ML20042F568

LER 89-021-01:on 891212,chlorine Monitors on One Train of Control Room Ventilation Inoperable for More than 11 H. Caused by Personnel Error.Operating Procedure for Chlorine Monitoring Sys Issued & Training provided.W/900504 Ltr

ML20042F568
Person / Time
Site:	Prairie Island
Issue date:	05/04/1990
From:	Hunstad A, Parker T NORTHERN STATES POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-89-021-01, LER-89-21-1, NUDOCS 9005090076
Download: ML20042F568 (5)

Text

- .. - _

. j Northem States Power Comparty )

I 414 Nicollet Mall Minneapohs, Minnesota 554011927  ;

Telephone (612) 3345500

• I J

May 4, 1990 10 CPR Part 50 Section 50.73 '

h Director of Nuclear Reactor Regulation V S Nuclear Regulatory Commission ,

Attn: Document Control Desk Washington. DC 20555 PRAIRIE ISLAND NUCLEAR GENERATING PLANT Dochet Nos. 50 282 License Nos, DPR 42 50 306 DPR 60 Both Chlorine Monitors on One Train of Control Room Ventilation Inoperable for More Than 11 Hours Due to Personnel Error An updated Licenseo Event Report for this occurrence is attached.

1 Please contact us if you require additional information related to this event.

/$hW h /

Thomas M Parker Manager Nuclear Support Services c: Regional Administrator - Region III, NRC NRR Project Manager, NRC Senior Resident Inspector, NRC MPCA Attn: Dr J W Ferman Attachment i i l

r-9005090076 900504 PDR ADOCK 05000282 S PDC g L

l t

C 8 0h4 3M v.8,I.UC4 tYJ a t GvL;.tav COMMi&&l04 APPIOvt0 0ut NO. 31604to.

ioci. .m>

tstewattD tumCEN Pin attp0Ntt to COMPLv win tuis UCENSEE EVENT REPORT (LER) Ee' *t ~,r's.Uc'AW1*0 fvIe*M.. fs'til A A~C- l i.Y.t'#,. ti '.IT0f.0

&~0 .i 0 4 A.Anot~t n v i vCa .

Patt mi Rt it0N J it 0, .4,.A a.kot ~t A~0.vou t. . A..Ci n(3 0~ 600DC

0. ' O mn

.ici4ity haut 11: 00Castsvu.5m m .Au6 4 Prairie Island Nuclear Generating Plant Unit 1 ol6IoIoIoi 2 18;2 3 lopl 0 ; 4

"'"' Bot h Chlorine Monitors on One Train of Contro.1 Room Ventilation Inoperable for More Than Eleven Hours Due to Personnel Error Evt.t DAtt 1 1 Lt R hvW8t h 166 #t' ORT Catt (76 OtMtm # ACsLitits INv04vt0101 EONT H DAY vt A8i fl&R 86,0, 6

, * #,v,a woNTM OAv vlam .acisite n was 00CKET 49wttatti Prairie Island Unit 2 olSio1oj0;3l0;6 1l2 1l2 8 9 Pj9 Oj 2 l1 0l1 0l5 0l4 9l 0 oi 6 t otoioi ; ;

vms e4 0 5 is aveuirtt 0 ev.IvA=t to v.a a60viaiut=n 0, s. C,n ( ,c . ., . ., . u. ., n u t,,,1 ,,,,0

=0D m N n mi., n .., a ni.nin i n,u i PO .as t.ninal .0 asten11 to.tSt.nlit.i M.tu.I nei t 8i n .o.i.in nm masi.im oo n anan.c _ Ot=8

,, gag .al+ge,,,

, n . nnisii g wnw>mm wns.,m<. iua, wo 1 n . inn i noi.ian* ni.i u . n.,

_ _ _ l n inn. an..imsi no ni.nin.i 46Cth80I CONTAit f em tMit LIR nti NIME fikt'M0ht NvW8th AmlaCO0t Arne A Henstad, Staff Engineer 6;1 ; 2 3;8; 8; i 1 ;1 ;2 ;1 COMPtitt Oht 4ths FOR B ACM C0urchtNT f alLumt OtactiS40 iN TMis atP0at nas Civ$t IYlttM C0ue04 TNT ""*%g lC. a j'p08,t,Aj,8 Cault Svit t M Coup 0Niset "QXglC- m e0mt ' '

p pa l I

i 1 I I I I I I I I I I f f I l I f I l 1 I l l 1 1 I l

.UP*t BWINTAL StPORT t R9t Ctt 0 ti.) MONtM C&Y vlam Sv8 mis 880N 48 8 IM vee rewm.we SK9tCTip Syg.HS$10k OA til No l l }

A.Cu ACv n ,-,, ,m . . . .-. ., ,,,i .y. . ,--... A - n e i On December 11, 1989, Unit 1 was at 83% power in coastdown for refueling, and Unit 2 was at 100% power. Due to malfunction, No.121 Chlorine Monitor had been in BYPASS (a 7 day LCO) since December 9th. At 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> on December lith, a I

computer alarm indicated malfunction of No.122 Chlorine Monitor, so it was also placed in BYPASS (a 6 hour6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> LCO with No. 121 Monitor out of service) and its l operation watched carefully. Af ter about an hour, it was decided to put No.122 Chlorine Monitor back in service. Telephone communication was established between the control room and the outplant operator at the monitor.

1 Miscommunication caused the control room operator to think the monitor had been returned to service when in fact it had not. At 0615 it was discovered that No.

122 Chlorine Monitor was still in BYPASS and the monitor was returned to NORMAL.

Both monitors were inoperable for over 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br />.

&C ..e. 4 e.

1 poem pe6a u.s. nucttAa a Gw6Atoav comemission p I

E APikIl. el3043 "l4*,,^',',00 ',,w*e D,M"0l'j;*,

• n,'* #"'t,' ,T,",2!

LICENSEE EVENT REPORT (LERI ,

TEXT CONTINUATION i*,j"',",l',*,181*yi n#iM!,'e'"'/l2 'w"' "Mi"f!  !

fa[tav[0Ing Tien N (3 90 it i

07 MANAG8 MENT AND tv0Gif.vt&SMe8*0tose.DC 2060L pacauty haut un 00Catt erwetta al Lik kuutta 141 PA06 (31

.... . unw.~ ::ce.n l

Prairie Island Unit 1 o ls jo jo jo l2 l8l2 8l 9 -

0l2l1 0l1 0l2 or 0 l4 rm ,, - . e w an.w nn EVENT DESCRIPTION On December 11, 1989, Unit I was at 83% power in coastdown for refueling, and Unit 2 was at 100% power. Due to a malfunction, No. 121 Chlorine Monitor (EIIS Component Identifier MON) had been in BYPASS (a 7 day Limiting Condition for Operation Action Statement) since December 9th. When the monitor is in BYPASS, it is prevented from initiating isolation of the control room upon detection of ,

chlorine. At 1900 hours0.022 days <br />0.528 hours <br />0.00314 weeks <br />7.2295e-4 months <br /> on December lith, a computer alarm indicated a potential malfunction of No.122 Chlorine Monitor, so it was also placed in BYPASS and its -

operation observed carefully. No. 121 and 122 Chlorine Monitors were currently the "high chlorine" initiation instrumentation for No. 122 Control Room Ventilation System. Since both instruments were in bypass no automatic actuations would have occurred and a 6 hour6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> Limiting Condition for Operation  :

Action Statement applied. After about an hour, it was decided to put No. 122 l Chlorine Monitor back in service. Telephone communication was established ,

l between the control room operator and the outplant operator at the ' monitor. '

Miscommunication caused the control room operator to think the monitor had been returned to service when in fact it had not. At 0615 on December 12th, it was discovered that No.122 Chlorine Monitor was still in BYPASS and the monitor was returned to NORMAL. Both monitors were jointly inoperable for over 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> with  ;

the outside air supply dampers remaining open.

CAUSE OF THE EVENT The cause of the event was miscommunication during a telephone conversation between a control room operator and an outplant operator. Contributing to the i miscommunication was inadequate operator knwledge of the chlorine monitoring '

l system, lack of an operating procedure, and the obscured location of the BYPASS-NORMAL switch. The outplant operator thought the control room operator had the ,

ability to return the monitor to service, whereas the control is local, and thought the control room operator had said that he had placed the switch in .

NORMAL. The outplant operator then verified that the monitor was functioning '

normally, without seeing the hidden BYPASS NORMAL switch.

Several causal factors intertcted synergistically to result in miscommunication.

The most significant .of f.'2se are inadequate training on the monitors, lack of procedures for the monitors, and poor human factors design of the monitors. ,

s f

~ +

, ,,,,,,,,,,, u. . caa. .t.~toa v co ='o-

,s a, ... t . . . . mt.

I RPie t 8. a/30/93 LICENSEE EVENT REPORT (LER) '.8,lg'JJ,0,Av;gRty" "'

!gil,,ff"',V ,7.".!",1 TEXT CONTINUATION E""A". ti"SniMf'tMt't!','!3;Pu'i' O'iMJ emptow'oInt t e0N DJ! t3 foe ONIC .

of manactwth? ANO evDGtt.na8Mimoton,0c 20603.

8ac4affhauttu DoGEtt NWutta G1 tin wuuttm to Pact tai vtaa je .- "

88QAM'b ! * : ,818dl Prairic Island Unit 1 o ls j o j o j o l 218l2 8l0 - 0 j2 l 1 -

0l1 0l3 or 0 l4 nxt , . ,, u.c w.muw nn ANALYSIS OF THE EVENT '

The event is reportable pursuant to 10CFR$0.73(a)(2)(1)(B). Tech Spec 3.13.E states:

"If both chlorine detection channels for one train of ventilation are in' operable then within six hours:

a. Pestore at least one channel to operable status, or

b. Operate the redundant ventilation system in the normal (non recirculation) mode and close the outside air supply dampers for the affected train of ventilation."

Both chlorine monitors on the affected ventilation train were inoperable for more than 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> while the outside air supply dampers remained open.

The function of the chlorine detection system is to monitor the air entering the ventilation ductwork leading to the control room and to isolate the control room from that air supply if a high concentration of chlorine is detected. The health and safety of the public were unaffected since no chlorine was present.

Additionally, the monitoring function of No. 122 Chlorine Monitor was still operable. Detection of high chlorine concentration would have alarmed (the alarm setpoint is at half the concentration of the isolation actuation setpoint),

prompting the control room operator to initiate isolation of the control room utilizing control room switches. '

CORRECTIVE ACTION Upon discovery, No. 122 Chlorine Monitor was returned to NORMAL.

An operating procedure for the chlorine monitoring system has been issued.

Additional training on the chlorine monitoring system has been provided.

l The obscured switches have been made apparent by improved labeling on the front doors of the monitors, i

i 4

O

,e em pomassesa W.S. huCL8Ah mtGWharont coumissiON g taPitss er3042 . J LICENSEE EVENT REPORT (LER) L'Is%.','Aq'R.M"o*'j;tli,T*,#t'.V ,"e'.",'"',',

TEXT CONTINUATION 'J"'",'4",l"C"&',Y0 !" !!'e'.", ','h!'v '"' "'E O! ]

UN'a','.9.".'o!!"."a'o?".o*,'!"Zil?"ia&P& e"?Ji Os use,a04wthf AND OWDG8T.WASMINGTON.DC 30603, paciLITV #aaMG 110 DOCK 47erVMethLIl ggg gggggg ggy pagg g3i >

T8&A I .:. . N k,', N,'b ?'",'j,*a **

Prairie Island Unit 1 rsafw - m . m , onc w asnn olsjojolol2;8l2 8; 9 ._

0[2l1 ._ 01 9 0;4 or 0f Because of various malfunctions of the chlorine monitoring system, system changes I are being made to improve the reliability. One of the changes already completed -i has modified the function of the BYPASS NORMAL switch on this train of control room ventilation (this is the train normally in service). At the time of the event, putting one of these monitors in BYPASS prevented the moni, tor from initiating isolation of the control room. Now putting one of the monitors for this train of ventilation in BYPASS puts that monitor into a trip condition, making up one half of a two out of two logic. Therefore, a repeat of the same miscommunication would not result in preventing the chlorine monitoring system .

from initiating isolation of the control room. ,

Verbal miscommunication, the designated cause of this event, will be entered into the causal factor analysis data base for trending. Additionally, Prairic Island's Error Reduction Task Force will be evaluating recommendations _ for a formalized approach to operator telephone communications and for outplant .

operator " team building" training.

FAILED COMPONENT IDENTIFICATION MDA Scientific Chlorine Detection System, Model 7040 FAN PREVIOUS SIMIIAR EVENTS There have been no previous similar events at Prairie Island.

I r

l 1

e

- &,