ML20059M858
| ML20059M858 | |
| Person / Time | |
|---|---|
| Site: | Diablo Canyon  |
| Issue date: | 08/18/1993 |
| From: | Rueger G PACIFIC GAS & ELECTRIC CO. |
| To: | |
| References | |
| OLA-2-I-MFP-034, OLA-2-I-MFP-34, NUDOCS 9311190387 | |
| Download: ML20059M858 (9) | |
Text
1
-. /Qfy) 'E3l~l~ WrinFT~ 3*
3: tit 9l% ko" Vesu$.,0 j Pacific Gas and Electric CampSny 77 Bae Sreet Gre;cn M Ruege l f, San Fran:.sco CA 94106 Sen:;r Mggem va l
- M'W h -]_'fg 373 -68 % j,fR,.y. i May 20, 1993 '93 0 48 '.35.28 ,
PG&E Letter No. DCL-93-125 ,;.
U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D.C. 20555 Re: Docket No. 50-275, OL-DPR-80 Diablo Canyon Unit 1 Licensee Event Report 1-92-031-00 Containment Ventilation Isolation Technical Specification 3.3.2 Not Met Due to Inadequate Procedure and Personnel Error Gentlemen:
Pursuant to 10 CFR 50.73(a)(2)(i)(B), PG&E is submitting the enclosed Licensee Event Report 1-92-031-00 regarding Techni:al Specification 3.3.2, Action 18 not being met when containment purges were performed while the containment isolation ventilation function was inoperable.
This event has in no way affected the health and safety of the public.
Sincerely, f
)&&V Gregary M. Rueger cc: Ann P. Hodgdon John B. Martin
._ Mary H; Miller - -
Sheri R. Peterson CPUC Diablo Distribution DCl-93-TP-N021 Enclosure l
l 6115S/85K/KAB/2246 NUCLEAR RCOULATORY COMYtSSION k + Nn 'gfj - h Of 5. No a w. ma jfEBcLMnYteialF/FcfRM C0 vn a w ti:n /
r mv.r . . . ,- . . _ _ .
w .c l
. M _ . _ _ _. K u G H.S _
I 9311190387 930818
~
g $ c ,u..rul)1Mkblbf I.~ *55 :ssi 0/? 0 S PDR ADOCK 05000275 b" - " "8 Re;, citer M4/5 ff/Cfl-- 3
LICENSEE EVENT REPORT (LER) b F AC8UTV NAME til OOCITY NUMPER (3) P AGE (31 3 ;
DIABLO CANYON UNIT 1 0l5l0l0l0! !7l5 1l'l 7 j mLE n CONTAINMENT VENTILATION ISOLATION TECHNICAL SPECIFICATION 3.3.2 NOT MET DUE TO INADEQUATE PROCEDURE AND PERSONNEL ERROR EVENT DATE Ill LER NUMBER 166 REPORT DAT E th OTHER F ACluY115 lNVOLVED (84 MOO DAY VR YM SEQ flat RE &lON MON DAY VR DOCKET NUMBER (Si 0 5 0 0 0 :
11 RQ G 07 92 92 -
0l3l1 -
0l0 05 20 93 T>es REPORT la SUEMITTED PUR8UANI TO THE REQUIRENIENTS OF 10 CFR. till 0 5 0 0 0 l
3 i LEYEL X 10 CFR 50.73!a)(?)(1)(8 -
0l0l0 OTHER -
(Specify in Abstract . below and in teat, NRC Form 366A) i UCENSET CONT ACT FOR THis LFR f t 2p if L E p*40NE NUMe f.9 DAVID P. SISK, SENIOR REGULATORY COMPLIANCE ENGINEER ^""
805 545-4420 COMPLETE ONE UNE FOR EACH COMPONENT F ALLURE DESCPUSED IN THl3 REPORT (13)
CQUSE SYSTEM COMPONENT MANUFAC. R OR A E CAUSE ST5 FEM COMPONENT MANUF AC. R Q E l
l lll lll l lIl lll l l lll lll l lll lll I
$UPPLEMENTAL REPORT EAPECTEO ll46 EXPECTED "" D*' '"*"
SUBMISSION l l YES (if yes, cornplete EXPECTED SUBMISSION DATE) lX l NO I QG5TRQCT (16)
On November 7, 1992, with Unit 1 in Mode 3 (Hot Standby), during startup following i the fifth refueling outage, the requirements of Technical Specification (TS) 3.3.2 i Action 18 for the Unit I containment ventilation isolation (CVI) function were not met when the containment purge valves were opened with the CVI function inoperable. The purge valves were opened similarly ca January 27,,1993, i Fc3ruary 7,1993, and March 26, 1993. '
These four events were preceded by implementation of a Unit I design change replacing the radiation monitors used for CVI initiation. On October 18, 1993, the CVI function was declared operable. This declaration was based on a surveillance test that was inadequate due to incomplete response time testing.
The inadequacy of the surveillance test was not discovered until April 20, 1993, following the same design change on Unit 2.
The surveillance test for the CVI function was reperformed and successfully completed on April 22, 1993, and the CVI function was declared operable on April 23, 1993.
j The root causes of this event are inadequate procedures and personnel error (cognitive). lo prevent recurrence of this event, the surveillance test procedures will be revised to more clearly state the requirements for response time testing of the CVI function.
l l 6115S/85K
.~ . - - . - . - - ..-. - .- - .-- -
LICENSEE EVENT REPORT (LER) TEXT CONTINUATION : -l wn ,- m n, ,-u , n -m t exar - m n-
== = ,
DIABLO CANYON UNIT 1 0l5l 0l0l 0l 2l 7l 5 92 -
0l3l1 -
0l0 ~ 2 - l *' l 7 -
itar (37)
I. Plant Conditions Unit I was.in various modes 'during startup followin'g the fifth refueling i outage-(IRS).
II. Descriotion of Event A. Summary:
On November 7,1992, the requiremen.tsf of Technical Specification (TS) .
3.3.2 ActionL18 for the Unit-1 containment ventilation isolation (CVI)_
function were not met when the' containment (NH) purge valves (VA)(V)- ~
were opened with the CVI? function-inoperable. The purge valves were; similarly opened.on January 27,51993, February 7, 1993, and March 26, 1993. !
B.
Background:
1 TS 3.3.2, Table' 3.3-3, Line 3.c.1 requires'.two channels of CVI: ;
automatic actuation logic and ' actuation relays (JM)(RLY). to be , ~
operable in Modes 1 (Power Operation), 2 (Startup),13 (Hot Standby),- i and 4 (Hot Shutdown). Action 18 of this TS allows plant _-operat_ ion with less than two channels operable, provided the containment purge
~
supply and exhaust valves (VA)(V) (RCV-11,12, FCV;660, . 661, 662, ~ 663, and 664) are maintained closed.
TS 4.3.2.2 requires that the response. time of each Engineered Safety Feature Actuation System (ESFAS) function- (including the CVI function)- i be demonstrated to be within'the -limits of Table 3.3-5."Enaineered.
Safety Features Response Times," at least once per 18 months.- Each g test 'shall include one Solid State-Prot:ition Syst& (JG)(SSPS) train ""* 1 such that both trains are tested at least once per 36 months, and at least one channel per ESFAS function.
TS 3.0.4 prohibits entry into an operational-mode when the conditions l of the Limiting Condition for Operation (LCO):are not met and the- !
associated Action requires a shutdown if.the LC011s not met. Entry ;
into an operational mode is permitted if conformance with'the Action ]
requirements permits contin ~ued operat' ion for:an unlimited period of i time, f STP.1-33A, " Reactor Trip and .ESF Response' Time Test," evaluates the:-
overall response times of various ESFAS features,fincluding CVI, to-fulfill TS 4.3.2.2. STPs' I-39-R44 A.C and 1-39-R44B'.C '" Containment:
Ventilation Exhaust Rad Monitor RM 44A(B) Time Respon'se Test:" collect the CVI response time data which are.then input to STP.I-33A for evaluation.
Response time testing may be excluded for components if their-responser 6115S/85K L
- - - , - . , , - . . , . . , , , , :,, , , ,+. . . - - -.-_,,,_,:-,. ~ = . . . . . - . - - . - - . . - - , -
. _ _ ~ .- . _. . .. _ - . __.
i LICENSEE EVENT REPORT (LER) TEXT CONTINUATION - ,
FACILITV NANC (1) DOCEET NUMDES (2) L(B N BfR (4) PAM W
\
DIABLO CANYON UNIT 1 0l5l0l0l0l2l7l5 92 -
0l3l1 -
0l0 3 l'l7
.I j
TEXT (17) l times are verified by other functional testing. .In order to justify exclusion, it must be demonstrable that changes.in response time beyond acceptable limits are accompanied by changes in_ performance l characteristics that 'are detectable- during routine functional. testing. )
l l
i All electro-magnetic relays have finite-response times associated.with the time required for the relay ~ coil (JM)(CL) to energize and actuate j the moveable contacts. This time is dependent on.the period required ;
for the magnetizing or inrush currents to reach steady state. The- 1 response. time for the relay _may be taken from the testing of a representative contact which changes state'on relay-actuation'. The response time for each individual contact is not. required-since the functionality of each contact is verified.during routine functional testing. The impact of response. time _ testing..is minimized by ensuring all components with finite response times are tested, without. '
duplicating testing paths.
l C. Event
Description:
'l On August 21, 1992, Design Change Package-(DCP) J-47031,. Revision 0, ,j was issued to replace plant vent radiation monitors.(IL)(MON) RM-ll, ;
RM-28A and B, RM-12, and RM-14A and B,.which provide a~CVI. function 1 with RM-44A and RM-448. In conjunction with the DCP, STPs'I-39-R44A.C l
and I-39-R44B.C were being written to test. the ' response times of the l new radiation monitors. The high radiation 1 output relay on~ each- of' l
the two new radiation monitors was connected directly to each of the two trains of the SSPS via separate contacts on the same relay. Since j no devices were added between the radiation monitors and the SSPS inputs, only two response time tests, one-for.'each train, would be required to fully test all components with finite- response times. in the CVI path (see Figure 1).
During reviews of Revision 0 of the DCP, Design' Engineering recognized )
that a failure in either RM-44A or RM-44B could potenti_ ally affect i both SSPS trains, A and B. The directly-connected output relays could not be qualified for contact-to-contact separation due to the close proximity of the contacts. Design. Engineering revised the DCP to add-two relays to isolate the two SSPS trains from each other for a postulated failure of one of the radiation monitors. Revision 1 of DCP J-47031 was initiated to include two. isolation relays with qualified contact separation. During discus'sions,with Design-Engineering, this change was noted by the writers of STPs 1-39-44A.C l and I-39-44B.C. As in Revision 0 of the design, two response time- !
tests were adequate to fully test all components with finite response l times in the CVI actuation paths.
4 Design Engineering determined that the desired type _.of isolation relay
! could not be obtained. Consequently, Revision 1 of the DCP was. issued j requiring a different type of relay. The change in the type of relay 6115S/85K
8 LICENSEE EVENT REPORT (LER) TEXT CONTINUATION
! DOCKET wMBEG (2) LEn NuMete 16) PAGE (3)
' FACILITY NAMC (3) m on vaan st myi DIABLO CANYON UNIT 1 Ol5l0l0l0l2l7l5 92 -
0l3l1 -
0l0 4 l"l 7 1[XT (17) required a change in configuration, resulting in a total of four new relays to achieve adequate separation (one from each radiation monitor to each train of SSPS). This configuration required additional response time testing, increasing the number of paths for response time tested from two to four.
On September 14, 1992, Revisien 1 to DCP J-47031 was approved. When Revision 1 was issued, the change in the number of relays was recognized by the procedure writers. However, they did.not recognize that this had an effect on the response time testing.
On October 14,.1992, STPs I-39-R44A.C and 1-39-R448.C were performed on two of the four response paths to demonstrate CVI operability prior to returning CVI to service, and to meet post-modification testing.
requirements t sociated with implamenting the design change.
On October 18, 1992, the CVI function was declared operable, based on the completion of the DCP acceptance tests, including I-39-R44A.C and I-39-R44B.C.
On November 3, 1992, based on the review of STP I-33A, all of the reactor trip and ESF response time channels required for Mode 4 entry were declared operable. This included the CVI channels from high radiation through SSPS, on the basis of the deficient data obtained from STPs I-39-R44A.C and I-39-R44B.C.
On November 4, 1992, Unit 2 entered Mode 4. This mode transition and subsequent transitions to Modes 3, 2, and 1, met the requirements of TS 3.0.4 since TS 3.3.2, Table 3.3-3, Action 18 does not require a plant shutdown.
On November 7, 1992, the requirements of TS 3.3.2 Action 18 for the .
Unit 1 CVI function were not met when the containment purge valves !i were opened with the CVI function inoperable. The purge valves were !
similarly opened on January 27, 1993, February 7, 1993, and March 26, l 1993. l l
On April 20, 1993, during preparations for STP I-33A on Unit 2, System Engineering identified the deficiencies with the November 3, 1992, STP l-33A for the CVI function (i.e., STPs I-39-R44A.C and I-39-R44B.C did not fully take into account the potential time delays introduced by the new output isolation relays installed as part of the radiation monitor design change).
On April 20, 1993, the Unit 1 Shift Foreman. declared the CVI function inoperable and entered TS 3.3.2 Action 18. The containment purge valves were verified to be closed, and were administratively tagged out to maintain them closed until the response time testing could be reperformed.
5115S/85K j
1 LICENSEE EVENT REPORT (LER) TEXT CONTINUATION- !,
FACILITV NAME (1) DOCKET NUMBER (2) ((O NUMS(R i 6) PAGE (3) .
w ug=y == r DIABLO CANYON UNIT 1 0l5l0l0l0l2l7l5 92 -
0l3l1 -
0 l0 '5 l'l7'-
1(af (17)
- On April 23, 1993, the response time. testing was successfully .
performed with revised STPs I-39-R44A.C.and I-39-R44B.C, and TS'3.3.2 1 was exited. .
D. Inoperable Structures, Components, or. System's that Contributed to the Event:
None.
E. Dates and Approximate Times for Major.0ccurrences:
- l. November 7, 1992: Event date. TS.3.3.2 Action 18 was exceeded >
when Unit 'l containment. purge valves were :
opened with the: CVI- function inoperable. The purge. valves were. similarly. opened on January-27, 1993, February 7, 1993; and March 26, 1993.
- 2. April 20, 1993: -Discovery date. LSystem Engineering identified.
deficiencies'with STP I-33A.. .
~
F. Other Systems or Secondary Functions Affected:
None.
G. Method of Discovery:
On April 20, 1993, System Engineering identif'ied the deficiencies in the November 3, 1992, performance of.STP I-33A on Unit I during evaluation of STP I-33A testing on Unit 21 H. Operator Actions:
On April 20, 1993, the Shift Foreman declared the CVI function inoperable and entered TS 3.3.2 Action'18. The contairment purge valves were verified'to be closed and were_ administratively tagged out to maintain'them closed until the response time testing could be reperformed.
I. Safety System Responses:
None.
III. Cause of the Event A. Immediate Cause:
~
The immediate cause of the CVI TS being exceeded was the incomplete response time testing of the CVI function.
6115S/85K
LICENSEE EVENT REPORT (LER) TEXT CONTINUATION
)muun,a m oocu r 'e m a m ua
.c m nu m DIABLO CANYON UNIT 1 0l5l0l0l0l2l7l5 92 -
0l3l1 -
0l0 6 l 'l 7 TEXT (17)
B. Root Cause:
The root causes of this event are inadequate procedures and personnel error (cognitive). STPs I-39-R44A.C and I-39-R44B.C did not indicate that they collect data required for the completion of STP I-33A and compliance with TS 3.3.2. In addition, STP I-33A did not include sufficient description of the requirements for response time testing.
IV. Analysis of the Event Following the design change that installed the radiation monitors and output isolation relays, each of the radiation monitors was functionally tested and verified to actuate each train of the SSPS with a CVI signal. Furthermore, response time testing performed after the discovery of the event verified that the CVI function met the required response times for all combinations of RM-44A, RM-44B, and the two trains of SSPS.
Therefore, the CVI would have functioned properly if called upon.
Consequently, this event did not adversely affect the health and safety of the public.
V. Corrective Actions A. Immediate Corrective Actions:
- 1. The containment purge valves were verified to be closed and were l
administratively tagged out to maintain them closed until response time testing could be reperformed.
- 2. The surveillance test procedures I-39-R44A.C and B.C were.
revised and complete response time 'esting was~ performed -
satisfactorily.
B. Corrective Actions to Prevent Recurrence:
- 1. STP I-33A will be revised to more clearly state the functional requirements of the response time program for all reactor trip i and ESF functions, including CVI. i l
- 2. Test procedures that collect response time data for input into l STP I-33A will be reviewed and revised as necessary to emphasize I that the data is required for STP I-33A and TS compliance. l
- 3. A lessons learned memorandum will be issued to applicable personnel informing them of the event.
G115S/85K
LICENSEE EVENT REPORT (LER) TEXT CONTINUATION h FOCILiff hAME (3) DOCKET NWB(R (2) LER NUMEER f6) PAGE (3) e stAn stoulmidt RE N 't DIABLO CANYON UNIT 1 0l5l0l0l0l2l7l5 92 -
0l3l1 -
0l0 7 l 'l 7 ftXT (17)
- VI. Additional Information A. Failed Components:
None.
B. Previous Similar LERs LER l-92-027-00, " Containment Ventilation Isolation Technical l Specification 3.3.2 Not Met Due to Personnel Error" This previous LER reported a violation of CVI TS 3.3.2 Action 18 l during the containment discharge on. November 7, 1992. -However, that event was due to a personnel-error in the incorrect positioning of a damper to bypass the CVI radiation monitors. Because the current-event occurred during the same time period but was not related to improper system or damper alignments, the. corrective actions from this
- previous LER did not prevent the current event.
l
- LER 2-85-004-00, " Containment Purge Valves Left Open With an SSPS l Channel Removed From Service for Testing" j l This previous LER reported a violation of' CVI.TS 3.3.2, Action 18 when i a containment purge was_ performed while SSPS Train A
- was removed from l service. A root cause of this event was that the STP lacked explicit ,
instructions concerning purge valve position. The associated corrective action to prevent recurrence was to revise the procedure to provide such instructions. Because the current event was not' caused by a misunderstanding of TS required purge _ valve positions, the corrective action from this previous LER did not prevent the current l event.
l l LER 1-87-029-00, " Missed Surveillance When the ESF Time: Response Test
! Frequency Was Not Followed Due to Misinterpretation of Technical Specification Requirement" l
i This previous LER reported exceeding the time interval requirement of TS 4.3.2 for time response testing of the safety injection. function, due to misinterpretation of the testing. frequency requirements. The corrective action for this event was:to revise STP I-33A to correct-l the required testing frequency. This corrective action-did not prevent the current event because.it was the elements of time; response testing that were not understood, not the testing frequency.-
l l
l 5115S/65K i
RADIATION CONTROI, ROOM SSPS MONITOR CONTROL ISOI.ATION RELAYS CVI MASTER SSPS DETECTOR SIGNAL SLAVE ROOM ADDED IN REV 1 INPUT RELAY PROCESSOR REl.AY I UNIT IRELAY l l l I I I I I I IK2 -14A1 IKu19 K503 KG22 IK2 t SAFEDUARDS I - '-
OUTPUT A518 If LRP I OUTPUT I t ) .
IK2-44A2 l I RELAY 1 I I i RM-44A i l TRAIN A RM-44B 1 l ]
CVI VALVES l
~
I I l i I I IK2-44B1 7 O O- LRP I _[ ,V:
SAFEDUARDS-OUTPUT A518 _liI V 1K2 1K2-44B2 K503 K622 lK819 I I OUTPUT I RELAY l i I I I I I FIGUllE 1
- - - _ _ _ _ _ - _ _ _ _ - _ _ _ _ _ - _ .