ML19332E892

LER 89-008-00:on 891103,during Refueling,Contractor Personnel Generated False Trip Signal While Investigating Wiring Discrepancy in Reactor Protection Sys Instrument Racks.Caused by Labeling Error.Supply changed.W/891204 Ltr

ML19332E892
Person / Time
Site:	Point Beach
Issue date:	12/04/1989
From:	Fay C WISCONSIN ELECTRIC POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
CON-NRC-89-150 LER-89-008-01, LER-89-8-1, VPNPD-89-632, NUDOCS 8912130112
Download: ML19332E892 (8)

Text

E 4 .

c p A ~

Wisconsin 1 Electnc POWER COMPANY 231 W Mchiocn PO. Box 2046. Milwoskee. WI 53201 (414)2?i-2345 VPNPD-89-632 10 CFR 50.73 NRC-89-150 December 4, 1989 U. S. NUCLEAR REGULATORY COMMISSION Document Control Desk Mail Station P1-137 Washington, D. C. 20555 Gentlemen:

• 0 DOCKET 50-301 ,

LICENSEE EVENT REPORT 89-008-00  :

INSTRUMENT BUS GROUND RESULTING IN SPURIOUS SAFEGUARDS ACTUATION i POINT BEACH NUCLEAR PLANT. UNIT 2 ,

Enclosed is Licensee Event Report 89-008-00 for Point Beach ,

Nuclear Plant, Unit 2. This report is provided in accordance with 10 CFR 50.73 (a) (2) (iv) , "Any event or condition that resultedLin-manual or autornatic actuation of an Engineered Safety Feature".

-This report describes the generation of a false reactor _ trip signal which occurred while investigating a wiring discrepancy i in the reactor protection system instrument racks.

If any further information is required, please contact us.

Very truly yours,

[ Ah , c

! C. W. Fy Vice President Nuclear Power Enclosure Copies to NRC Regional Administrator, Region III NRC Resident Inspector 8912130112 891204 PDR ADOCK0500ggg1 San ,

a.aa,wsunrantwennaan s

_ _ _ _ _ _ . _ _ _ _ _ - J.;

I g,w v..on.

"***'" 8'4'"

l

.t "**" ****

. UCENSEE EVENT REPORT (LER) sm,n mens m .ac.e,nameaei -=

Point Beach Nuclear Plant o15IoIoIo131011 i lod 7 I

- a ea i Instrument Bus Ground Fault Causes Spurious Safeguards Actuation i ev.w, ,e e ten . , = n i,, o, . .ciutiee =voive. ei M.e hav ve.a v4.h MOQ'86 , g

.W .s0.u t se Day vs.m .aciu v a ee Docat, v 0 misi

.x oisIel0l0l l l l

\

/. - ~

( 8 l9 1l1 03 hl9 0l 0l 8 0_j 0 1l2 0l4 8l 9

., w + n u 0 iEle t 0 1 0 i i i l v.di , e stemurn evneu.=, ,. ,.= no.um.m.,e n een s ,cm

%,,,,,,,,,- i

'"se m 4 , , , , , , , , , , , , , , , , n ,,

g _

= ==uma _

so mwm, _

m nwen.i _

tamm am 01010 mammm _

eeawm _

es nweem _ g a, g p,,;.gfy;,

'"waa" "*'*m "*"*"*" =

reemM:

""".w"".*,'

kM(m.pda . .. n,. **",,'",,

,,,,,,,m,,,, 1 tect Btt co.tv.c,909 v e Lta (131 viti, o.a vm.a AAla c000

, C. W. Fay, Vice President - Nuclear Power

[ 4 11 14 2 l 2l 11 -121811l 1

.n u . ..c. , . .a va. .e. ... .. , e .. , n.,

.nn. c ., =;;;;;e agga^!P c.v.i ...n co- o=. , =;;:::c aggaagt' -

i l i I l l l l l l 1 l i 1 -

1 111 I I I I I I i l 1 1 '

eu 6 .,.6 no , . en. n.i oo=,- e., vs..

sv s.io=

.M f# yes. same.e f M.f C,fD #64.ff88tO= Od ffJ NO l l l

,me, <u . . . . = .--.,n., .

On November 989, during refueling operations, contractor personnel generated a false trip signal while investigating a wiring-' discrepancy in the reactor protection system instrument racks. The reactor was defueled and the reactor trip breakers were open. Therefore, no safety related equipment started.

An original wire labeling error was considered the root cause of the event. ,

l Further analysis indicated two redundant chan els of boric acid storage tank T-6C level instrumentation (2 of 3 logic) had common i

instrument bus power supplies. Corrective action included changing l

the boric acid storage tank level instrument power supply and relabeling of the cable conductors.

l This event is reportable pursuant to 10 CFR 50.73(a)(2)(iv).

l met een see I

, , . u e =ve6au ..:vmoav ca... t

" **' UCENSEE EVENT CEPORT (LER) TEXT CONTINUAT60N < mono ous =o **-p

, u isswe

. .un n. .se.n u wea = 6..... e. . 1

.... tm,'- n*.i: l Point Beach Nuclear Plant O' O 17 olslejojo131011 '819 -

01019 -

ol n nI>

on. . .= =-amwnn EVENT DESCRIPTION  !

t On December 28, 1988, plant operators discovered a wiring l discrepancy for power supplies to instrumentation mounted in i cabinet 2-C128 and 2-C129. It was concluded that these two  ;

instrument cabinets had a power supply configuration different than the corresponding Unit 1 cabinets. In addition, boric acid tank level channel L-172 power supply, (physically in cabinet 2-Cl29)-was actually being powered via an extension cord i from cabinet 2-Cl28. The extension cord had been installed several >

years ago. The condition as found was evaluated not to be in  :

violation of the Technical Specifications or the Final Safety i Analysis Report; however, Modification Request (MR) 89-41 was .

, initiated to eliminate the inconsistencies between the two units.

l Hardware changes to be accomplished as part of MR 89-41 included:  ;

1. Swapping the power supplies for each cabinet to configure them the same as the corresponding Unit 1 instrument cabinets.

2. Remove the " jumper" (extension cord) for boric acid storage tank level instrument LQ-172, which existed between the two cabinets and power it from the instrument power in 2-C129.

3. Complete.some minor wiring changes within the individual cabinets.

Appropriate design controls and procedures were utilized for Modification Request 89-41. Since instruments controlling the interlock between reactor coolant pressure and residual heat removal

valve 2-720 are located within instrument cabinet 2-Cl28,(in effect, a loss of power to this cabinet would disable the remote, manual opening function of tnis valve.) the modification was scheduled to be completed concurrently with the refueling outage,Section XI, ten l

year inspection of valve 2-720. During this time period, power for the interlock and control circuitry would not be needed. A post-inspection stroke test was required to ensure operability of the valve. The modification and the valve work'were scheduled for implementation between November 1-3, 1989.

On October 27, 1989, operators opened supply breakers for both I

cabinets. Although several alarm responses were not received as

• expected, it was determined that all alarm responses would be more appropriately verified at the completion of the modification.

l l

eo.w seu e u s 0*o t ese oe 4 ue 46's W

wat see una y a noctsam as;og+.tpe, consenesson

. UCENSEE EVENT REPORT (LER) TEXT CONTINUATION peactic c~e *o >*-cios seines s sic

.or. .a .. . ooent avasia e si. v . . i., eau is.

m "t'd v "h*at.

Point Beach Nuclear Plant o is lo lo lo 13 l0 l1 Bl9 -

0l0l8 -

0IO 013 oF ) l7 ,

1921 m amo auss e seasser. amp eA8Bmap =8C 8p'* W W fih On November 3, an electrical contractor completed testing and terminations for conductors within cabinet 2-Cl28. Circuit breaker 2-YO4-6, thought to supply cabinet 2-Cl28, was closed.

Work recommenced on cabling in adjacent cabinet 2-C129. While testing the cable, the contractor had not reset the test instrument to the proper parameter (ohms vs. megohms). The not effect was an improper indication on the proper instrument scale. As work continued, the contractor received a minor electrical shock as one of the leads was exposed. Unknown to the contractor, the conductor was energized from circuit breaker 2-YO4-6.

Suspecting there may be a capacitance charge remaining in the conductor from the recent cable check or inductive current from parallel conductors, the wire was brushed across the cabinet ground to discharge it. This created a ground fault and subsequently sent a voltage spike to the instrument bus. The electrical perturbation resulted in the actuation of the Unit 2 "B" steam generator low feedwater flow reactor trip logic. The perturbation also generated:

1) a loss of power to nuclear instrument power range channel 2-NE-44;

2) Unit 2 containment ventilation isolation; 3) the loss of plant process computer video monitors No. 9, 11 and 12: and 4) loss of ,

Units 1 and 2 containment hydrogen monitors. Because Unit 2 was in '

refueling operations, the reactor trip breakers were open and no safety related equipment was started by the signal.

SYSTEM DESCRIPTION:

The Unit 2 120 volt AC instrument supply system consists of eight buses, divided among four channels. Each of the four channels is allocated two buses. Each channel is powered by one dedicated inverter and one backup inverter which is shared with the Unit 1 120 volt AC instrument supply.

Each bus supplies instrument power to a number of different instrument cabinets and control boards. Within each cabinet are instruments for various systems. Below is a list of instrument loops affected by instruments in cabinets 2-Cl28 and 2-C129.

Loop Description Function  !

Lll2 Volume Control Tank Level Indication / Control -

Ll72 Boric Acid Tank Level Indication / Control / Alarm T418 Reactor Vessel Flange Leakoff Indication / Alarm P420 Hot Leg RX Coolant Pressure Recorder / Control g.a. us, .9 s oro ,,,s o ua su ais

e.= vi avema u:vato., co imio  !

.. UCENSEE EVENT REPORT (LER) TEXT CONTINUATION u*aesseovewe ru.ee t

.. 5:'*t5 n o s at ut hautiw Docaat wuutth

• tem esunette 86 960t *

.e.. " M.g.a . @at.s Point Beach Nuclear Plant 0 p l0 lc lo l }9J .$ O -

Q90 --

Q0 l4 or 7l 1:n , . - w.c , . .no, [

Loop Description Function T421 Pressurizer Surge Line Temp. Indication / Alarm t T422 Loop A Press. Spray Line Temp. Indication / Alarm .

T423 Loop B Press. Spray Line Temp. Indication / Alarm '

T424 Pressurizer Liquid Temperature Indication / Alarm T425 Pressurizer Vapor Temperature Indication / Alarm T436 Press. Safety Valve Outlet Temp. Indication / Alarm ,

T437 Press. Safety Valve Outlet Temp. Indication / Alarm T438 Press. Power Rel. Valve Outlet Temp. Indication / Alarm 4 T439 Press. Rel. Tank Liquid Temp. Indication / Alarm T440 Press. Rel. Tank Pressure Indication / Alarm / Control L442 Pressurizer Relief Tank Level Indication / Alarm L447 Refueling Reactor Vessel Level Indication / Alarm  :

i T608 RCP Component Coolant Water Return Indication >

P923 SI Pump Discharge Pressure Indication P937 No. 2 Accumulator Pressure Indication / Alarm P941 No. 1 Accumulator Pressure Indication / Alarm The instruments and cabinets were manufactured by FoxBORO. The instrument buses and circuit breakers were manufactured by Westinghouse.

CAUSES AND CORRECTIVE ACTIONS:

The primary causes of the event were identified as follows:

1. The mislabeling of cable conductors ZQ2YO308-A and ZS2YO406-At Because of the as found labeling, power to instrument cabinet 2-C128 was incorrectly determined to be via instrument bus 2-YO3 and 2-C129 was incorrectly determined to be via instrument bus 2-YO4. In reality, cabinet 2-C128 was powered from bus 2-YO4 and cabinet 2-C129 was powered from bus 2-YO3.

In effect, the two units power supplies had been configured similarly -- per the design intent. Only the conductor labels were in error, t.

I After the labeling error was discovered, the engineering evaluation was conducted. Proper engineering change requests were completed and the power supplies were placed in there

" pre-event" configuration with correct labels. The power I

supply for boric acid storage tank level channel, LQ-172, was ,

-transferred to the cabinet supplied by instrument bus 2-YO3, to meet redundancy requirements and the extension cord was l removed.  :

i I

l l

l g 'o w ana .. s ce;. ;uim u+ icn

~

l - - . -- -

.m v. .ocw. uemo., ea ..

UCENSEE EVENT REPORT (LER) TEXT CONTINUATCN amoisc o.e =e w.eio.

. . mains. e si e t .. . n. == m .u=.. . ... . , .... +

vs.a " m ." M*.n Point Beach Nuclear Plant o ls l o l o l e l3 lo li gig -

olol n -

o lo ols o' nb

,.n,. . =ca ani.,m,

2. Inappropriate action by the contract personnel:

In this case the inappropriate action was to ground the energized conductor. A test device should have been utilized to determine if the conductor was suspected to be energized. The electrical contractor, as well as the engineers responsible for the oversight of contractor work, shall be notified that grounding low voltage conductors to test for current is an unacceptable work practice.

While it was not necessarily a primary caute, we believe that the refueling outage activities, which created a sense of urgency while

.tte actual work was being performed, may have been a contributing factor. During this event, the contractor was summoned with short notice before initiation of the planned activities. Manpower available to complete the assignment was limited at that time.

Maintenance and Operations were waiting on the completion of the contractor work activities to continue other refueling activities.

GENERIC CONCERNS AND SIMILAR OCCURRENCES:

There are no industry generic concerns. This event was evaluated ,

to be primarily a human performance issue. There have been three other events in which test equipment and human performance had abetted conditions to a safeguards signal. Those events are documented in Unit 2 Licensee Event Reports89-002 and 89-006 and Unit 1 Licensee Event Report 87-04.

REPORTABILITY:

This Licensee Event Report is provided to pursuant to:

10 CFR 50.73(a)(2)(iv) -- any event or condition that resulted in manual or automatic actuation of any Engineered Safety Feature (ESF), including the Reactor Protection System (RPS).

I l SAFETY ASSESSMENT:

The health and safety of plant employees and the general public were not affected during the event. The reactor was defueled to complete Section XI, ultrasonic inspection of reactor vessel

welds. The reactor trip breakers were open and no safety related L equipment was challenged (inadvertently started).

L l

i l

g. .. . -. . . o , . . . . . .= 4.

e .... no o a mucm. ae,uun o emissos

~'"

s, LICENSEE EVENT CEPORT (LER) TEXT C2NTINUATION cue se one se vn-er.

. . teals svo s&CMT t but s t* tocatt husetta 60 tit humatta ili tact (3-

.s" s'M p 2;g,'

Point Beach Nuclear Plant o p jololol pp .p A- pp R- p 9% or 7l nri.u . wncs- m..e The original nonconforming issue was reviewed with respect to the evidence analyzed after the event. With the exception of boric acid storage tank (BAST) level channel L-172, no instrument channels within either of the cabinets needed independent power supplies to meet safeguards logic redundancy requirements. As it was stated earlier, boric acid storage tank level instrument LQ-172 was being supplied from the adjacent instrument cabinet via an extension cord. It is not known when the extension cord was installed.

Because conditions were not as they appeared, two boric acid storage tank T-6C level channels were supplied via instrument bus 2-YO4 and one channel was supplied from 2-YO3 (vice one channel supplied from each of the three instrument bubes 2-YO1, 2-YO3 and 2-YO4). This decreased the level of redundancy from three independent channels to two independent channels.

During an accident event, the contents of BAST T-6C (boron concen-tration 20,000 ppm) would be pumped into the reactor vessel (via the safety injection pumps) to aid the reactor shutdown. After the boric acid storage tank is emptied, a tank low-low level signal would transfer safety injection pump suction from the BAST to the refueling water storcge tank. A 2/3 logic is needed from the three independent level channels to initiate the transfer signal.

If a fault (a single active failure) had occurred which would have opened the cabinet supply breaker (2-YO4-6) or the main breaker to instrument bus 2-YO4, two of the three boric acid storage tank level channels would have failed low (conservative). In effect, a ,

low-low (tank empty) signal would be sensed. Safety injection pump suction would immediately be transferred to the refueling water storage tank (2,000 ppm boric acid) as in a normal sequence of events. The 20,000 ppm boron solution injection would not occur.

The event most sensitive to the need for 20,000 ppm boron is the hypothesized steam line break accident. This event assumes the steam release results in an initial increase in steam flow which decreases during the accident as the steam pressure falls. The energy removal from the reactor coolant system causes a reduction of coolant temperature and pressure. With a negative moderator temperature coefficient, the cooldown results in a reduction of the core shutdown margin. If the most reactive control rod is assumed stuck in its fully withdrawn position, there exists a possibility that the core will become critical and return to power even with the remaining control rods inserted, l

l 1

osw anu 6.s a: .u + u n'o -

i -

..c . . m. vs =sca:.us wo., coa.. a

'" s.

LICENSEE EVENT REPORT (LER) TEXT CSNTINUATION 4x=o seove*o m . m.

smis e >o

'a&4'19 haut A pocas t wgensta it' 480 humasta 16' P&0513

'u* "tu;:~ 1%*n:

,s u Point Beach Nuclear Plant m . =:= .. : = sec ,n m ,, o,,

015101010l 3 Q 1 69 -

dQ8 -

Q0 l7 0F 7l The Final Safety Analysis Report (FSAR) describes methods in which boric acid can be injected into the core uti3izing the chemical and volume control system. FSAR Section 9.2, general design criteria 30 states, " Boric acid can be injected by one charging pump and one boric acid transfer pump at a rate which shuts the reactor down in less than 15 minutes...At least three separate and independent flow paths are available for reactor coolant boration; i.e., normal charging, alternate charging and charging via the reactor coolant pump labyrinth seals." Operations Critical Safety Procedure CSP-S.1, " Response to Nuclear Power Generation," identifies the plant conditions and cperations response to post accident criticality.

Also noteworthy is the fact that BAST T-6C is only one of three redundant tanks which are capable of supplying 20,000 ppm boron to the safety injection pumps or the charging pumps.

In summary, there was a minimal safety impact associated with having two boric acid storage tank level channels supplied from the same instrument bus. ,

k

)

g ,o= -. . . o:, ....... .,.. m

_