ML20115H853
| ML20115H853 | |
| Person / Time | |
|---|---|
| Site: | Limerick  |
| Issue date: | 04/02/1985 |
| From: | Cooney M PECO ENERGY CO., (FORMERLY PHILADELPHIA ELECTRIC |
| To: | Starostecki R NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION I) |
| References | |
| CON-#285-429 OL, NUDOCS 8504230292 | |
| Download: ML20115H853 (11) | |
Text
.
PHILADELPHIA ELECTRIC COMPANY
. e' 23O1 M ARKET STREET P.O. BOX 8699 PHILADELPHI A. PA.19101 (215) est sozo "N.?L"l'
. -,.":2':.".:~~::L.,
April 2, 1985 Re: Docket Mos. 50-352 50-353 Mr. Richard W. Starostecki, Director Division of Projects and Resident Programs U. S. Nuclear Regulatory Commission Region I G31 Park Avenue King of Prussia, Pa.
19406
Dear Mr. Starostecki:
SUBJECT:
Reportable Events at Limerick Generating Station
REFERENCES:
1)
R. U. Starostecki letter to S. L. Daltroff, dated February 11, 1985; Re: Inspection Report No. 50-352/
84-65 and 50-353/84-14.
- 2) M. J. Cooney Letter to R. W. Starostecki dated February 11, 1985; Re: Response to Reference 1.
Your letter of January 11, 1985, reference 1, contained an item of concern with regard to the number of personnel errors which you felt may be indicative of an adverse trend.
You further pointed out that, although these errors have not resulted in any immediate safety problems, the matter warrants management attention on the part of Philadelphia Electric Company.
Our response to this concern was contained in reference 2.
Philadelphia Electric Company senior management met with MRC Region I staff on February 22, 1985 to discuss Philadelphia Electric Company's investigations and corrective action programs to minimize future reportable events, in particular personnel The remainder of this letter describes those programs.
errors.
The details of the programs were presented to you at the meeting on February 22, 1985.
Philadelphia Electric Company believes that these corrective action programs have been effective in 8504230292 850402 ADOCK 050 g 2 PDR G
l
\\ f/
9 Mr. Richard W. Starostecki Page 2 reducing the number of reportable events as evidenced by the reduction that has occurred over the last few weeks.
Recognition of LER Trend In late November, 1984, the Station Superintendent became concerned with the adverse trend in reportable events, particularly those involving personnel error.
Although the station staff was involved in investigating these reportable events and identifying the appropriate corrective actions to prevent recurrence, he requested that the Nuclear Safety Section's Limerick Independent Safety Engineering Group (ISEG) evaluate these reportable events and develop corrective action recommendations to address common root causes.
Senior management was advised of the adverse trends.
The number of Suspected Licensee Event Reports (SLERs) was reported weekly beginning in early December to the Vice President
- Electric Production Department and to the Senior Vice President
- Nuclear Power.
It also was reported by the Station Superintendent to the Nuclear Review Board (NRB) at its meeting on December 14, 1984.
The NRD Chairman requested that the analysis performed on SLERS by the Nuclear Safety Section's Limerick ISEG be presented to the Nuclear Review Board.
Independent Safety Engineering Group Investigation One of the Limerick ISEG's functions is to examine Licensee Event Reports (LER's) and other design and operating experience information in order to identify areas for improving safety and to provide independent verification that activities are performed correctly and that human errors are reduced as much as practical.
A list of 37 questions was developed to investigate each SLER.
This list was developed out of a list of 70 questions contained in " Human Error Reporting Forms" developed by the Institute of Nuclear Power Operations in 1982 as a pilot program to investigate human errors.
All SLER's from the time of receipt of the licence to the end of 1984 were examined using these 37 questions.
Each event was examined to determine root cause, and these root causes were classified in a system consistent with the LER reporting guidance from the NRC.
However, for each of the 5 major cause categories (personnel error, design manufacturing construction /inctcllation deficiency, external causes, procedure deficiency and other) su bcategories were developed in order to provide better trending of the types of errors, deficiencies and causes.
These major categories and
gir. Richard W.
Starostecki Page 3 subcategories are consistent with the cause codes contained in a tracking and trending program being developed for application at Limerick and at Peach Bottom.
This Quality Assurance Tracking and Trending System is in the process of being computerized and implemented.
The analysis performed by the Limerick ISEG was submitted to the Limerick Plant Operating Review Committee and to the Nuclear Review Board for its review.
The attachment to our letter to you on February 11, 1985 (reference 2) provided a breakdown of SLER's in accordance with the categories and subcategories of the causes, and a detailed breakdown of the actual causes within each one of the subcategories.
This attachment was an update of the tables in the ISEG report presented to the Plant Operating Review Committee and the Nuclear Review Doard.
At the meeting of February 22, 1985, an updated listing of the cause of personnel errors and a detailed breakdown of these personnel errors were discussed.
This update is provided as Attachment 1 to this letter.
Included in the Limerick ISEG analysis were 8 recommendations.
These recommendations and the status of their implementation were discussed at the February 22, 1985 meeting. to this letter contains a listing of those recommendations and the status of their implementation.
The Limerick ISEG analysis is based on SLER's.
Upon further investigation by plant staff, an SLER may be determined to be not reportable as an LER.
Thus, the numbers of personnel errors and the numbers of other types of deficiencies identified in the Limerick ISEG's analysis and summaries do not correspond with the number of LER's submitted.
At the time of the meeting on February 22, 1985, 57 LER's had been submitted.
Of these, 23 were classified as personnel errors.
The ISEG's analysis agreed on 18 of the 23 classified as personnel error.
The five other events the ISEG classified as either design deficiencies or procedural deficiencies.
At the time of the meeting an additional 13 SLER's had occurred.
The Limerick ISEG's investigation of these SLER's identified 3 with personnel errors.
Of the 21 personnel errors, as classified by the Limerick ISEG, 8 involved licensed operators, 4 involved unlicensed operators, 1 involved a chemistry supervisor, 1 involved a maintenance worker and 7 involved I&C technicians.
Seventeen of the personnel errors
=
!,tr. Richard W.
Starostecki Page 4 involved workers, and the other 4 personnel errors involved first level supervision.
There has not been total agreement between the Limerick ISEG's classification of root causes and the classification reported in the LER.
The variation is about 10%.
Uc believe this is healthy and causes close examination of the root causes where differences appear.
These differences also illustrate the independence of the Limerick ISEG's views.
Following a presentation of the Limerick ISEG's analysis to the Nuclear Review Board at its meeting on January 15, 1985, Philadelphia Electric Company management requested that the Nuclear Safety Section examine the significance of the LER's.
This examination, which was reported at the February 22, 1985 meeting, revealed the following.
Conservative Reporting o
Three of the 57 LER's submitted occurred during initial conduct of surveillance testing, and a rationalization could have been made that these events were anticipated, and thus were not reportable because these tests were attempting to prove the initial operability of systems.
o Each reportable event has been reported as a single LER.
Examination of LER's submitted by other utilities reveals that one LER might describe coveral reportable events associated with one plant upset.
New LER Rule o
Thirty-nine of the 57 LER's submitted involved actuation of emergency safeguard features (prior to the new LER rule, which went into ef fect January 1, 1984, emergency safeguard feature actuations were not reportable).
Actuation of emergency safeguard features does not degrade system performance.
It merely provides confirmation that the system performs as designed.
o Eleven of the 39 LER's discussed above, involving omergency safeguard feature actuations, involved personnel errors.
Mr. Richard U.
Starostecki Page 5 Seriousness of Reportable Events o
No event resulted in serious degradation of safety barriers.
o No event resulted in the inability to:
- shutdown the reactor and maintain it in a safe shutdown condition
- remove residual heat
- control release of radioactivity
- mitigate the consequences of an accident.
Corrective Action Program The corrective action programs have addressed 3 areas:
- 1) modifications to eliminate recurring design deficienci6s, 2) actions to address personnel errors, and 3) programmatic improvements.
1)
Design, Manufacturing, Construction / Installation Deficiencies Of the 57 LER's submitted as of February 22, 1985, 23 had been classified-as design, manufacturing, construction /
installation deficiencies.
Six of these LER's were associated with Riley temperature switches inadvertently causing actuation of emergency safeguard features.
Five of the LER's were associated with the breaking of tapes on chlorine detectors.
Three LER's were associated with the reactor enclosure differential pressure sensors being too sensitive to gusts of wind.
Three of the LER's were associated with common process and reference leg valving disturbances.
These recurring design deficiencies were identified in the ISEG analysis and were the subject of ISEG's recommendations.
Three of these LER's were associated with reactor protection system static inverters.
Deficiencies associated with these static inverters are being addressed by assuring internal components within the RPS inverter cabinets are secure and by changing the source of the AC power supply to the RPS inverter to a power supply whose voltage levels are more stable than the previous power supply.
A two week outage in early February permitted many modifications to be perform;d to eliminate these recurring design deficiencies.
I r
Mr. Richard W. Starostecki Page 6 2)
Personnel Errors Corrective actions with regard to personnel errors were accomplished through the following activities.
o Plant personnel vere informed, via letters and group meetings, of the importance of following procedures and administrative controls and keeping shift supervision informed of work being performed.
In addition, plant personnel were told they must take sufficient time to properly evaluate and perform tasks independently of schedule needs.
o Simulator training was augmented with feedback from a reportable event in which the cooldown rate of the reactor exceeded the technical specification rate.
The simulator's logic for simulating cooldown of the reactor following a plant scram assumes normal decay heat levels.
Operators had not received training with regard to reactor cooldown rate when sufficient quantities of decay heat are not present, such as during initial plant startup.
The simulator training program has been changed to incorporate this experience.
o nequalification training for Senior Licensed Operators and Licensed Operators has begun to emphasize the final version of the Technical Specifications as issued with the Limerick 5% power license and to address the LER's that have occurred at Limerick.
Many of the Senior Licensed Operators and licensed operators went through initial licensing training and examination without the benefit of the Limerick 5% power license Technical Specifications since these Technical Specifications were issued just prior to the commencement of fuel loading.
o Phase 2 training, as delineated in the Final Safety Analysis Report, is now completed for non-licensed operators.
Non-licensed operators are now participating in a continuing training program.
o Philadelphia Electric Company's disciplinary policy has been applied to individuals when deemed appropriate by the Limerick Station Superintendent.
The basic objective of any disciplinary action in Philadelphia Electric Company is corrective rather than punitive.
It is to make the employee's aware that their actions are improper and that, if not corrected, their future with
Mr. Richard W. Starostecki Page 7 y
our company will be in jeopardy.
Whenever a decision.is made to take disciplinary action, many factors, such as the nature of the offense, the employees past record, and other. variables must-be.taken into consideration.
The implementation of this objective is a discipline policy followed throughout Philadelphia Electric Company
-based on the principle of progressive discipline in which' repeated minor offenses arc-dealt with increasingly severe measures and in which serious offenses are dealt with more severely at the outset.
There are six steps in this disciplinary process.
1.
Counseling 2.
Oral warning 3.
Written warning
~
4.
Reprimand 5.
Suspension 6.
Suspension with possible recommendation for termination The disciplinary actions taken with regard to employees involved in personnel errors at Limerick Generating Station have been to step 1, counseling, for all cases but one, in which step 2, an oral warning, was provided.
o An Operator Excellence Program was established in order to bring to the attention of all plant personnel the importance of minimizing reportable events, particularly reportable events associated with personnel error.
This program also has as an objective the establishment of l
competition among groups at the station in order to see i
which group would have the fewest errors.
The group with the fewest errors would be provided some type of award.
This is a one time program designed to immediately tackle the problem of personnel errors.
It will subsequently be replaced with a program aimed at improving the professionalism of plant personnel.
This professionalism program is in the early stages of development, but its goal is to improve operations i-through pride and exec 11ence.
i 3)
Programmatic Improvements i
Mr. Richard W.
Starostecki Page 8 The programmatic improvements instituted by the Station Superintendent include the following.
o A senior plant staff member has been assigned to personally investigate the cause of all SLERS.
This investigation is to identify promptly the corrective actions that need to be implemented and to bring this information to the attention of the senior plant staff.
(The Limerick ISEG's investigation of these SLERs remains independent of this senior staf f plant member's investigation.)
o The Plant Operating Review Committee is now reviewing the SLER's shortly after they occur in order to provide input as to the corrective actions and to judge whether proposed corrective actions are sufficient to prevent recurrence.
o A new procedure has been instituted which requires an operator to perform a daily check of the control room panels in order to determine if there are any readings or indications outside of Technical Specification requirements.
o A study was performed over several days by station staff as to the number of people in the control room during each hour of the day.
This study revealed that the number of people in the control room varied, with peaks occurring in late morning and early afternoon.
The Station Superintendent has determined that the absolute numbers of these peaks are unacceptable and unnecessary.
The shift supervisors and shift superintendents have been instructed to control access to the control room.
In addition, certain areas in the control room have been designated off limits for unauthorized personnel.
Lines on the floor and rope barriers identify these areas.
o A member of the plant staff performs a detailed review of control room logs several times a week to check if all actions that have been taken are in compliance with procedures and Technical Specficiations.
Results of Corrective Action Program There has been one reportable personnel error to date since the beginning of February,1985; however, there have been several LER's classified as design, manufacturing, construction / installation deficiencies, several classified as
,Mr. Richard W.
Starostecki Page 9 procedural deficiencies, and several classified as other deficiencies.
Philadelphia Electric Company believes the corrective action programs have halted the adverse trend in personnel errors.
We expect that more design, manufacturing, construction / installation deficiencies and some procedural deficiencies will be uncovered and corrected during startup testing.
We believe this is one of the objectives of the startup testing program.
V t 7,,t r, y your s,
/h Attachments cc:
J. T. Wiggins, Resident Site Inspector See Attached Service List
ATTACHMENT 1 LInERICK*ISEG. EVALUATION OF REPORTABLE PERSONNEL ERRORS TABLE 1 - CAUSES OF INCIDENTS
~
r 10/26/84 01/01/85 Cause to 12/31/84 to 02/16/85 Code Category Number Number A.
PERSONNEL ERROR 16 7
(total)
(total)
A-1 Failure to follow procedures,' rules, regulations...............................
5 1
A-2 Failure to use correct procedure..........
A-3 Failure to properly identify equipment....
4 A-4 Failure to properly communicate...........
1 A-5 Failure to observe changing conditions....
2 A-6 Failure to properly interpret information, results......................
4 A-7 Failure to perform required inspections / tests.........................
4 j
A-8 Failure to properly assess consequences of actions...................
4 1
l A-9 Other personnel errors....................
1 TABLE 2 - DETAILED BREAKDOWN OF CAUSES A.
Personnel Error....................................., (22 total)
A-1:
Failure to follow procedure, rules, regulations......
(7) a)
Failure to notify shift before working.
(2) i b). Procedure not followed.
(3)
{
c)
RWCU isolation while swapping.demineralizers.
(1) d)
Failure to comply with Tech. Spec.. time limits.
(1) l A-4:
Failure to properly communicate......................
(1)
/
i A-5:
Failure to observe changing conditions.
(2)
A-6:
Failure to properly interpret information............
(4) i A-7:
Failure to perform required inspections / tests........
(4) a)
Failure to properly follow l,
Technical Specifications........................
(3) b)
Incorrect evaluation of test results............
(1),
A-8 Failure to properly assess consequences of actions..
(4)
I 4)
Fire doors left open............................
(1) j b)
Did not check Technical Specifications while working on HPCI pressure transmitters......
(1) c)
Opened equalizing valve on refuel floor i
low Delta P Sensor..............................
(1) j d)
Reactor Enclosure low Delta P isolation..........
(1) l A-9:
Other Personnel Errors..............................
(1)
ATTACHMENT 2 LIMERICK ISEG RECOMMENDATIONS AND STATUS
~,' '., Continue to pursue the resolution of instrumentation problems l
associated with common instrument process and reference legs.
Support work for permanent head chambers worked during last outage.
Work to be continued during future outage periods until complete.
Use of portable head chambers is an adequate interim solution.
2.
Continue to pursue the resolution of problems associated with Riley temperature switches in the Reactor Water Clean-Up System portion of the Steam Leak Detection System.
The investigation in progress should be expedited to install necessary modifications as soon as possible, ie., before low power testing is completed.
Modification completed to. correct problem.
3.
Complete flow balancing on the Refuel Floor HVAC.
This should resolve the low differential pressure problems on this system.
Deferred to refueling outage.
Not presently part of secondary containment.
Should not have been reported.
4.
Engineering should investigate the low differential pressure problems on the Reactor Enclosure HVAC System that appear to be caused by outside wind effects on the outside air pressure sensors.
Interim administrative controls should not be allowed to become the permanent solution.
Modification completed to correct problem.
5.
Engineering should investigate the problem of breaking sample tapes on the Chlorine Detectors in the Control Enclosure HVAC System.
This has been responsible Mr at least five main control room isolations during the perio' at this study, and several other alarms.
This is an unacceptably nigh frequency for an operating plant and a permanent solution should be expedited.
l Increaspd surveillance by dedicated technician reduced.? frequency of occurance to manageable levels.
Engineeringsolutionstillbe[ng l
investigated.
6.
Emphasize to all plant staff, cont 'ctor, and vendor personnel that Shift Supervision must be inf.med of'all work to be l
performed on plant equipment.
l Recommended action completed.
7.
Re-emphasize to all plant personnel that all procedures and administrative controls must be followed.
Recommended action completed.
8.
Review, as time permits, check-off lists for all systems which have technical specifications for operability.
l Emergency Core Cooling Systems completed.
COL revieNs for otMer nysRPFn nRn Am p7@gRnnns