ML20214E335
| ML20214E335 | |
| Person / Time | |
|---|---|
| Site: | Turkey Point  |
| Issue date: | 03/06/1986 |
| From: | Trager E NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD) |
| To: | Black K, Booher H, Rowsome F NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD), Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML17342A418 | List: |
| References | |
| NUDOCS 8603120147 | |
| Download: ML20214E335 (20) | |
Text
h.
) 110 s
e pa neg'o UNITED STATES e
~g
[
g NUCLEAR REGULATORY COMMISSION D
g WASHINGTON, D. C. 20555
\\.*/
' 6 IN6 EAR MEMORANDUM FOR; Harold R. Booher, 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 Program Analysis Staff, NRR FROM:
Eugene A. Trager Nonreactor Assessment Staff Office for Analysis and Evaluation of Operational, Data
SUBJECT:
TRIP REPORT FOR TURKEY POINT SITE VISIT REGARDING WRONG UNIT / WRONG TRAIN EVENTS This memorandum documents the activities and findings of an NRC staff visit to the Turkey Point site on November 21-22, 1985. Members of the NRC team for this visit included A. Ramey-Smith (DHFT), D. Persinko (DHFT), and E. Trager (AE00). The site visit was conducted as part of the short-term effort to determine whether simple, low cost improvements can be identified and imple-mented 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 recomendations.
General Information Turkey Point units 3 and 4 are 725 MWe (net) W three-loop PWRs. The station is located on the shore of Biscayne Bay, 25 miles south of Miami, Florida, adja-cent to oil and gas fired units 1 and 2.
The major structures are two containment and turbine buildings with one auxiliary and control building. The reactor coolant systems of the two units are essentially identical and do not 1
share any components.
The licensee for Turkey Point is Florida Power and L'ight, and Bechtel was both the architect engineer and the contructor. Unit 3 achieved initial criticality 8(posta o/$7 XA
.weP
.' Harold R. Bocher et al. MAR 6 1986
~
on October _20, 1972, and began commercial operations in December-1972. Unit 4 achieved initial criticality on June 11, 1973, and began commercial operation in September 1973.
Site Visit Agenda The discussions and observations centered around five wrong unit / wrong train (WU/T) events at Turkey Point. The LER Numbers for the events are 250-83-007, 251-83-016, 250-84-003, 250-85-004, and 250-85-017.
During the site visit, 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 events. Enclosure 1 includes descriptions of the events, the licensee's conclusions regarding the events, and NRC staff observations.
Observations at Turkey Point The consnitmekt by the licensee to reduce the number of human errors at Turkey Point was indicated by a number of improvement programs that are currently underway. Turkey Point has had a Performance Enhancement Program (PEP) for several years. The PEP was responsible for a program to upgrade plant proce-dures and another to install embossed metal tags on all plant valves. Then a gQuality Improvement Program (QIP) was introduced about one and one-half years ago to upgrade plant activities with respect to human factors considerations.
The "Sy' tem Enhancement Coordinator" along with a six person QIP team have s
studied plant operations and have made recommendations about changes that are needed. QIP changes include an improved training drawing program and a plan to install more-legible Fiberglas tags on valves.
The licensee felt that the PEP was an adequate method of reducing errors and' decided not to take part in the INP0 Human Performance Evaluation System (HPES) pilot program.
Labeling and Identification Permanent 'name labels do not exist for all controls, displays, and other equipment that must be located, identified or manipulated by plant personnel.
although Turkey Point is working toward this goal. Licensee efforts to improve labeling began in April 1983 following a complete loss of the Auxiliary Feed-water (AFW) system for five days (wrong train event), and were further intensi-fied in October 1983 following another event in which the containment spray system for the wrong unit was isolated. That $100,000 was budgeted this year (1986) for labeling, color-coding and other identification improvements shows a commitment by the licensee (this type of work was never budgeted for previously).
As had happened on other site visits, the NRC team was reminded of the lack of standards for the labeling and identification of nuclear plants. Licensee staff noted that ANSI A13.1-1981 is' inadequate, because it has a limited number of colors (4) (while a nuclear plant might have more than 60 separa,te systems),
and suggested that a regulatory or industry initiative in this area would be helpful.
Turkey Point is attempting to use consistent nomenclature in procedures and labels, and to have drawings that show the "as built" rather than "as designed"
Harold R.,B:oh:r et al. MAR 6 1986 condition.
It was noted that every piece of equipment has been stencil painted and that new labels (like on the AFW system) have the unit number and unit color code and the train designation.
The NI channels are color-coded red, white, blue, and yellow for Channels I, II, III, and IV, respectively. - This is a 2" standard.
Labels currently exist that are not readible (lettering too small, view of label is obstructed, poor label contrast, not enough light available, etc.).
However, an effort is being made to correct this.
Operators on a training cycle who want to work overtime can do a walkdown of labeling against drawings.
This is repeated as time' permits (e.g., the fire protection system has been done twice).
The Turkey Point Quality Improvement Program.(QIP.) team developed " Visual Instructive Plant-[ VIP] QIP Recommended Plant Actions" to help reduce personnel errors.
A " Phase 1" project to label plant valves with embossed aluminum tags was completed 10/22/85 (>17,000 tags).
A " Phase 2" valve tagging project wil!
relabel all plant valves with a fiberglass tag that has improved legibility and durability compared with the Phase 1 tags (>1500 of these fiberglas tags were on order). The Phase 2 program uses color-coded Fiberglas tags that contain information on the valve number and nomenclature and the normal 100% plant operating position.
The current color-coding scheme for safety systems is as follows:
Unit 3 component is Tan.
(Tan has 3 letters)
Unit 4 component is Blue.
(Blue has 4. letters)
Common component is Orange.
These colors also permit the use of contrast in distinguishing between units.
Than is, labels are tan with black letters (dark on light) and Unit 4 labels are white on blue (light on dark).
This labeling scheme is illustrated by Exhibits 1.2 and 1.3 in Enclosure 1.
Turkey Point. began the use of " reverse-engraved" tags / labels / signs.
In this process a sheet of clear plastic is coated on the back surface with an appro-priate color (e.g., yellow for caution) plastic material.
Then the tag is engraved / grooved through the back and the grooves are filled with the lettering material (e.g., black plastic for the letters).
The result is a label that has a non glare, wear resistant surface, the lettering of which remains fairly constant at a wide range of viewing angles.
Exhibit 1 is a xerographic copy of a sample caution sign manufactured using this technique.
As part of the new labeling program, the large white labels with black lettering used on electrical switchgear panels are being modified to include unit color-code information.
A small disk of polyester tape color-coded to match the unit identity is attached to the white surface of the label.
When possible, the component number is in one-inch high lettering.
The. appropriate unit number (i.e., 3 or 4) is the first character in the component number.
In addition, all 4KV, 480V load center, 480V motor control center, lighting panel, and field component labeling will be upgraded.
Exhibit 2 shows the appearance of the labeling that is being replaced.
Exhibits 3 and 4 illustrate the improved labeling.
Harold R. Bocher et al..
- CAUIl0N -
..cs.
BINDING ~0R SHAKING'0F14160 VOLT CUBICLE D00RS MAY CAUSE AMAVE TYPE ACTION
~
- .i THAT CAN TRIP PROTECTIVE RELAYING RESULTING IN A POSSIBLE UNIT TRIP Exhibit 1.
Pemanent Information Tag (Warning Sign; Black Letters on Yellow Background)
. ' \\ 'l
+
ff a
t_
6 J h
ys 1 O
LABELS s,
ee y
f
.j 6
Exhibit 2.
Electrical Panels Prior to Labeling Upgrade.
(Most Labels are Small with White Lettering on Black Bacrground)
\\
' ~G(
- 9) '
3
);.'
r l
l
'5,
\\
j l
i
' ' 1,..
f,#'"
(
d; k
8.
g 30604 he M
~f
~
~L ss
.tj< M
-[.
DISKS l
D Sg t
i e Q,.
pe.,
70
- s..
l J
[c 4
5-Exhibit 3.
Exhibit 4.
i Upgraded Electrical Panel Labeling.
Another Example of Upgraded Electrical (Labels are large with Black Lettering Panel Labeling, on White Background; Label contains a disk of material that matches the color code of the unit, i.e., Tan for Unit 3, Blue for Unit 4, and Orange for Station Common) i
M 6 1986
' Harold R. BoohGr et al. Turkey Point has begun a program of " Area Information Striping" to help get people to the right location and to reduce wrong unit errors.
In this program color-coded stripes are applied to walls and electrical panels.
In addition, equipment bases / pedestals and/or curbing around equipment is painted with color-coded paint.
As a short-term measure, color-coded tape was applied (e.g., to electrical panels), but the tape is removed as components / areas are painted.
The Turkey Point QIP team has recommended that a list of power feeds or other specific hazards be developed for each enclosed area and that the listing be posted at the entrance (s) to the area to provide information during a fire or other emergency.
In addition, the team recommended that general equipment location maps be posted at the entrance to portions of certain low access areas (e.g., inside containment) to aid personnel in locating equipment.
Exhibit 5 shows a map posted in the emergency diesel generator (EDG) building. A similar map is posted on the doors entering the building along with yellow caution signs listing electrical power hazards and combustible liquids.
Turkey Point is ekperimenting with methods for identifying different piping systems (such as? nitrogen or condensate).
A polyester sheet material such as Hostaphan (high strength, temperature resistent) could be secured to surfaces with double-faced tape or with stainless-steel tie wraps.
(An operator would know a tag was missing if he saw just the tie wrap.)
Turkey Point makes use of both temporary and permanent information tags.
Permanent tags are used for information that is important, but is used infre-quently, for example, what to do and check immediately following a trip (while getting out the manuals and procedures).
Administrative Procedure 0103.36, i
" Control of Operator Aids and Temporary Information Tags," references INP0
" good practices." Permanent information tags are not covered by this proce-dure. The backgrounds of the permanent information tags are yellow for caution, red for danger, and white for information.
Exhibit 1 shows a yellow caution sign and Exhibit 6 shows a permanent white tag with information on 4160V bus indicating lights.
[The unit designation sign ("THINK 3 THREE") is not in accordance with the Turkey Point contrast standard discussed above, i.e., unit 3 designation signs should have dark (black) letters on light (tan) background.]
Procedures Work is generally done in accordance with written procedures, clearance orders, and plant work orders. Whether a procedure must be brought to the work site depends on the task and the experience level of the person performing the task.
The procedure must be brought along when a step is to be signed off.
For example, when containment spray is to be isolated, the operator must bring along a clearance order when closing and tagging valves, and the CR0 must have the signed off clearance order before accepting that the action has been completed.
There are plans to modify procedures to include a check-off or '
verification that the correct tags are in place during system alignment checks and maintenance.
Plant procedures used to be common to both units and people performed the applicable portions (a "3" or "4" would be crossed-out, as applicable).
Procedures are now unit specific, but are not color coded.
'7 I
r
.,l'}3}
~
~
4,i:' \\,,Q r
{u/~.~ll
'.Q<
w%
99s Ly.m
-;Q.g
'~*
r na oi.
Q%
7
~
%$,%g 4Q
~:.
'~ y l&
- N
., ;'- 4
/
- ee N
., dI.
a#
Xhibig $'
locatt*n Map p steg on Egg Doop, and gg,,,
Exhibit S Pa nt in %dt10n y09 I b te)
m.:.., s-
,.,.~. x
.m.. _,1.,,.
,. m _. -
Harbid Rl Bioher et al. -
A procedure upgrade program (PUP) is used to control the quality and technical content of procedures.
All departments are responsible for the procedures for which they are technically cognizant.
Operations procedures are being reviewed and changed to a new format that should help to reduce the potential for human error (50% were revised by November 1985 and the remaining will be revised by November 1986).
In the new format, procedures are required to be written to the entry-level person and have less print per page, one action per step, and cautions and warnings before rather than after the applicable steps.
A review will also be made of
. maintenance procedures, health physics and chemistry procedures, etc., with s
the intention of making them " user-friendly."
Turkey Point does not currently have administrative procedures covering all of the projects that'have been completed and that are planned.
However, the QIP team has recommended plant procedures for the labeling and identification program and the drawing maintenance program.
The system labeling program would be in accordance with standards derived from ANSI A13.1-1981.
The current plant administrative procedure will have to be revised to cover the new labeling program.
Replacerant be required for all tags lost, missing, or damaged during maintenance.
The QIP team has recommended an improved drawing maintenance program.
The program would identify differences in units, unit layouts, etc., when the differences might result in an error or delay of maintenance activities. When procedures are " walked down" there would be 'a requirement to check labeling and report deficiencies.
Shift Manning Scheme The operations staff per shift for the two units is as follows:
1 Shift Supervisor (Senior Control Room Operator) 1 Assistant Nuclear Plant Supervisor (ANPS) 1 Watch Engineer 3 Control Room Operators (Licensed SR0s or CR0s)*
2 Nuclear Operators (Auxiliary Building) 2 Turbine Operators (Turbine Area:
AFW, DG, Condensate, Feedwater, etc.)
1 Auxiliary Equipment Operator (Waterfront, Water Treatment) 1 Senior Nuclear Plant Operator 1 Shift Technical Advisor 1 Shift Technician 1 Shift Quality Control (when needed)
Operations staff work 8-hour shifts rotating with the clock.
Personnel fre-quently work overtime.
The formal briefings and shift turnovers were considered to be a help in rotating NE0's.
- This permits' breaks.
, wm s
Har,old R.' B=her et al. MAR 6 1986 There was no evidence that this shift manning scheme increases or decreases the potential for this type of error.
Comparable information from other sites
'is necessary to permit a comparison.
Communications The licensee has no formal program in this area.
However, three required, formal procedures (pre-shift briefings, shift turnover, and shift checklist) are intended to reduce errors due to miscommunications between shifts.
Planning and Control of Work As noted above, formal pre-shift briefings, formal shift turnovers, and formal shift checklists are intended to prevent miscommunications.
Turkey Point does all routine maintenance and does not ordinarily use much contractor labor, however, certain jobs are contracted out.
For example, 300 cutside employees were required for steam generator repair work. When such work is done, an " Area Coordinator" or a '! Room Coordinator" is designated (the exact term depends on the importance of the equipment in the area) to ensure that contractor employees know they are being watched.
This incidently has resulted in an increase in productivity.
The Coordinator, who is usually a contractor (e.g., Bechtel) supervisory employee, is expected to want work performed correctly.
System status is monitored on the equipment out of service (005) log, shift log sheets, shift check lists, and status boards (e.g., grease pencil board in the control room).
Verification of Work Turkey Point seems to have developed a program adequate to ensure the indepen-dent verification of work when required.
Administrative procedure 0-ADM-031, Rev. 7/12/85, includes definitions, responsibilities and procedures for inde-pendent verification of work-by operations and maintenance.
The procedure l
lists the. evolutions and the plant systems that require independent verification.
i Training i
Turkey Point issued a "watchstanders'/ operators' guide" to all operations personnel for training and retraining.
The guide (October 1985 revision) discusses in a philosophical manner the thinking and attitude that are necessary to doing a good job.
The guide urges operators to "be here now" by asking themselves the following types of questions:
" 1.
Am I on the right unit?
2.
Am I at the right location for the task I am about to perform?
3.
Do I understand the consequences of the action I am about to perform?"
Documents like this operators' guide appear to be a' good idea because it increases awareness of the potential for and consequences of errors.
. ~. - _ _ _ _ _ ~ _ -
MAR 6 1986
. ' Harold R.' Booher et al. The print improvement program is considered to be a significant operator training aid.
Previously a drawing might contain four different " North" arrows.
Now, P&ID's used in training are being revised to improve accuracy,
' clarity and consistency.
Factors such as equipment layout and scale are given greater attention.
It has been found that.this improves efficiency in work in the plant in almost every area from design to operations and maintenance.
It also helps with ALARA, because exposure time is reduced.
i F
The most recent LERs are now being covered in the monthly requalification training.
Feedback of Operating Experience 1
l Turkey Point Administrative Procedure 0103.15, dated November 27, 1985 (revi-sion received subsequent to site visit), governs the feedback of nuclear plant operating experience to plant staff. The procedure describes a program for the i.
screening, evaluation, and feedback of information from INP0 (50ER, SERs, l
08MRs, etc.), NRC (IENs), In-house Event Reports [IHEs; LERs, Possible Report-able Occurrences (PR0s), Nuclear log entries, etc.], and others. The Plant i
Technical Review Board (PTRB) reviews plant operating and design data as well as. operating experience feedback items and makes detailed recommendations to management regarding these. The PTRB' meets once per month, is chaired by'the
. Technical Department Supervisor, and has representatives from the Technical, Operations, Maintenance, Training, Quality Assurance, and Quality Control Departments, and the Operating Experience Feedback (OEF) Coordinator. Approved i
action items are tracked until completed.
The NRC team did not have an oppor-tunity to observe or review the activities of the PTRB or review the implemen-tation of the information feedback program.
Turkey Point Administr'ative Procedure 0103.6, dated 10/25/85, governs the '
evaluation and reporting of in-house events (IHEs) that fall under the criteria established in the-LER rule 10 CFR 50.73.
Anon-SiteReg&lationandCompliance i
(ORC) Supervisor is notified of operating events that may or~may not be I
reportable. The ORC Supervisor is responsible for investigating Potential i
Reportable Occurrences (PR0s) and recommending that an event or condition is reportabTe, generating an LER for each. reportable occurrence, and ensuring LERs
.are reviewed by the Plant Nuclear Safety Committee and forwarded to Nuclear i
Energy-Licensing for further' processing.
The Plant Manager-Nuclear determines j
whether a PRO constitutes a reportable occurrence.
Information concerning events is entered into a Potential Reportable Occurrences Determination Work-sheet and there is a PRO Status Log to monitor action on events. The NRC team i
did not have an opportunity to observe or review IHE evaluation and reporting activities.
When preparing LERs the licensing group at the site investigates personnel errors to determine root cause.
There is no procedure or instruction on performing activities such as interviewing or collecting evidence at the site of an incident. There is a trending of personnel errors at licensee j
headquarters.
l Licensee staff were not sure how many times wrong unit / train event's occurred at Turkey Point, although one did recall an incident in which the wrong turbine cooling water supply was valved out.
A computerized data base to be imple-i
NAR 4 198g Harald R Bo;h2r et al..
mented in 1986 for all reportable and non-reportable events should make this information available in the future.
Exit Meeting Prior to leaving the FP&L site, the NRC team expressed its appreciation to the Turkey Point staff for their cooperation ~in planning the visit, coordinating the tour and discussions, and providing available information.
When asked about their opinions regarding the most significant contributing factors, the Turkey Point staff responded with (1) poor labeling, (2) changes (opportunities for error are introduced when there are changes to what has existed and what people expect), and (3) lack of vigilance (people trying to work in " automatic").
Appreciatiori was also expressed to the NRC Resident Inspectors at the site.
If you have any questions regarding this report, please contact me on X24495.
p, Eugene A. Trager Nonreactor Assessment Staff Office for Analysis and Evaluation of Operational Data
Enclosure:
As Stated cc:
C. J. Heltemes F. J. Hebdon J. Funches V. Brownlee C. Cwalina R. Brewer D. MacDonald 9
_._ - _ -_ _ _._ -_ ~. ~..._ _ _.._.._ -
MAR G' L ENCLOSURE 1 Wrong Unit / Wrong Train Events at Turkey Point 1.
LER 250-83-007 - Auxiliary Feedwater System Isolated (Wrong Train)
This event was thoroughly investigated by the NRC and the license and was~1ater determined to be an abnormal occurrence. The event description and cause and corrective action information included in the special study report AE00/S401 was as following:
"While operating at 100% power, a Nuclear Turbine Operator (NTO) discovered that all steam supply. lines to Unit 3 auxiliary feedwater pumps B and C were closed.
Auxiliary feedwater pump A was out of service.
All auxiliary feedwater pumps had been inoperable from 4/14/83 to 4/19/83 (5. days).
" Evidence indicated that the NTO closed isolation valves in the steam supply lines rather than valves in the redundant steam supply lines.
The verification sign off for the NTOs action was not understood or performed.
Numerous system alignment checks over next five days missed identifying the incorrect alignment.
"Upon discovery, the steam supply line isolation' valves were opened and "B" and "C'! pumps tested on a daily basis.
Instructions issued to stress independent verification importance.. Logsheets revised to improve detail and clarity."
AE00/5401 also included the following on significance from both the safety and human factors perspectives:
,"For 5 days, all AFW was valved out to unit 3.
Major safety significance in that plant capability to respond to events requiring essential secondary cooling was negated until the condition was recogni_ zed and action was taken to manufally restore AFW.
" Highly significant, not only because of the misalignment error but also because the independent verification was not understood or performed'and because AFW alignment checks on all shifts for the next 5 days did not identify the error.
Independent verification of the rehanging of the tags, which may have detected the operator error, was not performed because the requirement was not completely understood by the i
personnel. The licensee had implemented independent verification for tagging operations but left it ambiguous as to the need for independent verification during tempqrary lifts of tags for testing and restoration.
The twice per shift NTO AFW system checks were not propbrly detailed in
~
MAR 6 192 instructions.
As part of actions taken to prevent recurrence, the licensee issued instructions to operators emphasizing the need for independent verification and revised tue NTO logsheets to improve detail and clarity."
During the site visit the NRC team was reminded that there was not total agreement on the cause of this event.
However, there was agreement that there were a number of factors at the time of the event that might have contributed to the event.
The NTO had worked for 1 to 1-1/2 years.
He did well on tests but on earlier occasions had trouble getting to the correct piece of equipment.
The AFW system had been modified during a maintenance outage. At that time there was no requirement that drawings be up-to-date when a system is returned to service.
Now there is a requirement that revised drawings be available prior to accepting a system back into service.
The valves to be checked were at an upper elevation.
Poor lighting may have been a problem.
It is also possible that the involved personnel who performed the later verifications were just looking at the locks rather than the valve position.
No valve tags wer,e on the valves ~that were involved.
The training for this job was in the classroom and the worker did not have to demonstrate knowledge of the system in the plant itself.
The twice per shift checks were apparently not psycho-logically independent, in that once the alignment was checked as correct, subsequent checks may have been less thorough.
2.
LER 251-83-016 - Containment Spray Isolated (Wrong Unit)
The event description and cause and corrective action information included in report AE00/S401 was as follows:
"On October 4,1983, a manual isolation valve on the discharge side of each of the redundant centainment spray pumps for unit 4 was found in the locked closed position.
Unit 4 was operating at power and unit 3 was proceeding,
from hot to cold shutdown for a refueling outage.
x
~
m s
1 MAP n ta~
"In the process of proceeding from hot to cold shutdown for unit 3, an operator was dispatched on October 2 to lockout several valves as required by procedures.
Included in the
-list of valves were manual valves (891A and B) on the discharge side of unit 3 redundant containment spray pumps.
The operator closed and locked the valves on the discharge
~
side of the unit 4 containment spray pumps, but did not initial the tag out sheet, as required.
(He did not have the tag out sheet with him). Subsequent to this action, the operator was isolating a sample line and was exposed to contaminated mist from a leak. The operator was relieved to receive medical attention. His replacement proceeded to-the correct valves, unit 3, and noted they were open. He closed, locked and tagged the valves in a'ccordance with procedures and was apparently unaware that the valve on the unit 4 discharge lines were in the closed position.
"On October 4, 1983, the licensee's technical staff were performing monthly periodic tests of the unit 4 containment spray pumps and noted that the discharge valves were in the
' closed position.
"The Resident Inspector and Region II Office were notified by the licensee. A walkdown of the accessible portions of all safety-related flowpaths was performed to verify that each valve (manual, power operated and automatic) in the flowpaths were in their correct positions."
AE0D S401 also incl.uded the following discussion of significance, both from the safety and human factors standpoints:
"Significant [from the safety standpoint] because containment spray was unavailable for over 28 hours3.240741e-4 days <br />0.00778 hours <br />4.62963e-5 weeks <br />1.0654e-5 months <br /> and personnel were unaware of this condition until it was discovered during a periodic test.
"The design basis for containment heat removal for Turkey Point 3 and 4 considers simultaneous operation of one spray pump and 2 of 3 emergency containment coolers. This is the basis for containment pressure transient calculations in Section 14 of that FSAR. However, the operation of either of the spray pumps or two of the three emergency containment coolers will provide heat removal capability to maintain the post accident containment pressure below the design value.
[This was a mitigating factor.]
"Significant [from a human factors standpoint, because] the operator was considered experienced and competent and had a good operating record.
In fact, the operator claimed responsibility when the condition was discovered.
(Otherwise the cause would have been " unknown.")
l m
c i. -
"The operator had just begun work on the Sunday dayshift and went to the wrong' set of valves 891 A and B and closed, locked and tagged them.
(These valves are numbered the same on both units.) The operator did not bring the tag out sheet along and was unable to sign that he believed he had closed valves on unit 3.
"This error was not caught, because there was no documentation that the operator closed the (what he believed to be) unit 3 valves. The licensee now strictly enforces a requirement to bring the tag out sheet to the job site and immediately initial when an action has been taken."
After this event the licensee began to use color-coded locks with different keys for the locks of different units. The licensee later extended this program to the control of all locked safety system valves.
The current color-coding scheme for safety systems is as follows:
Unit 3 component is Tan Unit 4 component is Blue Common component is Orange.
During the site visit the licensee staff repeated that the man who made the error was considered competent, had one to one-and-one-half years experience as an operator, and was fresh at the time he made the error.
He had been doing work on unit 4 and apparently had a "mindset," when he was told to work on unit 3.
At the end of the passageway (on route from the control room) you ' turn right for the unit 4 containment spray pump (CSP) area and left for the unit 3 CSP area. Although the CSP areas have different physical features, these were not noticed by the operator, and once he made the wrong turn he proceeded to close the wrong valves. Afterward he was sprayed and contaminated when trying to close a sample line that had been left open earlier by a chemistry technician, and he immediately went to be decontaminat e.
In addition, the. shutdown procedure in use at the time was the same for both units (i.e., there was nothing in the procedure that would have helped prevent this error).
The positions of all safety related valves are checked every two weeks.
In addition, there is a formal, written shift turnover that discusses what was done on each unit and what will be done.
3.
LER 250-84-003 - Operator tripped wrong reactor trip breaker (Wrong Train)
The event is described in the text of the LER as follows:
"On January 9, 1984, at 2:16 p.m., the unit 3 reactor tripped from 63 percent power while escalating to full power following a previous unit trip (LER 250-84-002).
HAR 6,
" Instrument and Control Department requested a clearance from Operations to replace a relay in Rack 37 of the Reactor Protection System Train B.
Following instructions for Operating Procedure 1004.2, Reactor Protection Periodic Test, a licensed operator proceeded to Rack 36 (Reactor Protection -
Train A) and closed Reactor Trip Bypass Breaker B.
The procedure then instructed the operator to go to Rack 41 (Train B) and trip the Reactor Trip Breaker B, but instead he stayed at Rack 36 and mistakenly tripped Reactor Trip Breaker-A.
This caused a reactor trip since Reactor Trip Bypass Breaker A was open.
Following the trip, a post-trip review was performed and with no further problems, the unit was stabilized and reactor start-up commenced.
"- To prevent recurrence, the following corrective actions will be taken:
1)
Review of this incident during the requalification sessions for licensed operators.
2)
Review of procedure for human factors considerations.
This will include such things as additional sign offs for closing and locking the rack door, proceeding to another rack, and unlocking and opening the door for that rack."
The licensed operator who made this error was interviewed during the site visit.
The operator had lost his train of thought during a delay in performing the procedure.
However, the operating procedure in use
-at that time did not include adequate detail. A comparison of the procedure in use at that time (revision 12/15/83) and the one in use now (revision 8/14/85) shows that work that was formerly described in about ten steps is now described in more than forty steps and substeps.
(Note that some of the added substeps were made necessary by the shunt trip modification.) The more detailed steps now make it possible to know exactly what has been done and what remains to be done.
It was noted that once the cabinet door is open, it is not possible to tell the identity.of the rack.
Exhibit 1.1 shows the current panel exterior labeling.
4.
LER 250-85-004 - Wrong Unit MG Sets Removed from Service The following information on the event description, cause, and corrective actions was included in the LER:
" Event:
On January 29,1985, at 11:14 a.m., while unit 3 was at 100% power and unit 4 was proceeding from hot shutdown to cold shutdown conditions under administrative controls, the non-licensed Operations personnel who had been instructed to remove the A and B MG sets of unit 4 from service, inadvertently entered the unit 3 motor control center room and removed-A and B MG sets for unit 3 from service instead, Since the MG sets supply power the control rod system, isolating this power source resulted in the de-energization of the control rod drive mechanisms which allowed the control rods to drop to full insertion, i.e..,
a reactor trip....
i c; j l
i a,
P I
l
~
f l
Exhibit 1.1 Improved Electrical Panel Labeling i
.e
,------m-w
. W5 i1?<
"Cause of Event:
The reactor trip was the result of personnel oversight when the unit 3 MG sets were inadvertently removed from service rather than those for unit 4.
Room markings distinguishing unit 3's from unit 4's motor control center rooms had recently been removed for installation of new Appendix R fire doors and painting of these doors.
" Corrective Actions:
The licensed and non-licensed personnel involved were individually counseled concerning this event.
The room markings were reposted.
The rod drive MG sets control cabinets' identification was enhanced. Although-removing the MG sets from service on a shutdown unit is not a critical evolution, the consequences of performing the evolution on the opposite unit must. be considered.
Instructions were issued to Operations personnel concerning the use of qualified personnel on critical evolutions. The event will be reviewed in operator requalification class."
During the site visit this event was discussed with licensee personnel.
The non-licensed person involved in the event was a trainee who was just observing the actions of the licensed person. The licensee's staff indicated that the involved licensed person did not have a good
" attitude" exhibited by his lack of concern following the trip.
The trainee felt that the signs would have helped prevent the error.
The revised labeling for the MG sets is shown in Exhibits 1.2 and 1.3.
5.
LER 250-85'017 - ESF Actuation (Wrong Train)
The following information on the event was included in the LER:
" Event:
On June 17, 1985, while unit 3 was in a scheduled refueling outage and unit 4 was at hot standby conditions for unit 3 safeguards testing, a spurious safety injection signal was generated in the unit 3 Engineered Safety Features system which actuated train "B" safeguards equipment.
At
,the time of the event, operating personnel were searching
-for a ground on one of two unit 3 125 volt DC buses by opening and reclosing breakers using an off-normal operating procedure.
Upon opening and reclosing breaker 3D23-9, a spurious safety injection signal was generated which initiated the start of unit 3 train "B" safeguards equipment, including the 3B and 4B safety injection pumps.
Although the train B" safety injection pumps started, no resultant safety injection flow was delivered to the unit 3 reactor coolant system (RCS) because the RCS inlet valves were procedurally closed for refueling' activities.
Following the safety injection actuation, the proper operation of safeguards equipment was verified using an emergency operating procedure.
All equipment actuated by t
the Engineered Safety Features Actuation signals functioned as designed.
O
i r.
[,, (' y a
4 nw l
~~-
l ggggg ff ),
- 4 cc
. h.L W.;1 GT' 4
s
)Q, m..,....
C '. -
= 'h NTRQL Q
6:, i..;2.
.h;:
~~
l s.,
e (fif.;*
Ilh' Ni l
.W
.i
.t.u. < -
l 3
i'I 8
a
,h.
.;#@g*
ih,.[
h m.yg
- I"(
- f i.
c c,
5 l
p s
r; 't l
1 5
b)l g
l
,..'.?
T l
m 4-l ll 2 E
u l
ft ! ',
l Mil
- r = + ~ sw r..
..w Exhibit 1.2 Improved MG Set Control Panel Exhibit 1.3 Improved MG Set Control Panel l
Labeling with Clear Unit labeling with Clear Unit i
Identification (Black Letters Identification (White Letters on Tan Background for Unit 3).
on Blue Background for Unit 4).
.' MAR 6 198g "Cause of Event:
The cause of the safeguards actuation was the inadvertent use of the wrong page from the appendix of a procedure being used to identify the source of a ground on the 3D23 DC bus. While using the page applicable to bus 3D01, the operator was opening and closing the corresponding breakers powered from the redundant DC bus 3023.
" Corrective Actions:
The following corrective actions were taken after the event.
1)
Using an emergency procedure, safeguards equipment were verified to have actuated properly upon initiation of the safety injection signal.
2)
Safeguards equipment which had actuated were returned to their normal operating status when the cause of the safeguards initiation was verified.
3)
The operator involved in the event was counseled on the exercise of greater care in using plant procedures."
This event was discussed during the site visit. The same breaker is on both battery buses. The procedure had a different procedure page for each battery bus. On the correct procedure page(s) it says you cannot open that breaker when searching for grounds. On the (incorrect) page that was used it said you could (and the bus would
" fail safe").
There were a number of possible contributing factors that were not discussed in the LER.
The procedure required that the person knew that 3001 is on panel 3A. This might not be clear because of the following association of buses and panels:
Bus Panel 3001 3A 4001 4B 3D23 3B 4023 4A The old procedure gave the engineering number only.
The revised procedure gives the engineering number and the panel number.
New breaker labels had been ordered six or eight months earlier. The new labels were not installed.
d