%ef80s" roe!E'SEE oE,$R%o'?
TEXT CONTINUATION ""LM"J'e"gyE"'w$^*fa"g*D$ 75'A%"@y'l THE FAPERWORK PEDUCTION PRCk:ECT (3153 01 Ciel, OFFICE OF MANAGEMENT AND BJDGET, WASHINGTON, DC 20503.
F ACILITY NAME (1) DOCP ET NUMBER (2) EIR NUMBER (6) FAGE (3)
SEQJENhAL REv4 TON NUMBER NUVBE R 05000 OF RIVER BEND STATION 458 93 003 01 6 11 un w n ,v.=. . ,, a. .. .~ eon., avc n,~ m o n the manual mode, but rather was considered to be an enhancement to the design configuration of the airlocks. Themfore, when it appeared that the configuration of the plant should be changed, a timely design change was not implemented.
.A review of maintenance work orders (MWOs) indicates that since 1985 cormctive and preventative maintenance on the airlocks was not given adequate priority to restore them to their original design mode (i.e., automatic operation). It appears that maintenance was not provided to the extent necessary to preclude handle lock solenoid and hydraulic system problems. Therefore, adequate management attention to long standing equipment maintenance problems was not applied. Frequent discussions with the airlock vendor indicate that the airlock was designed for about 2000 cycles per year; however, the airlocks at RBS are cycled more than 2000 times in one month.
ROOT CAUSE The root cause of this event consists of five causal factors, as determined by a cause and effect task analysis:
- 1) The original design was inadequate for the level of service at RBS.
- 2) The 1986 review of the design requirements versus the licensing basis was inadequate.
- 3) The 10CFR50.59 evaluation (ISEE) for CR 90-0934 was not performed in accordance with the guidance of NSAC 125 as identified in engineering procedure ENG-3-004, " Safety and l Environmental Evaluations."
- 4) When it appeared that the configuration of the plant should be changed, a timely design change was not implemented. ,
- 5) Adequate management attention to long standing equipment maintenance problems was not l applied. 1 i
A review of previously submitted LERs revealed none similar; however, the event described above that occurred in August 1986 is similar to the occurrences in this report. As stated previously, CR 86-1296 was initiated in August 1986 to address a condition in the upper containment airlock (13RB*DRAl). In this event, the outer airlock door was ajar with the seals inflated and the inner door was opened. This condition existed for approximately two minutes. This event meets the criteria for reportability under 10CFR50.73.
I wC rwu w m l
NAC FORM 366A U.S. NUCLEAR REGULATORY COMMISSloN APPROVED BY OMB NO. 3150-0104 15 sa EXPIRES 5/31/95 (ST1 MATED BURDEN PER AESPONSE TO COMPLY WN TH$
LICENSEE EVENT REPORT (LER) *lNSEG'"*#EEnlE$$EoOJT2 c
TEXT CONTINUATION Q $ 5 ? $ ' "OMM E "'W! 7,$ a7 "a"C % e s ," s h'$
THE PAPERWORK REDUCTION PROJECT (3 5001fA), OFFaCE OF M ANGEMENT AND BUDGET, WASHINGTON. DC 20S03 F ACIUTY HAME p) DOCF ET NUMBLR (2) LER NUMBER 16) PAGE (3) bEGJLNT AL hlW5CN NUMBE R NUMBER 05000 CF RIVER BEND STATION 458 93 ~ 003 01 7 11 un m .L;.v.w.w..- - .w we w m on CORRECTIVE ACTION Upon identifying the first incident, STP-057-0401 " Primary Containment Airlock Door Interlock Test" was perfonned to verify operation of the mechanical interlocks. Maintenance Work Order !
No. R174728 was generated by the Shift Supervisor to inspect the interlock malfunction following the second incident. As a compensatory measure, GSU took actions consistent with the Action -
Statement of TS 3.6.1.4, " Primary Containment Airlocks."
Prompt modification request (PMR)93-009 has been completed. The handwheel lock solenoid has been modified to energize directly from the airlock door open/close limit switches during manual operation of the door, with control power supplied to the airlock. This will prohibit movement of the door handwheel when the door is outside its frame. This change ensures that the mechanical l interlocks are maintained even if an attempt is made to misoperate the equipment. In addition, prohibiting movement of the handwheel will ensure that the seals are not inflated when the door is open, which could result in personal injury and/or damage to the seals.
To ensure safe and proper operation of the airlock doors, detailed operating instructions were posted at each airlock door as requimd by PMR 93-0009. These instructions contain appropriate caution statements warning individuals of the potential personnel injury consequences of improper operation of the door.
A high perfonnance team has been organized to thoroughly review the airlock design. With assistance from the airlock vendor this team will perfonn a detailed design review of the River Bend airlocks. This design review will concentrate on the past and present airlock design, operational and maintenance problems. Following evaluation, necessary modifications of the airlocks will be initiated to enhance their perfonnance. These modifications will be implemented no later than the fifth refueling outage (RF-5).
Several corrective actions were identified during the enforcement conference with the NRC Region IV on April 21,1993 and in subsequent discussions with the NRC Senior Resident Inspector. A case study training class using the details of the airlock incidents has been developed to provide additional tmining and guidance in the perfonnance of 10CFR50.59 evaluations. This training has been provided to all appropriate personnel in the Engineering Department as well as numerous inidividuals in other depanments. Other individuals on site who are qualified to perform 10CFR50.59 evaluation reviews and those who review dispositions of condition reports will be provided vith this case study training as well. The expected completion date for this additional NRC FOAU %64 (S th
1 l
NRC FORM 366A U.S. NVCLEAR REGULATORY COMMISSION APPROVED BY OMB NO. 3150-0104 is or, EXPIRES 5/31/95 EST; MATED BJRDEN PER RESPONSE TO COMPL Y WTH THIS LICENSEE EVENT REPORT (LER) $$"T*S$"es'. S'a T RIEN $ $ 4# b E ! E 'SR
" B CH $B N TEXT CONTINUATION 7$$' $"S*c7^3$E?w NcT 0 THE PAPERWORK REDUCTION PROJECT (31540104 OFFICE OF MANAGEMENT AND BUDGE'T. WASHINGTON, DC 20503 FACluTY NAME 0) DOCKET NUMBER (2) LER NUMBER tel PAGE (3) bEQUENiaAL REWON NJMBER NUVBER 05000 OF
~
RIVER BEND STATION 458 93 ~ 003 01 8 11 rw m . .o.=. v.w.w. .m-n.ac m o n training is September 1,1993. In addition, clarification will be provided in the procedures related to 10CFR50.59 evaluation to reDect conclusions from the training session, Previous 10CFR50.59 evaluations will be reviewed for conditions similar to the airlock situation. ^
This review will encompass modification requests (MR) and condition repons (CR) that have been completed since 1985. If similar conditions are identified, the 10CFR50.59 evaluation will be reviewed in detail and, if necessary, re-performed using the guidance provided by the case study training. The initial reviews of the CRs and MRs are expected to be complete by July 31,1993.
Depending on the quantity of additional 10CFR50.59 evaluations required it is expected that they will also be completed by July 31, 1993. Included in this completion of additional 10CFR50.59 evaluations will be engineering correspondence that in the past has provided guidance to operations on various operating scenarios in the plant.
A review has been perfonned of all outstanding maintenance work orders (MWO) to identify any items of potential concem that should be reviewed under 10CFR50.59. Fourteen items were identified that require further review. These reviews will also be completed by the July 31,1993 date. Maintenance is developing an upgraded work control program that will incorporate this type of review for degraded conditions and potential safety implications that may require a 10CFR50.59 evaluation. The expected implementation date for the upgraded work control program is July 31, 1993.
In order to reduce the rate of usage of the airlocks, operations will develop a plan to reduce the entries to less than 1000 per month per airlock. This plan will be fully implemented by December 1,1993. Integral to this plan is the resolution of fire protection issues within the containment building. This will allow removal of firewatches from containment which represents almost half of the current entries. In order to ensure reliable functioning of the interlock system the Opemtions Depanment has also revised STP-057-0401 to include testing of the handle lock solenoid as pan of the Technical Specification required interlock system. This frequency of this STP has been administratively increased to once per week for a minimum of six months. In addition, the Maintenance Depanment has revised the monthly preventive maintenance task for the airlock door to include additional checks on the interlocic system. <
The description in the USAR of the operation of the airlocks was revised with PMR 93-0009.
Additional clarification of both the USAR and the Tech Specs will be provided with the modifications for the pennanent upgrade to the airlock system.
The drywell airlock has also been reworked to ensure that the handwheel locking solenoid and the N4'; FORM 366A (5 02)
t NRC FOPM 366A, U.S. NUCLEAR REGULATORY COMMISSION APPROVED BY OMB NO. 3150-0104 g EXPIRES 5/31/95 ESNATED BJRDEN PER RESPONSE TO COMPLY WiM TH13 LICENSEE EVENT REPORT (LER) E Z E E Ea" f 2 $">RoEN $ $ # a O u % %ol
" 0 TEXT CONTINtJATION $$$$70*c"'M,$["U"$"N3M*c'$Y,$,"$f$
THE PAPERWVORK REDUCTON PROJECT (3150-01M;, OFF6CE OF MANAGEMENT AND BUDGET. WASHINGTON. DC 20503.
F ACILITY NAME (y l DOCpf7 NUMBER (2) LER NUMBER (4 PAGE (3)
YEAR NUMBER NUMBER 05000 OF.
RIVER BEND STATION 458 93 - 003 01 9 11
,w en..w....w.v..m n n.w ma m <m mechanical interlock system are fully functional. This was completed during the current forced outage (FO 93-01).
On April 23,1993, the Plant Manager held a briefing on airlock issues for plant staff personnel, and '
other River Bend Nuclear Group personnel who hold at least a Supervisory position.
SAFETY ASSESSMENT Technical Specification 3.6.1.1, " Primary Containment Integrity - Operating" provides one hour to restore containment integrity. The three incidents described in this repon, two on February 25, 1993 and one in 1986, were momentary breaches due to seal deflation and thus were well within the one hour time limit.
GSU has performed a safety assessment of the two incidents that occurred on 2/25/93 in which the containment airlock on the 171' elevation was operated improperly.
In our analysis we have used a PRA approach as well as a mechanistic approach to assess the safety significance of the condition reponed. Both of these approaches are presented in detail below. In summary, the condition reported is not safety significant from either a PRA or a mechanistic basis.
PRA Approach Based on security computer records for the card reader at the airlock cage, there are approximately 40768 entries into the containment or exits from the containment per year via the airlock on the 171' level. We have conservatively assumed that it would take 45 seconds to open and close each door of the airlock. In other words, it takes 90 seconds to make one entry or exit through the 171' airlock.
We also assumed that there were only two mis-operations of the containment airlock in the last year.
This was determined by reviewing past condition reports for similar events. Consequently, there have been 2 mis-operations per 40768 entries or exits on a per year basis. As a result the probability of mis-operating an airlock door is:
2 mis-operations = 2.5E-5 (2 x 40768) door operations Therefore, if the frequency of a large LOCA is 1.0E-4 per year (or 3.17E-12 per second) then the I
I oC FoRv 3aA a m
MRC FORM 366A U.S. NUCLEAR REGULATORY COMMISSION APPROVED BY oMB NO. 3150-0104 (S g EXP!RES 5/31/95 ESTIMATED BJRDEN PER RESPONSE TO COMPLY WTH TH'S LICENSEE EVENT REPORT (LER) '7"TEs C u "REo'E$'ERitWIEAIElo O dn017 TEXT CONTINt)ATION $$$CE*S8cOu "*'u'$$"'w"*f* , 3$"*"e ". 'A$ S ."$^$
THE PAPERWORK REDUCTION PRCLIECT (3150-c100, OrrtCE OF W ANAGEMENT AND BUDGET. WASHINGTON, DC PD503 f ACluTY NAME 0) DOCfL ET NUMBE R (2) LER NUMBER (4) PAGE (3) 1.EQukNT4A6.. ftE VSON NUMBER NUMBER 05000 OF
~
RIVER BEND STATION 458 93 003 01 10 11 ww , ,w. ,o. . v.w.a. w. .am e nu-am o n probability of a large LOCA occurring when an airlock door is mis-operated is:
3.17E-12 x 4,5 sec x (2 x 4Dl68) door ops x 2,,5E ,5 = 2.9E-10 sec door ops year 1 year If we compare this to the NRC's safety goal for large releases (1.0E-6 per year), then it is clear that the probability of a large LOCA concurrent with the mis-operation of the airlock doors is 3000 times less than the safety goal. From a PRA perspective, this event is not safety significant.
Mechanistic Aparpach The containntent airlocks constitute potential annulus bypass leakage paths between primary and secondary containments and are included as such in Table 3.6.1.3-1 of the Technical Specifications.
Also included in this table are the 36" primary containment purge isolation valves, which provide analogous communication paths between primary and secondary containment.
According to Technical Specification 3.6.1,9, each of the 36" primary purge isolation valves may be opened for purge system operation. However, purge system operation is limited to a total of 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br /> per 365 days.
The rationale behind the 1000-hours-per-365-days limit is that the probability of a LOCA occurring any time during any one purge system operation is very small. (Note that this argument is similar to the PRA approach discussed above.) It is reasonable to use this rationale to cover any :
combination of annulus bypass pathways, since the net impact is the same. If it can be sijown that I during any 365-day period, the total number of hours during which the purge valves and the airlocks l were open comes to a total of less than or equal to 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br />, then we can consider that the intent :
of the Technical Specifications has been met with regard to open pathways between the primary and secondary containments during normal plant operation. ;
I To do this we compare the number of times either airlock door is open (N) to the number of times l they are allowed to be open. The allowed time either door can be opened is based upon the time the 36" purge valves are estimated opened for a given year. Based on the recorded times for purge valve openings in STP-000-001 from January 1,1992 through March 18,1993, the estimated open time is 784 hours0.00907 days <br />0.218 hours <br />0.0013 weeks <br />2.98312e-4 months <br /> per year.
The basis for the Technical Specification value of 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br /> is the NRC's calculated probability of a LOCA and the dose released. The dose released is related to the h;akage flownte and the length uRC rOav mA e n, I
f i
U.S. NUCLEAR REGULATORY COMMISSION APPROVED BY OMB NO. 3150-0104 NRC FORM 366A g ,,; EXPIRES 5/31/95 ESTIMATED BURDEN PER RESPONSE TO COMPLY W'TH TH'S LICENSEE EVENT REPORT (LER) *Z7JsiEoTsa"l"Ri!STLIElo7isFEll C
^ "
TEXT CONTINUATION Ev"E'e"'IM $$"'w'7N"E'd',S'Asi fy?$
T THE FAPERWORd. REDUCTION PRGJECT (3150-0104l, OFFCE OF MANAGEVENT AND BJDGET, WASHINGTON, DC 20503.
F ACILITY NAME (1) DOCKET NUMBER pp LIR NUMBER tal PAGE0)
LEQuENTA NV@ON NUMBER NUMBER 05000 OF RIVER BEND STATION 458 93 003 01 11 11 um, ,,,,, ,,.<. .. ,.w.w. .an~ .mm a- xs o n of time the leak occurs. In our panicular case that length of time is built into the number of times the airlock doors are opened a year.
To find the leak rate, calculation G13.18.14.0*129-0 was performed. It consisted of two parts, one ,
part determined the leakage through the 36" purge valves and the other part determined the leakage through deflated seals on one airlock door while the other door is opened. DBA-LOCA conditions were assumed in both parts of the calculation. For the purge valves, the DBA-LOCA flow rate was ,
found to be 2.54E5 scfm, or 14 times the airlock leakage rate of 1.77E4 scfm.
Since one airlock door operation is assumed to take 45 seconds, then 80 door operations are possible per hour. Also, since purge valves were estimated to be open for 784 hours0.00907 days <br />0.218 hours <br />0.0013 weeks <br />2.98312e-4 months <br /> per year, then the airlocks could leak for 216 hours0.0025 days <br />0.06 hours <br />3.571429e-4 weeks <br />8.2188e-5 months <br /> (1000 hrs - 784 hrs). Since the purge valve flow rate is 14 times greater than the airlock leak rate, then I hour with the purge valves open is equal to 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br /> of airlock door operations. Therefore, the equivalent time for airlock door operations is 3024 hours0.035 days <br />0.84 hours <br />0.005 weeks <br />0.00115 months <br /> (216 hours0.0025 days <br />0.06 hours <br />3.571429e-4 weeks <br />8.2188e-5 months <br /> x 14). Then, the number of airlock door operations could be as high as 241920 (3024 x ;
- 80) without exceeding the 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br /> of annulus bypass leakage presumed in Technical Specification 3.6.1.9.
Because the estimated number of airlock door operations (81536 per year) was 1/3 the allowed number of operations we were bounded by the Technical Specification limit of 1000 hours0.0116 days <br />0.278 hours <br />0.00165 weeks <br />3.805e-4 months <br /> for annulus bypass leakage. Therefore, the safety significance of this condition is low from a mechanistic perspective.
Note: Energy indastry identification codes are indicated in the text as (*XX*). !
l Nac soRu asA is 8:n
.