ML20141H855

Trip Rept of 851112-15 Site Visits Re Wrong Unit/Wrong Train Events.Visit Conducted as Part of short-term Effort to Determine If Low Cost Improvements Can Be Identified & Implemented to Reduce Frequency of Events

ML20141H855
Person / Time
Site:	Cook, LaSalle, 05000000
Issue date:	01/10/1986
From:	Persinko D, Rameysmith A Office of Nuclear Reactor Regulation
To:	Booher H Office of Nuclear Reactor Regulation
Shared Package
ML17324A714	List: ML20141H855
References
NUDOCS 8601170628
Download: ML20141H855 (40)

Text

- . _ _ - -

[ u.

\ umisa:Tavas NUCLEAR REGULATORY COMMISSION 9

l f 5 WASMineG7008. D. C. 20666

• g ...* Ju 10 BBS ..

MEMORANDUM FOR: Harold R. Bocher Chief Maintenance and Training Branch Division of Human Factors Technology, NRR Frank H. Rowsome, Chief Human Factors Issues Branch Division of Human Factors Technology, NRR Kathleen M. Black. Chief Nonreactor Assessment Staff Office for Analysis and Evaluation of Operational Data James E. Lyons. Chief Technical and Operations Support Branch Planning and Prcgram Analysis Staff, NRR FROM: Ann Ramey-Smith, Engineering Psychologist Human Factors Issues Branch Division of Hur&n Factors Technology NRR Drew Persinko Maintenance and Surveillance Engineer

- Maintenance and Training Branch Division of Human Factors Technology NRR

SUBJECT:

TRIP REPORT FOR LASALLE AND D. C. COOK SITE VISITS REGARDING WRONG UNIT / WRONG TRAIN EVENTS This memorandum documents the activities and findings of an NRC staff visit to the LaSalle and D. C. Cook sites on November 12-15, 1985. Members of the NRC team for this visit included A. Ramey-Smith (DHFT) and D. Persinko (DHFT). The site visit was conducted as part of the short-tem effort to l detemine whether simple, low cost improvements can be identified and l implemented to reduce the frequency of wrong unit / wrong train events occurring at nuclear power reactor facilities. Upon completion of all site visits, the factors contributing to the events will be evaluated and a report issued which discusses causes and reconnendations. Long term assessment will be addressed as part of the Maintenance and Surveillance Program Plan being conducted by DHFT.

General Information ,

The LaSalle site is located 11 miles southeast of Ottawa. Illinoist There are two reactors. LaSalle 1 and LaSalle 2. located at the site, each of which has a maximum dependable capacity (net) of 1078 MWe. LaSalle I was placed 60.}. J

[-

. .s

'V

. into comercial operation on January 1.1984, and LaSalle 2 on ,

October 19, 1984. Both units are General Electric BWRs and the architect / i engineer for both units was Sargent and Lundy. The Ifeensee is Comonwealth j Edison. l l

The D. C. Cook site is located 11 miles south of Benton Harbor. Michigan, and l has two reactors. Cook 1 and 2. Both units are Westinghouse PWRs and the l architect / engineer for both units was the American Electric Power Service '

Corporation. Cook I has a maximum dependable capacity (net) of 1020 MWe and  ;

was placed into comercial operation on August 27, 1975. Cook 2 has a -

maximum dependable capacity (net) of 1060 MWe and went into comercial l

operation on July 1 '1978. The licensee is Indiana and Michigan Electric.

l Site Visit Agenda i

The discussions and in-plant observations centered around six wrong I j unit / wrong train events that occurred at LaSalle between 1983 and 1985, and i four that occurred at Cook between 1981 and 1985. The LER numbers for these events at LaSalle are 373-83-140 and 84-017,84-072. 85-012 ard 85-020 and at i

Cook are 315-81-005,83-009, 83-048, and 84-014. Also discussed at Cook was

_ the occurrence discussed in the Daily Report of 4/25/85. During both site visits, the NRC team inspected the locations of the reported wrong unit / wrong train events to the extent possible, and discussed the events with plant management as well as many of the individuals directly involved with the ,

i event. Enclosures 1 and 2 provide the sequence of events resulting in the LERs at LaSalle and Cook. respectively, the licensee's conclusions regarding L the event, and NRC staff observations.

Observations at LaSalle l

Color Codine. LaSalle utilizes a color coding scheme of light tan for l Unit Z and white for Unit I which are painted on the floors of the respective ,

units. The team notes that because of the similarities in the two colors.

one cannot easily distinguish between the colors to determine unit except at the dividing line between the units where the two colors meet and provide a discernible contrast.  !

plant labeline. Component labels do not correspond to the unit color coding

! schemebutrather,denotesystems(e.g., safety,nonsafety, water. air).

Metal embossed tags attached to valves are used for valve identification  :

(Figure 1) with p'astic engraved tags and black stenciled labels used to  !

identify electrical cabinets and other components (e.g., chain operators)

(Figures 2 and 3). Stenciled black labels are also used on the outside of doors throughout the plant to aid in the identification and location of equipment located beyond the door (Figure 4). Additionally, because of' the '

similarity of equipment in the electrical equipment room, a map showing the location of the equipment within the room is presented as one enters the room  :

} (Figure 5). Below the map is a listing of the panel numbers and a description of the panels.  ;

1 1

L 4

I l

' . 'rs $0 _ , W . ' . & < .

. . . >,. r. . -

A. V e

' .1 gi :722e ~ g .tl0 . c

~

y f ;{ ..,fe

. r.

a-

~

1. l, R $

1

. q\ b . . , W/ j '

. ,, s

^

l -% <

. e- Is -

r

' ,' M  ;

NS-y cx%^ -- n % = ,'=

e f, +v t-i\ + k>%'y

,c. i _.

4 -

h I4. Y k* Y j '

, y %_ Q,X 7.J'..r w;) .** * * .' +P( s A --

r. - . .

gc i

ikC2 M M Figure 1. LaSalle: Valve Identification Tag I

I t

l ,. ,

f n .1 -

35v. 3 .

.M. 2re r ~~

1 "M. ..EL1P..L r.

p.,uj. !g. . ..

- ~ '

j - . ,

1

' * ' ' + .

1 4

f Figure 2. LaSalle: Cabinet Labeling i

i 4

-.-..use. . .am I

.- z

-. . mz .

g si em~ m ein nous e e 'j '

,,l 9' M @y. *e

n. - i

- .) 3 .t

.. u- -

I J; I [

- [. I '

i

_ ,. w -

~

• h ., g . [I TECH. SPEC.

e- flRE DDDR

c. 4 i'

+ :sc . h- KEEP CLD5ED i

AT ALL TIMES i r ., .

'.' , . .~-d  ; ., l '

. :n

> US i

Figure 3. LaSalle: Chain Operator Figure 4. LaSalle: Labeling of Roons Labeling

/

o

, r. .. , , ,

"1 *

%s

,- e e.'i.1 4.m , . 4 l

! ,A;h M

{. - e. 2)

,t F .  :

)Q. l  : .;-.:'

. , y -

i h1,R ~

. . .. . i

-  ;. f43d!"~a. wu 'sy~

rip @:.g'h, f t- 9 h, .p,.

4

@. 2. *T(3 ?7-- - f,:d ,

g. - pt.n: m .- -

'N ( Cl h,

. ' ' .{

u

.g, .. . ,

3 j ' 1 -

f *d t

---

. t.J; j  ? ; i  ;; @ Esp

= _ , ,

.a.,. ,

j.

l -- = -

s'n 'S.,j] g g:j E '

Figure 5. LaSalle: Map of Electrical Panel Location in Electrical l Equipment Room.

g ',e i

s 1

s. ,

h i) I Qg td <. s

r s l '

~

\ \ )

' Except for.a few operating procedo'res, most operating procedures apply to l

'both units with Unit 2 component designations listed after Unit l'., ,

l When asked their opinion regarding' contributing factors to WU/WT events, a Commonwealth Edison staff member believed that communications breakdowns and a lack of attention were thie 'o'ominant contributors. I s . i Because Commonwealth Edison is' the licensee for both Dresden and LaSalle, T ,

matters directed from the corporate headquarters such as the investigation of \

events involving human error and training are handled identically for both aj' -

LaSalle and Dresden and have been previously described in the staff's WU/WT ,

trip report for Dresden (memo dated June 11, 1985, from Virgilio to Edison,N -

Regan,Bocher, Black). #

q.; g Observations at D. C. Cook l t

Philosophy. As stated by D. C. Cook personnel, a goal for the plant is to reduce the probability of human error and minimize their effects when they do I occur. The programs discussed below provide examples of,how D. C. Cook is l trying to implement this philosophy. s ./ Aq.x s

Color Coding. A color code is being implemented ta' differentiate tietween the two units, with Unit 1 being orange and Unit 2 being blue. The cabinets and panels are being painted, doorways are color coded and signs are provided. l and tank bands have been painted (see Figures 6, 7 and 8). Labelstare color '

coded by train and channel for safety grade equipment; channel 1 is orange, -

IL s

(

channel procedures 2 isareblue, channel color coded by 3,isprotection white, and channel 4 is yellow.s channel. t Y ,5 ~

A s  ;

Plant labeline. The D. C. Cook staff indicated that ev' erything' on flow prints and everything that has a human interface will be labeled.. Cook has  ;( l selected aluminum tags for labelleg components as follows: orange - Unft 1 l mechanical components outside of containment; blue - Unit 2 mechanical % 1 components outside containment; and, green - mechanical components on shared I systems outside of containments. Stainless steel tags will be used for labeling mechanical components inside both containments or in adverse ,

environments outside containments. The readability of the aluminum tags has been enhanced by increasing the contrast between the tagslettering and the tag background. This was done by fjling the raised letters to remove the\

color and contrast with the coloreditag backgrounds (see Figure 9). The same technique would not work with the stainless \ steel tags, resulting in what the

'NRC staff considered tags with very poor readability. ,,;

' >v .

s .

Each component is assigned a uniq've mark number. The first digit,is the unit designation; in the second series of digits', the first letter is the sy' stem, the second is the type of device (e.g;c fflan instrument), and the third l 1etter woule. indicate whether, for exawle; the component is a switch.' The e '

faci 11ty dmbase contains the official noun name of devices which kill be used on al.1 procedures and drawings. A list of standard abbreviations was s

\

' % , , ,, - ,J' 1

O 3 ,

.+

g ~

l 1W l

- \

.u

. l

-~

l9 ,

t x

i 4

Figure 6. Color-coded cabinets at DC Cook. Unit I cabinet is orange. Unit 2 is blue.

i l

- c 1 -

'b

. ::2. .

j .

r 1 . ,,,,

l [, ,

j SaM I, Figure 7. Color-coded doorway and entrance sign at DC Cook. t Entr3vay is blue. ,, l

-s j

I

, l 4

v e

.- i

.- i-

. t. . .s ,s .

Jy. _. . .

. . . . . ..p gy' ,.

. .s;g

1. 4

.fi e

' 3 d ~ ;. .,., . . ,  ;

1 4

Figure 8. Color-coded tank band at DC Cook. Band is orange.

l 4

l l

l l

. , l o

l l

i t

I 1

l i

i I

I 3

Nm d.

j

).I ,

n 3

0

!i ,

l

.i Y

s-a Figure 9. Color-coded component label with paint filed off '

lettering to improve readability. Also shown is a color-coded valve seal.

0 9

, ./

~

-~,w,--------... . ---- . _ _ _ _ - _ _ - - -------,-_----,-------,--_,n - - , -- -- --- _- - -

developed during conduct of the control room design review. The Cook personnel saw the plant labeling process as an on-going, iterative process.

A procedure exists for maintaining labels, and plant maintenance procedures include a check-off for the adequacy of labels on the equipment being maintained. An operator aids procedure has also been developed to control the quality of operator aids.

D. C. Cook personnel also described an effort underway to provide pipe

_ labeling on all plant systems. These labels (see Figure 10) are color coded according to unit and train and include flow arrows. The Cook staff indicated that this pipe labeling has been approved for use on stainless steel piping, but had not yet been approved for use in containment.

Electrical cabinets will be labeled inside as well as outside per a 4 commitment to IMPO. The Cook staff indicated that a circuit directory will also be prominently placed inside the cabinet door or nearby if no door is

.present.

! During the tour of the facility, the NRC staff was stunned by the gross lack of permanent labels in some areas of the plant and yet impressed at the stark

' ~

difference between those areas that had been upgraded and those yet to be done. Components in areas where permanent labels had not yet been installed were identified with duct tape and magic marker lettering.

+ Print Program. In mid-1982, D. C. Cook initiated its Regulatory Performance Improvement Program (RPIP). As part of that effort, a crew of plant personnel including those with drafting experience. started " walking down" prints. The following were the major aspects of that program: 1) to identify sequence of connection errors in which notes and references guide j the user to the wrong next drawing; 2) identify components on the flow prints without unique component numbers and input the unique number to the facility database; and 3) clarification of drawings to identify those that are not easy to understand. To assist instrumentation people and aid in labeling. Engineering Control Procedures were developed to augment infomation on flow prints. A unique numbering convention was developed.

Previous to establishment of the numbering scheme, construction numbers were used, a practice that did not address the users' needs. The Cook staff indicated that the present numbering scheme is much more user friendly.

Sealed Valve Program. Cook has instituted a sealed valve program to protect and give administrative control over manual valves. If a valve can inhibit a safety injection flow path, it gets a color coded (by unit) safety seal that is uniquely numbered. (Fire protection seals are red.) Periodically, the positions of these valves are verified. A procedure or clearance is required to change the position of these valves. ,

Comouterized Clearance Permits. D. C. Cook personnel indicated that a computerized clearance permit system was being estabibbed. At the present time, repetitive clearances have been computerized. Eventually, all

i l

l i *

..; ( (K2 ut Esw M ,,; ,

l1.wE ' " .*vn Figure 10. Color-coded piping labels at DC Cook. The borders of these labels are orange.

4 I

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

clearances will be incorporated into a computer system which will have the capability to check the clearance against the facility data base %nd will print out tags.

Miscellaneous. The D. C. Cook staff provided a hand-out that sumarized the activities completed or underway to reduce the frequency of events in which human error is a major contributor. That list is provided as Enclosure 3 to

. this report.

Exit Meetinos At the exit meetings at both sites,'the NRC team expressed its appreciation to th'e LaSalle and Cook staffs for their cooperation in planning the visit, c,ordinating the tour and discussions and providing pertinent information.

These actions made the site visits inforinative and productive.

The team would like to note that the staff at Cook was extremely well prepared for our visit. The handouts, containing labeling samples supplemented by the well-prepared presentations, assisted the staff greatly

- in understanding the W/WT events at Cook and Cook's initiatives to reduce human errors.

The team also expressed its appreciation to the Resident Inspectors at LaSalle and Cook fer *. heir assistance. Mike Jordon, the Senior Resident Inspector at LaSalle, was extreinely helpful in arranging discussions with the licensee.

N I

Ann Ramey-Smith, Engineering Psychologist Human Factors Issues Branch Division of Human Factors Technology, NRR Aud?

Drew Persinko, Maintenance and Surveillance Engineer Maintenance and Training Branch Division of Human Factors Technology, NRR

Enclosures:

As stated .

~

cc: See next page ,

4

6-cc: J. Funches -

F. Hebdon ~

G. Cwalina E. Trager V. Benaroya A. Bournia D. Wigginton mio 4

e 9

,D L.

I-Enclosure 1 WU/WTEVENTSATLASALLE(DOCKETNO.50-373 Unit 1)

(DOCKET NO. 50-374 Unit 2)--

I. LER 373-83-140 Incorrect Jumper Installation (Wrong Unit)

A. The following event information'was provided by the licensee in the LER:

'"On November 16, 1983 during removal of unit separation jumpers prior to performing LES-PC-02 and LES-PC-10, an incorrect jumper installation was identified. The jumper was installed on September 15, 1983 in accordance with LAP 240-3. Attachment 8. Jumper and Block installation form.

~

"The jumper should have been installed between terminals CC-79 and CC-80 t

in Panel IPA 14J to prevent a Unit 2 isolation signal from automatically starting the. Standby Gas Treatment System. The jumper was actually installed between terminals CC-79 and CC-80 in Panel 2PA14J which prevented Unit I and Unit 2 Standby Gas Treatment Systems from automatically starting on a Unit 1 Division 1 isolation signal. The jumper was innediately removed and a Deviation Report was initiated.

"LaSalle Unit I was in Cold Shutdown when the incorrectly placed jumper was identified. During the time that the jumper was in place, LaSalle Unit I was in start-up testing at power levels between Cold Shutdown and 100%.

" Electrician installing the jumper failed to note that the jumper form, j LAP 240-3 Attachment B, specified that the jumper was to be placed in j Panel IPA 14J instead of Panel 2PA14J. He was in the process of placing l numerous jumpers in the Unit 2 panels to prevent spurious starts of shared safety related equipment caused by invalid signals 'from Unit 2 (Unit 2 is in a pre-fuel load status), -

. o

'F "The operator that verified the jumper installation did not understand l the. significance of independent verification with respect to this job '

and failed to notice the "1" on the panel EPN. In addition, the operator that verified the jumper installation was with the electrician when the jumper was being installed which defeats the purpose of-independent verification. '!

"A Daily Order book entry was made by the Unit Operating Engineer on  ;

November 28, 1983 outlining the problem that had occurred and directing

.the Shift Engineers to begin implementing a time separation for independent verification of jumpers insnediately.  !

"A letter is being issued by the Station Superintendent to all .i department heads re-emphasizing the requirement for independent verification of items as specified in LAP 240-6 Temporary System Changes j and LAP 900-4, Equipment Out-of-Service Procedure.

"A discussion was held with the individuals involved with respect to ensuring that jumpers are installed in the correct location and the  !

necessity for independent verification.

"A discussion was conducted by the Master Electrician on November 29, 1983 with the Electrical Maintenance Department with respect to this event. The importance of ensuring that jumpers are properly installed and for independent verifications was stressed to all individuals." )

8. .NRC Discussions with Plant Staff During the site visit, the NRC team spoke with a member of the LaSalle staff about the event. From that discussion, it was learned that the incident occurred at a time when Unit I was in operation and Unit.2 was under construction. The valves in question were part of a fossnon isolation group whereby the valves from both units close upon a signal frem either unit. The purpose of the jumpers was to prevent a Unit 2 a -

l t .-

signal from affecting Unit 1. The logic is such that in order to bypass Unit 1 valves from being operated by a Unit 2 signal, the Jumper should have been installed in a Unit 1 electrical panel. The panels, located in separate rooms, were labeled on the front but it was unclear whether they were also labeled in the back where the jumpers were installed.

The electrician involved was considered to be experienced. The independent verification performed was actually dual verification due to a lack of understanding about " independent" verification by the independent verifier.

Currently, a separate procedure is used for each unit and the procedures now provide unit number cautions. Additionally, the panels are

- identified on the backside as well as the front. Panel identification also ext:ts on the inside of the rear door so that one will know the panel number even with the door open.

The NRC team was able to view the panels where the jumper was inadvertently installed and the panel where it should have been installed.

II. LER 373-84-071 Inadvertent Closure of Reactor Water Cleanup Outboard IsolationValve(WrongComponent)

A. The following event information was provided by the licensee in the LER:

"On 10/24/84 at 1330, the Unit 1 Reactor Water Cleanup system (RWCU, CE)

Outboard Isolation Valve,1G33-F004, inadvertently cicsed, due to a Group 5 Isolation Signal (JM) generated during the logic testing of the Primary Containment Isolation System (PCIS). At the time,of the.

occurrence Unit 1 was in Cold Shutdown. LES-PC-10. Primary Containment Isolation Manual Initiation Logic Test, required that specified control '

power breakers to valves be opened prior to testing the PCIS Manual i- _ _ _ _ _ _ _ ________._ ._ _ _

7 Initiation logic, to prevent spurious isolation of the systems during testing. These breakers were specified by the Shif t Forema'n on Attachment A of LES-PC-10. During the testing of the Division 1 PCIS Manual Initiation logic, the RWCU Outboard Isolation Valve, IG33-F004, unerpectedly isolated upon PCIS initiation.

"The RWCU Outboard Isolation Valve closed on the manual initiation signal because the wrong control power breaker had been opened.

Attachment A of LES-PC-10 had specified that breaker A4 at MCC 134X-1 be opened for 1G33-F004, instead of the correct breaker E5 at MCC 135X-1.

The reason for the wrong breaker being identified on the Attachment is due to the closeness in the equipment number of 1633-F004, instead of the Attachment is due to the closeness in the equipment number of

- 1633-2001-4 which is fed from MCC 134X-1 A4, and an error made when the control power checklist was initially developed.

"OperatingEquipmentAttendants(Non-LicensedOperators)havebeen instructed to look at the equipment name labels identified on the MCC breakers, as well as the MCC number and breaker identified by a checklist when removing and returning equipment from service.

Additional emphasis has been placed on the importance of independence of the second verification."

8. NRC Discussions with Plant Staff From discussions with the plant staff, it was learned that the shift foreman who generated the list of valve breakers to be operated during this surveillance mistakenly listed the power supply for valve number 1G33-2004 instead of 1G33-F004. The error was attributed to the closeness of the two numbers. The list had been generated from .

electrical drawings each time the surveillance was to be performed; however, subsequent to the event, a standardized list has ,beien developed. The plant staff member indicated that developing standardized outage lists is still in its infancy at LaSalle.

l

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

~

i

, -5 e

The itRC team did not view these breakers because physical location was not a contributor to this incident. -

III. LER 374-84-017 Reactor Scram on Loss of Feedwater (Wrong Component) l f

A. The following'information was provided by the licensee in the LER: j "The Unit 2 Reactor was scrammed manually on 5-3-84 at 2340 when it j

• becameapparentthattheMotorDrivenReactorFeedPump($J)couldnot  !

be restarted and reactor vessel level could not be maintained above 12.5  :

inches. At the time of the event, the unit was proceeding with a normal  !

l shutdown to obtain data for the Start-Up Test Program. l l -

"The NSO (Licensed Operator) on the unit decided that at that point in

' l the shutdown, it was convenient to close all the Reactor Feed Pump

, Warming Line Valves. These valves provide a flow path to the Reactor  :

Vessel even though the discharge valve on the Reactor Feed Pump is l l closed. With any Warming Line Valve open on any of the Reactor Feed Pumps. it provides enough make-up to the vessel that make-up would l exceed blowdown during low power operation and vessel level would rise ,

j at a rapid rate, which is not desirable. Therefore, the N50 on the unit

{

instructedanEquipmentAttendant(EA)(Non-LicensedOperator)toclose the weming line valves on all three feedwater pumps. The warming line >

! valves on the 2A and 28 TDRFP were correctly valved out. Immediately i after, the Equipment Attendant valved out the Balancing Line Valve 2CB037 on the Motor Driven Reactor Feed Pump instead of the Warning Line  !

! Valve 2FWO37. The Motor Driven Reactor Feed Pump tripped and could not ,

be restarted. Both valves, the Balancing Line Valve 2C8037 and the [

l Warming Line Valve 2FWO37, are in the same room and about five (5) feet I l apart. . .

[

l -

"The Equipment Attendant who closed the Balancing Line Valvar 2CB037  !

instead of the Warming Line Valve 2FWO37 indicated that he thought he  !

was on the correct valve because the number "37" caught his eye on the  !

I i

l

p

(

6-v valve Equipment Part Number. He did not read the entire ngun name of the valve that was indicated on the associated valve tag. T5e Equipment Attendant also thought he was on the correct valve when he read the number "37" on the valve tag because he had just closed the Waming Line Valves on the A and B Turbine Driven Reactor Feed Pumps and there was similarity associated with the piping and valves between the Turbine Driven Reactor Feed Pumps and the Motor Driven Reactor Feed Pump. The Equipment Attendant was not given any indication that a Balancing Line Valve 2CB037 existed or the consequences if it were closed."

The following corrective actions were taken:

"1. A Caution Card was placed insnediately on the Balancing Line Valve 2C8037 with instructions indicating that it should not be closed unless the Motor Driven Reactor Feed Pump is to be taken out of service (forbothUnits1and2).

2. Work Request L36421 was initiated so that repairs could be made to the Motor Driven Reactor Feed Pump. (Repairshavebeencompleted l and the Motor Driven Reactor Feed Pump is back in operation.)

3. Signs were placed on both Unit 1 and 2 Balancing Line Valves 1CB037 and 2C8037 indicating that the valves should not be closed unless the pump is out of service. This replaced the Caution Cards. I

4. AIR 01-84 67076 wassubmittedtoensurepersonnel(Equipment Attendants and License Personnel and all new Eeuipment Attendants) were trained on the purpose of the Motor Driven Reactor Feed Pump Balancing Line.

l

.e e

i e

i l

(

5. The following procedures were reviewed to ensure that ,the Reactor [

Feed Pump Waming Lines were in the proper position for the plant t condition: [

l

a. ' LOP-FW-01M U-1 FW Mechanical Checklist  ;

b. LOP-FW-02M U-? FW Mechanical Check 11st

c. LOP-FW-01 Feedwater System Filling and Venting

d. LOP-FW-02 Feedwater System Draining ,

e. LOP-FW-03 Start-Up of Meter Driven Reactor Feed Pump - .

Dange submitted I LOP FW-06 Shutdown of Motor Driven Reactor Feed Pump to f.

l l Not Standby - Change submitted  :

l g. LOP-FW-04 5 tart-Up of a Unit 1 TORFP - Change submitted  !

h. LOP-FW-05 Shutdown of a TORFP - Change submitted I i 1. LOP-FW-07 Preparation for TORFP 5/U with Main Steam l

! System Pressurized l

j. LOP-FW-08 Condensate Recirc to Condenser via Heater Drain  !

!. System l l k. LOP-FW-09 TORFP tahaust Duct Isolation During Unit  !

! Operation I 1. LOP-FW-11 Start-up of Unit TORFP - Change submitted I l m. LGP-1-1 Nomal Unit Start-up - Change submitted '

l n. LGP-1-2 Unit Start-Up to Not Standby - Change submitted

! o. LGP-1-3 Unit Start up from Not Standby to Power  !

Operation - Change submitted ,

p. LGP-2-1 Normal Unit Shutdown - Change submitted j

q. LGP-2-2 Shutdown to Not Standby - Change submitted l l r. LGP-3-2 Reactor Scram - Change submitted l

The above procedure changes will be tracked by AIR 01-84 47077. In f

addition, the procedures were reviewed to ensure correct reference i and use of the Balancing Line Valves." c.'

i l \

i

r .

l, -s-B. MC Discussions with plant Staff ,

plant staff told the MC team that this is not the first time this event has happened. The waming lines which were to be closed wem the reactor feedwater pump casing to reduce themel stresses in the casing

' whicharedevelopedfromthehightemperature(approx.350*F) developed in the reactor feed pump. The balancing lines, one of which was incorrectly closed. keep the meter driven reactor feedweter pump thrust

! bearing in position.

At the time of the event, the valves were labeled with metal embossed identification tags; it is unclear whether these tags were the new or-l old tags. Theequipmentattendant(EA)indicatedthatthiswasthe

- first time he performed this action and the training he received did not

_ cover this plant action and specifically, did not cover the differences between the waming and balancing lines. The operator was considered a l good operator.

As a result of this event, this plant action is covered in classroom and on-the-job training. Caution cards have been installed on pump control switches which indicate when the waming valves are closed and signs warning not to close the balancing valves witt. the pump running have been installed on the valves. The procedures in use did not say specifically which valve to close and required operators to look on drawings for the valves whereas now, procedures call out the valve to be closed. Additionally, instead of specifying FWO37 to be closed, another valve (FW115)inserieswithFWO37hasbeenspecified. The FW115 detegnation does not resemble FW037 and this valve is physically located outside of the room containing the warming line valves.

The MC team was able to view valve FW115: however, due to'rediological consideration, was not able to view the warming line valves'or balancing line valves involved in this incident.

l l

L.

l

-g-w-

The NRC team observed that metal embossed tags are difficul,t to read p unless lighting conditions are good due to very little contrast between j the raised identification numbers and the background.

! IV. LER 374-84-072 VR ! solation Damper Closure on Wrong Unit i

A. The following information was provided by the licensee in the LER:

i "On10/29/84at1952hoursanEquipmentAttendant(nonitcensedoperator) closed the Unit 2 Reactor Building Ventilation (VA VR) isolation  !

l dampers 2VR04YB and 2VR0$YA. This caused the Unit 2 Reactor Building  ;

I Ventilation system to shutdown. The Equipment Attendant had been ,

instructed to open the Unit 1 Reactar Building Ventilation dampers

~

IVR04YA 1VR04YB. IVR05YA. and IVR05YS but inadvertently closed dampers l

on the other unit. The Unit 2 dampers were reopened and the Unit 2 i

Reactor Building Ventilation system was restarted at 1953 hours0.0226 days <br />0.543 hours <br />0.00323 weeks <br />7.431165e-4 months <br />. ,

"At the time of the incident Unit I was in cold shutdown and Unit 2 was l

in the "Run" mode at 735 power. The Reactor Building differential  :

pressure with respect to outdoors was being maintained by Unit 2 VR.

l. Unit 1 VR had been shutdown so that LaSalle Operating Procedure l l LOP-RP-01 could be performed. This was used so electrical power could l l be transferred from Reactor Protection System (EE. RPS) "B" Motor (

Generator Set to Alternate Power. As a precaution against momentarily ,

losing power to Unit 1 RP5 the Unit 1 VR system was shutdown and ,

secondary containment isolation dampers IVR04YA. IVR04Y8. IVR05YA. and IVR05YS closed. If power were to be momentarily lost to the damper l actuators during the transfer the isolation dampers would fall closed and shut.

i

! "After the power has been successfully transferred the Unit 1 '

Control RoomOperator(N50)instructedanEquipmentAttendant(EA)'.toopenthe f

Unit 1 VR isolation dampers so the system could be restarted. The EA inadvertently went to the Unit 2 isolation damper control panel 2PL27JB

located in the Auxiliary Building (NF) and closed dampers 2VR04YA and j L

2VR05VA. The dampers promptly closed shut causing Unit 2 VR to l i . shutdown. The EA still in the Auxiliary Building called the NSO who l infomed him that he had closed the dampers on Unit 2 instead of opening ,

the dampers on Unit 1. The N$0' told the EA to open the Unit 2 dampers ,

i and then go over to the Unit 1 isolation damper control panel and open l l

the Unit 1 dampers. The EA opened the Unit 2 dampers and Unit 2 VR was i

restarted at 1953 hours0.0226 days <br />0.543 hours <br />0.00323 weeks <br />7.431165e-4 months <br />. ine EA then went over to the Unit 1 VR damper .

I control panels 1pt27JA and 1pL27J8 and opened dampers IVR04YA IVR04Y8  !

! IVR05VA. and IVR05YS as originally instructed. Unit 1 VR was then l I started up at approximately 2000 hours0.0231 days <br />0.556 hours <br />0.00331 weeks <br />7.61e-4 months <br />. l 1

"The station initiated an investigation and detemined the incident was the result of an error made by the Equipment Attendant when he closed l L the dampers on the wrong unit. The damper control panels were labeled j correctly'and did not contribute to the error made." l r

! 8. NRC. Discussions with Plant Staff  !

As stated in Section A the EA was supposed to open Unit 1 VR isolation L dampers, pe* verbal instructions from the control room operators, and instead, went to Unit 2 and closed the Unit 2 VR isolation dampers. The L dampers are operated locally from the auxiliary building, which is

! comon to both units. The controls for both units are on the same elevation with Unit 2 controls located on the north end of the auxiliary building and Unit I controls on the south end. The outside of the damper control box is shown in Figure 2 and upon opening the door, the j controls which the operator inadvertently closed are shown in Figure 11.

l The operator manipulated two Unit 2 control room damper boxes which were l approximately ten feet apart. At the time of the incident, the boxes were labelled with the plastic identification tags shown in Figure 2.

i No independent verification was performed as the control roo's operator I caught the error from a warning light in the control room shortly after the incident occurred. No explanation into the cause was put forth l

l \

r 1

P a g y.A.y:s y ; .4.) s e ,,es..+.vi,.,p g p, eA x.e 4

..g ,

($. c$a.S 5' g-A

. k' 1 g , T.

p

' c Q.

. am o *.. , . .

. ,.g l

y .

~

(. .. ' 6 .

i

%' -j s mi i l

.. wr

! (:- i. " . .

t .

i Figure 11. LaSalle: LER 374-84-072 Inside of Damper Centrol Box. i i

4 l

1

\

i i

i ,

1 l

i t

I I

_~ _. - -___ - _ __ - ______

L

. 11 other than the operator must have had in his mind that he was required to change position of the dampers, whereupon after entering the wrong unit, the change in position closed the dampers. The NRC team observed the color coding described above which was painted ento the floor of the auxiliary building at this level.

4 V. LER 374-85-012 InadvertentRHFShutdownCoolingIsolation(Wrong

! Component)

A. "At-1103 hours on March 31, 1985, with LaSalle Unit 2 in Cold Shutdown, a Residual Heat Removal (RNR, 80) Shutdown Cooling isolation occurred during the performance of surveillance procedure, LIS-RH-10. "LPCS/RHR

. Injection Line Integrity Monitor Calibration and Functional Test.

~

i "A Group 6. Division 2, Primary Containment Isolation (PCIS, JM) occurred on an apparent RHR suction high flow signal. The RHR Suction Inboard Isolation Valve, 2E12-F009, automatically closed. All other Group 6. Division 2, isolation valves had been previously closed. The l l

l operating RNR pump, 2E12-C002A, tripped as required on the closure of l l valve 2E12-F009.

l' L

f "The Instrument Maintenance Technician performing surveillance LIS-RH-10 ,

promptly notified the Unit Nuclear Station Operator (N50) of an L . instrument valving error. This was identified as the cause of the i

isolation; RHR Shutdown Cooling piping integrity was verified, and the l RHR loop was returned to operation at 1107 hours0.0128 days <br />0.308 hours <br />0.00183 weeks <br />4.212135e-4 months <br /> on March 31, 1985. j "The 2E12-F009 isolation occurred due to the valving out of instrument 2E31-N0128. Under LIS-RH-10 instrument 2E12-N0298 was to be valved out l

!- but the Instrument Maintenance Technician performing the surveillance incorrectly traced the piping from instrument 2E12-N1298 and without 1

l l looking at the tag on the instrument isolation valve, mistak'enly l 1solated a line to instrument 2E31-N0128.

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

1 12-v- i "2E31-N0128 is a Barton 288A Differential Pressure Indicating Switch (DPIS) and is physically located next to 2E12-N0298, anothef Barton 288A DPIS. 2E31-N0128 senses differential pressure caused by flow through the RHR Shutdown Cooling suction piping. When the low pressure sensing line to 2E31-N0128 was mistakenly isolated, a lower than actual pressure was sensed at the low side of 2E31-N0128. With the high pressure side unchanged, a high differential pressure (high flow) condition was simulated and 2E31-N0128 actuated.

"The Instrument Maintenance Technician's incorrect tracing of the piping

• from instrument 2E12-N0298 and failure to read the valve tag before moving the valve caused the March 31, 1985 occurrence.

~

"An investigation followed this occurrence and a meeting was set up between the Instrument Maintenance Technician involved, the Foreman in charge of the work, and the Assistant Superintendent of Maintenance.

The events of this occurrence were discussed and the need for attention to detail was stressed.

"The Instrument Maintenance Department will be trained on this occurrence with a focus on positive identification of instruments to be valved out. The completion of this training is being tracked by AIR 01-85-67055."

8. NRC Discussions with Plant Staff The NRC team was unable to view the instrument lines involved in this event due to radiological considerations; however, the team was able to view a typical instrument rack located on the CRD system. Through discussions with the plant staff, the team learned that the instruments on the rack sit close together; however, the area is not very crowded.

Although the instruments and the valves are labeled, the valve labeling is not used because they are not considered to be trustworthy. As a j result, the instrument maintenance technicians are instructed to hand trace the lines, a process which is done frequently. The instrument i

maintenance technician, in this case, was relatively new with ,

less than 1 year experience, and may have been nervous due to the high consequence oferror(reactortrip). As a result of the event, training now covers hand tracing.

i The team notes that although the LER attributes not looking at the instrument valve tag as a contributor, such practice was comunon because l the tags were considered unreliable. Had the tags been maintained and considered reliable by the technicians, they may have been used whereupon, a discrepancy between hand tracing and the valve tag most likely would cause a technician to check out the discrepancy before proceeding with his intended functions. The plant staff indicated an intent to eventually upgrade instrument valve tags so that they are

- considered reliable.

VI. LER 374-85-20 Missed Service Water Sample (Wrong Component) 1 A. The following information was provided by the licensee in the LER:

" Unit 2 Service Water (KG) PRM (IL) was declared inoperable due to low sample flow at approximately 1830 hours0.0212 days <br />0.508 hours <br />0.00303 weeks <br />6.96315e-4 months <br /> on April 23, 1985. A sample was not obtained until 0620 hours0.00718 days <br />0.172 hours <br />0.00103 weeks <br />2.3591e-4 months <br />. April 24, 1985. This was contrary to Technical Specification 3.3.7.10 which requires samples to be taken at least once every 8 hours9.259259e-5 days <br />0.00222 hours <br />1.322751e-5 weeks <br />3.044e-6 months <br />. At the time of this event Unit 2 was in ,

Mode 4 at 0% power.

CAUSE An investigation was conducted as a result of the missed sample from the i-Unit 2 service Water PRM. The Unit 2 "A" RHR (BI) Service. Water.PRM was mistakenly sampled instead of the Service Water PRM. -

.f ,

"According to a Unit 2 log entry made at 1830 hours0.0212 days <br />0.508 hours <br />0.00303 weeks <br />6.96315e-4 months <br /> on April 23, 1985,  !

the Unit 2 Service Water monitor was declared inoperable due to a low  :

i

. ~ _ _ _ - -

14 r  ;

flow alarm with no adjustment for higher flow. The Center Desk Log entry ,

made at the same time essentia11y said the same thing. ".. l l "According to a log entry made at 1830 hours0.0212 days <br />0.508 hours <br />0.00303 weeks <br />6.96315e-4 months <br /> on April 23, 1985, by a Rad l ChemTechnician(RCT),theUnit2ServiceWaterPRMwasdeclaredINOPas of 1830 hours0.0212 days <br />0.508 hours <br />0.00303 weeks <br />6.96315e-4 months <br />, l

"The responsible Red Chem Foreman was coming on shift and was taking a turnover when the call came in. The RCT who took the call showed the l l Foreman the log entry and was acknowledged. About 5 minutes later, the l Foreman called the hot lab and spoke to the RCT on duty. Direction was given to the RCT to sample and analyze the Unit 2 "A" RNR Service Water PRM instead of the Unit 2 Service Water PRM. This was done, and the Unit 2 Service Water PRM sample was not taken.

l "During subsequent shift turnover around 0605 hours0.007 days <br />0.168 hours <br />0.001 weeks <br />2.302025e-4 months <br /> on April 24, 1985,

! the mistake was discovered by the oncoming Red Chem Foreman. An RCT was immediately dispatched to take a Unit 2 service water sample. This was completed at 0620 hours0.00718 days <br />0.172 hours <br />0.00103 weeks <br />2.3591e-4 months <br />.

"The cause of this missed sample could be attributed to failure to pay attention to detail. Ineffective comunication was also a contributing factor.

"The responsible Red Chem Foreman was counseled by the Rad Chem Supervisor as to the Foreman's responsibilities when comunicating a f verbal message. In addition, the Rad Chem Supervisor will issue a department memo providing additional guidance for effective verbal f communication. This will be tracked by AIR 374-200-85 00066. ,

8. NRC Discussions with Plant Staff -

.f Through discussions with the plant staf f. It was learned that RHR >

service water sampling was routinely performed whereas service water l

m 35-radiation monitoring was not. As a result, the rad ches foreman developed a mindset and transmitted instructions to monito/ the RNR service water system.

t i

e e

s 8

e

_m

ENCLOSURE 2

?

l WRONG UNIT / WRONG TRAIN EVENTS AT D. C. COOK ,

I

1. LER-315-81-005 - Safety Injection Yalve treaker (Wrong Unit)

The following event infomation was provided by the licensee during the site <

visit:

"An auxiliary eouipment operator with approntmately 11/2 years of  ;

experience was dispatched by the Unit Two supervisor to hang a clearance on a series of motor operated valves for the U 't Two  !

Safety Injection pump. The operator then mistakenly went to Unit  !

One and opened the breakers for the specified valves. This error

~

was immediately recognized by the Unit One control room supervisor  !

l

and power was r1 stored in approntmately two minutes.

i o

• f

'This event occurred on the afternoon shif t at about 10:30. The operator involved had not worked any overtime during the time

, 1

! surrounding the event. The area this event occurred in was the l t

! Aust11ery Building lasement. This area is a well lit, controlled i (radiologically) and spacious area in which to work. At this 1

time the Motor Control Centers involved were the same color on l both units. The only labelling that entsted at this time was black component tags for each breaker and these were not con-l sistently available for all breakers.

"As a result of this event an operations memo was sent to all l shif ts concerning operator attentiveness. The intended f6nction I

of the ICC$ system was not inhibited by this event."

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

In discussions during the site visit, the licensee's staff indicated

~

that the auxiliary equipment operator (AE0) had been interrupted during the task of hanging the clearance tags. The AE0 had gone to the Waste Disposal System (WDS) because an alarm had sounded there. He spent five to ten minutes handling the situation there before returning to the tag hanging task. The breaker panel that he mistakenly went to (Unit 1) is located near the WSD. The AE0 realized his mistake as he walked to the Unit 2 area to perform his next task

. and realized that he had manipulated the breakers on Unit I rather than Unit 2.

i It was the opinion of the licensee staff involved or familiar with the

! event that the interuption of the flow of work was the major contributor to

~

l the error. Another possible contributor discussed include the reliance on magic mart labels at that time, many of which did not provide complete unit or system  !

l , designations.

2. LER-315-83-00g - Spray Additive Tank outlet Valve (Wrong Unit)

The following event infomation was proviJed by the licensee during the site visit:

"During a routine tour an aux 111ery equipment operator discovered that the Unit 1 spray additive tank outlet valve was sealed closed. Further investigation of this event disclosed that this valve was inadvertently left I closed following the perfomance of 1-OHP 4030.5TP1007 (Unit One Containment

$ pray System surveillance.) The AE0 who performed the independent verifica-tion initially looked at the wrong unit's valve. Realizing his' mistake he checked the correct unit's seal number but failed to verify the,fralve position. i This in turn caused the loss of both trains of Containment $ pray to utilize the Spray Additive tank.

l l

l

F- , n

!- "Several reasons can be traced to explain this event. The Spray

, additional tank room contains both unit one's and unit two's

! spray additive tank. The AE0 involved in the initial lineup had been working on other major surveillance procedures during the l shift. The Reactor Operator involved was'.' assigned to the other Unit and crossed units to work on the lineup and the general work l load of the shift was heavier than normal.

"As a result of this event, procedires were split up and re-scheduled to aid reducing work load, 'na operations memo on shift manpowerwasdistributed,andlabellinhenhancementsofthearea were made." -

~

! In discussions during the site visit, the licensee's staff indicated that the procedures being run at the time of the event were time consuming'.

and involved as many as five people to run one pump. As indicated, that procedure was subsequently broken into smaller procedures. Other. '

potential contributors to this event exist as well. The second verifiel ,

t was to verify valve position and seal number. The independant verifier .,.

originally went to the Unit 2 additive tank rather than the Unit 1 tank. /

He became concerned that a mistake had been made when hanging the other tags. Then realtning his error, he then went to the Unit 1 tank and verified that the valves had the correct seal number. What he did not check was whether the valve was in the correct position (which it was not). The course of events that led to the valve not being in the correct position '

was described as follows. The valves in question are difficult' to operate .

in that it is difficult to " break the seal"; but when the stal-[s broken, ,

the valve stem suddenly spins free. When the AE0 involved in the originel

x .g

- t

-y '

l

, , . i

, valve #Iineup manipulated the vb1ve the sudden free spin of the valve caused '

v l

e'

'ths AE0'to loose !)is balance somewhat and the seals in his pocket to sp111 '

f on the floor. He picked'up the seals and placed the seal on tk valve with-

-out completing the tesE of opening the valve. Another potential contributor t

to the errors invcTve'd in this event is the independent verification process i

-. 4 7 l 4

in plate at the titw. The licensee's staff indicated that at the time of the l event 1 / seined that there was as much emphasis put on recording the seal number t . .

f. ,

as checking the position durini verification tasks. Now the independent  !

verifics[1on practice has been changed such that seal numbers are recorded on a weekly seal tour, rather than when the independent verification is being l perfomed. This has shifted the emphasis back to verifying posititon.  :

3. LER-315-83-048 - C ntainment $ pray Heat Exchanger (Wrong Train)

The following ebnt information was provided by the licensee during the site visit: / -

/ .

"$urvef11ance test procedure 2-OHp 4030.5Tp.007 requires that the f

inlet and outlet v&lves be tagged closed on the Containment Spray I

, Heat Cachanger under test.

^

/*Two operators (one AE0 and one R0) utilizing the independent verific'ation system closed the inlet valve on the heat exchanger

, not' under test. Since the outlet valve on the heat exchanger l' urgerftest was closed, both trains were simultaneously inoperable.

,~ ,.'/

., /  ;,-

"%e Containnent $ pray-Heat Exchangers are located on the 609'

/, #

.i l elevation via the Aux Building. The heat exchanger rooms'are  !

1 /. /

' mirror images; that is, both rooms have exactly the same components l

i .- -

i

.y ,,

. . i F

but are positioned opposite of each other. y e o <

^

?  %

o .

"The valves are labelled and sealed in their respective positions.

. ~

s e 1 i "Several factors contrib ted to this event taking place: *J-i A ., _ 's

1) The Auxiliary Equipment,0perator. involved was inexperienced 9 (first time on the job) on the.Intrve111ance test..

2) The work load on the shif t ws; heavy and the second verifica- .

tion was performed by an operator assigned to another job. He  ;

st

)

sided the AE0 in the valve manipulatica inadvertently adding to the error.

s s

~

3) A third verification was perfomed by the Unit Supervisor > ,

A I who discovered the error and corrected the problem. ,

~ l The time that both heat exchangers were out of service was approxi- l mately 15 minutes."

Discussions with the licensee's staff indicated that the AE0 who manipulated the wrong valve was tired, was on the first night .

of the 12:00 midnight to 8:00 a.m. shif t, and had not had much sleep the night before. The error occured at approximately 4:00 e.m. In addition, the individuals involved mere in a hurry and "way behind ,

u.

schedule."

The valve in question is ins'ide containment and requires the use of u a reach rod to manipulate it. The operator invol'v'ed had traced out the reach .. '

1 rod but to the wrong valve. The operator manipulated the valve and t,en h '

l

& (

.went to check whether the valve had moved. The position of the valve was hard l

-.. - , q to see from the platform and there was no rising stem on the valve. It did not I

seem to the operator that the valve had moved so he checked the same valve on L

2

- , , - , ~ . , . , . , - - . , , , , . . . - , - - - - - - .

r . - , - nm n,, , - , yme,,-,.,,,.-- .nm,---.,--,.y '.,'p-,,,.y.-.rw e 6-,n,-ws.,

g 44 9 E%

9 s

x

(; <

the other unit. When the valve from the other unit was comared to the q valve in question, the two valve positions were different. The operator assumed that the valve he had manipulated was open. It did not occur to the operator that he was on the wrong train. -

Labeling available to the operator could have been improved. The reach rod had a tag on it, but a more easily readable large, magic marker label 4 was printed on the wall. This label did not contain the train designation.

In talking to the operator who was doing the dual verification, he indicated Yhat he does not believe that he looked at the label because he had done the procedure before. He also was not there when the other operator, who had T. , . hurried on ahead, manipulated the valve. Consequently, he had not provided i

' s a dual verification. The independent verification perfomed later did find e,N the error.

i

4. LER-315-84-014 - Safety Injection Pumps (Wrong Train)

The following event information was provided by the licensee during the

. site visit:

"An Auxiliary Equipment Operator while in the process of performing a surveillance test (1-0HP 4030.STP.005) on the South safety injec-tion pump inadvertently isolated the North safety injection pump.

. Since the Reactor Operator had isolated the South pump in the control room, this rendered both safety injection pumps inoperable.

"The safety injection pumps are located in two adjoining rooms and are designated North and South. At the time this surveillance -

. was conducted, this area was contaminated and required anti-contami-nation clothing.

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

f "The event was caused by personnel error primarily due to the breakdown in comunication.

~

1) The Reactor Operator involved in performing this test was doing so for the first time. Evidence of ineffective comunication between the R0 and AE0 which included uncertainty as te which pump was to be isolated led to the occurrence of this event.

' 2) Also the procedure utilized did not differentiate between trains except for brackets which were placed around South / West train com-ponents.

3) As a result this procedure has been divided into independent .

train procedures. This will aid in minimizing the chances for

_ crossing trains during surveillance testing."

The event was discussed with plant personnel familiar with or involved

, in it. potential contributors to the wrong-train error that were discussed

. are as follows. It was a very hot nignt with inside temperatures estimated at about 100 degrees, and the AE0 had to wear anticontamination clothing. It was 4:30 or 5:00 a.m., and the AE0 had not had a break. Because of the temperature, valves were turning hard. The control room operator had never run this procedure before, a very long and complex procedure. The AE0 had also never l run this procedure. General practice is for the AE0 to complete his/her tour prior to running any procedures. Before the AE0 had completed tour, the control roon operator called the AE0 from the control room to begin manipulating valves as called for in the procedure. The procedure involved is referred to as a

" double star" procedure which means that the AE0 must carry a copy of,the procedure when working on that procedure. The AE0 diti not have a copy of the procedure so returned to the control room to get a copy. The control room operator did not mark up the procedure, which entails customizing it for the train being tested, by crossing out the component identifiers for the other train.

, - , - . , . - , , , , --.w.r, ,m-e,w,,,m c q ,; e,,,m,,,,,,v.,m,m._e4,-,.m,,,,,,- __,,_,.-m,emmman,,w__w,.,---ene,,,--,,wn-

The control room operator had apparently been anxious about running this procedure. He was trying to hurry the AE0 along, but the valves.were turning hard which slowed the work. The AE0 pot to a point in t,he procedure

~

that had not been marked-up to indicate the proper train. Rather than call the control room operator, the AE0 assumed which valve was supposed to be manipulated and turned the wrong one (i.e. the North train). The AE0 had worked on the North train the night before, but the procedure being run the night of the event was for the South train. The error was discovered when the AE0 called the control room operator and told which valve had been operated.

The NRC staff agrees with the licensee that the major contributor to this event viss poor communication. That poor cosumnication was evidenced by the lack of discussion of the procedure before the procedure was initiated, the lack of properly marked-up procedures for the use of the AE0, the.

frustrated and harried atmosphere developed as a result of verbal comununications between the AE0 and control room operator, and the AE0's decision not to call the control room operator to get clarification of the valve to be manipulated.

5. Event Report C/R 02-04-85-836 - Containment Spray (Wrong Unit)

The following event information was provided by the licensee during the ,

site visit:

l l

e e

a v ~ev , e, , , ,em - w -- w e e. m-g,e, .me--wma y,-,,,.,,-ww. men.-w w w a mm m w-,w, , ,. ww--- ,w ww vww, w me w-*wwm eo- m m e _ _ -one--w-

~

"An Auxiliary Equipment Operator was assigned to place a clearance on the Unit 1 West Containment Spray pump. The operator ,

correctly placed the tags on the Control Room switches end the l pump breaker in the switchgear room. The operator then proceeded )

to the auxiliary building to hang the remainder of the tags, once in the hallway which separates the Unit 1 and Unit 2  !

Containment spray pumps the operator went to the Unit 2 side via

the Unit 1 side. The operator then closed the manual valves via  !

the remote operators which protrude into the hallway. Then the l operator went into the pump room to hang a tag on a motor l

operated valve handwheel. After the operator entered the room he .

noticed that the motor operated valve was open when it should  ;

have been closed since the valve was closed prior to hanging the tag on the control switch in the Control Room. The operator not realizing his error, went to the control Room informed the Unit Supervisor of a possible limit switch problem. The Unit l Supervisor asked the operator if he went to the right Unit. The i operator was sure he had and so it was decided to write a Job Order on the limit switch. When the operator went to hang the l Job Order, he discovered his error. At this point, approximately 1 15 minutes after the error, the Control Room was informed and the Unit 2 valves were opened.

_ "The hallway where the error occurred was well lighted and was not  !

in a contamination area. The entrance ways to the pump rooms are j painted blue and orange, and the valves were clearly labelled.

" A contributing factor to this error is that the operator was working non-routina hours in that he was coming in at 0330 and working until 1130 as an extra person during what normally would have been a training week. "

• I 1

1 l

9'

'f i

- - - - - - - - - - , - ., -- --m --

,-. . r, .r..~ . , , _ , _ . , - - . . , ,, y _ www-m w

,, i .

Enclosum 3

• Provided by D. C. Cook ENHANCEMENTS TO PRECLUDE HUMAN ERROR TYPE EVENTS

1. Color coding'the Plant units. (Unit 1 orange, Unit 2 blue.)

' Carried out through tags, labels, job orders, cons ~idering for procedure.

2. Independent verification Program. ,

3. Electrical / wiring drawing enhancement.

4. Revision to piping flow diagrams being redrawn for clarity.

Creating many accuratescontrolled drawings for where none i existed. (Fire protection, Plant air, turbine fluid control, etc.)

5. Procedures being identified by color (Unit 2 cover sheet yellow). Considering changing cover sheet color to i

correspond to unit color.

i 6. Job order revised with a '2' superimposed on the Job Order

- for Unit 2 jobs.

7. Labeling and walkdown program, Operator aids program.

S. Full page change sheets instead of pen and ink.

9. Reducing complex procedures into several more distinct and

applicable ones.

i

10. Personnel Error Forms for Condition Reports. Opportunity to i F

discover contributing factors such as human factors 4 contribution.

11. Opening of Condition Report process to identification of drawing discrepancies. Statistics available.

12. Prohibition of shift people assigned to one unit from working on other unit except under prescribed conditions (certain i

tours and independent verification are cross unit) .

13. Issue of Unit One and Unit Two procedures versus 12.

14. Facility data base effort.

15. Computerised Clearance Permit Program and creation of ,

computerized " standard" clearance boundaries for, standard repeat type jobs. (saves time and decreases opportunity '

for error.)

{

F '

l

• l ENHANCEMENTS TO PRECLUDE HUMAN ERROR TYPE EVENTS

16. Creation of master clearance on which all work needing to be

'done is collected, rather than issuing / hanging individual clearance / tags for each work item. (Equal protecti~on, saves time, reduces chance for error.)

17. Increased mobility of persornal among departments particularly licensed operetors transferred to Planning, QA, etc. for benefit of that t. raining / experience to others efforts.

18. Use of Department outage coordinators and much more attention put on planning / coordinating jobs to reduce confusion / error.

19. Use of experienced revision SROs or Operations Outage Coordinators. Reduces outage load on Shift Supervisors and

- improves coordination of outage activities with operating

. unit, Tech. Specs., etc.

20. (Future) Definition within Plant systems of ' maintenance

- areas" to which job orders will be keyed by computerized job order system and for which standard clearance boundarias will be computerised. .

21. Direction given "If it isn't labeled, confirmable on an approved drawing-1 a) cannot operate it, put at least a temporary tag on i b) canr?.t use for isolation point" '

I 6

,e O

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